Appendix

1. VO2max

1.1 Overall

1.1.1 Forest plot

1.1.2 R output

##                            SMD            95%-CI %W(fixed) %W(random)
## Bækkerud 2016           0.5820 [-0.3903; 1.5542]       1.4        1.8
## Beetham 2019           -0.6337 [-1.7518; 0.4845]       1.0        1.5
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       1.7        2.0
## Ciolac 2010             0.7968 [-0.0715; 1.6650]       1.7        2.0
## Cocks 2013             -0.5676 [-1.5671; 0.4320]       1.3        1.7
## Conraads 2015           0.1953 [-0.1026; 0.4933]      14.5        4.1
## Currie 2015             0.5330 [-0.3833; 1.4494]       1.5        1.9
## Earnest 2013           -0.0715 [-0.7221; 0.5791]       3.0        2.7
## Fisher 2015            -0.7440 [-1.5960; 0.1079]       1.8        2.1
## Gillen 2016            -0.4021 [-1.3117; 0.5075]       1.6        1.9
## Gorostiaga 1991         1.4458 [ 0.1749; 2.7166]       0.8        1.2
## Grieco 2013             0.6057 [-0.2523; 1.4638]       1.7        2.1
## Helgerud 2007           1.0281 [ 0.0955; 1.9607]       1.5        1.9
## Helgerud 2007           0.9016 [-0.0184; 1.8216]       1.5        1.9
## Henriksson 1976        -0.5694 [-1.9102; 0.7715]       0.7        1.1
## Honkala 2017 (Healthy)  0.8299 [ 0.0579; 1.6019]       2.2        2.3
## Honkala 2017 (T2D)      1.7112 [ 0.5592; 2.8633]       1.0        1.4
## Keating 2014            0.3091 [-0.5316; 1.1498]       1.8        2.1
## Keteyian 2014           0.6606 [-0.1020; 1.4232]       2.2        2.3
## Kim 2015                0.9018 [ 0.1242; 1.6793]       2.1        2.3
## Klonizakis 2014         0.1765 [-0.7729; 1.1259]       1.4        1.8
## Lunt 2014               0.9567 [ 0.0748; 1.8385]       1.7        2.0
## Lunt 2014               0.5858 [-0.2686; 1.4401]       1.8        2.1
## Macpherson 2011        -0.0240 [-0.9005; 0.8526]       1.7        2.0
## Madssen 2014            0.5664 [-0.1090; 1.2418]       2.8        2.6
## Martins 2016           -0.0365 [-0.7538; 0.6809]       2.5        2.5
## Matsuo 2014             0.6957 [-0.0960; 1.4874]       2.0        2.3
## Matsuo 2015             0.8715 [ 0.0342; 1.7088]       1.8        2.1
## Mitranun 2014           0.9929 [ 0.2078; 1.7781]       2.1        2.3
## Molmen-Hansen 2011      0.6999 [ 0.1724; 1.2275]       4.6        3.2
## Motiani 2017            0.4668 [-0.3123; 1.2460]       2.1        2.3
## Nalcakan 2014          -0.2565 [-1.2750; 0.7620]       1.2        1.7
## Nie 2017                0.1324 [-0.5857; 0.8505]       2.5        2.5
## O’Leary 2018           -0.0213 [-0.8979; 0.8552]       1.7        2.0
## Ramos 2016a             0.7144 [ 0.0977; 1.3311]       3.4        2.8
## Ramos 2016b             0.8823 [ 0.1552; 1.6095]       2.4        2.5
## Robinson 2015           0.0000 [-0.6279; 0.6279]       3.3        2.8
## Rognmo 2004             0.6529 [-0.3245; 1.6302]       1.3        1.8
## Sandvei 2012            0.2446 [-0.5765; 1.0658]       1.9        2.2
## Sawyer 2016             0.3610 [-0.5704; 1.2925]       1.5        1.9
## Scribbans 2014         -0.1028 [-1.0039; 0.7984]       1.6        2.0
## Shepherd 2013          -0.5676 [-1.5671; 0.4320]       1.3        1.7
## Sjöros 2018             0.7485 [-0.1373; 1.6343]       1.6        2.0
## Skleryk 2013           -0.1275 [-1.1085; 0.8535]       1.3        1.8
## Tjønna 2008             1.0772 [ 0.1042; 2.0502]       1.4        1.8
## Trapp 2008              0.3351 [-0.3856; 1.0558]       2.5        2.5
## Winn 2018              -0.2185 [-1.2014; 0.7644]       1.3        1.8
## Wisløff 2007            4.5911 [ 2.8296; 6.3526]       0.4        0.7
## 
## Number of studies combined: k = 48
## 
##                         SMD           95%-CI    z  p-value
## Fixed effect model   0.3794 [0.2661; 0.4927] 6.56 < 0.0001
## Random effects model 0.4017 [0.2381; 0.5653] 4.81 < 0.0001
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1456 [0.1246; 0.5265]; tau = 0.3816 [0.3529; 0.7256];
##  I^2 = 47.1% [25.8%; 62.3%]; H = 1.38 [1.16; 1.63]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  88.86   47  0.0002
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

1.1.3 Sensitivity analysis

1.1.3.1 Forest plot

1.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                                      SMD           95%-CI  p-value   tau^2     tau    I^2
## Omitting Bækkerud 2016            0.3799 [0.2229; 0.5370] < 0.0001  0.1147  0.3387  41.3%
## Omitting Beetham 2019             0.3963 [0.2424; 0.5503] < 0.0001  0.1048  0.3238  39.2%
## Omitting Burgomaster 2008         0.3960 [0.2406; 0.5513] < 0.0001  0.1083  0.3290  39.8%
## Omitting Ciolac 2010              0.3749 [0.2183; 0.5316] < 0.0001  0.1127  0.3357  40.8%
## Omitting Cocks 2013               0.3979 [0.2440; 0.5518] < 0.0001  0.1040  0.3225  38.9%
## Omitting Conraads 2015            0.3922 [0.2300; 0.5543] < 0.0001  0.1250  0.3536  40.3%
## Omitting Currie 2015              0.3805 [0.2233; 0.5378] < 0.0001  0.1151  0.3393  41.3%
## Omitting Earnest 2013             0.3959 [0.2390; 0.5527] < 0.0001  0.1109  0.3331  40.0%
## Omitting Fisher 2015              0.4048 [0.2537; 0.5559] < 0.0001  0.0934  0.3057  36.3%
## Omitting Gillen 2016              0.3973 [0.2426; 0.5520] < 0.0001  0.1061  0.3257  39.3%
## Omitting Gorostiaga 1991          0.3718 [0.2175; 0.5262] < 0.0001  0.1070  0.3270  39.7%
## Omitting Grieco 2013              0.3788 [0.2215; 0.5361] < 0.0001  0.1148  0.3388  41.2%
## Omitting Helgerud 2007            0.3715 [0.2159; 0.5271] < 0.0001  0.1093  0.3307  40.1%
## Omitting Helgerud 2007            0.3737 [0.2175; 0.5298] < 0.0001  0.1114  0.3338  40.5%
## Omitting Henriksson 1976          0.3921 [0.2373; 0.5469] < 0.0001  0.1088  0.3298  40.2%
## Omitting Honkala 2017 (Healthy)   0.3727 [0.2161; 0.5293] < 0.0001  0.1117  0.3342  40.4%
## Omitting Honkala 2017 (T2D)       0.3660 [0.2138; 0.5182] < 0.0001  0.0993  0.3150  37.9%
## Omitting Keating 2014             0.3848 [0.2272; 0.5424] < 0.0001  0.1157  0.3402  41.4%
## Omitting Keteyian 2014            0.3767 [0.2192; 0.5341] < 0.0001  0.1144  0.3383  41.0%
## Omitting Kim 2015                 0.3711 [0.2149; 0.5273] < 0.0001  0.1102  0.3319  40.1%
## Omitting Klonizakis 2014          0.3869 [0.2298; 0.5440] < 0.0001  0.1147  0.3387  41.3%
## Omitting Lunt 2014                0.3719 [0.2160; 0.5278] < 0.0001  0.1101  0.3318  40.2%
## Omitting Lunt 2014                0.3791 [0.2218; 0.5365] < 0.0001  0.1150  0.3391  41.2%
## Omitting Macpherson 2011          0.3911 [0.2344; 0.5479] < 0.0001  0.1129  0.3360  40.8%
## Omitting Madssen 2014             0.3783 [0.2201; 0.5364] < 0.0001  0.1159  0.3405  41.1%
## Omitting Martins 2016             0.3937 [0.2367; 0.5507] < 0.0001  0.1122  0.3350  40.4%
## Omitting Matsuo 2014              0.3762 [0.2189; 0.5334] < 0.0001  0.1140  0.3376  40.9%
## Omitting Matsuo 2015              0.3729 [0.2165; 0.5293] < 0.0001  0.1113  0.3337  40.4%
## Omitting Mitranun 2014            0.3691 [0.2135; 0.5246] < 0.0001  0.1080  0.3287  39.6%
## Omitting Molmen-Hansen 2011       0.3721 [0.2141; 0.5302] < 0.0001  0.1136  0.3371  40.2%
## Omitting Motiani 2017             0.3814 [0.2235; 0.5392] < 0.0001  0.1160  0.3406  41.3%
## Omitting Nalcakan 2014            0.3931 [0.2374; 0.5487] < 0.0001  0.1102  0.3320  40.3%
## Omitting Nie 2017                 0.3896 [0.2318; 0.5473] < 0.0001  0.1151  0.3393  41.1%
## Omitting O’Leary 2018             0.3911 [0.2344; 0.5478] < 0.0001  0.1129  0.3361  40.8%
## Omitting Ramos 2016a              0.3732 [0.2156; 0.5308] < 0.0001  0.1134  0.3368  40.5%
## Omitting Ramos 2016b              0.3705 [0.2142; 0.5268] < 0.0001  0.1100  0.3317  40.0%
## Omitting Robinson 2015            0.3943 [0.2369; 0.5517] < 0.0001  0.1127  0.3358  40.4%
## Omitting Rognmo 2004              0.3788 [0.2219; 0.5356] < 0.0001  0.1143  0.3381  41.2%
## Omitting Sandvei 2012             0.3862 [0.2286; 0.5439] < 0.0001  0.1156  0.3400  41.3%
## Omitting Sawyer 2016              0.3837 [0.2264; 0.5410] < 0.0001  0.1154  0.3397  41.4%
## Omitting Scribbans 2014           0.3923 [0.2359; 0.5486] < 0.0001  0.1119  0.3344  40.6%
## Omitting Shepherd 2013            0.3979 [0.2440; 0.5518] < 0.0001  0.1040  0.3225  38.9%
## Omitting Sjöros 2018              0.3761 [0.2193; 0.5330] < 0.0001  0.1134  0.3367  40.9%
## Omitting Skleryk 2013             0.3916 [0.2354; 0.5478] < 0.0001  0.1118  0.3344  40.7%
## Omitting Tjønna 2008              0.3714 [0.2160; 0.5267] < 0.0001  0.1089  0.3300  40.0%
## Omitting Trapp 2008               0.3845 [0.2263; 0.5427] < 0.0001  0.1166  0.3414  41.4%
## Omitting Winn 2018                0.3930 [0.2372; 0.5488] < 0.0001  0.1105  0.3325  40.4%
## Omitting Wisløff 2007             0.3585 [0.2220; 0.4950] < 0.0001  0.0513  0.2264  24.1%
##                                                                                          
## Pooled estimate                   0.4017 [0.2381; 0.5653] < 0.0001  0.1456  0.3816  47.1%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

1.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

1.2 Subgroups

1.2.1 Overall

1.2.1.1 Forest plot

1.2.1.2 R output
##                           SMD            95%-CI     meta-analysis
##                        0.4017 [ 0.2381; 0.5653]           Overall
## Healthy                0.2461 [ 0.0354; 0.4568]        Population
## Overweight/obese       0.1786 [-0.1067; 0.4640]        Population
## Cardiac Rehabilitation 0.7734 [ 0.2385; 1.3082]        Population
## Metabolic Syndrome     0.6192 [ 0.2461; 0.9922]        Population
## T2D                    1.0085 [ 0.4805; 1.5365]        Population
## < 30 y                 0.1405 [-0.1045; 0.3855]               Age
## 30 - 50 y              0.4251 [ 0.1858; 0.6644]               Age
## > 50 y                 0.6006 [ 0.3120; 0.8892]               Age
## < 5 weeks              0.4000 [ 0.0756; 0.7244] Training Duration
## 5 - 10 weeks           0.2799 [ 0.0420; 0.5178] Training Duration
## > 10 weeks             0.4893 [ 0.2237; 0.7549] Training Duration
## < 0.5                  0.4651 [ 0.1920; 0.7382]         Men Ratio
## > 0.5                  0.3426 [ 0.1539; 0.5313]         Men Ratio
## Running                0.6490 [ 0.4087; 0.8893]  Type of Exercise
## Cycling                0.1894 [ 0.0189; 0.3598]  Type of Exercise
## < 30%                  0.4076 [ 0.2175; 0.5977]   Baseline Values
## 30 - 60%               0.2537 [-0.0956; 0.6030]   Baseline Values
## > 60%                  0.4704 [ 0.0077; 0.9330]   Baseline Values
## HIIT                   0.4978 [ 0.3119; 0.6837]      Type of HIIE
## SIT                    0.1794 [-0.0802; 0.4390]      Type of HIIE
## 
## Number of studies combined: k = 48
## 
##                         SMD           95%-CI    z  p-value
## Random effects model 0.4017 [0.2381; 0.5653] 4.81 < 0.0001
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1456; tau = 0.3816; I^2 = 47.1% [25.8%; 62.3%]; H = 1.38 [1.16; 1.63]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  88.86   47  0.0002
## 
## Results for meta-analyses (random effects model):
##                     k    SMD           95%-CI  tau^2    tau     Q   I^2
## Overall            48 0.4017 [0.2381; 0.5653] 0.1456 0.3816 88.86 47.1%
## Population         48 0.4017 [0.2381; 0.5653] 0.1456 0.3816 88.86 47.1%
## Age                48 0.4017 [0.2381; 0.5653] 0.1456 0.3816 88.86 47.1%
## Training Duration  48 0.4017 [0.2381; 0.5653] 0.1456 0.3816 88.86 47.1%
## Men Ratio          48 0.4017 [0.2381; 0.5653] 0.1456 0.3816 88.86 47.1%
## Type of Exercise   48 0.4017 [0.2381; 0.5653] 0.1456 0.3816 88.86 47.1%
## Baseline Values    48 0.4017 [0.2381; 0.5653] 0.1456 0.3816 88.86 47.1%
## Type of HIIE       48 0.4017 [0.2381; 0.5653] 0.1456 0.3816 88.86 47.1%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

1.2.2 Population

1.2.2.1 Forest plot

1.2.2.2 R output
##                            SMD            95%-CI %W(fixed) %W(random)             population
## Bækkerud 2016           0.5820 [-0.3903; 1.5542]       1.4        1.8       Overweight/obese
## Beetham 2019           -0.6337 [-1.7518; 0.4845]       1.0        1.5       Overweight/obese
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       1.7        2.0                Healthy
## Ciolac 2010             0.7968 [-0.0715; 1.6650]       1.7        2.0                Healthy
## Cocks 2013             -0.5676 [-1.5671; 0.4320]       1.3        1.7                Healthy
## Conraads 2015           0.1953 [-0.1026; 0.4933]      14.5        4.1 Cardiac Rehabilitation
## Currie 2015             0.5330 [-0.3833; 1.4494]       1.5        1.9 Cardiac Rehabilitation
## Earnest 2013           -0.0715 [-0.7221; 0.5791]       3.0        2.7       Overweight/obese
## Fisher 2015            -0.7440 [-1.5960; 0.1079]       1.8        2.1       Overweight/obese
## Gillen 2016            -0.4021 [-1.3117; 0.5075]       1.6        1.9                Healthy
## Gorostiaga 1991         1.4458 [ 0.1749; 2.7166]       0.8        1.2                Healthy
## Grieco 2013             0.6057 [-0.2523; 1.4638]       1.7        2.1                Healthy
## Helgerud 2007           1.0281 [ 0.0955; 1.9607]       1.5        1.9                Healthy
## Helgerud 2007           0.9016 [-0.0184; 1.8216]       1.5        1.9                Healthy
## Henriksson 1976        -0.5694 [-1.9102; 0.7715]       0.7        1.1                Healthy
## Honkala 2017 (Healthy)  0.8299 [ 0.0579; 1.6019]       2.2        2.3                Healthy
## Honkala 2017 (T2D)      1.7112 [ 0.5592; 2.8633]       1.0        1.4                    T2D
## Keating 2014            0.3091 [-0.5316; 1.1498]       1.8        2.1       Overweight/obese
## Keteyian 2014           0.6606 [-0.1020; 1.4232]       2.2        2.3 Cardiac Rehabilitation
## Kim 2015                0.9018 [ 0.1242; 1.6793]       2.1        2.3 Cardiac Rehabilitation
## Klonizakis 2014         0.1765 [-0.7729; 1.1259]       1.4        1.8                Healthy
## Lunt 2014               0.9567 [ 0.0748; 1.8385]       1.7        2.0       Overweight/obese
## Lunt 2014               0.5858 [-0.2686; 1.4401]       1.8        2.1       Overweight/obese
## Macpherson 2011        -0.0240 [-0.9005; 0.8526]       1.7        2.0                Healthy
## Madssen 2014            0.5664 [-0.1090; 1.2418]       2.8        2.6 Cardiac Rehabilitation
## Martins 2016           -0.0365 [-0.7538; 0.6809]       2.5        2.5       Overweight/obese
## Matsuo 2014             0.6957 [-0.0960; 1.4874]       2.0        2.3                Healthy
## Matsuo 2015             0.8715 [ 0.0342; 1.7088]       1.8        2.1     Metabolic Syndrome
## Mitranun 2014           0.9929 [ 0.2078; 1.7781]       2.1        2.3                    T2D
## Molmen-Hansen 2011      0.6999 [ 0.1724; 1.2275]       4.6        3.2       Overweight/obese
## Motiani 2017            0.4668 [-0.3123; 1.2460]       2.1        2.3                Healthy
## Nalcakan 2014          -0.2565 [-1.2750; 0.7620]       1.2        1.7                Healthy
## Nie 2017                0.1324 [-0.5857; 0.8505]       2.5        2.5                Healthy
## O’Leary 2018           -0.0213 [-0.8979; 0.8552]       1.7        2.0                Healthy
## Ramos 2016a             0.7144 [ 0.0977; 1.3311]       3.4        2.8     Metabolic Syndrome
## Ramos 2016b             0.8823 [ 0.1552; 1.6095]       2.4        2.5     Metabolic Syndrome
## Robinson 2015           0.0000 [-0.6279; 0.6279]       3.3        2.8     Metabolic Syndrome
## Rognmo 2004             0.6529 [-0.3245; 1.6302]       1.3        1.8 Cardiac Rehabilitation
## Sandvei 2012            0.2446 [-0.5765; 1.0658]       1.9        2.2                Healthy
## Sawyer 2016             0.3610 [-0.5704; 1.2925]       1.5        1.9       Overweight/obese
## Scribbans 2014         -0.1028 [-1.0039; 0.7984]       1.6        2.0                Healthy
## Shepherd 2013          -0.5676 [-1.5671; 0.4320]       1.3        1.7                Healthy
## Sjöros 2018             0.7485 [-0.1373; 1.6343]       1.6        2.0                    T2D
## Skleryk 2013           -0.1275 [-1.1085; 0.8535]       1.3        1.8       Overweight/obese
## Tjønna 2008             1.0772 [ 0.1042; 2.0502]       1.4        1.8     Metabolic Syndrome
## Trapp 2008              0.3351 [-0.3856; 1.0558]       2.5        2.5                Healthy
## Winn 2018              -0.2185 [-1.2014; 0.7644]       1.3        1.8       Overweight/obese
## Wisløff 2007            4.5911 [ 2.8296; 6.3526]       0.4        0.7 Cardiac Rehabilitation
## 
## Number of studies combined: k = 48
## 
##                         SMD           95%-CI    z  p-value
## Fixed effect model   0.3794 [0.2661; 0.4927] 6.56 < 0.0001
## Random effects model 0.4017 [0.2381; 0.5653] 4.81 < 0.0001
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1456 [0.1246; 0.5265]; tau = 0.3816 [0.3529; 0.7256];
##  I^2 = 47.1% [25.8%; 62.3%]; H = 1.38 [1.16; 1.63]
## 
## Quantifying residual heterogeneity:
##  I^2 = 36.0% [7.5%; 55.7%]; H = 1.25 [1.04; 1.50]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  88.86   47  0.0002
## 
## Results for subgroups (fixed effect model):
##                          k    SMD            95%-CI     Q   I^2
## Healthy                 21 0.2504 [ 0.0570; 0.4439] 23.47 14.8%
## Overweight/obese        12 0.2063 [-0.0270; 0.4396] 15.67 29.8%
## Cardiac Rehabilitation   7 0.4372 [ 0.2096; 0.6647] 21.47 72.1%
## Metabolic Syndrome       5 0.6018 [ 0.2771; 0.9266]  5.15 22.3%
## T2D                      3 1.0085 [ 0.4805; 1.5365]  1.44  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups 11.26    4  0.0238
## Within groups  67.21   43  0.0105
## 
## Results for subgroups (random effects model):
##                          k    SMD            95%-CI  tau^2    tau
## Healthy                 21 0.2461 [ 0.0354; 0.4568] 0.0357 0.1889
## Overweight/obese        12 0.1786 [-0.1067; 0.4640] 0.0738 0.2717
## Cardiac Rehabilitation   7 0.7734 [ 0.2385; 1.3082] 0.3336 0.5776
## Metabolic Syndrome       5 0.6192 [ 0.2461; 0.9922] 0.0405 0.2012
## T2D                      3 1.0085 [ 0.4805; 1.5365]      0      0
## 
## Test for subgroup differences (random effects model):
##                      Q d.f. p-value
## Between groups   12.56    4  0.0136
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
1.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 48; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.1359 (SE = 0.0690)
## tau (square root of estimated tau^2 value):             0.3686
## I^2 (residual heterogeneity / unaccounted variability): 43.71%
## H^2 (unaccounted variability / sampling variability):   1.78
## R^2 (amount of heterogeneity accounted for):            6.68%
## 
## Test for Residual Heterogeneity:
## QE(df = 43) = 76.3860, p-val = 0.0013
## 
## Test of Moderators (coefficients 2:5):
## QM(df = 4) = 10.8469, p-val = 0.0283
## 
## Model Results:
## 
##                               estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                         0.2487  0.1283   1.9377  0.0527  -0.0029  0.5002  . 
## .byvarOverweight/obese         -0.0757  0.2077  -0.3645  0.7155  -0.4827  0.3313    
## .byvarCardiac Rehabilitation    0.4675  0.2426   1.9271  0.0540  -0.0080  0.9430  . 
## .byvarMetabolic Syndrome        0.4132  0.2692   1.5351  0.1248  -0.1144  0.9407    
## .byvarT2D                       0.8365  0.3683   2.2713  0.0231   0.1147  1.5583  * 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
1.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

1.2.3 Age

1.2.3.1 Forest plot

1.2.3.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_age
## Bækkerud 2016           0.5820 [-0.3903; 1.5542]       1.4        1.8    30 - 50 y
## Beetham 2019           -0.6337 [-1.7518; 0.4845]       1.0        1.5       > 50 y
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       1.7        2.0       < 30 y
## Ciolac 2010             0.7968 [-0.0715; 1.6650]       1.7        2.0       < 30 y
## Cocks 2013             -0.5676 [-1.5671; 0.4320]       1.3        1.7       < 30 y
## Conraads 2015           0.1953 [-0.1026; 0.4933]      14.5        4.1       > 50 y
## Currie 2015             0.5330 [-0.3833; 1.4494]       1.5        1.9       > 50 y
## Earnest 2013           -0.0715 [-0.7221; 0.5791]       3.0        2.7    30 - 50 y
## Fisher 2015            -0.7440 [-1.5960; 0.1079]       1.8        2.1       < 30 y
## Gillen 2016            -0.4021 [-1.3117; 0.5075]       1.6        1.9       < 30 y
## Gorostiaga 1991         1.4458 [ 0.1749; 2.7166]       0.8        1.2       < 30 y
## Grieco 2013             0.6057 [-0.2523; 1.4638]       1.7        2.1       < 30 y
## Helgerud 2007           1.0281 [ 0.0955; 1.9607]       1.5        1.9       < 30 y
## Helgerud 2007           0.9016 [-0.0184; 1.8216]       1.5        1.9       < 30 y
## Henriksson 1976        -0.5694 [-1.9102; 0.7715]       0.7        1.1       < 30 y
## Honkala 2017 (Healthy)  0.8299 [ 0.0579; 1.6019]       2.2        2.3    30 - 50 y
## Honkala 2017 (T2D)      1.7112 [ 0.5592; 2.8633]       1.0        1.4    30 - 50 y
## Keating 2014            0.3091 [-0.5316; 1.1498]       1.8        2.1    30 - 50 y
## Keteyian 2014           0.6606 [-0.1020; 1.4232]       2.2        2.3       > 50 y
## Kim 2015                0.9018 [ 0.1242; 1.6793]       2.1        2.3       > 50 y
## Klonizakis 2014         0.1765 [-0.7729; 1.1259]       1.4        1.8       > 50 y
## Lunt 2014               0.9567 [ 0.0748; 1.8385]       1.7        2.0    30 - 50 y
## Lunt 2014               0.5858 [-0.2686; 1.4401]       1.8        2.1    30 - 50 y
## Macpherson 2011        -0.0240 [-0.9005; 0.8526]       1.7        2.0       < 30 y
## Madssen 2014            0.5664 [-0.1090; 1.2418]       2.8        2.6       > 50 y
## Martins 2016           -0.0365 [-0.7538; 0.6809]       2.5        2.5    30 - 50 y
## Matsuo 2014             0.6957 [-0.0960; 1.4874]       2.0        2.3       < 30 y
## Matsuo 2015             0.8715 [ 0.0342; 1.7088]       1.8        2.1    30 - 50 y
## Mitranun 2014           0.9929 [ 0.2078; 1.7781]       2.1        2.3       > 50 y
## Molmen-Hansen 2011      0.6999 [ 0.1724; 1.2275]       4.6        3.2       > 50 y
## Motiani 2017            0.4668 [-0.3123; 1.2460]       2.1        2.3    30 - 50 y
## Nalcakan 2014          -0.2565 [-1.2750; 0.7620]       1.2        1.7       < 30 y
## Nie 2017                0.1324 [-0.5857; 0.8505]       2.5        2.5       < 30 y
## O’Leary 2018           -0.0213 [-0.8979; 0.8552]       1.7        2.0       < 30 y
## Ramos 2016a             0.7144 [ 0.0977; 1.3311]       3.4        2.8       > 50 y
## Ramos 2016b             0.8823 [ 0.1552; 1.6095]       2.4        2.5       > 50 y
## Robinson 2015           0.0000 [-0.6279; 0.6279]       3.3        2.8       > 50 y
## Rognmo 2004             0.6529 [-0.3245; 1.6302]       1.3        1.8       > 50 y
## Sandvei 2012            0.2446 [-0.5765; 1.0658]       1.9        2.2       < 30 y
## Sawyer 2016             0.3610 [-0.5704; 1.2925]       1.5        1.9    30 - 50 y
## Scribbans 2014         -0.1028 [-1.0039; 0.7984]       1.6        2.0       < 30 y
## Shepherd 2013          -0.5676 [-1.5671; 0.4320]       1.3        1.7       < 30 y
## Sjöros 2018             0.7485 [-0.1373; 1.6343]       1.6        2.0    30 - 50 y
## Skleryk 2013           -0.1275 [-1.1085; 0.8535]       1.3        1.8    30 - 50 y
## Tjønna 2008             1.0772 [ 0.1042; 2.0502]       1.4        1.8       > 50 y
## Trapp 2008              0.3351 [-0.3856; 1.0558]       2.5        2.5       < 30 y
## Winn 2018              -0.2185 [-1.2014; 0.7644]       1.3        1.8    30 - 50 y
## Wisløff 2007            4.5911 [ 2.8296; 6.3526]       0.4        0.7       > 50 y
## 
## Number of studies combined: k = 48
## 
##                         SMD           95%-CI    z  p-value
## Fixed effect model   0.3794 [0.2661; 0.4927] 6.56 < 0.0001
## Random effects model 0.4017 [0.2381; 0.5653] 4.81 < 0.0001
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1456 [0.1246; 0.5265]; tau = 0.3816 [0.3529; 0.7256];
##  I^2 = 47.1% [25.8%; 62.3%]; H = 1.38 [1.16; 1.63]
## 
## Quantifying residual heterogeneity:
##  I^2 = 37.5% [10.5%; 56.3%]; H = 1.26 [1.06; 1.51]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  88.86   47  0.0002
## 
## Results for subgroups (fixed effect model):
##             k    SMD            95%-CI     Q   I^2
## < 30 y     19 0.1453 [-0.0602; 0.3509] 25.13 28.4%
## 30 - 50 y  14 0.4194 [ 0.1920; 0.6469] 14.28  9.0%
## > 50 y     15 0.4845 [ 0.3141; 0.6548] 32.56 57.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  6.50    2  0.0388
## Within groups  71.97   45  0.0065
## 
## Results for subgroups (random effects model):
##             k    SMD            95%-CI  tau^2    tau
## < 30 y     19 0.1405 [-0.1045; 0.3855] 0.0831 0.2883
## 30 - 50 y  14 0.4251 [ 0.1858; 0.6644] 0.0187 0.1369
## > 50 y     15 0.6006 [ 0.3120; 0.8892] 0.1644 0.4054
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   6.03    2  0.0492
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
1.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 48; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.1149 (SE = 0.0614)
## tau (square root of estimated tau^2 value):             0.3390
## I^2 (residual heterogeneity / unaccounted variability): 40.88%
## H^2 (unaccounted variability / sampling variability):   1.69
## R^2 (amount of heterogeneity accounted for):            21.09%
## 
## Test for Residual Heterogeneity:
## QE(df = 46) = 77.8116, p-val = 0.0023
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 10.6036, p-val = 0.0011
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.3055  0.2307  -1.3240  0.1855  -0.7576  0.1467     
## age        0.0169  0.0052   3.2563  0.0011   0.0067  0.0271  ** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
1.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

1.2.4 Training Duration

1.2.4.1 Forest plot

1.2.4.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_duration
## Bækkerud 2016           0.5820 [-0.3903; 1.5542]       1.4        1.8      5 - 10 weeks
## Beetham 2019           -0.6337 [-1.7518; 0.4845]       1.0        1.5        > 10 weeks
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       1.7        2.0      5 - 10 weeks
## Ciolac 2010             0.7968 [-0.0715; 1.6650]       1.7        2.0        > 10 weeks
## Cocks 2013             -0.5676 [-1.5671; 0.4320]       1.3        1.7      5 - 10 weeks
## Conraads 2015           0.1953 [-0.1026; 0.4933]      14.5        4.1        > 10 weeks
## Currie 2015             0.5330 [-0.3833; 1.4494]       1.5        1.9        > 10 weeks
## Earnest 2013           -0.0715 [-0.7221; 0.5791]       3.0        2.7      5 - 10 weeks
## Fisher 2015            -0.7440 [-1.5960; 0.1079]       1.8        2.1      5 - 10 weeks
## Gillen 2016            -0.4021 [-1.3117; 0.5075]       1.6        1.9        > 10 weeks
## Gorostiaga 1991         1.4458 [ 0.1749; 2.7166]       0.8        1.2      5 - 10 weeks
## Grieco 2013             0.6057 [-0.2523; 1.4638]       1.7        2.1         < 5 weeks
## Helgerud 2007           1.0281 [ 0.0955; 1.9607]       1.5        1.9      5 - 10 weeks
## Helgerud 2007           0.9016 [-0.0184; 1.8216]       1.5        1.9      5 - 10 weeks
## Henriksson 1976        -0.5694 [-1.9102; 0.7715]       0.7        1.1      5 - 10 weeks
## Honkala 2017 (Healthy)  0.8299 [ 0.0579; 1.6019]       2.2        2.3         < 5 weeks
## Honkala 2017 (T2D)      1.7112 [ 0.5592; 2.8633]       1.0        1.4         < 5 weeks
## Keating 2014            0.3091 [-0.5316; 1.1498]       1.8        2.1        > 10 weeks
## Keteyian 2014           0.6606 [-0.1020; 1.4232]       2.2        2.3      5 - 10 weeks
## Kim 2015                0.9018 [ 0.1242; 1.6793]       2.1        2.3      5 - 10 weeks
## Klonizakis 2014         0.1765 [-0.7729; 1.1259]       1.4        1.8         < 5 weeks
## Lunt 2014               0.9567 [ 0.0748; 1.8385]       1.7        2.0        > 10 weeks
## Lunt 2014               0.5858 [-0.2686; 1.4401]       1.8        2.1        > 10 weeks
## Macpherson 2011        -0.0240 [-0.9005; 0.8526]       1.7        2.0      5 - 10 weeks
## Madssen 2014            0.5664 [-0.1090; 1.2418]       2.8        2.6        > 10 weeks
## Martins 2016           -0.0365 [-0.7538; 0.6809]       2.5        2.5        > 10 weeks
## Matsuo 2014             0.6957 [-0.0960; 1.4874]       2.0        2.3      5 - 10 weeks
## Matsuo 2015             0.8715 [ 0.0342; 1.7088]       1.8        2.1      5 - 10 weeks
## Mitranun 2014           0.9929 [ 0.2078; 1.7781]       2.1        2.3      5 - 10 weeks
## Molmen-Hansen 2011      0.6999 [ 0.1724; 1.2275]       4.6        3.2        > 10 weeks
## Motiani 2017            0.4668 [-0.3123; 1.2460]       2.1        2.3         < 5 weeks
## Nalcakan 2014          -0.2565 [-1.2750; 0.7620]       1.2        1.7      5 - 10 weeks
## Nie 2017                0.1324 [-0.5857; 0.8505]       2.5        2.5        > 10 weeks
## O’Leary 2018           -0.0213 [-0.8979; 0.8552]       1.7        2.0      5 - 10 weeks
## Ramos 2016a             0.7144 [ 0.0977; 1.3311]       3.4        2.8        > 10 weeks
## Ramos 2016b             0.8823 [ 0.1552; 1.6095]       2.4        2.5        > 10 weeks
## Robinson 2015           0.0000 [-0.6279; 0.6279]       3.3        2.8         < 5 weeks
## Rognmo 2004             0.6529 [-0.3245; 1.6302]       1.3        1.8      5 - 10 weeks
## Sandvei 2012            0.2446 [-0.5765; 1.0658]       1.9        2.2      5 - 10 weeks
## Sawyer 2016             0.3610 [-0.5704; 1.2925]       1.5        1.9      5 - 10 weeks
## Scribbans 2014         -0.1028 [-1.0039; 0.7984]       1.6        2.0      5 - 10 weeks
## Shepherd 2013          -0.5676 [-1.5671; 0.4320]       1.3        1.7      5 - 10 weeks
## Sjöros 2018             0.7485 [-0.1373; 1.6343]       1.6        2.0         < 5 weeks
## Skleryk 2013           -0.1275 [-1.1085; 0.8535]       1.3        1.8         < 5 weeks
## Tjønna 2008             1.0772 [ 0.1042; 2.0502]       1.4        1.8        > 10 weeks
## Trapp 2008              0.3351 [-0.3856; 1.0558]       2.5        2.5        > 10 weeks
## Winn 2018              -0.2185 [-1.2014; 0.7644]       1.3        1.8         < 5 weeks
## Wisløff 2007            4.5911 [ 2.8296; 6.3526]       0.4        0.7        > 10 weeks
## 
## Number of studies combined: k = 48
## 
##                         SMD           95%-CI    z  p-value
## Fixed effect model   0.3794 [0.2661; 0.4927] 6.56 < 0.0001
## Random effects model 0.4017 [0.2381; 0.5653] 4.81 < 0.0001
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1456 [0.1246; 0.5265]; tau = 0.3816 [0.3529; 0.7256];
##  I^2 = 47.1% [25.8%; 62.3%]; H = 1.38 [1.16; 1.63]
## 
## Quantifying residual heterogeneity:
##  I^2 = 41.8% [17.2%; 59.2%]; H = 1.31 [1.10; 1.56]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  88.86   47  0.0002
## 
## Results for subgroups (fixed effect model):
##                k    SMD           95%-CI     Q   I^2
## < 5 weeks      9 0.3875 [0.1030; 0.6720] 10.15 21.2%
## 5 - 10 weeks  22 0.2850 [0.0957; 0.4743] 32.45 35.3%
## > 10 weeks    17 0.4167 [0.2528; 0.5807] 34.77 54.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  1.09    2  0.5793
## Within groups  77.38   45  0.0019
## 
## Results for subgroups (random effects model):
##                k    SMD           95%-CI  tau^2    tau
## < 5 weeks      9 0.4000 [0.0756; 0.7244] 0.0518 0.2277
## 5 - 10 weeks  22 0.2799 [0.0420; 0.5178] 0.1124 0.3353
## > 10 weeks    17 0.4893 [0.2237; 0.7549] 0.1507 0.3882
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.34    2  0.5104
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
1.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 48; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.1424 (SE = 0.0685)
## tau (square root of estimated tau^2 value):             0.3774
## I^2 (residual heterogeneity / unaccounted variability): 46.30%
## H^2 (unaccounted variability / sampling variability):   1.86
## R^2 (amount of heterogeneity accounted for):            2.20%
## 
## Test for Residual Heterogeneity:
## QE(df = 46) = 85.6575, p-val = 0.0003
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 2.3382, p-val = 0.1262
## 
## Model Results:
## 
##           estimate      se    zval    pval    ci.lb   ci.ub 
## intrcpt     0.1407  0.1897  0.7420  0.4581  -0.2310  0.5125    
## duration    0.0301  0.0197  1.5291  0.1262  -0.0085  0.0686    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
1.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

1.2.5 Men Ratio

1.2.5.1 Forest plot

1.2.5.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_men_ratio
## Bækkerud 2016           0.5820 [-0.3903; 1.5542]       1.4        1.8              < 0.5
## Beetham 2019           -0.6337 [-1.7518; 0.4845]       1.0        1.5              > 0.5
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       1.7        2.0              < 0.5
## Ciolac 2010             0.7968 [-0.0715; 1.6650]       1.7        2.0              < 0.5
## Cocks 2013             -0.5676 [-1.5671; 0.4320]       1.3        1.7              > 0.5
## Conraads 2015           0.1953 [-0.1026; 0.4933]      14.5        4.1              > 0.5
## Currie 2015             0.5330 [-0.3833; 1.4494]       1.5        1.9              > 0.5
## Earnest 2013           -0.0715 [-0.7221; 0.5791]       3.0        2.7              > 0.5
## Fisher 2015            -0.7440 [-1.5960; 0.1079]       1.8        2.1              > 0.5
## Gillen 2016            -0.4021 [-1.3117; 0.5075]       1.6        1.9              > 0.5
## Gorostiaga 1991         1.4458 [ 0.1749; 2.7166]       0.8        1.2              < 0.5
## Grieco 2013             0.6057 [-0.2523; 1.4638]       1.7        2.1              < 0.5
## Helgerud 2007           1.0281 [ 0.0955; 1.9607]       1.5        1.9              > 0.5
## Helgerud 2007           0.9016 [-0.0184; 1.8216]       1.5        1.9              > 0.5
## Henriksson 1976        -0.5694 [-1.9102; 0.7715]       0.7        1.1              > 0.5
## Honkala 2017 (Healthy)  0.8299 [ 0.0579; 1.6019]       2.2        2.3              > 0.5
## Honkala 2017 (T2D)      1.7112 [ 0.5592; 2.8633]       1.0        1.4              > 0.5
## Keating 2014            0.3091 [-0.5316; 1.1498]       1.8        2.1              < 0.5
## Keteyian 2014           0.6606 [-0.1020; 1.4232]       2.2        2.3              > 0.5
## Kim 2015                0.9018 [ 0.1242; 1.6793]       2.1        2.3              > 0.5
## Klonizakis 2014         0.1765 [-0.7729; 1.1259]       1.4        1.8              < 0.5
## Lunt 2014               0.9567 [ 0.0748; 1.8385]       1.7        2.0              < 0.5
## Lunt 2014               0.5858 [-0.2686; 1.4401]       1.8        2.1              < 0.5
## Macpherson 2011        -0.0240 [-0.9005; 0.8526]       1.7        2.0              > 0.5
## Madssen 2014            0.5664 [-0.1090; 1.2418]       2.8        2.6              > 0.5
## Martins 2016           -0.0365 [-0.7538; 0.6809]       2.5        2.5              < 0.5
## Matsuo 2014             0.6957 [-0.0960; 1.4874]       2.0        2.3              > 0.5
## Matsuo 2015             0.8715 [ 0.0342; 1.7088]       1.8        2.1              > 0.5
## Mitranun 2014           0.9929 [ 0.2078; 1.7781]       2.1        2.3              < 0.5
## Molmen-Hansen 2011      0.6999 [ 0.1724; 1.2275]       4.6        3.2              > 0.5
## Motiani 2017            0.4668 [-0.3123; 1.2460]       2.1        2.3              > 0.5
## Nalcakan 2014          -0.2565 [-1.2750; 0.7620]       1.2        1.7              > 0.5
## Nie 2017                0.1324 [-0.5857; 0.8505]       2.5        2.5              < 0.5
## O’Leary 2018           -0.0213 [-0.8979; 0.8552]       1.7        2.0              > 0.5
## Ramos 2016a             0.7144 [ 0.0977; 1.3311]       3.4        2.8              > 0.5
## Ramos 2016b             0.8823 [ 0.1552; 1.6095]       2.4        2.5              > 0.5
## Robinson 2015           0.0000 [-0.6279; 0.6279]       3.3        2.8              < 0.5
## Rognmo 2004             0.6529 [-0.3245; 1.6302]       1.3        1.8              > 0.5
## Sandvei 2012            0.2446 [-0.5765; 1.0658]       1.9        2.2              < 0.5
## Sawyer 2016             0.3610 [-0.5704; 1.2925]       1.5        1.9              < 0.5
## Scribbans 2014         -0.1028 [-1.0039; 0.7984]       1.6        2.0              > 0.5
## Shepherd 2013          -0.5676 [-1.5671; 0.4320]       1.3        1.7              > 0.5
## Sjöros 2018             0.7485 [-0.1373; 1.6343]       1.6        2.0              > 0.5
## Skleryk 2013           -0.1275 [-1.1085; 0.8535]       1.3        1.8              > 0.5
## Tjønna 2008             1.0772 [ 0.1042; 2.0502]       1.4        1.8              < 0.5
## Trapp 2008              0.3351 [-0.3856; 1.0558]       2.5        2.5              < 0.5
## Winn 2018              -0.2185 [-1.2014; 0.7644]       1.3        1.8              < 0.5
## Wisløff 2007            4.5911 [ 2.8296; 6.3526]       0.4        0.7              < 0.5
## 
## Number of studies combined: k = 48
## 
##                         SMD           95%-CI    z  p-value
## Fixed effect model   0.3794 [0.2661; 0.4927] 6.56 < 0.0001
## Random effects model 0.4017 [0.2381; 0.5653] 4.81 < 0.0001
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1456 [0.1246; 0.5265]; tau = 0.3816 [0.3529; 0.7256];
##  I^2 = 47.1% [25.8%; 62.3%]; H = 1.38 [1.16; 1.63]
## 
## Quantifying residual heterogeneity:
##  I^2 = 41.1% [16.4%; 58.5%]; H = 1.30 [1.09; 1.55]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  88.86   47  0.0002
## 
## Results for subgroups (fixed effect model):
##         k    SMD           95%-CI     Q   I^2
## < 0.5  19 0.4114 [0.2140; 0.6087] 33.14 45.7%
## > 0.5  29 0.3414 [0.2025; 0.4804] 45.00 37.8%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.32    1  0.5701
## Within groups  78.15   46  0.0022
## 
## Results for subgroups (random effects model):
##         k    SMD           95%-CI  tau^2    tau
## < 0.5  19 0.4651 [0.1920; 0.7382] 0.1632 0.4040
## > 0.5  29 0.3426 [0.1539; 0.5313] 0.0922 0.3037
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.52    1  0.4693
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
1.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 48; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.1477 (SE = 0.0697)
## tau (square root of estimated tau^2 value):             0.3844
## I^2 (residual heterogeneity / unaccounted variability): 47.21%
## H^2 (unaccounted variability / sampling variability):   1.89
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 46) = 87.1306, p-val = 0.0002
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.3509, p-val = 0.2451
## 
## Model Results:
## 
##            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt      0.5981  0.1885   3.1722  0.0015   0.2286  0.9677  ** 
## men_ratio   -0.3022  0.2600  -1.1623  0.2451  -0.8119  0.2074     
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
1.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

1.2.6 Type of Exercise

1.2.6.1 Forest plot

1.2.6.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) type_exercise
## Bækkerud 2016           0.5820 [-0.3903; 1.5542]       1.4        1.8       Running
## Beetham 2019           -0.6337 [-1.7518; 0.4845]       1.0        1.5       Running
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       1.7        2.0       Cycling
## Ciolac 2010             0.7968 [-0.0715; 1.6650]       1.7        2.0       Running
## Cocks 2013             -0.5676 [-1.5671; 0.4320]       1.3        1.7       Cycling
## Conraads 2015           0.1953 [-0.1026; 0.4933]      14.5        4.1       Cycling
## Currie 2015             0.5330 [-0.3833; 1.4494]       1.5        1.9       Cycling
## Earnest 2013           -0.0715 [-0.7221; 0.5791]       3.0        2.7       Running
## Fisher 2015            -0.7440 [-1.5960; 0.1079]       1.8        2.1       Cycling
## Gillen 2016            -0.4021 [-1.3117; 0.5075]       1.6        1.9       Cycling
## Gorostiaga 1991         1.4458 [ 0.1749; 2.7166]       0.8        1.2       Cycling
## Grieco 2013             0.6057 [-0.2523; 1.4638]       1.7        2.1       Cycling
## Helgerud 2007           1.0281 [ 0.0955; 1.9607]       1.5        1.9       Running
## Helgerud 2007           0.9016 [-0.0184; 1.8216]       1.5        1.9       Running
## Henriksson 1976        -0.5694 [-1.9102; 0.7715]       0.7        1.1       Cycling
## Honkala 2017 (Healthy)  0.8299 [ 0.0579; 1.6019]       2.2        2.3       Cycling
## Honkala 2017 (T2D)      1.7112 [ 0.5592; 2.8633]       1.0        1.4       Cycling
## Keating 2014            0.3091 [-0.5316; 1.1498]       1.8        2.1       Cycling
## Keteyian 2014           0.6606 [-0.1020; 1.4232]       2.2        2.3       Running
## Kim 2015                0.9018 [ 0.1242; 1.6793]       2.1        2.3       Running
## Klonizakis 2014         0.1765 [-0.7729; 1.1259]       1.4        1.8       Cycling
## Lunt 2014               0.9567 [ 0.0748; 1.8385]       1.7        2.0       Running
## Lunt 2014               0.5858 [-0.2686; 1.4401]       1.8        2.1       Running
## Macpherson 2011        -0.0240 [-0.9005; 0.8526]       1.7        2.0       Cycling
## Madssen 2014            0.5664 [-0.1090; 1.2418]       2.8        2.6       Running
## Martins 2016           -0.0365 [-0.7538; 0.6809]       2.5        2.5       Cycling
## Matsuo 2014             0.6957 [-0.0960; 1.4874]       2.0        2.3       Cycling
## Matsuo 2015             0.8715 [ 0.0342; 1.7088]       1.8        2.1       Cycling
## Mitranun 2014           0.9929 [ 0.2078; 1.7781]       2.1        2.3       Running
## Molmen-Hansen 2011      0.6999 [ 0.1724; 1.2275]       4.6        3.2       Running
## Motiani 2017            0.4668 [-0.3123; 1.2460]       2.1        2.3       Cycling
## Nalcakan 2014          -0.2565 [-1.2750; 0.7620]       1.2        1.7       Cycling
## Nie 2017                0.1324 [-0.5857; 0.8505]       2.5        2.5       Cycling
## O’Leary 2018           -0.0213 [-0.8979; 0.8552]       1.7        2.0       Cycling
## Ramos 2016a             0.7144 [ 0.0977; 1.3311]       3.4        2.8       Running
## Ramos 2016b             0.8823 [ 0.1552; 1.6095]       2.4        2.5       Running
## Robinson 2015           0.0000 [-0.6279; 0.6279]       3.3        2.8       Cycling
## Rognmo 2004             0.6529 [-0.3245; 1.6302]       1.3        1.8       Running
## Sandvei 2012            0.2446 [-0.5765; 1.0658]       1.9        2.2       Running
## Sawyer 2016             0.3610 [-0.5704; 1.2925]       1.5        1.9       Cycling
## Scribbans 2014         -0.1028 [-1.0039; 0.7984]       1.6        2.0       Cycling
## Shepherd 2013          -0.5676 [-1.5671; 0.4320]       1.3        1.7       Cycling
## Sjöros 2018             0.7485 [-0.1373; 1.6343]       1.6        2.0       Cycling
## Skleryk 2013           -0.1275 [-1.1085; 0.8535]       1.3        1.8       Cycling
## Tjønna 2008             1.0772 [ 0.1042; 2.0502]       1.4        1.8       Running
## Trapp 2008              0.3351 [-0.3856; 1.0558]       2.5        2.5       Cycling
## Winn 2018              -0.2185 [-1.2014; 0.7644]       1.3        1.8       Running
## Wisløff 2007            4.5911 [ 2.8296; 6.3526]       0.4        0.7       Running
## 
## Number of studies combined: k = 48
## 
##                         SMD           95%-CI    z  p-value
## Fixed effect model   0.3794 [0.2661; 0.4927] 6.56 < 0.0001
## Random effects model 0.4017 [0.2381; 0.5653] 4.81 < 0.0001
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1456 [0.1246; 0.5265]; tau = 0.3816 [0.3529; 0.7256];
##  I^2 = 47.1% [25.8%; 62.3%]; H = 1.38 [1.16; 1.63]
## 
## Quantifying residual heterogeneity:
##  I^2 = 28.7% [0.0%; 50.5%]; H = 1.18 [1.00; 1.42]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  88.86   47  0.0002
## 
## Results for subgroups (fixed effect model):
##           k    SMD           95%-CI     Q   I^2
## Running  20 0.6330 [0.4521; 0.8139] 31.77 40.2%
## Cycling  28 0.1898 [0.0438; 0.3358] 32.74 17.5%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups 13.96    1  0.0002
## Within groups  64.51   46  0.0371
## 
## Results for subgroups (random effects model):
##           k    SMD           95%-CI  tau^2    tau
## Running  20 0.6490 [0.4087; 0.8893] 0.1159 0.3404
## Cycling  28 0.1894 [0.0189; 0.3598] 0.0346 0.1861
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   9.35    1  0.0022
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
1.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 48; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0998 (SE = 0.0577)
## tau (square root of estimated tau^2 value):             0.3159
## I^2 (residual heterogeneity / unaccounted variability): 37.56%
## H^2 (unaccounted variability / sampling variability):   1.60
## R^2 (amount of heterogeneity accounted for):            31.48%
## 
## Test for Residual Heterogeneity:
## QE(df = 46) = 73.6718, p-val = 0.0059
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 9.3390, p-val = 0.0022
## 
## Model Results:
## 
##                       estimate      se    zval    pval    ci.lb   ci.ub 
## intrcpt                 0.1981  0.1013  1.9543  0.0507  -0.0006  0.3967   . 
## type_exerciseRunning    0.4769  0.1561  3.0560  0.0022   0.1710  0.7827  ** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
1.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

1.2.7 Baseline Values

1.2.7.1 Forest plot

1.2.7.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_bsln
## Bækkerud 2016           0.5820 [-0.3903; 1.5542]       1.4        1.8      30 - 60%
## Beetham 2019           -0.6337 [-1.7518; 0.4845]       1.0        1.5         < 30%
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       1.7        2.0      30 - 60%
## Ciolac 2010             0.7968 [-0.0715; 1.6650]       1.7        2.0         < 30%
## Cocks 2013             -0.5676 [-1.5671; 0.4320]       1.3        1.7      30 - 60%
## Conraads 2015           0.1953 [-0.1026; 0.4933]      14.5        4.1         < 30%
## Currie 2015             0.5330 [-0.3833; 1.4494]       1.5        1.9         < 30%
## Earnest 2013           -0.0715 [-0.7221; 0.5791]       3.0        2.7         < 30%
## Fisher 2015            -0.7440 [-1.5960; 0.1079]       1.8        2.1         < 30%
## Gillen 2016            -0.4021 [-1.3117; 0.5075]       1.6        1.9         < 30%
## Gorostiaga 1991         1.4458 [ 0.1749; 2.7166]       0.8        1.2      30 - 60%
## Grieco 2013             0.6057 [-0.2523; 1.4638]       1.7        2.1         < 30%
## Helgerud 2007           1.0281 [ 0.0955; 1.9607]       1.5        1.9         > 60%
## Helgerud 2007           0.9016 [-0.0184; 1.8216]       1.5        1.9         > 60%
## Henriksson 1976        -0.5694 [-1.9102; 0.7715]       0.7        1.1         > 60%
## Honkala 2017 (Healthy)  0.8299 [ 0.0579; 1.6019]       2.2        2.3         < 30%
## Honkala 2017 (T2D)      1.7112 [ 0.5592; 2.8633]       1.0        1.4         < 30%
## Keating 2014            0.3091 [-0.5316; 1.1498]       1.8        2.1         < 30%
## Keteyian 2014           0.6606 [-0.1020; 1.4232]       2.2        2.3         < 30%
## Kim 2015                0.9018 [ 0.1242; 1.6793]       2.1        2.3         < 30%
## Klonizakis 2014         0.1765 [-0.7729; 1.1259]       1.4        1.8         < 30%
## Lunt 2014               0.9567 [ 0.0748; 1.8385]       1.7        2.0         < 30%
## Lunt 2014               0.5858 [-0.2686; 1.4401]       1.8        2.1         < 30%
## Macpherson 2011        -0.0240 [-0.9005; 0.8526]       1.7        2.0      30 - 60%
## Madssen 2014            0.5664 [-0.1090; 1.2418]       2.8        2.6         < 30%
## Martins 2016           -0.0365 [-0.7538; 0.6809]       2.5        2.5         < 30%
## Matsuo 2014             0.6957 [-0.0960; 1.4874]       2.0        2.3      30 - 60%
## Matsuo 2015             0.8715 [ 0.0342; 1.7088]       1.8        2.1         < 30%
## Mitranun 2014           0.9929 [ 0.2078; 1.7781]       2.1        2.3         < 30%
## Molmen-Hansen 2011      0.6999 [ 0.1724; 1.2275]       4.6        3.2      30 - 60%
## Motiani 2017            0.4668 [-0.3123; 1.2460]       2.1        2.3         < 30%
## Nalcakan 2014          -0.2565 [-1.2750; 0.7620]       1.2        1.7         < 30%
## Nie 2017                0.1324 [-0.5857; 0.8505]       2.5        2.5         < 30%
## O’Leary 2018           -0.0213 [-0.8979; 0.8552]       1.7        2.0      30 - 60%
## Ramos 2016a             0.7144 [ 0.0977; 1.3311]       3.4        2.8         < 30%
## Ramos 2016b             0.8823 [ 0.1552; 1.6095]       2.4        2.5         < 30%
## Robinson 2015           0.0000 [-0.6279; 0.6279]       3.3        2.8         < 30%
## Rognmo 2004             0.6529 [-0.3245; 1.6302]       1.3        1.8      30 - 60%
## Sandvei 2012            0.2446 [-0.5765; 1.0658]       1.9        2.2         > 60%
## Sawyer 2016             0.3610 [-0.5704; 1.2925]       1.5        1.9         < 30%
## Scribbans 2014         -0.1028 [-1.0039; 0.7984]       1.6        2.0         > 60%
## Shepherd 2013          -0.5676 [-1.5671; 0.4320]       1.3        1.7      30 - 60%
## Sjöros 2018             0.7485 [-0.1373; 1.6343]       1.6        2.0         < 30%
## Skleryk 2013           -0.1275 [-1.1085; 0.8535]       1.3        1.8         < 30%
## Tjønna 2008             1.0772 [ 0.1042; 2.0502]       1.4        1.8         > 60%
## Trapp 2008              0.3351 [-0.3856; 1.0558]       2.5        2.5         < 30%
## Winn 2018              -0.2185 [-1.2014; 0.7644]       1.3        1.8         < 30%
## Wisløff 2007            4.5911 [ 2.8296; 6.3526]       0.4        0.7         < 30%
## 
## Number of studies combined: k = 48
## 
##                         SMD           95%-CI    z  p-value
## Fixed effect model   0.3794 [0.2661; 0.4927] 6.56 < 0.0001
## Random effects model 0.4017 [0.2381; 0.5653] 4.81 < 0.0001
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1456 [0.1246; 0.5265]; tau = 0.3816 [0.3529; 0.7256];
##  I^2 = 47.1% [25.8%; 62.3%]; H = 1.38 [1.16; 1.63]
## 
## Quantifying residual heterogeneity:
##  I^2 = 42.2% [17.8%; 59.4%]; H = 1.32 [1.10; 1.57]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  88.86   47  0.0002
## 
## Results for subgroups (fixed effect model):
##            k    SMD           95%-CI     Q   I^2
## < 30%     32 0.3681 [0.2358; 0.5004] 56.93 45.5%
## 30 - 60%  10 0.2967 [0.0268; 0.5667] 14.05 35.9%
## > 60%      6 0.4760 [0.0866; 0.8654]  6.93 27.9%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.56    2  0.7559
## Within groups  77.91   45  0.0017
## 
## Results for subgroups (random effects model):
##            k    SMD            95%-CI  tau^2    tau
## < 30%     32 0.4076 [ 0.2175; 0.5977] 0.1259 0.3549
## 30 - 60%  10 0.2537 [-0.0956; 0.6030] 0.1104 0.3322
## > 60%      6 0.4704 [ 0.0077; 0.9330] 0.0927 0.3045
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.73    2  0.6951
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
1.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 48; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.1523 (SE = 0.0708)
## tau (square root of estimated tau^2 value):             0.3903
## I^2 (residual heterogeneity / unaccounted variability): 48.09%
## H^2 (unaccounted variability / sampling variability):   1.93
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 46) = 88.6112, p-val = 0.0002
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0009, p-val = 0.9755
## 
## Model Results:
## 
##                estimate      se    zval    pval    ci.lb   ci.ub 
## intrcpt          0.3997  0.1131  3.5347  0.0004   0.1781  0.6213  *** 
## bsln_adjusted    0.0001  0.0033  0.0307  0.9755  -0.0064  0.0066      
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
1.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

1.2.8 Type of HIIE

1.2.8.1 Forest plot

1.2.8.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) HIIE
## Bækkerud 2016           0.5820 [-0.3903; 1.5542]       1.4        1.8 HIIT
## Beetham 2019           -0.6337 [-1.7518; 0.4845]       1.0        1.5 HIIT
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       1.7        2.0  SIT
## Ciolac 2010             0.7968 [-0.0715; 1.6650]       1.7        2.0 HIIT
## Cocks 2013             -0.5676 [-1.5671; 0.4320]       1.3        1.7  SIT
## Conraads 2015           0.1953 [-0.1026; 0.4933]      14.5        4.1 HIIT
## Currie 2015             0.5330 [-0.3833; 1.4494]       1.5        1.9 HIIT
## Earnest 2013           -0.0715 [-0.7221; 0.5791]       3.0        2.7 HIIT
## Fisher 2015            -0.7440 [-1.5960; 0.1079]       1.8        2.1  SIT
## Gillen 2016            -0.4021 [-1.3117; 0.5075]       1.6        1.9  SIT
## Gorostiaga 1991         1.4458 [ 0.1749; 2.7166]       0.8        1.2  SIT
## Grieco 2013             0.6057 [-0.2523; 1.4638]       1.7        2.1 HIIT
## Helgerud 2007           1.0281 [ 0.0955; 1.9607]       1.5        1.9 HIIT
## Helgerud 2007           0.9016 [-0.0184; 1.8216]       1.5        1.9  SIT
## Henriksson 1976        -0.5694 [-1.9102; 0.7715]       0.7        1.1 HIIT
## Honkala 2017 (Healthy)  0.8299 [ 0.0579; 1.6019]       2.2        2.3  SIT
## Honkala 2017 (T2D)      1.7112 [ 0.5592; 2.8633]       1.0        1.4  SIT
## Keating 2014            0.3091 [-0.5316; 1.1498]       1.8        2.1 HIIT
## Keteyian 2014           0.6606 [-0.1020; 1.4232]       2.2        2.3 HIIT
## Kim 2015                0.9018 [ 0.1242; 1.6793]       2.1        2.3 HIIT
## Klonizakis 2014         0.1765 [-0.7729; 1.1259]       1.4        1.8 HIIT
## Lunt 2014               0.9567 [ 0.0748; 1.8385]       1.7        2.0 HIIT
## Lunt 2014               0.5858 [-0.2686; 1.4401]       1.8        2.1  SIT
## Macpherson 2011        -0.0240 [-0.9005; 0.8526]       1.7        2.0  SIT
## Madssen 2014            0.5664 [-0.1090; 1.2418]       2.8        2.6 HIIT
## Martins 2016           -0.0365 [-0.7538; 0.6809]       2.5        2.5  SIT
## Matsuo 2014             0.6957 [-0.0960; 1.4874]       2.0        2.3 HIIT
## Matsuo 2015             0.8715 [ 0.0342; 1.7088]       1.8        2.1 HIIT
## Mitranun 2014           0.9929 [ 0.2078; 1.7781]       2.1        2.3 HIIT
## Molmen-Hansen 2011      0.6999 [ 0.1724; 1.2275]       4.6        3.2 HIIT
## Motiani 2017            0.4668 [-0.3123; 1.2460]       2.1        2.3  SIT
## Nalcakan 2014          -0.2565 [-1.2750; 0.7620]       1.2        1.7  SIT
## Nie 2017                0.1324 [-0.5857; 0.8505]       2.5        2.5 HIIT
## O’Leary 2018           -0.0213 [-0.8979; 0.8552]       1.7        2.0 HIIT
## Ramos 2016a             0.7144 [ 0.0977; 1.3311]       3.4        2.8 HIIT
## Ramos 2016b             0.8823 [ 0.1552; 1.6095]       2.4        2.5 HIIT
## Robinson 2015           0.0000 [-0.6279; 0.6279]       3.3        2.8 HIIT
## Rognmo 2004             0.6529 [-0.3245; 1.6302]       1.3        1.8 HIIT
## Sandvei 2012            0.2446 [-0.5765; 1.0658]       1.9        2.2  SIT
## Sawyer 2016             0.3610 [-0.5704; 1.2925]       1.5        1.9 HIIT
## Scribbans 2014         -0.1028 [-1.0039; 0.7984]       1.6        2.0  SIT
## Shepherd 2013          -0.5676 [-1.5671; 0.4320]       1.3        1.7  SIT
## Sjöros 2018             0.7485 [-0.1373; 1.6343]       1.6        2.0  SIT
## Skleryk 2013           -0.1275 [-1.1085; 0.8535]       1.3        1.8  SIT
## Tjønna 2008             1.0772 [ 0.1042; 2.0502]       1.4        1.8 HIIT
## Trapp 2008              0.3351 [-0.3856; 1.0558]       2.5        2.5  SIT
## Winn 2018              -0.2185 [-1.2014; 0.7644]       1.3        1.8 HIIT
## Wisløff 2007            4.5911 [ 2.8296; 6.3526]       0.4        0.7 HIIT
## 
## Number of studies combined: k = 48
## 
##                         SMD           95%-CI    z  p-value
## Fixed effect model   0.3794 [0.2661; 0.4927] 6.56 < 0.0001
## Random effects model 0.4017 [0.2381; 0.5653] 4.81 < 0.0001
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1456 [0.1246; 0.5265]; tau = 0.3816 [0.3529; 0.7256];
##  I^2 = 47.1% [25.8%; 62.3%]; H = 1.38 [1.16; 1.63]
## 
## Quantifying residual heterogeneity:
##  I^2 = 37.6% [10.9%; 56.2%]; H = 1.27 [1.06; 1.51]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  88.86   47  0.0002
## 
## Results for subgroups (fixed effect model):
##        k    SMD            95%-CI     Q   I^2
## HIIT  29 0.4500 [ 0.3131; 0.5869] 44.96 37.7%
## SIT   19 0.1760 [-0.0276; 0.3795] 28.71 37.3%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  4.79    1  0.0286
## Within groups  73.68   46  0.0059
## 
## Results for subgroups (random effects model):
##        k    SMD            95%-CI  tau^2    tau
## HIIT  29 0.4978 [ 0.3119; 0.6837] 0.0891 0.2986
## SIT   19 0.1794 [-0.0802; 0.4390] 0.1225 0.3500
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   3.82    1  0.0506
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
1.2.8.3 Meta-regression
## 
## Mixed-Effects Model (k = 48; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.1355 (SE = 0.0666)
## tau (square root of estimated tau^2 value):             0.3681
## I^2 (residual heterogeneity / unaccounted variability): 45.14%
## H^2 (unaccounted variability / sampling variability):   1.82
## R^2 (amount of heterogeneity accounted for):            6.92%
## 
## Test for Residual Heterogeneity:
## QE(df = 46) = 83.8556, p-val = 0.0005
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 3.8516, p-val = 0.0497
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb    ci.ub 
## intrcpt    0.5250  0.1036   5.0692  <.0001   0.3220   0.7280  *** 
## HIIESIT   -0.3336  0.1700  -1.9626  0.0497  -0.6668  -0.0004    * 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
1.2.8.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

2. Flow-mediated Dilation

2.1 Overall

2.1.1 Forest plot

2.1.2 R output

##                        SMD            95%-CI %W(fixed) %W(random)
## Bækkerud 2016      -0.6077 [-1.5818; 0.3663]       4.3        9.6
## Conraads 2015       0.0860 [-0.2114; 0.3833]      46.2       14.7
## Jo 2020             0.9166 [ 0.2099; 1.6233]       8.2       11.7
## Klonizakis 2014     0.0650 [-0.8829; 1.0128]       4.6        9.8
## Madssen 2014       -0.2313 [-0.8960; 0.4335]       9.3       12.1
## Mitranun 2014       0.2375 [-0.5059; 0.9809]       7.4       11.4
## Molmen-Hansen 2011  1.3059 [ 0.7421; 1.8697]      12.9       12.9
## Sawyer 2016         1.4943 [ 0.4494; 2.5393]       3.7        9.1
## Tjønna 2008         1.8981 [ 0.8055; 2.9906]       3.4        8.7
## 
## Number of studies combined: k = 9
## 
##                         SMD           95%-CI    z p-value
## Fixed effect model   0.3768 [0.1746; 0.5791] 3.65  0.0003
## Random effects model 0.5370 [0.0485; 1.0255] 2.15  0.0312
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3999 [0.1350; 2.2069]; tau = 0.6324 [0.3675; 1.4856];
##  I^2 = 77.7% [57.7%; 88.2%]; H = 2.12 [1.54; 2.92]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  35.88    8 < 0.0001
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

2.1.3 Sensitivity analysis

2.1.3.1 Forest plot

2.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                                  SMD            95%-CI p-value   tau^2     tau    I^2
## Omitting Bækkerud 2016        0.6282 [ 0.1385; 1.1178]  0.0119  0.3489  0.5907  76.2%
## Omitting Conraads 2015        0.5936 [ 0.0329; 1.1543]  0.0380  0.4683  0.6843  73.9%
## Omitting Jo 2020              0.4670 [-0.0521; 0.9861]  0.0779  0.3960  0.6293  77.3%
## Omitting Klonizakis 2014      0.5674 [ 0.0508; 1.0840]  0.0313  0.4023  0.6343  78.6%
## Omitting Madssen 2014         0.6192 [ 0.1033; 1.1352]  0.0187  0.3871  0.6222  76.5%
## Omitting Mitranun 2014        0.5564 [ 0.0233; 1.0894]  0.0408  0.4273  0.6537  78.8%
## Omitting Molmen-Hansen 2011   0.3851 [-0.0674; 0.8376]  0.0953  0.2594  0.5094  67.3%
## Omitting Sawyer 2016          0.4265 [-0.0574; 0.9104]  0.0841  0.3412  0.5841  76.0%
## Omitting Tjønna 2008          0.3951 [-0.0647; 0.8549]  0.0922  0.2976  0.5455  73.5%
##                                                                                      
## Pooled estimate               0.5370 [ 0.0485; 1.0255]  0.0312  0.3999  0.6324  77.7%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

2.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

2.2 Subgroups

2.2.1 Overall

2.2.1.1 Forest plot

2.2.1.2 R output
##                           SMD            95%-CI     meta-analysis
##                        0.5370 [ 0.0485; 1.0255]           Overall
## Healthy                0.0619 [-0.8860; 1.0098]        Population
## Overweight/obese       0.7341 [-0.4521; 1.9203]        Population
## Cardiac Rehabilitation 0.0336 [-0.2378; 0.3051]        Population
## Metabolic Syndrome     1.2483 [ 0.3728; 2.1239]        Population
## T2D                    0.2306 [-0.5131; 0.9742]        Population
## 30 - 50 y              0.4118 [-1.5480; 2.3717]               Age
## > 50 y                 0.5410 [ 0.0365; 1.0454]               Age
## < 5 weeks              0.0619 [-0.8860; 1.0098] Training Duration
## 5 - 10 weeks           0.4879 [-0.2648; 1.2405] Training Duration
## > 10 weeks             0.6571 [-0.1522; 1.4665] Training Duration
## < 0.5                  0.5499 [-0.2552; 1.3550]         Men Ratio
## > 0.5                  0.4976 [-0.1690; 1.1642]         Men Ratio
## Running                0.5593 [-0.1117; 1.2302]  Type of Exercise
## Cycling                0.4101 [-0.3234; 1.1436]  Type of Exercise
## < 6 %                  0.7518 [-0.0275; 1.5311]   Baseline Values
## > 6 %                  0.3359 [-0.3868; 1.0587]   Baseline Values
## 
## Number of studies combined: k = 9
## 
##                         SMD           95%-CI    z p-value
## Random effects model 0.5370 [0.0485; 1.0255] 2.15  0.0312
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3999; tau = 0.6324; I^2 = 77.7% [57.7%; 88.2%]; H = 2.12 [1.54; 2.92]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  35.88    8 < 0.0001
## 
## Results for meta-analyses (random effects model):
##                     k    SMD           95%-CI  tau^2    tau     Q   I^2
## Overall             9 0.5370 [0.0485; 1.0255] 0.3999 0.6324 35.88 77.7%
## Population          9 0.5370 [0.0485; 1.0255] 0.3999 0.6324 35.88 77.7%
## Age                 9 0.5370 [0.0485; 1.0255] 0.3999 0.6324 35.88 77.7%
## Training Duration   9 0.5370 [0.0485; 1.0255] 0.3999 0.6324 35.88 77.7%
## Men Ratio           9 0.5370 [0.0485; 1.0255] 0.3999 0.6324 35.88 77.7%
## Type of Exercise    9 0.5370 [0.0485; 1.0255] 0.3999 0.6324 35.88 77.7%
## Baseline Values     9 0.5370 [0.0485; 1.0255] 0.3999 0.6324 35.88 77.7%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

2.2.2 Population

2.2.2.1 Forest plot

2.2.2.2 R output
##                        SMD            95%-CI %W(fixed) %W(random)             population
## Bækkerud 2016      -0.6077 [-1.5818; 0.3663]       4.3        9.6       Overweight/obese
## Conraads 2015       0.0860 [-0.2114; 0.3833]      46.2       14.7 Cardiac Rehabilitation
## Jo 2020             0.9166 [ 0.2099; 1.6233]       8.2       11.7     Metabolic Syndrome
## Klonizakis 2014     0.0650 [-0.8829; 1.0128]       4.6        9.8                Healthy
## Madssen 2014       -0.2313 [-0.8960; 0.4335]       9.3       12.1 Cardiac Rehabilitation
## Mitranun 2014       0.2375 [-0.5059; 0.9809]       7.4       11.4                    T2D
## Molmen-Hansen 2011  1.3059 [ 0.7421; 1.8697]      12.9       12.9       Overweight/obese
## Sawyer 2016         1.4943 [ 0.4494; 2.5393]       3.7        9.1       Overweight/obese
## Tjønna 2008         1.8981 [ 0.8055; 2.9906]       3.4        8.7     Metabolic Syndrome
## 
## Number of studies combined: k = 9
## 
##                         SMD           95%-CI    z p-value
## Fixed effect model   0.3768 [0.1746; 0.5791] 3.65  0.0003
## Random effects model 0.5370 [0.0485; 1.0255] 2.15  0.0312
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3999 [0.1350; 2.2069]; tau = 0.6324 [0.3675; 1.4856];
##  I^2 = 77.7% [57.7%; 88.2%]; H = 2.12 [1.54; 2.92]
## 
## Quantifying residual heterogeneity:
##  I^2 = 71.5% [28.0%; 88.7%]; H = 1.87 [1.18; 2.98]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  35.88    8 < 0.0001
## 
## Results for subgroups (fixed effect model):
##                          k    SMD            95%-CI     Q   I^2
## Healthy                  1 0.0619 [-0.8860; 1.0098]  0.00    --
## Overweight/obese         3 0.9260 [ 0.4812; 1.3709] 11.49 82.6%
## Cardiac Rehabilitation   2 0.0336 [-0.2378; 0.3051]  0.70  0.0%
## Metabolic Syndrome       2 1.1586 [ 0.5598; 1.7575]  1.85 45.9%
## T2D                      1 0.2306 [-0.5131; 0.9742]  0.00    --
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups 19.10    4  0.0008
## Within groups  14.05    4  0.0072
## 
## Results for subgroups (random effects model):
##                          k    SMD            95%-CI  tau^2    tau
## Healthy                  1 0.0619 [-0.8860; 1.0098]     --     --
## Overweight/obese         3 0.7341 [-0.4521; 1.9203] 0.8979 0.9476
## Cardiac Rehabilitation   2 0.0336 [-0.2378; 0.3051]      0      0
## Metabolic Syndrome       2 1.2483 [ 0.3728; 2.1239] 0.1935 0.4398
## T2D                      1 0.2306 [-0.5131; 0.9742]     --     --
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   7.74    4  0.1016
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
2.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 9; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.3798 (SE = 0.3944)
## tau (square root of estimated tau^2 value):             0.6163
## I^2 (residual heterogeneity / unaccounted variability): 73.88%
## H^2 (unaccounted variability / sampling variability):   3.83
## R^2 (amount of heterogeneity accounted for):            5.02%
## 
## Test for Residual Heterogeneity:
## QE(df = 4) = 15.3125, p-val = 0.0041
## 
## Test of Moderators (coefficients 2:5):
## QM(df = 4) = 4.5928, p-val = 0.3317
## 
## Model Results:
## 
##                               estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                         0.0650  0.7834   0.0829  0.9339  -1.4705  1.6004    
## .byvarOverweight/obese          0.7148  0.8965   0.7973  0.4253  -1.0424  2.4719    
## .byvarCardiac Rehabilitation   -0.1214  0.9142  -0.1327  0.8944  -1.9132  1.6705    
## .byvarMetabolic Syndrome        1.2685  0.9524   1.3319  0.1829  -0.5981  3.1351    
## .byvarT2D                       0.1725  1.0665   0.1618  0.8715  -1.9178  2.2629    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
2.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

2.2.3 Age

2.2.3.1 Forest plot

2.2.3.2 R output
##                        SMD            95%-CI %W(fixed) %W(random) category_age
## Bækkerud 2016      -0.6077 [-1.5818; 0.3663]       4.3        9.6    30 - 50 y
## Conraads 2015       0.0860 [-0.2114; 0.3833]      46.2       14.7       > 50 y
## Jo 2020             0.9166 [ 0.2099; 1.6233]       8.2       11.7       > 50 y
## Klonizakis 2014     0.0650 [-0.8829; 1.0128]       4.6        9.8       > 50 y
## Madssen 2014       -0.2313 [-0.8960; 0.4335]       9.3       12.1       > 50 y
## Mitranun 2014       0.2375 [-0.5059; 0.9809]       7.4       11.4       > 50 y
## Molmen-Hansen 2011  1.3059 [ 0.7421; 1.8697]      12.9       12.9       > 50 y
## Sawyer 2016         1.4943 [ 0.4494; 2.5393]       3.7        9.1    30 - 50 y
## Tjønna 2008         1.8981 [ 0.8055; 2.9906]       3.4        8.7       > 50 y
## 
## Number of studies combined: k = 9
## 
##                         SMD           95%-CI    z p-value
## Fixed effect model   0.3768 [0.1746; 0.5791] 3.65  0.0003
## Random effects model 0.5370 [0.0485; 1.0255] 2.15  0.0312
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3999 [0.1350; 2.2069]; tau = 0.6324 [0.3675; 1.4856];
##  I^2 = 77.7% [57.7%; 88.2%]; H = 2.12 [1.54; 2.92]
## 
## Quantifying residual heterogeneity:
##  I^2 = 78.9% [58.6%; 89.2%]; H = 2.18 [1.55; 3.04]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  35.88    8 < 0.0001
## 
## Results for subgroups (fixed effect model):
##             k    SMD            95%-CI     Q   I^2
## 30 - 50 y   2 0.3396 [-0.3801; 1.0592]  7.37 86.4%
## > 50 y      7 0.3659 [ 0.1546; 0.5772] 25.77 76.7%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f.  p-value
## Between groups  0.00    1   0.9452
## Within groups  33.14    7 < 0.0001
## 
## Results for subgroups (random effects model):
##             k    SMD            95%-CI  tau^2    tau
## 30 - 50 y   2 0.4118 [-1.5480; 2.3717] 1.7285 1.3147
## > 50 y      7 0.5410 [ 0.0365; 1.0454] 0.3295 0.5740
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.02    1  0.9005
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
2.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 9; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.3793 (SE = 0.3034)
## tau (square root of estimated tau^2 value):             0.6158
## I^2 (residual heterogeneity / unaccounted variability): 77.00%
## H^2 (unaccounted variability / sampling variability):   4.35
## R^2 (amount of heterogeneity accounted for):            5.16%
## 
## Test for Residual Heterogeneity:
## QE(df = 7) = 30.4306, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.7035, p-val = 0.4016
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    1.8217  1.5525   1.1734  0.2406  -1.2211  4.8644    
## age       -0.0241  0.0287  -0.8388  0.4016  -0.0803  0.0322    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
2.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

2.2.4 Training Duration

2.2.4.1 Forest plot

2.2.4.2 R output
##                        SMD            95%-CI %W(fixed) %W(random) category_duration
## Bækkerud 2016      -0.6077 [-1.5818; 0.3663]       4.3        9.6      5 - 10 weeks
## Conraads 2015       0.0860 [-0.2114; 0.3833]      46.2       14.7        > 10 weeks
## Jo 2020             0.9166 [ 0.2099; 1.6233]       8.2       11.7      5 - 10 weeks
## Klonizakis 2014     0.0650 [-0.8829; 1.0128]       4.6        9.8         < 5 weeks
## Madssen 2014       -0.2313 [-0.8960; 0.4335]       9.3       12.1        > 10 weeks
## Mitranun 2014       0.2375 [-0.5059; 0.9809]       7.4       11.4      5 - 10 weeks
## Molmen-Hansen 2011  1.3059 [ 0.7421; 1.8697]      12.9       12.9        > 10 weeks
## Sawyer 2016         1.4943 [ 0.4494; 2.5393]       3.7        9.1      5 - 10 weeks
## Tjønna 2008         1.8981 [ 0.8055; 2.9906]       3.4        8.7        > 10 weeks
## 
## Number of studies combined: k = 9
## 
##                         SMD           95%-CI    z p-value
## Fixed effect model   0.3768 [0.1746; 0.5791] 3.65  0.0003
## Random effects model 0.5370 [0.0485; 1.0255] 2.15  0.0312
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3999 [0.1350; 2.2069]; tau = 0.6324 [0.3675; 1.4856];
##  I^2 = 77.7% [57.7%; 88.2%]; H = 2.12 [1.54; 2.92]
## 
## Quantifying residual heterogeneity:
##  I^2 = 81.4% [62.7%; 90.8%]; H = 2.32 [1.64; 3.29]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  35.88    8 < 0.0001
## 
## Results for subgroups (fixed effect model):
##                k    SMD            95%-CI     Q   I^2
## < 5 weeks      1 0.0619 [-0.8860; 1.0098]  0.00    --
## 5 - 10 weeks   4 0.4979 [ 0.0800; 0.9158]  9.25 67.6%
## > 10 weeks     4 0.3391 [ 0.1000; 0.5782] 23.07 87.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f.  p-value
## Between groups  0.83    2   0.6615
## Within groups  32.32    6 < 0.0001
## 
## Results for subgroups (random effects model):
##                k    SMD            95%-CI  tau^2    tau
## < 5 weeks      1 0.0619 [-0.8860; 1.0098]     --     --
## 5 - 10 weeks   4 0.4879 [-0.2648; 1.2405] 0.3933 0.6271
## > 10 weeks     4 0.6571 [-0.1522; 1.4665] 0.5619 0.7496
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.90    2  0.6366
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
2.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 9; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.4532 (SE = 0.3554)
## tau (square root of estimated tau^2 value):             0.6732
## I^2 (residual heterogeneity / unaccounted variability): 79.93%
## H^2 (unaccounted variability / sampling variability):   4.98
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 7) = 34.8759, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.4855, p-val = 0.2229
## 
## Model Results:
## 
##           estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    -0.3034  0.7395  -0.4102  0.6816  -1.7528  1.1461    
## duration    0.0868  0.0712   1.2188  0.2229  -0.0528  0.2264    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
2.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

2.2.5 Men Ratio

2.2.5.1 Forest plot

2.2.5.2 R output
##                        SMD            95%-CI %W(fixed) %W(random) category_men_ratio
## Bækkerud 2016      -0.6077 [-1.5818; 0.3663]       4.3        9.6              < 0.5
## Conraads 2015       0.0860 [-0.2114; 0.3833]      46.2       14.7              > 0.5
## Jo 2020             0.9166 [ 0.2099; 1.6233]       8.2       11.7              > 0.5
## Klonizakis 2014     0.0650 [-0.8829; 1.0128]       4.6        9.8              < 0.5
## Madssen 2014       -0.2313 [-0.8960; 0.4335]       9.3       12.1              > 0.5
## Mitranun 2014       0.2375 [-0.5059; 0.9809]       7.4       11.4              < 0.5
## Molmen-Hansen 2011  1.3059 [ 0.7421; 1.8697]      12.9       12.9              > 0.5
## Sawyer 2016         1.4943 [ 0.4494; 2.5393]       3.7        9.1              < 0.5
## Tjønna 2008         1.8981 [ 0.8055; 2.9906]       3.4        8.7              < 0.5
## 
## Number of studies combined: k = 9
## 
##                         SMD           95%-CI    z p-value
## Fixed effect model   0.3768 [0.1746; 0.5791] 3.65  0.0003
## Random effects model 0.5370 [0.0485; 1.0255] 2.15  0.0312
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3999 [0.1350; 2.2069]; tau = 0.6324 [0.3675; 1.4856];
##  I^2 = 77.7% [57.7%; 88.2%]; H = 2.12 [1.54; 2.92]
## 
## Quantifying residual heterogeneity:
##  I^2 = 78.7% [58.3%; 89.1%]; H = 2.17 [1.55; 3.03]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  35.88    8 < 0.0001
## 
## Results for subgroups (fixed effect model):
##         k    SMD           95%-CI     Q   I^2
## < 0.5   5 0.4588 [0.0382; 0.8794] 14.15 71.7%
## > 0.5   4 0.3351 [0.1037; 0.5664] 18.74 84.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f.  p-value
## Between groups  0.26    1   0.6135
## Within groups  32.89    7 < 0.0001
## 
## Results for subgroups (random effects model):
##         k    SMD            95%-CI  tau^2    tau
## < 0.5   5 0.5499 [-0.2552; 1.3550] 0.5991 0.7740
## > 0.5   4 0.4976 [-0.1690; 1.1642] 0.3784 0.6152
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.01    1  0.9219
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
2.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 9; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.4713 (SE = 0.3532)
## tau (square root of estimated tau^2 value):             0.6865
## I^2 (residual heterogeneity / unaccounted variability): 77.56%
## H^2 (unaccounted variability / sampling variability):   4.46
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 7) = 31.1980, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0857, p-val = 0.7698
## 
## Model Results:
## 
##            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt      0.7020  0.6113   1.1483  0.2508  -0.4962  1.9002    
## men_ratio   -0.3077  1.0514  -0.2927  0.7698  -2.3684  1.7529    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
2.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

2.2.6 Type of Exercise

2.2.6.1 Forest plot

2.2.6.2 R output
##                        SMD            95%-CI %W(fixed) %W(random) type_exercise
## Bækkerud 2016      -0.6077 [-1.5818; 0.3663]       4.3        9.6       Running
## Conraads 2015       0.0860 [-0.2114; 0.3833]      46.2       14.7       Cycling
## Jo 2020             0.9166 [ 0.2099; 1.6233]       8.2       11.7       Running
## Klonizakis 2014     0.0650 [-0.8829; 1.0128]       4.6        9.8       Cycling
## Madssen 2014       -0.2313 [-0.8960; 0.4335]       9.3       12.1       Running
## Mitranun 2014       0.2375 [-0.5059; 0.9809]       7.4       11.4       Running
## Molmen-Hansen 2011  1.3059 [ 0.7421; 1.8697]      12.9       12.9       Running
## Sawyer 2016         1.4943 [ 0.4494; 2.5393]       3.7        9.1       Cycling
## Tjønna 2008         1.8981 [ 0.8055; 2.9906]       3.4        8.7       Running
## 
## Number of studies combined: k = 9
## 
##                         SMD           95%-CI    z p-value
## Fixed effect model   0.3768 [0.1746; 0.5791] 3.65  0.0003
## Random effects model 0.5370 [0.0485; 1.0255] 2.15  0.0312
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3999 [0.1350; 2.2069]; tau = 0.6324 [0.3675; 1.4856];
##  I^2 = 77.7% [57.7%; 88.2%]; H = 2.12 [1.54; 2.92]
## 
## Quantifying residual heterogeneity:
##  I^2 = 75.9% [51.7%; 88.0%]; H = 2.04 [1.44; 2.88]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  35.88    8 < 0.0001
## 
## Results for subgroups (fixed effect model):
##           k    SMD            95%-CI     Q   I^2
## Running   6 0.5946 [ 0.2934; 0.8957] 23.32 78.6%
## Cycling   3 0.1725 [-0.1016; 0.4467]  5.70 64.9%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  4.13    1  0.0422
## Within groups  29.02    7  0.0001
## 
## Results for subgroups (random effects model):
##           k    SMD            95%-CI  tau^2    tau
## Running   6 0.5593 [-0.1117; 1.2302] 0.5379 0.7334
## Cycling   3 0.4101 [-0.3234; 1.1436] 0.2716 0.5211
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.09    1  0.7687
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
2.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 9; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.5099 (SE = 0.3668)
## tau (square root of estimated tau^2 value):             0.7141
## I^2 (residual heterogeneity / unaccounted variability): 77.82%
## H^2 (unaccounted variability / sampling variability):   4.51
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 7) = 31.5548, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0302, p-val = 0.8620
## 
## Model Results:
## 
##                       estimate      se    zval    pval    ci.lb   ci.ub 
## intrcpt                 0.4755  0.4724  1.0065  0.3142  -0.4505  1.4015    
## type_exerciseRunning    0.1007  0.5791  0.1738  0.8620  -1.0343  1.2356    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
2.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

2.2.7 Baseline Values

2.2.7.1 Forest plot

2.2.7.2 R output
##                        SMD            95%-CI %W(fixed) %W(random) category_bsln
## Bækkerud 2016      -0.6077 [-1.5818; 0.3663]       4.3        9.6         > 6 %
## Conraads 2015       0.0860 [-0.2114; 0.3833]      46.2       14.7         < 6 %
## Jo 2020             0.9166 [ 0.2099; 1.6233]       8.2       11.7         > 6 %
## Klonizakis 2014     0.0650 [-0.8829; 1.0128]       4.6        9.8         > 6 %
## Madssen 2014       -0.2313 [-0.8960; 0.4335]       9.3       12.1         > 6 %
## Mitranun 2014       0.2375 [-0.5059; 0.9809]       7.4       11.4         < 6 %
## Molmen-Hansen 2011  1.3059 [ 0.7421; 1.8697]      12.9       12.9         > 6 %
## Sawyer 2016         1.4943 [ 0.4494; 2.5393]       3.7        9.1         < 6 %
## Tjønna 2008         1.8981 [ 0.8055; 2.9906]       3.4        8.7         < 6 %
## 
## Number of studies combined: k = 9
## 
##                         SMD           95%-CI    z p-value
## Fixed effect model   0.3768 [0.1746; 0.5791] 3.65  0.0003
## Random effects model 0.5370 [0.0485; 1.0255] 2.15  0.0312
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3999 [0.1350; 2.2069]; tau = 0.6324 [0.3675; 1.4856];
##  I^2 = 77.7% [57.7%; 88.2%]; H = 2.12 [1.54; 2.92]
## 
## Quantifying residual heterogeneity:
##  I^2 = 78.2% [57.0%; 88.9%]; H = 2.14 [1.52; 3.00]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  35.88    8 < 0.0001
## 
## Results for subgroups (fixed effect model):
##         k    SMD           95%-CI     Q   I^2
## < 6 %   4 0.2767 [0.0168; 0.5367] 13.33 77.5%
## > 6 %   5 0.4991 [0.1751; 0.8230] 18.71 78.6%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f.  p-value
## Between groups  1.10    1   0.2941
## Within groups  32.04    7 < 0.0001
## 
## Results for subgroups (random effects model):
##         k    SMD            95%-CI  tau^2    tau
## < 6 %   4 0.7518 [-0.0275; 1.5311] 0.4600 0.6782
## > 6 %   5 0.3359 [-0.3868; 1.0587] 0.5238 0.7237
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.59    1  0.4432
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
2.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 9; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.4646 (SE = 0.3631)
## tau (square root of estimated tau^2 value):             0.6816
## I^2 (residual heterogeneity / unaccounted variability): 79.25%
## H^2 (unaccounted variability / sampling variability):   4.82
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 7) = 33.7369, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 2.2409, p-val = 0.1344
## 
## Model Results:
## 
##                estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt          1.9301  0.9650   2.0001  0.0455   0.0387  3.8214  * 
## bsln_adjusted   -0.2105  0.1406  -1.4970  0.1344  -0.4861  0.0651    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
2.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

3. BMI

3.1 Overall

3.1.1 Forest plot

3.1.2 R output

##                            SMD            95%-CI %W(fixed) %W(random)
## Abdelbasset 2020       -0.0910 [-0.7957; 0.6138]       3.7        3.7
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.9        1.9
## Conraads 2015          -0.1764 [-0.4743; 0.1214]      20.8       20.8
## Currie 2015             0.0405 [-0.8602; 0.9411]       2.3        2.3
## Eguchi 2012             0.0492 [-0.8275; 0.9259]       2.4        2.4
## Fisher 2015            -0.1130 [-0.9380; 0.7121]       2.7        2.7
## Gillen 2016             0.3289 [-0.5777; 1.2354]       2.2        2.2
## Grieco 2013            -0.0678 [-0.9073; 0.7716]       2.6        2.6
## Honkala 2017 (Healthy)  0.1477 [-0.5941; 0.8895]       3.4        3.4
## Honkala 2017 (T2D)      0.0775 [-0.9106; 1.0656]       1.9        1.9
## Jo 2020                -0.1235 [-0.7964; 0.5494]       4.1        4.1
## Lunt 2014              -0.0548 [-0.8923; 0.7827]       2.6        2.6
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.6        2.6
## Madssen 2014           -0.0471 [-0.7098; 0.6156]       4.2        4.2
## Maillard 2016           0.1751 [-0.8067; 1.1570]       1.9        1.9
## Matsuo 2014            -0.3770 [-1.1526; 0.3985]       3.1        3.1
## Mitranun 2014           0.7236 [-0.0411; 1.4883]       3.2        3.2
## Moreira 2008            0.0000 [-0.9800; 0.9800]       1.9        1.9
## Motiani 2017            0.1618 [-0.6082; 0.9318]       3.1        3.1
## Nalcakan 2014          -0.0869 [-1.1018; 0.9279]       1.8        1.8
## Nie 2017               -0.1134 [-0.8313; 0.6044]       3.6        3.6
## Ramos 2016b             0.3328 [-0.3662; 1.0319]       3.8        3.8
## Robinson 2015           0.0659 [-0.5622; 0.6939]       4.7        4.7
## Sandvei 2012            0.0636 [-0.7547; 0.8819]       2.8        2.8
## Sawyer 2016             0.0000 [-0.9239; 0.9239]       2.2        2.2
## Shepherd 2013           0.0000 [-0.9800; 0.9800]       1.9        1.9
## Sjöros 2018             0.0000 [-0.8564; 0.8564]       2.5        2.5
## Skleryk 2013            0.0000 [-0.9800; 0.9800]       1.9        1.9
## Tjønna 2008            -0.1783 [-1.0908; 0.7341]       2.2        2.2
## Winn 2018              -0.4417 [-1.4336; 0.5501]       1.9        1.9
## 
## Number of studies combined: k = 30
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## Random effects model -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  8.91   29  0.9999
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

3.1.3 Sensitivity analysis

3.1.3.1 Forest plot

3.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                                       SMD            95%-CI p-value   tau^2     tau   I^2
## Omitting Abdelbasset 2020         -0.0160 [-0.1546; 0.1226]  0.8213  0.0000  0.0000  0.0%
## Omitting Cocks 2013               -0.0190 [-0.1564; 0.1183]  0.7858  0.0000  0.0000  0.0%
## Omitting Conraads 2015             0.0227 [-0.1302; 0.1755]  0.7712  0.0000  0.0000  0.0%
## Omitting Currie 2015              -0.0200 [-0.1576; 0.1176]  0.7756  0.0000  0.0000  0.0%
## Omitting Eguchi 2012              -0.0203 [-0.1579; 0.1174]  0.7726  0.0000  0.0000  0.0%
## Omitting Fisher 2015              -0.0162 [-0.1540; 0.1217]  0.8184  0.0000  0.0000  0.0%
## Omitting Gillen 2016              -0.0263 [-0.1639; 0.1112]  0.7076  0.0000  0.0000  0.0%
## Omitting Grieco 2013              -0.0174 [-0.1552; 0.1204]  0.8043  0.0000  0.0000  0.0%
## Omitting Honkala 2017 (Healthy)   -0.0243 [-0.1626; 0.1140]  0.7305  0.0000  0.0000  0.0%
## Omitting Honkala 2017 (T2D)       -0.0204 [-0.1577; 0.1168]  0.7704  0.0000  0.0000  0.0%
## Omitting Jo 2020                  -0.0143 [-0.1532; 0.1245]  0.8396  0.0000  0.0000  0.0%
## Omitting Lunt 2014                -0.0177 [-0.1556; 0.1201]  0.8007  0.0000  0.0000  0.0%
## Omitting Lunt 2014                -0.0192 [-0.1570; 0.1186]  0.7850  0.0000  0.0000  0.0%
## Omitting Madssen 2014             -0.0175 [-0.1564; 0.1215]  0.8053  0.0000  0.0000  0.0%
## Omitting Maillard 2016            -0.0223 [-0.1596; 0.1150]  0.7505  0.0000  0.0000  0.0%
## Omitting Matsuo 2014              -0.0077 [-0.1458; 0.1304]  0.9129  0.0000  0.0000  0.0%
## Omitting Mitranun 2014            -0.0421 [-0.1803; 0.0961]  0.5506  0.0000  0.0000  0.0%
## Omitting Moreira 2008             -0.0190 [-0.1564; 0.1183]  0.7858  0.0000  0.0000  0.0%
## Omitting Motiani 2017             -0.0243 [-0.1625; 0.1138]  0.7301  0.0000  0.0000  0.0%
## Omitting Nalcakan 2014            -0.0175 [-0.1547; 0.1197]  0.8024  0.0000  0.0000  0.0%
## Omitting Nie 2017                 -0.0153 [-0.1538; 0.1232]  0.8290  0.0000  0.0000  0.0%
## Omitting Ramos 2016b              -0.0322 [-0.1708; 0.1065]  0.6494  0.0000  0.0000  0.0%
## Omitting Robinson 2015            -0.0228 [-0.1621; 0.1165]  0.7487  0.0000  0.0000  0.0%
## Omitting Sandvei 2012             -0.0209 [-0.1588; 0.1170]  0.7660  0.0000  0.0000  0.0%
## Omitting Sawyer 2016              -0.0191 [-0.1566; 0.1184]  0.7855  0.0000  0.0000  0.0%
## Omitting Shepherd 2013            -0.0190 [-0.1564; 0.1183]  0.7858  0.0000  0.0000  0.0%
## Omitting Sjöros 2018              -0.0192 [-0.1569; 0.1186]  0.7852  0.0000  0.0000  0.0%
## Omitting Skleryk 2013             -0.0190 [-0.1564; 0.1183]  0.7858  0.0000  0.0000  0.0%
## Omitting Tjønna 2008              -0.0152 [-0.1527; 0.1223]  0.8282  0.0000  0.0000  0.0%
## Omitting Winn 2018                -0.0111 [-0.1483; 0.1262]  0.8745  0.0000  0.0000  0.0%
##                                                                                          
## Pooled estimate                   -0.0183 [-0.1543; 0.1176]  0.7919  0.0000  0.0000  0.0%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

3.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

3.2 Subgroups

3.2.1 Overall

3.2.1.1 Forest plot

3.2.1.2 R output
##                            SMD            95%-CI     meta-analysis
##                        -0.0183 [-0.1543; 0.1176]           Overall
## Healthy                 0.0043 [-0.2490; 0.2576]        Population
## Overweight/obese       -0.0766 [-0.4179; 0.2647]        Population
## Cardiac Rehabilitation -0.1378 [-0.3979; 0.1223]        Population
## Metabolic Syndrome      0.0446 [-0.3093; 0.3984]        Population
## T2D                     0.1769 [-0.1974; 0.5512]        Population
## < 30 y                 -0.0554 [-0.3412; 0.2303]               Age
## 30 - 50 y               0.0078 [-0.2694; 0.2851]               Age
## > 50 y                 -0.0150 [-0.2013; 0.1713]               Age
## < 5 weeks               0.0208 [-0.2691; 0.3108] Training Duration
## 5 - 10 weeks           -0.0003 [-0.2605; 0.2600] Training Duration
## > 10 weeks             -0.0457 [-0.2367; 0.1453] Training Duration
## < 0.5                   0.0352 [-0.2045; 0.2749]         Men Ratio
## > 0.5                  -0.0442 [-0.2093; 0.1209]         Men Ratio
## Cycling                -0.0480 [-0.2075; 0.1115]  Type of Exercise
## Running                 0.0592 [-0.2008; 0.3193]  Type of Exercise
## BMI < 25 kg/m²         -0.1022 [-0.4470; 0.2425]   Baseline Values
## BMI 25 - 30 kg/m²      -0.0185 [-0.2024; 0.1654]   Baseline Values
## BMI > 30 kg/m²          0.0247 [-0.2246; 0.2740]   Baseline Values
## HIIT                   -0.0479 [-0.2091; 0.1133]      Type of HIIE
## SIT                     0.0535 [-0.1998; 0.3067]      Type of HIIE
## 
## Number of studies combined: k = 30
## 
##                          SMD            95%-CI     z p-value
## Random effects model -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## 
## Quantifying heterogeneity:
##  tau^2 = 0; tau = 0; I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  8.91   29  0.9999
## 
## Results for meta-analyses (random effects model):
##                     k     SMD            95%-CI tau^2 tau    Q  I^2
## Overall            30 -0.0183 [-0.1543; 0.1176]     0   0 8.91 0.0%
## Population         30 -0.0183 [-0.1543; 0.1176]     0   0 8.91 0.0%
## Age                30 -0.0183 [-0.1543; 0.1176]     0   0 8.91 0.0%
## Training Duration  30 -0.0183 [-0.1543; 0.1176]     0   0 8.91 0.0%
## Men Ratio          30 -0.0183 [-0.1543; 0.1176]     0   0 8.91 0.0%
## Type of Exercise   30 -0.0183 [-0.1543; 0.1176]     0   0 8.91 0.0%
## Baseline Values    30 -0.0183 [-0.1543; 0.1176]     0   0 8.91 0.0%
## Type of HIIE       30 -0.0183 [-0.1543; 0.1176]     0   0 8.91 0.0%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

3.2.2 Population

3.2.2.1 Forest plot

3.2.2.2 R output
##                            SMD            95%-CI %W(fixed) %W(random)             population
## Abdelbasset 2020       -0.0910 [-0.7957; 0.6138]       3.7        3.7                    T2D
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.9        1.9                Healthy
## Conraads 2015          -0.1764 [-0.4743; 0.1214]      20.8       20.8 Cardiac Rehabilitation
## Currie 2015             0.0405 [-0.8602; 0.9411]       2.3        2.3 Cardiac Rehabilitation
## Eguchi 2012             0.0492 [-0.8275; 0.9259]       2.4        2.4                Healthy
## Fisher 2015            -0.1130 [-0.9380; 0.7121]       2.7        2.7       Overweight/obese
## Gillen 2016             0.3289 [-0.5777; 1.2354]       2.2        2.2                Healthy
## Grieco 2013            -0.0678 [-0.9073; 0.7716]       2.6        2.6                Healthy
## Honkala 2017 (Healthy)  0.1477 [-0.5941; 0.8895]       3.4        3.4                Healthy
## Honkala 2017 (T2D)      0.0775 [-0.9106; 1.0656]       1.9        1.9                    T2D
## Jo 2020                -0.1235 [-0.7964; 0.5494]       4.1        4.1     Metabolic Syndrome
## Lunt 2014              -0.0548 [-0.8923; 0.7827]       2.6        2.6       Overweight/obese
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.6        2.6       Overweight/obese
## Madssen 2014           -0.0471 [-0.7098; 0.6156]       4.2        4.2 Cardiac Rehabilitation
## Maillard 2016           0.1751 [-0.8067; 1.1570]       1.9        1.9                    T2D
## Matsuo 2014            -0.3770 [-1.1526; 0.3985]       3.1        3.1                Healthy
## Mitranun 2014           0.7236 [-0.0411; 1.4883]       3.2        3.2                    T2D
## Moreira 2008            0.0000 [-0.9800; 0.9800]       1.9        1.9       Overweight/obese
## Motiani 2017            0.1618 [-0.6082; 0.9318]       3.1        3.1                Healthy
## Nalcakan 2014          -0.0869 [-1.1018; 0.9279]       1.8        1.8                Healthy
## Nie 2017               -0.1134 [-0.8313; 0.6044]       3.6        3.6                Healthy
## Ramos 2016b             0.3328 [-0.3662; 1.0319]       3.8        3.8     Metabolic Syndrome
## Robinson 2015           0.0659 [-0.5622; 0.6939]       4.7        4.7     Metabolic Syndrome
## Sandvei 2012            0.0636 [-0.7547; 0.8819]       2.8        2.8                Healthy
## Sawyer 2016             0.0000 [-0.9239; 0.9239]       2.2        2.2       Overweight/obese
## Shepherd 2013           0.0000 [-0.9800; 0.9800]       1.9        1.9                Healthy
## Sjöros 2018             0.0000 [-0.8564; 0.8564]       2.5        2.5                    T2D
## Skleryk 2013            0.0000 [-0.9800; 0.9800]       1.9        1.9       Overweight/obese
## Tjønna 2008            -0.1783 [-1.0908; 0.7341]       2.2        2.2     Metabolic Syndrome
## Winn 2018              -0.4417 [-1.4336; 0.5501]       1.9        1.9       Overweight/obese
## 
## Number of studies combined: k = 30
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## Random effects model -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  8.91   29  0.9999
## 
## Results for subgroups (fixed effect model):
##                          k     SMD            95%-CI    Q  I^2
## Healthy                 11  0.0043 [-0.2490; 0.2576] 1.78 0.0%
## Overweight/obese         7 -0.0766 [-0.4179; 0.2647] 0.57 0.0%
## Cardiac Rehabilitation   3 -0.1378 [-0.3979; 0.1223] 0.28 0.0%
## Metabolic Syndrome       4  0.0446 [-0.3093; 0.3984] 1.06 0.0%
## T2D                      5  0.1769 [-0.1974; 0.5512] 2.56 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 2.12    4  0.7136
## Within groups  6.25   25  0.9999
## 
## Results for subgroups (random effects model):
##                          k     SMD            95%-CI tau^2 tau
## Healthy                 11  0.0043 [-0.2490; 0.2576]     0   0
## Overweight/obese         7 -0.0766 [-0.4179; 0.2647]     0   0
## Cardiac Rehabilitation   3 -0.1378 [-0.3979; 0.1223]     0   0
## Metabolic Syndrome       4  0.0446 [-0.3093; 0.3984]     0   0
## T2D                      5  0.1769 [-0.1974; 0.5512]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   2.12    4  0.7136
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
3.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 30; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0476)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 25) = 6.6874, p-val = 0.9999
## 
## Test of Moderators (coefficients 2:5):
## QM(df = 4) = 2.2274, p-val = 0.6940
## 
## Model Results:
## 
##                               estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                         0.0047  0.1292   0.0361  0.9712  -0.2486  0.2579    
## .byvarOverweight/obese         -0.0853  0.2168  -0.3936  0.6938  -0.5102  0.3396    
## .byvarCardiac Rehabilitation   -0.1431  0.1852  -0.7726  0.4398  -0.5061  0.2199    
## .byvarMetabolic Syndrome        0.0405  0.2220   0.1826  0.8551  -0.3945  0.4756    
## .byvarT2D                       0.1793  0.2305   0.7782  0.4364  -0.2723  0.6310    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
3.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

3.2.3 Age

3.2.3.1 Forest plot

3.2.3.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_age
## Abdelbasset 2020       -0.0910 [-0.7957; 0.6138]       3.7        3.7       > 50 y
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.9        1.9       < 30 y
## Conraads 2015          -0.1764 [-0.4743; 0.1214]      20.8       20.8       > 50 y
## Currie 2015             0.0405 [-0.8602; 0.9411]       2.3        2.3       > 50 y
## Eguchi 2012             0.0492 [-0.8275; 0.9259]       2.4        2.4       > 50 y
## Fisher 2015            -0.1130 [-0.9380; 0.7121]       2.7        2.7       < 30 y
## Gillen 2016             0.3289 [-0.5777; 1.2354]       2.2        2.2       < 30 y
## Grieco 2013            -0.0678 [-0.9073; 0.7716]       2.6        2.6       < 30 y
## Honkala 2017 (Healthy)  0.1477 [-0.5941; 0.8895]       3.4        3.4    30 - 50 y
## Honkala 2017 (T2D)      0.0775 [-0.9106; 1.0656]       1.9        1.9    30 - 50 y
## Jo 2020                -0.1235 [-0.7964; 0.5494]       4.1        4.1       > 50 y
## Lunt 2014              -0.0548 [-0.8923; 0.7827]       2.6        2.6    30 - 50 y
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.6        2.6    30 - 50 y
## Madssen 2014           -0.0471 [-0.7098; 0.6156]       4.2        4.2       > 50 y
## Maillard 2016           0.1751 [-0.8067; 1.1570]       1.9        1.9       > 50 y
## Matsuo 2014            -0.3770 [-1.1526; 0.3985]       3.1        3.1       < 30 y
## Mitranun 2014           0.7236 [-0.0411; 1.4883]       3.2        3.2       > 50 y
## Moreira 2008            0.0000 [-0.9800; 0.9800]       1.9        1.9    30 - 50 y
## Motiani 2017            0.1618 [-0.6082; 0.9318]       3.1        3.1    30 - 50 y
## Nalcakan 2014          -0.0869 [-1.1018; 0.9279]       1.8        1.8       < 30 y
## Nie 2017               -0.1134 [-0.8313; 0.6044]       3.6        3.6       < 30 y
## Ramos 2016b             0.3328 [-0.3662; 1.0319]       3.8        3.8       > 50 y
## Robinson 2015           0.0659 [-0.5622; 0.6939]       4.7        4.7       > 50 y
## Sandvei 2012            0.0636 [-0.7547; 0.8819]       2.8        2.8       < 30 y
## Sawyer 2016             0.0000 [-0.9239; 0.9239]       2.2        2.2    30 - 50 y
## Shepherd 2013           0.0000 [-0.9800; 0.9800]       1.9        1.9       < 30 y
## Sjöros 2018             0.0000 [-0.8564; 0.8564]       2.5        2.5    30 - 50 y
## Skleryk 2013            0.0000 [-0.9800; 0.9800]       1.9        1.9    30 - 50 y
## Tjønna 2008            -0.1783 [-1.0908; 0.7341]       2.2        2.2       > 50 y
## Winn 2018              -0.4417 [-1.4336; 0.5501]       1.9        1.9    30 - 50 y
## 
## Number of studies combined: k = 30
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## Random effects model -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  8.91   29  0.9999
## 
## Results for subgroups (fixed effect model):
##             k     SMD            95%-CI    Q  I^2
## < 30 y      9 -0.0554 [-0.3412; 0.2303] 1.39 0.0%
## 30 - 50 y  10  0.0078 [-0.2694; 0.2851] 1.01 0.0%
## > 50 y     11 -0.0150 [-0.2013; 0.1713] 5.87 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.10    2  0.9512
## Within groups  8.27   27  0.9998
## 
## Results for subgroups (random effects model):
##             k     SMD            95%-CI tau^2 tau
## < 30 y      9 -0.0554 [-0.3412; 0.2303]     0   0
## 30 - 50 y  10  0.0078 [-0.2694; 0.2851]     0   0
## > 50 y     11 -0.0150 [-0.2013; 0.1713]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.10    2  0.9512
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
3.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 30; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0409)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 28) = 8.7767, p-val = 0.9998
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.1380, p-val = 0.7103
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.1037  0.2401  -0.4319  0.6658  -0.5742  0.3669    
## age        0.0018  0.0049   0.3715  0.7103  -0.0078  0.0115    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
3.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

3.2.4 Training Duration

3.2.4.1 Forest plot

3.2.4.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_duration
## Abdelbasset 2020       -0.0910 [-0.7957; 0.6138]       3.7        3.7      5 - 10 weeks
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.9        1.9      5 - 10 weeks
## Conraads 2015          -0.1764 [-0.4743; 0.1214]      20.8       20.8        > 10 weeks
## Currie 2015             0.0405 [-0.8602; 0.9411]       2.3        2.3        > 10 weeks
## Eguchi 2012             0.0492 [-0.8275; 0.9259]       2.4        2.4        > 10 weeks
## Fisher 2015            -0.1130 [-0.9380; 0.7121]       2.7        2.7      5 - 10 weeks
## Gillen 2016             0.3289 [-0.5777; 1.2354]       2.2        2.2        > 10 weeks
## Grieco 2013            -0.0678 [-0.9073; 0.7716]       2.6        2.6         < 5 weeks
## Honkala 2017 (Healthy)  0.1477 [-0.5941; 0.8895]       3.4        3.4         < 5 weeks
## Honkala 2017 (T2D)      0.0775 [-0.9106; 1.0656]       1.9        1.9         < 5 weeks
## Jo 2020                -0.1235 [-0.7964; 0.5494]       4.1        4.1      5 - 10 weeks
## Lunt 2014              -0.0548 [-0.8923; 0.7827]       2.6        2.6        > 10 weeks
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.6        2.6        > 10 weeks
## Madssen 2014           -0.0471 [-0.7098; 0.6156]       4.2        4.2        > 10 weeks
## Maillard 2016           0.1751 [-0.8067; 1.1570]       1.9        1.9        > 10 weeks
## Matsuo 2014            -0.3770 [-1.1526; 0.3985]       3.1        3.1      5 - 10 weeks
## Mitranun 2014           0.7236 [-0.0411; 1.4883]       3.2        3.2      5 - 10 weeks
## Moreira 2008            0.0000 [-0.9800; 0.9800]       1.9        1.9        > 10 weeks
## Motiani 2017            0.1618 [-0.6082; 0.9318]       3.1        3.1         < 5 weeks
## Nalcakan 2014          -0.0869 [-1.1018; 0.9279]       1.8        1.8      5 - 10 weeks
## Nie 2017               -0.1134 [-0.8313; 0.6044]       3.6        3.6        > 10 weeks
## Ramos 2016b             0.3328 [-0.3662; 1.0319]       3.8        3.8        > 10 weeks
## Robinson 2015           0.0659 [-0.5622; 0.6939]       4.7        4.7         < 5 weeks
## Sandvei 2012            0.0636 [-0.7547; 0.8819]       2.8        2.8      5 - 10 weeks
## Sawyer 2016             0.0000 [-0.9239; 0.9239]       2.2        2.2      5 - 10 weeks
## Shepherd 2013           0.0000 [-0.9800; 0.9800]       1.9        1.9      5 - 10 weeks
## Sjöros 2018             0.0000 [-0.8564; 0.8564]       2.5        2.5         < 5 weeks
## Skleryk 2013            0.0000 [-0.9800; 0.9800]       1.9        1.9         < 5 weeks
## Tjønna 2008            -0.1783 [-1.0908; 0.7341]       2.2        2.2        > 10 weeks
## Winn 2018              -0.4417 [-1.4336; 0.5501]       1.9        1.9         < 5 weeks
## 
## Number of studies combined: k = 30
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## Random effects model -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  8.91   29  0.9999
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI    Q  I^2
## < 5 weeks      8  0.0208 [-0.2691; 0.3108] 1.05 0.0%
## 5 - 10 weeks  10 -0.0003 [-0.2605; 0.2600] 4.37 0.0%
## > 10 weeks    12 -0.0457 [-0.2367; 0.1453] 2.79 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.17    2  0.9195
## Within groups  8.20   27  0.9998
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI tau^2 tau
## < 5 weeks      8  0.0208 [-0.2691; 0.3108]     0   0
## 5 - 10 weeks  10 -0.0003 [-0.2605; 0.2600]     0   0
## > 10 weeks    12 -0.0457 [-0.2367; 0.1453]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.17    2  0.9195
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
3.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 30; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0407)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 28) = 8.9111, p-val = 0.9998
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0037, p-val = 0.9518
## 
## Model Results:
## 
##           estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    -0.0095  0.1621  -0.0583  0.9535  -0.3271  0.3082    
## duration   -0.0010  0.0162  -0.0604  0.9518  -0.0328  0.0308    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
3.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

3.2.5 Men Ratio

3.2.5.1 Forest plot

3.2.5.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_men_ratio
## Abdelbasset 2020       -0.0910 [-0.7957; 0.6138]       3.7        3.7              > 0.5
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.9        1.9              > 0.5
## Conraads 2015          -0.1764 [-0.4743; 0.1214]      20.8       20.8              > 0.5
## Currie 2015             0.0405 [-0.8602; 0.9411]       2.3        2.3              > 0.5
## Eguchi 2012             0.0492 [-0.8275; 0.9259]       2.4        2.4              > 0.5
## Fisher 2015            -0.1130 [-0.9380; 0.7121]       2.7        2.7              > 0.5
## Gillen 2016             0.3289 [-0.5777; 1.2354]       2.2        2.2              > 0.5
## Grieco 2013            -0.0678 [-0.9073; 0.7716]       2.6        2.6              < 0.5
## Honkala 2017 (Healthy)  0.1477 [-0.5941; 0.8895]       3.4        3.4              > 0.5
## Honkala 2017 (T2D)      0.0775 [-0.9106; 1.0656]       1.9        1.9              > 0.5
## Jo 2020                -0.1235 [-0.7964; 0.5494]       4.1        4.1              > 0.5
## Lunt 2014              -0.0548 [-0.8923; 0.7827]       2.6        2.6              < 0.5
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.6        2.6              < 0.5
## Madssen 2014           -0.0471 [-0.7098; 0.6156]       4.2        4.2              > 0.5
## Maillard 2016           0.1751 [-0.8067; 1.1570]       1.9        1.9              < 0.5
## Matsuo 2014            -0.3770 [-1.1526; 0.3985]       3.1        3.1              > 0.5
## Mitranun 2014           0.7236 [-0.0411; 1.4883]       3.2        3.2              < 0.5
## Moreira 2008            0.0000 [-0.9800; 0.9800]       1.9        1.9              < 0.5
## Motiani 2017            0.1618 [-0.6082; 0.9318]       3.1        3.1              > 0.5
## Nalcakan 2014          -0.0869 [-1.1018; 0.9279]       1.8        1.8              > 0.5
## Nie 2017               -0.1134 [-0.8313; 0.6044]       3.6        3.6              < 0.5
## Ramos 2016b             0.3328 [-0.3662; 1.0319]       3.8        3.8              > 0.5
## Robinson 2015           0.0659 [-0.5622; 0.6939]       4.7        4.7              < 0.5
## Sandvei 2012            0.0636 [-0.7547; 0.8819]       2.8        2.8              < 0.5
## Sawyer 2016             0.0000 [-0.9239; 0.9239]       2.2        2.2              < 0.5
## Shepherd 2013           0.0000 [-0.9800; 0.9800]       1.9        1.9              > 0.5
## Sjöros 2018             0.0000 [-0.8564; 0.8564]       2.5        2.5              > 0.5
## Skleryk 2013            0.0000 [-0.9800; 0.9800]       1.9        1.9              > 0.5
## Tjønna 2008            -0.1783 [-1.0908; 0.7341]       2.2        2.2              < 0.5
## Winn 2018              -0.4417 [-1.4336; 0.5501]       1.9        1.9              < 0.5
## 
## Number of studies combined: k = 30
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## Random effects model -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  8.91   29  0.9999
## 
## Results for subgroups (fixed effect model):
##         k     SMD            95%-CI    Q  I^2
## < 0.5  12  0.0352 [-0.2045; 0.2749] 4.25 0.0%
## > 0.5  18 -0.0442 [-0.2093; 0.1209] 3.83 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.29    1  0.5929
## Within groups  8.09   28  0.9999
## 
## Results for subgroups (random effects model):
##         k     SMD            95%-CI tau^2 tau
## < 0.5  12  0.0352 [-0.2045; 0.2749]     0   0
## > 0.5  18 -0.0442 [-0.2093; 0.1209]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.29    1  0.5929
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
3.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 30; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0406)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 28) = 8.6288, p-val = 0.9998
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.2860, p-val = 0.5928
## 
## Model Results:
## 
##            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt      0.0663  0.1728   0.3839  0.7010  -0.2724  0.4051    
## men_ratio   -0.1218  0.2277  -0.5348  0.5928  -0.5681  0.3246    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
3.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

3.2.6 Type of Exercise

3.2.6.1 Forest plot

3.2.6.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) type_exercise
## Abdelbasset 2020       -0.0910 [-0.7957; 0.6138]       3.7        3.7       Cycling
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.9        1.9       Cycling
## Conraads 2015          -0.1764 [-0.4743; 0.1214]      20.8       20.8       Cycling
## Currie 2015             0.0405 [-0.8602; 0.9411]       2.3        2.3       Cycling
## Eguchi 2012             0.0492 [-0.8275; 0.9259]       2.4        2.4       Cycling
## Fisher 2015            -0.1130 [-0.9380; 0.7121]       2.7        2.7       Cycling
## Gillen 2016             0.3289 [-0.5777; 1.2354]       2.2        2.2       Cycling
## Grieco 2013            -0.0678 [-0.9073; 0.7716]       2.6        2.6       Cycling
## Honkala 2017 (Healthy)  0.1477 [-0.5941; 0.8895]       3.4        3.4       Cycling
## Honkala 2017 (T2D)      0.0775 [-0.9106; 1.0656]       1.9        1.9       Cycling
## Jo 2020                -0.1235 [-0.7964; 0.5494]       4.1        4.1       Running
## Lunt 2014              -0.0548 [-0.8923; 0.7827]       2.6        2.6       Running
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.6        2.6       Running
## Madssen 2014           -0.0471 [-0.7098; 0.6156]       4.2        4.2       Running
## Maillard 2016           0.1751 [-0.8067; 1.1570]       1.9        1.9       Cycling
## Matsuo 2014            -0.3770 [-1.1526; 0.3985]       3.1        3.1       Cycling
## Mitranun 2014           0.7236 [-0.0411; 1.4883]       3.2        3.2       Running
## Moreira 2008            0.0000 [-0.9800; 0.9800]       1.9        1.9       Cycling
## Motiani 2017            0.1618 [-0.6082; 0.9318]       3.1        3.1       Cycling
## Nalcakan 2014          -0.0869 [-1.1018; 0.9279]       1.8        1.8       Cycling
## Nie 2017               -0.1134 [-0.8313; 0.6044]       3.6        3.6       Cycling
## Ramos 2016b             0.3328 [-0.3662; 1.0319]       3.8        3.8       Running
## Robinson 2015           0.0659 [-0.5622; 0.6939]       4.7        4.7       Cycling
## Sandvei 2012            0.0636 [-0.7547; 0.8819]       2.8        2.8       Running
## Sawyer 2016             0.0000 [-0.9239; 0.9239]       2.2        2.2       Cycling
## Shepherd 2013           0.0000 [-0.9800; 0.9800]       1.9        1.9       Cycling
## Sjöros 2018             0.0000 [-0.8564; 0.8564]       2.5        2.5       Cycling
## Skleryk 2013            0.0000 [-0.9800; 0.9800]       1.9        1.9       Cycling
## Tjønna 2008            -0.1783 [-1.0908; 0.7341]       2.2        2.2       Running
## Winn 2018              -0.4417 [-1.4336; 0.5501]       1.9        1.9       Running
## 
## Number of studies combined: k = 30
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## Random effects model -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  8.91   29  0.9999
## 
## Results for subgroups (fixed effect model):
##           k     SMD            95%-CI    Q  I^2
## Cycling  21 -0.0480 [-0.2075; 0.1115] 3.06 0.0%
## Running   9  0.0592 [-0.2008; 0.3193] 4.84 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.47    1  0.4910
## Within groups  7.90   28  0.9999
## 
## Results for subgroups (random effects model):
##           k     SMD            95%-CI tau^2 tau
## Cycling  21 -0.0480 [-0.2075; 0.1115]     0   0
## Running   9  0.0592 [-0.2008; 0.3193]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.47    1  0.4910
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
3.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 30; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0406)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 28) = 8.4313, p-val = 0.9999
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.4834, p-val = 0.4869
## 
## Model Results:
## 
##                       estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                -0.0479  0.0814  -0.5886  0.5562  -0.2074  0.1116    
## type_exerciseRunning    0.1082  0.1556   0.6953  0.4869  -0.1968  0.4131    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
3.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

3.2.7 Baseline Values

3.2.7.1 Forest plot

3.2.7.2 R output
##                            SMD            95%-CI %W(fixed) %W(random)     category_bsln
## Abdelbasset 2020       -0.0910 [-0.7957; 0.6138]       3.7        3.7    BMI > 30 kg/m²
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.9        1.9    BMI < 25 kg/m²
## Conraads 2015          -0.1764 [-0.4743; 0.1214]      20.8       20.8 BMI 25 - 30 kg/m²
## Currie 2015             0.0405 [-0.8602; 0.9411]       2.3        2.3 BMI 25 - 30 kg/m²
## Eguchi 2012             0.0492 [-0.8275; 0.9259]       2.4        2.4 BMI 25 - 30 kg/m²
## Fisher 2015            -0.1130 [-0.9380; 0.7121]       2.7        2.7 BMI 25 - 30 kg/m²
## Gillen 2016             0.3289 [-0.5777; 1.2354]       2.2        2.2 BMI 25 - 30 kg/m²
## Grieco 2013            -0.0678 [-0.9073; 0.7716]       2.6        2.6 BMI 25 - 30 kg/m²
## Honkala 2017 (Healthy)  0.1477 [-0.5941; 0.8895]       3.4        3.4 BMI 25 - 30 kg/m²
## Honkala 2017 (T2D)      0.0775 [-0.9106; 1.0656]       1.9        1.9    BMI > 30 kg/m²
## Jo 2020                -0.1235 [-0.7964; 0.5494]       4.1        4.1    BMI < 25 kg/m²
## Lunt 2014              -0.0548 [-0.8923; 0.7827]       2.6        2.6    BMI > 30 kg/m²
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.6        2.6    BMI > 30 kg/m²
## Madssen 2014           -0.0471 [-0.7098; 0.6156]       4.2        4.2 BMI 25 - 30 kg/m²
## Maillard 2016           0.1751 [-0.8067; 1.1570]       1.9        1.9    BMI > 30 kg/m²
## Matsuo 2014            -0.3770 [-1.1526; 0.3985]       3.1        3.1    BMI < 25 kg/m²
## Mitranun 2014           0.7236 [-0.0411; 1.4883]       3.2        3.2 BMI 25 - 30 kg/m²
## Moreira 2008            0.0000 [-0.9800; 0.9800]       1.9        1.9 BMI 25 - 30 kg/m²
## Motiani 2017            0.1618 [-0.6082; 0.9318]       3.1        3.1 BMI 25 - 30 kg/m²
## Nalcakan 2014          -0.0869 [-1.1018; 0.9279]       1.8        1.8    BMI < 25 kg/m²
## Nie 2017               -0.1134 [-0.8313; 0.6044]       3.6        3.6 BMI 25 - 30 kg/m²
## Ramos 2016b             0.3328 [-0.3662; 1.0319]       3.8        3.8    BMI > 30 kg/m²
## Robinson 2015           0.0659 [-0.5622; 0.6939]       4.7        4.7    BMI > 30 kg/m²
## Sandvei 2012            0.0636 [-0.7547; 0.8819]       2.8        2.8    BMI < 25 kg/m²
## Sawyer 2016             0.0000 [-0.9239; 0.9239]       2.2        2.2    BMI > 30 kg/m²
## Shepherd 2013           0.0000 [-0.9800; 0.9800]       1.9        1.9    BMI < 25 kg/m²
## Sjöros 2018             0.0000 [-0.8564; 0.8564]       2.5        2.5    BMI > 30 kg/m²
## Skleryk 2013            0.0000 [-0.9800; 0.9800]       1.9        1.9    BMI > 30 kg/m²
## Tjønna 2008            -0.1783 [-1.0908; 0.7341]       2.2        2.2 BMI 25 - 30 kg/m²
## Winn 2018              -0.4417 [-1.4336; 0.5501]       1.9        1.9    BMI > 30 kg/m²
## 
## Number of studies combined: k = 30
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## Random effects model -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  8.91   29  0.9999
## 
## Results for subgroups (fixed effect model):
##                     k     SMD            95%-CI    Q  I^2
## BMI < 25 kg/m²      6 -0.1022 [-0.4470; 0.2425] 0.68 0.0%
## BMI 25 - 30 kg/m²  13 -0.0185 [-0.2024; 0.1654] 5.63 0.0%
## BMI > 30 kg/m²     11  0.0247 [-0.2246; 0.2740] 1.71 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.34    2  0.8429
## Within groups  8.03   27  0.9998
## 
## Results for subgroups (random effects model):
##                     k     SMD            95%-CI tau^2 tau
## BMI < 25 kg/m²      6 -0.1022 [-0.4470; 0.2425]     0   0
## BMI 25 - 30 kg/m²  13 -0.0185 [-0.2024; 0.1654]     0   0
## BMI > 30 kg/m²     11  0.0247 [-0.2246; 0.2740]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.34    2  0.8429
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
3.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 30; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0398)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 28) = 8.8474, p-val = 0.9998
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0674, p-val = 0.7952
## 
## Model Results:
## 
##                estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt         -0.1614  0.5558  -0.2905  0.7715  -1.2508  0.9279    
## bsln_adjusted    0.0050  0.0192   0.2596  0.7952  -0.0327  0.0427    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
3.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

3.2.8 Type of HIIE

3.2.8.1 Forest plot

3.2.8.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) HIIE
## Abdelbasset 2020       -0.0910 [-0.7957; 0.6138]       3.7        3.7 HIIT
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.9        1.9  SIT
## Conraads 2015          -0.1764 [-0.4743; 0.1214]      20.8       20.8 HIIT
## Currie 2015             0.0405 [-0.8602; 0.9411]       2.3        2.3 HIIT
## Eguchi 2012             0.0492 [-0.8275; 0.9259]       2.4        2.4 HIIT
## Fisher 2015            -0.1130 [-0.9380; 0.7121]       2.7        2.7  SIT
## Gillen 2016             0.3289 [-0.5777; 1.2354]       2.2        2.2  SIT
## Grieco 2013            -0.0678 [-0.9073; 0.7716]       2.6        2.6 HIIT
## Honkala 2017 (Healthy)  0.1477 [-0.5941; 0.8895]       3.4        3.4  SIT
## Honkala 2017 (T2D)      0.0775 [-0.9106; 1.0656]       1.9        1.9  SIT
## Jo 2020                -0.1235 [-0.7964; 0.5494]       4.1        4.1 HIIT
## Lunt 2014              -0.0548 [-0.8923; 0.7827]       2.6        2.6 HIIT
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.6        2.6  SIT
## Madssen 2014           -0.0471 [-0.7098; 0.6156]       4.2        4.2 HIIT
## Maillard 2016           0.1751 [-0.8067; 1.1570]       1.9        1.9 HIIT
## Matsuo 2014            -0.3770 [-1.1526; 0.3985]       3.1        3.1 HIIT
## Mitranun 2014           0.7236 [-0.0411; 1.4883]       3.2        3.2 HIIT
## Moreira 2008            0.0000 [-0.9800; 0.9800]       1.9        1.9 HIIT
## Motiani 2017            0.1618 [-0.6082; 0.9318]       3.1        3.1  SIT
## Nalcakan 2014          -0.0869 [-1.1018; 0.9279]       1.8        1.8  SIT
## Nie 2017               -0.1134 [-0.8313; 0.6044]       3.6        3.6 HIIT
## Ramos 2016b             0.3328 [-0.3662; 1.0319]       3.8        3.8 HIIT
## Robinson 2015           0.0659 [-0.5622; 0.6939]       4.7        4.7 HIIT
## Sandvei 2012            0.0636 [-0.7547; 0.8819]       2.8        2.8  SIT
## Sawyer 2016             0.0000 [-0.9239; 0.9239]       2.2        2.2 HIIT
## Shepherd 2013           0.0000 [-0.9800; 0.9800]       1.9        1.9  SIT
## Sjöros 2018             0.0000 [-0.8564; 0.8564]       2.5        2.5  SIT
## Skleryk 2013            0.0000 [-0.9800; 0.9800]       1.9        1.9  SIT
## Tjønna 2008            -0.1783 [-1.0908; 0.7341]       2.2        2.2 HIIT
## Winn 2018              -0.4417 [-1.4336; 0.5501]       1.9        1.9 HIIT
## 
## Number of studies combined: k = 30
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## Random effects model -0.0183 [-0.1543; 0.1176] -0.26  0.7919
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  8.91   29  0.9999
## 
## Results for subgroups (fixed effect model):
##        k     SMD            95%-CI    Q  I^2
## HIIT  18 -0.0479 [-0.2091; 0.1133] 7.21 0.0%
## SIT   12  0.0535 [-0.1998; 0.3067] 0.73 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.44    1  0.5082
## Within groups  7.93   28  0.9999
## 
## Results for subgroups (random effects model):
##        k     SMD            95%-CI tau^2 tau
## HIIT  18 -0.0479 [-0.2091; 0.1133]     0   0
## SIT   12  0.0535 [-0.1998; 0.3067]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.44    1  0.5082
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
3.2.8.3 Meta-regression
## 
## Mixed-Effects Model (k = 30; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0404)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 28) = 8.4565, p-val = 0.9999
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.4583, p-val = 0.4984
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.0482  0.0822  -0.5860  0.5579  -0.2093  0.1130    
## HIIESIT    0.1037  0.1532   0.6770  0.4984  -0.1965  0.4038    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
3.2.8.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

4. Body Mass

4.1 Overall

4.1.1 Forest plot

4.1.2 R output

##                            SMD            95%-CI %W(fixed) %W(random)
## Bækkerud 2016           0.0834 [-0.8694; 1.0362]       1.5        1.5
## Beetham 2019           -0.7117 [-1.8362; 0.4129]       1.1        1.1
## Burgomaster 2008        0.1633 [-0.7147; 1.0413]       1.8        1.8
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.4        1.4
## Conraads 2015          -0.1303 [-0.4278; 0.1673]      15.7       15.7
## Currie 2015            -0.0220 [-0.9226; 0.8785]       1.7        1.7
## Earnest 2013            0.1102 [-0.5407; 0.7611]       3.3        3.3
## Eguchi 2012             0.0711 [-0.8057; 0.9479]       1.8        1.8
## Fisher 2015            -0.1077 [-0.9327; 0.7173]       2.0        2.0
## Gillen 2016            -0.0859 [-0.9869; 0.8150]       1.7        1.7
## Gorostiaga 1991         0.2781 [-0.8590; 1.4151]       1.1        1.1
## Granata 2015           -0.0737 [-0.9549; 0.8075]       1.8        1.8
## Granata 2015           -0.1189 [-1.0437; 0.8058]       1.6        1.6
## Grieco 2013            -0.1097 [-0.9495; 0.7302]       2.0        2.0
## Helgerud 2007          -0.1347 [-1.0122; 0.7428]       1.8        1.8
## Helgerud 2007          -0.2857 [-1.1667; 0.5953]       1.8        1.8
## Honkala 2017 (Healthy)  0.0925 [-0.6487; 0.8337]       2.5        2.5
## Honkala 2017 (T2D)      0.0204 [-0.9674; 1.0081]       1.4        1.4
## Jo 2020                -0.0456 [-0.7179; 0.6268]       3.1        3.1
## Keating 2014           -0.2150 [-1.0532; 0.6231]       2.0        2.0
## Klonizakis 2014         0.0572 [-0.8906; 1.0050]       1.5        1.5
## Macpherson 2011         0.1198 [-0.7576; 0.9971]       1.8        1.8
## Maillard 2016           0.1024 [-0.8783; 1.0830]       1.4        1.4
## Martins 2016            0.1874 [-0.5314; 0.9063]       2.7        2.7
## Matsuo 2014            -0.3160 [-1.0895; 0.4575]       2.3        2.3
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       2.2        2.2
## Mitranun 2014           0.2274 [-0.5158; 0.9705]       2.5        2.5
## Moreira 2008           -0.0139 [-0.9939; 0.9661]       1.4        1.4
## Nalcakan 2014          -0.0992 [-1.1142; 0.9158]       1.4        1.4
## Nie 2017               -0.0519 [-0.7693; 0.6655]       2.7        2.7
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       3.9        3.9
## Ramos 2016b             0.3042 [-0.3941; 1.0025]       2.9        2.9
## Robinson 2015          -0.0622 [-0.6903; 0.5658]       3.5        3.5
## Rognmo 2004            -0.1673 [-1.1213; 0.7867]       1.5        1.5
## Sandvei 2012           -0.0307 [-0.8489; 0.7875]       2.1        2.1
## Sawyer 2016             0.0090 [-0.9149; 0.9330]       1.6        1.6
## Scribbans 2014          0.0000 [-0.9005; 0.9005]       1.7        1.7
## Shepherd 2013          -0.0169 [-0.9969; 0.9631]       1.4        1.4
## Sjöros 2018             0.0162 [-0.8401; 0.8726]       1.9        1.9
## Tjønna 2008            -0.1325 [-1.0442; 0.7792]       1.7        1.7
## Trapp 2008              0.2121 [-0.5056; 0.9298]       2.7        2.7
## Winn 2018              -0.5813 [-1.5818; 0.4191]       1.4        1.4
## Zapata-Lamana 2018     -0.7184 [-1.4827; 0.0459]       2.4        2.4
## 
## Number of studies combined: k = 43
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## Random effects model -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  10.66   42  1.0000
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

4.1.3 Sensitivity analysis

4.1.3.1 Forest plot

4.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                                       SMD            95%-CI p-value   tau^2     tau   I^2
## Omitting Bækkerud 2016            -0.0496 [-0.1685; 0.0693]  0.4135  0.0000  0.0000  0.0%
## Omitting Beetham 2019             -0.0408 [-0.1595; 0.0778]  0.5000  0.0000  0.0000  0.0%
## Omitting Burgomaster 2008         -0.0514 [-0.1705; 0.0677]  0.3977  0.0000  0.0000  0.0%
## Omitting Cocks 2013               -0.0483 [-0.1672; 0.0705]  0.4254  0.0000  0.0000  0.0%
## Omitting Conraads 2015            -0.0323 [-0.1609; 0.0962]  0.6221  0.0000  0.0000  0.0%
## Omitting Currie 2015              -0.0481 [-0.1671; 0.0709]  0.4283  0.0000  0.0000  0.0%
## Omitting Earnest 2013             -0.0529 [-0.1729; 0.0671]  0.3874  0.0000  0.0000  0.0%
## Omitting Eguchi 2012              -0.0498 [-0.1689; 0.0693]  0.4127  0.0000  0.0000  0.0%
## Omitting Fisher 2015              -0.0465 [-0.1657; 0.0728]  0.4450  0.0000  0.0000  0.0%
## Omitting Gillen 2016              -0.0470 [-0.1661; 0.0720]  0.4386  0.0000  0.0000  0.0%
## Omitting Gorostiaga 1991          -0.0509 [-0.1696; 0.0677]  0.4000  0.0000  0.0000  0.0%
## Omitting Granata 2015             -0.0472 [-0.1663; 0.0719]  0.4370  0.0000  0.0000  0.0%
## Omitting Granata 2015             -0.0466 [-0.1655; 0.0724]  0.4431  0.0000  0.0000  0.0%
## Omitting Grieco 2013              -0.0465 [-0.1657; 0.0727]  0.4447  0.0000  0.0000  0.0%
## Omitting Helgerud 2007            -0.0461 [-0.1652; 0.0729]  0.4476  0.0000  0.0000  0.0%
## Omitting Helgerud 2007            -0.0435 [-0.1626; 0.0756]  0.4738  0.0000  0.0000  0.0%
## Omitting Honkala 2017 (Healthy)   -0.0512 [-0.1707; 0.0683]  0.4011  0.0000  0.0000  0.0%
## Omitting Honkala 2017 (T2D)       -0.0486 [-0.1675; 0.0702]  0.4228  0.0000  0.0000  0.0%
## Omitting Jo 2020                  -0.0477 [-0.1676; 0.0721]  0.4350  0.0000  0.0000  0.0%
## Omitting Keating 2014             -0.0444 [-0.1636; 0.0748]  0.4651  0.0000  0.0000  0.0%
## Omitting Klonizakis 2014          -0.0492 [-0.1682; 0.0697]  0.4170  0.0000  0.0000  0.0%
## Omitting Macpherson 2011          -0.0506 [-0.1697; 0.0685]  0.4047  0.0000  0.0000  0.0%
## Omitting Maillard 2016            -0.0498 [-0.1686; 0.0691]  0.4119  0.0000  0.0000  0.0%
## Omitting Martins 2016             -0.0540 [-0.1736; 0.0656]  0.3762  0.0000  0.0000  0.0%
## Omitting Matsuo 2014              -0.0415 [-0.1609; 0.0779]  0.4958  0.0000  0.0000  0.0%
## Omitting Matsuo 2015              -0.0487 [-0.1680; 0.0706]  0.4237  0.0000  0.0000  0.0%
## Omitting Mitranun 2014            -0.0546 [-0.1741; 0.0649]  0.3708  0.0000  0.0000  0.0%
## Omitting Moreira 2008             -0.0481 [-0.1670; 0.0707]  0.4273  0.0000  0.0000  0.0%
## Omitting Nalcakan 2014            -0.0470 [-0.1658; 0.0718]  0.4380  0.0000  0.0000  0.0%
## Omitting Nie 2017                 -0.0476 [-0.1672; 0.0721]  0.4359  0.0000  0.0000  0.0%
## Omitting Ramos 2016a              -0.0496 [-0.1699; 0.0708]  0.4196  0.0000  0.0000  0.0%
## Omitting Ramos 2016b              -0.0577 [-0.1775; 0.0620]  0.3445  0.0000  0.0000  0.0%
## Omitting Robinson 2015            -0.0471 [-0.1673; 0.0730]  0.4418  0.0000  0.0000  0.0%
## Omitting Rognmo 2004              -0.0459 [-0.1648; 0.0730]  0.4492  0.0000  0.0000  0.0%
## Omitting Sandvei 2012             -0.0480 [-0.1673; 0.0712]  0.4300  0.0000  0.0000  0.0%
## Omitting Sawyer 2016              -0.0486 [-0.1675; 0.0704]  0.4236  0.0000  0.0000  0.0%
## Omitting Scribbans 2014           -0.0485 [-0.1675; 0.0706]  0.4248  0.0000  0.0000  0.0%
## Omitting Shepherd 2013            -0.0481 [-0.1670; 0.0708]  0.4277  0.0000  0.0000  0.0%
## Omitting Sjöros 2018              -0.0489 [-0.1680; 0.0703]  0.4215  0.0000  0.0000  0.0%
## Omitting Tjønna 2008              -0.0463 [-0.1653; 0.0727]  0.4458  0.0000  0.0000  0.0%
## Omitting Trapp 2008               -0.0547 [-0.1743; 0.0649]  0.3702  0.0000  0.0000  0.0%
## Omitting Winn 2018                -0.0406 [-0.1594; 0.0782]  0.5030  0.0000  0.0000  0.0%
## Omitting Zapata-Lamana 2018       -0.0319 [-0.1513; 0.0876]  0.6010  0.0000  0.0000  0.0%
##                                                                                          
## Pooled estimate                   -0.0494 [-0.1674; 0.0686]  0.4117  0.0000  0.0000  0.0%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

4.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

4.2 Subgroups

4.2.1 Overall

4.2.1.1 Forest plot

4.2.1.2 R output
##                            SMD            95%-CI     meta-analysis
##                        -0.0494 [-0.1674; 0.0686]           Overall
## Healthy                -0.0184 [-0.2123; 0.1754]        Population
## Overweight/obese       -0.1437 [-0.4112; 0.1238]        Population
## Cardiac Rehabilitation -0.1222 [-0.3931; 0.1487]        Population
## Metabolic Syndrome      0.0164 [-0.2681; 0.3008]        Population
## T2D                     0.1033 [-0.3337; 0.5402]        Population
## < 30 y                 -0.0838 [-0.2816; 0.1140]               Age
## 30 - 50 y               0.0035 [-0.2481; 0.2552]               Age
## > 50 y                 -0.0438 [-0.2250; 0.1374]               Age
## < 5 weeks              -0.0635 [-0.3438; 0.2169] Training Duration
## 5 - 10 weeks           -0.0117 [-0.2071; 0.1837] Training Duration
## > 10 weeks             -0.0701 [-0.2444; 0.1042] Training Duration
## < 0.5                  -0.0370 [-0.2247; 0.1507]         Men Ratio
## > 0.5                  -0.0546 [-0.2063; 0.0971]         Men Ratio
## Running                -0.0345 [-0.2554; 0.1865]  Type of Exercise
## Cycling                -0.0529 [-0.1925; 0.0867]  Type of Exercise
## BMI < 25 kg/m²         -0.0233 [-0.2409; 0.1943]   Baseline Values
## BMI 25 - 30 kg/m²      -0.0662 [-0.2492; 0.1169]   Baseline Values
## BMI > 30 kg/m²         -0.0457 [-0.2647; 0.1733]   Baseline Values
## HIIT                   -0.0817 [-0.2239; 0.0605]      Type of HIIE
## SIT                     0.0276 [-0.1838; 0.2389]      Type of HIIE
## 
## Number of studies combined: k = 43
## 
##                          SMD            95%-CI     z p-value
## Random effects model -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## 
## Quantifying heterogeneity:
##  tau^2 = 0; tau = 0; I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  10.66   42  1.0000
## 
## Results for meta-analyses (random effects model):
##                     k     SMD            95%-CI tau^2 tau     Q  I^2
## Overall            43 -0.0494 [-0.1674; 0.0686]     0   0 10.66 0.0%
## Population         43 -0.0494 [-0.1674; 0.0686]     0   0 10.66 0.0%
## Age                43 -0.0494 [-0.1674; 0.0686]     0   0 10.66 0.0%
## Training Duration  43 -0.0494 [-0.1674; 0.0686]     0   0 10.66 0.0%
## Men Ratio          43 -0.0494 [-0.1674; 0.0686]     0   0 10.66 0.0%
## Type of Exercise   43 -0.0494 [-0.1674; 0.0686]     0   0 10.66 0.0%
## Baseline Values    43 -0.0494 [-0.1674; 0.0686]     0   0 10.66 0.0%
## Type of HIIE       43 -0.0494 [-0.1674; 0.0686]     0   0 10.66 0.0%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

4.2.2 Population

4.2.2.1 Forest plot

4.2.2.2 R output
##                            SMD            95%-CI %W(fixed) %W(random)             population
## Bækkerud 2016           0.0834 [-0.8694; 1.0362]       1.5        1.5       Overweight/obese
## Beetham 2019           -0.7117 [-1.8362; 0.4129]       1.1        1.1       Overweight/obese
## Burgomaster 2008        0.1633 [-0.7147; 1.0413]       1.8        1.8                Healthy
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.4        1.4                Healthy
## Conraads 2015          -0.1303 [-0.4278; 0.1673]      15.7       15.7 Cardiac Rehabilitation
## Currie 2015            -0.0220 [-0.9226; 0.8785]       1.7        1.7 Cardiac Rehabilitation
## Earnest 2013            0.1102 [-0.5407; 0.7611]       3.3        3.3       Overweight/obese
## Eguchi 2012             0.0711 [-0.8057; 0.9479]       1.8        1.8                Healthy
## Fisher 2015            -0.1077 [-0.9327; 0.7173]       2.0        2.0       Overweight/obese
## Gillen 2016            -0.0859 [-0.9869; 0.8150]       1.7        1.7                Healthy
## Gorostiaga 1991         0.2781 [-0.8590; 1.4151]       1.1        1.1                Healthy
## Granata 2015           -0.0737 [-0.9549; 0.8075]       1.8        1.8                Healthy
## Granata 2015           -0.1189 [-1.0437; 0.8058]       1.6        1.6                Healthy
## Grieco 2013            -0.1097 [-0.9495; 0.7302]       2.0        2.0                Healthy
## Helgerud 2007          -0.1347 [-1.0122; 0.7428]       1.8        1.8                Healthy
## Helgerud 2007          -0.2857 [-1.1667; 0.5953]       1.8        1.8                Healthy
## Honkala 2017 (Healthy)  0.0925 [-0.6487; 0.8337]       2.5        2.5                Healthy
## Honkala 2017 (T2D)      0.0204 [-0.9674; 1.0081]       1.4        1.4                    T2D
## Jo 2020                -0.0456 [-0.7179; 0.6268]       3.1        3.1     Metabolic Syndrome
## Keating 2014           -0.2150 [-1.0532; 0.6231]       2.0        2.0       Overweight/obese
## Klonizakis 2014         0.0572 [-0.8906; 1.0050]       1.5        1.5                Healthy
## Macpherson 2011         0.1198 [-0.7576; 0.9971]       1.8        1.8                Healthy
## Maillard 2016           0.1024 [-0.8783; 1.0830]       1.4        1.4                    T2D
## Martins 2016            0.1874 [-0.5314; 0.9063]       2.7        2.7       Overweight/obese
## Matsuo 2014            -0.3160 [-1.0895; 0.4575]       2.3        2.3                Healthy
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       2.2        2.2     Metabolic Syndrome
## Mitranun 2014           0.2274 [-0.5158; 0.9705]       2.5        2.5                    T2D
## Moreira 2008           -0.0139 [-0.9939; 0.9661]       1.4        1.4       Overweight/obese
## Nalcakan 2014          -0.0992 [-1.1142; 0.9158]       1.4        1.4                Healthy
## Nie 2017               -0.0519 [-0.7693; 0.6655]       2.7        2.7                Healthy
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       3.9        3.9     Metabolic Syndrome
## Ramos 2016b             0.3042 [-0.3941; 1.0025]       2.9        2.9     Metabolic Syndrome
## Robinson 2015          -0.0622 [-0.6903; 0.5658]       3.5        3.5     Metabolic Syndrome
## Rognmo 2004            -0.1673 [-1.1213; 0.7867]       1.5        1.5 Cardiac Rehabilitation
## Sandvei 2012           -0.0307 [-0.8489; 0.7875]       2.1        2.1                Healthy
## Sawyer 2016             0.0090 [-0.9149; 0.9330]       1.6        1.6       Overweight/obese
## Scribbans 2014          0.0000 [-0.9005; 0.9005]       1.7        1.7                Healthy
## Shepherd 2013          -0.0169 [-0.9969; 0.9631]       1.4        1.4                Healthy
## Sjöros 2018             0.0162 [-0.8401; 0.8726]       1.9        1.9                    T2D
## Tjønna 2008            -0.1325 [-1.0442; 0.7792]       1.7        1.7     Metabolic Syndrome
## Trapp 2008              0.2121 [-0.5056; 0.9298]       2.7        2.7                Healthy
## Winn 2018              -0.5813 [-1.5818; 0.4191]       1.4        1.4       Overweight/obese
## Zapata-Lamana 2018     -0.7184 [-1.4827; 0.0459]       2.4        2.4       Overweight/obese
## 
## Number of studies combined: k = 43
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## Random effects model -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  10.66   42  1.0000
## 
## Results for subgroups (fixed effect model):
##                          k     SMD            95%-CI    Q  I^2
## Healthy                 20 -0.0184 [-0.2123; 0.1754] 2.04 0.0%
## Overweight/obese        10 -0.1437 [-0.4112; 0.1238] 5.23 0.0%
## Cardiac Rehabilitation   3 -0.1222 [-0.3931; 0.1487] 0.06 0.0%
## Metabolic Syndrome       6  0.0164 [-0.2681; 0.3008] 0.81 0.0%
## T2D                      4  0.1033 [-0.3337; 0.5402] 0.16 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 1.53    4  0.8219
## Within groups  8.30   38  1.0000
## 
## Results for subgroups (random effects model):
##                          k     SMD            95%-CI tau^2 tau
## Healthy                 20 -0.0184 [-0.2123; 0.1754]     0   0
## Overweight/obese        10 -0.1437 [-0.4112; 0.1238]     0   0
## Cardiac Rehabilitation   3 -0.1222 [-0.3931; 0.1487]     0   0
## Metabolic Syndrome       6  0.0164 [-0.2681; 0.3008]     0   0
## T2D                      4  0.1033 [-0.3337; 0.5402]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.53    4  0.8219
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
4.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 43; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0407)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 38) = 9.0142, p-val = 1.0000
## 
## Test of Moderators (coefficients 2:5):
## QM(df = 4) = 1.6460, p-val = 0.8005
## 
## Model Results:
## 
##                               estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                        -0.0191  0.0989  -0.1931  0.8469  -0.2129  0.1747    
## .byvarOverweight/obese         -0.1328  0.1684  -0.7884  0.4305  -0.4629  0.1973    
## .byvarCardiac Rehabilitation   -0.1044  0.1699  -0.6142  0.5391  -0.4374  0.2287    
## .byvarMetabolic Syndrome        0.0357  0.1756   0.2036  0.8387  -0.3084  0.3799    
## .byvarT2D                       0.1262  0.2439   0.5175  0.6048  -0.3517  0.6041    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
4.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

4.2.3 Age

4.2.3.1 Forest plot

4.2.3.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_age
## Bækkerud 2016           0.0834 [-0.8694; 1.0362]       1.5        1.5    30 - 50 y
## Beetham 2019           -0.7117 [-1.8362; 0.4129]       1.1        1.1       > 50 y
## Burgomaster 2008        0.1633 [-0.7147; 1.0413]       1.8        1.8       < 30 y
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.4        1.4       < 30 y
## Conraads 2015          -0.1303 [-0.4278; 0.1673]      15.7       15.7       > 50 y
## Currie 2015            -0.0220 [-0.9226; 0.8785]       1.7        1.7       > 50 y
## Earnest 2013            0.1102 [-0.5407; 0.7611]       3.3        3.3    30 - 50 y
## Eguchi 2012             0.0711 [-0.8057; 0.9479]       1.8        1.8       > 50 y
## Fisher 2015            -0.1077 [-0.9327; 0.7173]       2.0        2.0       < 30 y
## Gillen 2016            -0.0859 [-0.9869; 0.8150]       1.7        1.7       < 30 y
## Gorostiaga 1991         0.2781 [-0.8590; 1.4151]       1.1        1.1       < 30 y
## Granata 2015           -0.0737 [-0.9549; 0.8075]       1.8        1.8       < 30 y
## Granata 2015           -0.1189 [-1.0437; 0.8058]       1.6        1.6       < 30 y
## Grieco 2013            -0.1097 [-0.9495; 0.7302]       2.0        2.0       < 30 y
## Helgerud 2007          -0.1347 [-1.0122; 0.7428]       1.8        1.8       < 30 y
## Helgerud 2007          -0.2857 [-1.1667; 0.5953]       1.8        1.8       < 30 y
## Honkala 2017 (Healthy)  0.0925 [-0.6487; 0.8337]       2.5        2.5    30 - 50 y
## Honkala 2017 (T2D)      0.0204 [-0.9674; 1.0081]       1.4        1.4    30 - 50 y
## Jo 2020                -0.0456 [-0.7179; 0.6268]       3.1        3.1       > 50 y
## Keating 2014           -0.2150 [-1.0532; 0.6231]       2.0        2.0    30 - 50 y
## Klonizakis 2014         0.0572 [-0.8906; 1.0050]       1.5        1.5       > 50 y
## Macpherson 2011         0.1198 [-0.7576; 0.9971]       1.8        1.8       < 30 y
## Maillard 2016           0.1024 [-0.8783; 1.0830]       1.4        1.4       > 50 y
## Martins 2016            0.1874 [-0.5314; 0.9063]       2.7        2.7    30 - 50 y
## Matsuo 2014            -0.3160 [-1.0895; 0.4575]       2.3        2.3       < 30 y
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       2.2        2.2    30 - 50 y
## Mitranun 2014           0.2274 [-0.5158; 0.9705]       2.5        2.5       > 50 y
## Moreira 2008           -0.0139 [-0.9939; 0.9661]       1.4        1.4    30 - 50 y
## Nalcakan 2014          -0.0992 [-1.1142; 0.9158]       1.4        1.4       < 30 y
## Nie 2017               -0.0519 [-0.7693; 0.6655]       2.7        2.7       < 30 y
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       3.9        3.9       > 50 y
## Ramos 2016b             0.3042 [-0.3941; 1.0025]       2.9        2.9       > 50 y
## Robinson 2015          -0.0622 [-0.6903; 0.5658]       3.5        3.5       > 50 y
## Rognmo 2004            -0.1673 [-1.1213; 0.7867]       1.5        1.5       > 50 y
## Sandvei 2012           -0.0307 [-0.8489; 0.7875]       2.1        2.1       < 30 y
## Sawyer 2016             0.0090 [-0.9149; 0.9330]       1.6        1.6    30 - 50 y
## Scribbans 2014          0.0000 [-0.9005; 0.9005]       1.7        1.7       < 30 y
## Shepherd 2013          -0.0169 [-0.9969; 0.9631]       1.4        1.4       < 30 y
## Sjöros 2018             0.0162 [-0.8401; 0.8726]       1.9        1.9    30 - 50 y
## Tjønna 2008            -0.1325 [-1.0442; 0.7792]       1.7        1.7       > 50 y
## Trapp 2008              0.2121 [-0.5056; 0.9298]       2.7        2.7       < 30 y
## Winn 2018              -0.5813 [-1.5818; 0.4191]       1.4        1.4    30 - 50 y
## Zapata-Lamana 2018     -0.7184 [-1.4827; 0.0459]       2.4        2.4       < 30 y
## 
## Number of studies combined: k = 43
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## Random effects model -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  10.66   42  1.0000
## 
## Results for subgroups (fixed effect model):
##             k     SMD            95%-CI    Q  I^2
## < 30 y     19 -0.0838 [-0.2816; 0.1140] 4.54 0.0%
## 30 - 50 y  11  0.0035 [-0.2481; 0.2552] 1.82 0.0%
## > 50 y     13 -0.0438 [-0.2250; 0.1374] 3.18 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.29    2  0.8653
## Within groups  9.54   40  1.0000
## 
## Results for subgroups (random effects model):
##             k     SMD            95%-CI tau^2 tau
## < 30 y     19 -0.0838 [-0.2816; 0.1140]     0   0
## 30 - 50 y  11  0.0035 [-0.2481; 0.2552]     0   0
## > 50 y     13 -0.0438 [-0.2250; 0.1374]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.29    2  0.8653
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
4.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 43; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0359)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 41) = 10.6031, p-val = 1.0000
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0571, p-val = 0.8112
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.0880  0.1723  -0.5106  0.6096  -0.4257  0.2497    
## age        0.0009  0.0038   0.2389  0.8112  -0.0066  0.0084    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
4.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

4.2.4 Training Duration

4.2.4.1 Forest plot

4.2.4.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_duration
## Bækkerud 2016           0.0834 [-0.8694; 1.0362]       1.5        1.5      5 - 10 weeks
## Beetham 2019           -0.7117 [-1.8362; 0.4129]       1.1        1.1        > 10 weeks
## Burgomaster 2008        0.1633 [-0.7147; 1.0413]       1.8        1.8      5 - 10 weeks
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.4        1.4      5 - 10 weeks
## Conraads 2015          -0.1303 [-0.4278; 0.1673]      15.7       15.7        > 10 weeks
## Currie 2015            -0.0220 [-0.9226; 0.8785]       1.7        1.7        > 10 weeks
## Earnest 2013            0.1102 [-0.5407; 0.7611]       3.3        3.3      5 - 10 weeks
## Eguchi 2012             0.0711 [-0.8057; 0.9479]       1.8        1.8        > 10 weeks
## Fisher 2015            -0.1077 [-0.9327; 0.7173]       2.0        2.0      5 - 10 weeks
## Gillen 2016            -0.0859 [-0.9869; 0.8150]       1.7        1.7        > 10 weeks
## Gorostiaga 1991         0.2781 [-0.8590; 1.4151]       1.1        1.1      5 - 10 weeks
## Granata 2015           -0.0737 [-0.9549; 0.8075]       1.8        1.8         < 5 weeks
## Granata 2015           -0.1189 [-1.0437; 0.8058]       1.6        1.6         < 5 weeks
## Grieco 2013            -0.1097 [-0.9495; 0.7302]       2.0        2.0         < 5 weeks
## Helgerud 2007          -0.1347 [-1.0122; 0.7428]       1.8        1.8      5 - 10 weeks
## Helgerud 2007          -0.2857 [-1.1667; 0.5953]       1.8        1.8      5 - 10 weeks
## Honkala 2017 (Healthy)  0.0925 [-0.6487; 0.8337]       2.5        2.5         < 5 weeks
## Honkala 2017 (T2D)      0.0204 [-0.9674; 1.0081]       1.4        1.4         < 5 weeks
## Jo 2020                -0.0456 [-0.7179; 0.6268]       3.1        3.1      5 - 10 weeks
## Keating 2014           -0.2150 [-1.0532; 0.6231]       2.0        2.0        > 10 weeks
## Klonizakis 2014         0.0572 [-0.8906; 1.0050]       1.5        1.5         < 5 weeks
## Macpherson 2011         0.1198 [-0.7576; 0.9971]       1.8        1.8      5 - 10 weeks
## Maillard 2016           0.1024 [-0.8783; 1.0830]       1.4        1.4        > 10 weeks
## Martins 2016            0.1874 [-0.5314; 0.9063]       2.7        2.7        > 10 weeks
## Matsuo 2014            -0.3160 [-1.0895; 0.4575]       2.3        2.3      5 - 10 weeks
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       2.2        2.2      5 - 10 weeks
## Mitranun 2014           0.2274 [-0.5158; 0.9705]       2.5        2.5      5 - 10 weeks
## Moreira 2008           -0.0139 [-0.9939; 0.9661]       1.4        1.4        > 10 weeks
## Nalcakan 2014          -0.0992 [-1.1142; 0.9158]       1.4        1.4      5 - 10 weeks
## Nie 2017               -0.0519 [-0.7693; 0.6655]       2.7        2.7        > 10 weeks
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       3.9        3.9        > 10 weeks
## Ramos 2016b             0.3042 [-0.3941; 1.0025]       2.9        2.9        > 10 weeks
## Robinson 2015          -0.0622 [-0.6903; 0.5658]       3.5        3.5         < 5 weeks
## Rognmo 2004            -0.1673 [-1.1213; 0.7867]       1.5        1.5      5 - 10 weeks
## Sandvei 2012           -0.0307 [-0.8489; 0.7875]       2.1        2.1      5 - 10 weeks
## Sawyer 2016             0.0090 [-0.9149; 0.9330]       1.6        1.6      5 - 10 weeks
## Scribbans 2014          0.0000 [-0.9005; 0.9005]       1.7        1.7      5 - 10 weeks
## Shepherd 2013          -0.0169 [-0.9969; 0.9631]       1.4        1.4      5 - 10 weeks
## Sjöros 2018             0.0162 [-0.8401; 0.8726]       1.9        1.9         < 5 weeks
## Tjønna 2008            -0.1325 [-1.0442; 0.7792]       1.7        1.7        > 10 weeks
## Trapp 2008              0.2121 [-0.5056; 0.9298]       2.7        2.7        > 10 weeks
## Winn 2018              -0.5813 [-1.5818; 0.4191]       1.4        1.4         < 5 weeks
## Zapata-Lamana 2018     -0.7184 [-1.4827; 0.0459]       2.4        2.4        > 10 weeks
## 
## Number of studies combined: k = 43
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## Random effects model -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  10.66   42  1.0000
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI    Q  I^2
## < 5 weeks      9 -0.0635 [-0.3438; 0.2169] 1.20 0.0%
## 5 - 10 weeks  19 -0.0117 [-0.2071; 0.1837] 2.12 0.0%
## > 10 weeks    15 -0.0701 [-0.2444; 0.1042] 6.30 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.21    2  0.9021
## Within groups  9.62   40  1.0000
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI tau^2 tau
## < 5 weeks      9 -0.0635 [-0.3438; 0.2169]     0   0
## 5 - 10 weeks  19 -0.0117 [-0.2071; 0.1837]     0   0
## > 10 weeks    15 -0.0701 [-0.2444; 0.1042]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.21    2  0.9021
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
4.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 43; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0356)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 41) = 10.6587, p-val = 1.0000
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0016, p-val = 0.9685
## 
## Model Results:
## 
##           estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    -0.0546  0.1449  -0.3769  0.7062  -0.3386  0.2294    
## duration    0.0006  0.0144   0.0395  0.9685  -0.0277  0.0288    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
4.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

4.2.5 Men Ratio

4.2.5.1 Forest plot

4.2.5.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_men_ratio
## Bækkerud 2016           0.0834 [-0.8694; 1.0362]       1.5        1.5              < 0.5
## Beetham 2019           -0.7117 [-1.8362; 0.4129]       1.1        1.1              > 0.5
## Burgomaster 2008        0.1633 [-0.7147; 1.0413]       1.8        1.8              < 0.5
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.4        1.4              > 0.5
## Conraads 2015          -0.1303 [-0.4278; 0.1673]      15.7       15.7              > 0.5
## Currie 2015            -0.0220 [-0.9226; 0.8785]       1.7        1.7              > 0.5
## Earnest 2013            0.1102 [-0.5407; 0.7611]       3.3        3.3              > 0.5
## Eguchi 2012             0.0711 [-0.8057; 0.9479]       1.8        1.8              > 0.5
## Fisher 2015            -0.1077 [-0.9327; 0.7173]       2.0        2.0              > 0.5
## Gillen 2016            -0.0859 [-0.9869; 0.8150]       1.7        1.7              > 0.5
## Gorostiaga 1991         0.2781 [-0.8590; 1.4151]       1.1        1.1              < 0.5
## Granata 2015           -0.0737 [-0.9549; 0.8075]       1.8        1.8              < 0.5
## Granata 2015           -0.1189 [-1.0437; 0.8058]       1.6        1.6              < 0.5
## Grieco 2013            -0.1097 [-0.9495; 0.7302]       2.0        2.0              < 0.5
## Helgerud 2007          -0.1347 [-1.0122; 0.7428]       1.8        1.8              > 0.5
## Helgerud 2007          -0.2857 [-1.1667; 0.5953]       1.8        1.8              > 0.5
## Honkala 2017 (Healthy)  0.0925 [-0.6487; 0.8337]       2.5        2.5              > 0.5
## Honkala 2017 (T2D)      0.0204 [-0.9674; 1.0081]       1.4        1.4              > 0.5
## Jo 2020                -0.0456 [-0.7179; 0.6268]       3.1        3.1              > 0.5
## Keating 2014           -0.2150 [-1.0532; 0.6231]       2.0        2.0              < 0.5
## Klonizakis 2014         0.0572 [-0.8906; 1.0050]       1.5        1.5              < 0.5
## Macpherson 2011         0.1198 [-0.7576; 0.9971]       1.8        1.8              > 0.5
## Maillard 2016           0.1024 [-0.8783; 1.0830]       1.4        1.4              < 0.5
## Martins 2016            0.1874 [-0.5314; 0.9063]       2.7        2.7              < 0.5
## Matsuo 2014            -0.3160 [-1.0895; 0.4575]       2.3        2.3              > 0.5
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       2.2        2.2              > 0.5
## Mitranun 2014           0.2274 [-0.5158; 0.9705]       2.5        2.5              < 0.5
## Moreira 2008           -0.0139 [-0.9939; 0.9661]       1.4        1.4              < 0.5
## Nalcakan 2014          -0.0992 [-1.1142; 0.9158]       1.4        1.4              > 0.5
## Nie 2017               -0.0519 [-0.7693; 0.6655]       2.7        2.7              < 0.5
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       3.9        3.9              > 0.5
## Ramos 2016b             0.3042 [-0.3941; 1.0025]       2.9        2.9              > 0.5
## Robinson 2015          -0.0622 [-0.6903; 0.5658]       3.5        3.5              < 0.5
## Rognmo 2004            -0.1673 [-1.1213; 0.7867]       1.5        1.5              > 0.5
## Sandvei 2012           -0.0307 [-0.8489; 0.7875]       2.1        2.1              < 0.5
## Sawyer 2016             0.0090 [-0.9149; 0.9330]       1.6        1.6              < 0.5
## Scribbans 2014          0.0000 [-0.9005; 0.9005]       1.7        1.7              > 0.5
## Shepherd 2013          -0.0169 [-0.9969; 0.9631]       1.4        1.4              > 0.5
## Sjöros 2018             0.0162 [-0.8401; 0.8726]       1.9        1.9              > 0.5
## Tjønna 2008            -0.1325 [-1.0442; 0.7792]       1.7        1.7              < 0.5
## Trapp 2008              0.2121 [-0.5056; 0.9298]       2.7        2.7              < 0.5
## Winn 2018              -0.5813 [-1.5818; 0.4191]       1.4        1.4              < 0.5
## Zapata-Lamana 2018     -0.7184 [-1.4827; 0.0459]       2.4        2.4              < 0.5
## 
## Number of studies combined: k = 43
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## Random effects model -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  10.66   42  1.0000
## 
## Results for subgroups (fixed effect model):
##         k     SMD            95%-CI    Q  I^2
## < 0.5  20 -0.0370 [-0.2247; 0.1507] 5.99 0.0%
## > 0.5  23 -0.0546 [-0.2063; 0.0971] 3.82 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.02    1  0.8863
## Within groups  9.81   41  1.0000
## 
## Results for subgroups (random effects model):
##         k     SMD            95%-CI tau^2 tau
## < 0.5  20 -0.0370 [-0.2247; 0.1507]     0   0
## > 0.5  23 -0.0546 [-0.2063; 0.0971]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.02    1  0.8863
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
4.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 43; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0356)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 41) = 10.6419, p-val = 1.0000
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0183, p-val = 0.8923
## 
## Model Results:
## 
##            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt     -0.0339  0.1297  -0.2610  0.7941  -0.2880  0.2203    
## men_ratio   -0.0244  0.1801  -0.1354  0.8923  -0.3775  0.3287    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
4.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

4.2.6 Type of Exercise

4.2.6.1 Forest plot

4.2.6.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) type_exercise
## Bækkerud 2016           0.0834 [-0.8694; 1.0362]       1.5        1.5       Running
## Beetham 2019           -0.7117 [-1.8362; 0.4129]       1.1        1.1       Running
## Burgomaster 2008        0.1633 [-0.7147; 1.0413]       1.8        1.8       Cycling
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.4        1.4       Cycling
## Conraads 2015          -0.1303 [-0.4278; 0.1673]      15.7       15.7       Cycling
## Currie 2015            -0.0220 [-0.9226; 0.8785]       1.7        1.7       Cycling
## Earnest 2013            0.1102 [-0.5407; 0.7611]       3.3        3.3       Running
## Eguchi 2012             0.0711 [-0.8057; 0.9479]       1.8        1.8       Cycling
## Fisher 2015            -0.1077 [-0.9327; 0.7173]       2.0        2.0       Cycling
## Gillen 2016            -0.0859 [-0.9869; 0.8150]       1.7        1.7       Cycling
## Gorostiaga 1991         0.2781 [-0.8590; 1.4151]       1.1        1.1       Cycling
## Granata 2015           -0.0737 [-0.9549; 0.8075]       1.8        1.8       Cycling
## Granata 2015           -0.1189 [-1.0437; 0.8058]       1.6        1.6       Cycling
## Grieco 2013            -0.1097 [-0.9495; 0.7302]       2.0        2.0       Cycling
## Helgerud 2007          -0.1347 [-1.0122; 0.7428]       1.8        1.8       Running
## Helgerud 2007          -0.2857 [-1.1667; 0.5953]       1.8        1.8       Running
## Honkala 2017 (Healthy)  0.0925 [-0.6487; 0.8337]       2.5        2.5       Cycling
## Honkala 2017 (T2D)      0.0204 [-0.9674; 1.0081]       1.4        1.4       Cycling
## Jo 2020                -0.0456 [-0.7179; 0.6268]       3.1        3.1       Running
## Keating 2014           -0.2150 [-1.0532; 0.6231]       2.0        2.0       Cycling
## Klonizakis 2014         0.0572 [-0.8906; 1.0050]       1.5        1.5       Cycling
## Macpherson 2011         0.1198 [-0.7576; 0.9971]       1.8        1.8       Cycling
## Maillard 2016           0.1024 [-0.8783; 1.0830]       1.4        1.4       Cycling
## Martins 2016            0.1874 [-0.5314; 0.9063]       2.7        2.7       Cycling
## Matsuo 2014            -0.3160 [-1.0895; 0.4575]       2.3        2.3       Cycling
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       2.2        2.2       Cycling
## Mitranun 2014           0.2274 [-0.5158; 0.9705]       2.5        2.5       Running
## Moreira 2008           -0.0139 [-0.9939; 0.9661]       1.4        1.4       Cycling
## Nalcakan 2014          -0.0992 [-1.1142; 0.9158]       1.4        1.4       Cycling
## Nie 2017               -0.0519 [-0.7693; 0.6655]       2.7        2.7       Cycling
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       3.9        3.9       Running
## Ramos 2016b             0.3042 [-0.3941; 1.0025]       2.9        2.9       Running
## Robinson 2015          -0.0622 [-0.6903; 0.5658]       3.5        3.5       Cycling
## Rognmo 2004            -0.1673 [-1.1213; 0.7867]       1.5        1.5       Running
## Sandvei 2012           -0.0307 [-0.8489; 0.7875]       2.1        2.1       Running
## Sawyer 2016             0.0090 [-0.9149; 0.9330]       1.6        1.6       Cycling
## Scribbans 2014          0.0000 [-0.9005; 0.9005]       1.7        1.7       Cycling
## Shepherd 2013          -0.0169 [-0.9969; 0.9631]       1.4        1.4       Cycling
## Sjöros 2018             0.0162 [-0.8401; 0.8726]       1.9        1.9       Cycling
## Tjønna 2008            -0.1325 [-1.0442; 0.7792]       1.7        1.7       Running
## Trapp 2008              0.2121 [-0.5056; 0.9298]       2.7        2.7       Cycling
## Winn 2018              -0.5813 [-1.5818; 0.4191]       1.4        1.4       Running
## Zapata-Lamana 2018     -0.7184 [-1.4827; 0.0459]       2.4        2.4       Cycling
## 
## Number of studies combined: k = 43
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## Random effects model -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  10.66   42  1.0000
## 
## Results for subgroups (fixed effect model):
##           k     SMD            95%-CI    Q  I^2
## Running  13 -0.0345 [-0.2554; 0.1865] 4.21 0.0%
## Cycling  30 -0.0529 [-0.1925; 0.0867] 5.60 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.02    1  0.8900
## Within groups  9.81   41  1.0000
## 
## Results for subgroups (random effects model):
##           k     SMD            95%-CI tau^2 tau
## Running  13 -0.0345 [-0.2554; 0.1865]     0   0
## Cycling  30 -0.0529 [-0.1925; 0.0867]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.02    1  0.8900
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
4.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 43; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0355)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 41) = 10.6471, p-val = 1.0000
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0132, p-val = 0.9087
## 
## Model Results:
## 
##                       estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                -0.0538  0.0712  -0.7552  0.4501  -0.1933  0.0858    
## type_exerciseRunning    0.0153  0.1333   0.1147  0.9087  -0.2459  0.2765    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
4.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

4.2.7 Baseline Values

4.2.7.1 Forest plot

4.2.7.2 R output
##                            SMD            95%-CI %W(fixed) %W(random)     category_bsln
## Bækkerud 2016           0.0834 [-0.8694; 1.0362]       1.5        1.5    BMI > 30 kg/m²
## Beetham 2019           -0.7117 [-1.8362; 0.4129]       1.1        1.1    BMI > 30 kg/m²
## Burgomaster 2008        0.1633 [-0.7147; 1.0413]       1.8        1.8    BMI < 25 kg/m²
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.4        1.4    BMI < 25 kg/m²
## Conraads 2015          -0.1303 [-0.4278; 0.1673]      15.7       15.7 BMI 25 - 30 kg/m²
## Currie 2015            -0.0220 [-0.9226; 0.8785]       1.7        1.7 BMI 25 - 30 kg/m²
## Earnest 2013            0.1102 [-0.5407; 0.7611]       3.3        3.3    BMI > 30 kg/m²
## Eguchi 2012             0.0711 [-0.8057; 0.9479]       1.8        1.8 BMI 25 - 30 kg/m²
## Fisher 2015            -0.1077 [-0.9327; 0.7173]       2.0        2.0 BMI 25 - 30 kg/m²
## Gillen 2016            -0.0859 [-0.9869; 0.8150]       1.7        1.7 BMI 25 - 30 kg/m²
## Gorostiaga 1991         0.2781 [-0.8590; 1.4151]       1.1        1.1    BMI < 25 kg/m²
## Granata 2015           -0.0737 [-0.9549; 0.8075]       1.8        1.8    BMI < 25 kg/m²
## Granata 2015           -0.1189 [-1.0437; 0.8058]       1.6        1.6    BMI < 25 kg/m²
## Grieco 2013            -0.1097 [-0.9495; 0.7302]       2.0        2.0 BMI 25 - 30 kg/m²
## Helgerud 2007          -0.1347 [-1.0122; 0.7428]       1.8        1.8    BMI < 25 kg/m²
## Helgerud 2007          -0.2857 [-1.1667; 0.5953]       1.8        1.8    BMI < 25 kg/m²
## Honkala 2017 (Healthy)  0.0925 [-0.6487; 0.8337]       2.5        2.5 BMI 25 - 30 kg/m²
## Honkala 2017 (T2D)      0.0204 [-0.9674; 1.0081]       1.4        1.4    BMI > 30 kg/m²
## Jo 2020                -0.0456 [-0.7179; 0.6268]       3.1        3.1    BMI < 25 kg/m²
## Keating 2014           -0.2150 [-1.0532; 0.6231]       2.0        2.0 BMI 25 - 30 kg/m²
## Klonizakis 2014         0.0572 [-0.8906; 1.0050]       1.5        1.5    BMI < 25 kg/m²
## Macpherson 2011         0.1198 [-0.7576; 0.9971]       1.8        1.8    BMI < 25 kg/m²
## Maillard 2016           0.1024 [-0.8783; 1.0830]       1.4        1.4    BMI > 30 kg/m²
## Martins 2016            0.1874 [-0.5314; 0.9063]       2.7        2.7    BMI > 30 kg/m²
## Matsuo 2014            -0.3160 [-1.0895; 0.4575]       2.3        2.3    BMI < 25 kg/m²
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       2.2        2.2 BMI 25 - 30 kg/m²
## Mitranun 2014           0.2274 [-0.5158; 0.9705]       2.5        2.5 BMI 25 - 30 kg/m²
## Moreira 2008           -0.0139 [-0.9939; 0.9661]       1.4        1.4 BMI 25 - 30 kg/m²
## Nalcakan 2014          -0.0992 [-1.1142; 0.9158]       1.4        1.4    BMI < 25 kg/m²
## Nie 2017               -0.0519 [-0.7693; 0.6655]       2.7        2.7 BMI 25 - 30 kg/m²
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       3.9        3.9    BMI > 30 kg/m²
## Ramos 2016b             0.3042 [-0.3941; 1.0025]       2.9        2.9    BMI > 30 kg/m²
## Robinson 2015          -0.0622 [-0.6903; 0.5658]       3.5        3.5    BMI > 30 kg/m²
## Rognmo 2004            -0.1673 [-1.1213; 0.7867]       1.5        1.5 BMI 25 - 30 kg/m²
## Sandvei 2012           -0.0307 [-0.8489; 0.7875]       2.1        2.1    BMI < 25 kg/m²
## Sawyer 2016             0.0090 [-0.9149; 0.9330]       1.6        1.6    BMI > 30 kg/m²
## Scribbans 2014          0.0000 [-0.9005; 0.9005]       1.7        1.7    BMI < 25 kg/m²
## Shepherd 2013          -0.0169 [-0.9969; 0.9631]       1.4        1.4    BMI < 25 kg/m²
## Sjöros 2018             0.0162 [-0.8401; 0.8726]       1.9        1.9    BMI > 30 kg/m²
## Tjønna 2008            -0.1325 [-1.0442; 0.7792]       1.7        1.7 BMI 25 - 30 kg/m²
## Trapp 2008              0.2121 [-0.5056; 0.9298]       2.7        2.7    BMI < 25 kg/m²
## Winn 2018              -0.5813 [-1.5818; 0.4191]       1.4        1.4    BMI > 30 kg/m²
## Zapata-Lamana 2018     -0.7184 [-1.4827; 0.0459]       2.4        2.4    BMI > 30 kg/m²
## 
## Number of studies combined: k = 43
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## Random effects model -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  10.66   42  1.0000
## 
## Results for subgroups (fixed effect model):
##                     k     SMD            95%-CI    Q  I^2
## BMI < 25 kg/m²     16 -0.0233 [-0.2409; 0.1943] 1.86 0.0%
## BMI 25 - 30 kg/m²  14 -0.0662 [-0.2492; 0.1169] 1.24 0.0%
## BMI > 30 kg/m²     13 -0.0457 [-0.2647; 0.1733] 6.64 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.09    2  0.9571
## Within groups  9.74   40  1.0000
## 
## Results for subgroups (random effects model):
##                     k     SMD            95%-CI tau^2 tau
## BMI < 25 kg/m²     16 -0.0233 [-0.2409; 0.1943]     0   0
## BMI 25 - 30 kg/m²  14 -0.0662 [-0.2492; 0.1169]     0   0
## BMI > 30 kg/m²     13 -0.0457 [-0.2647; 0.1733]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.09    2  0.9571
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
4.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 43; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0351)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 41) = 10.5637, p-val = 1.0000
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0965, p-val = 0.7561
## 
## Model Results:
## 
##                estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt          0.1000  0.4849   0.2063  0.8365  -0.8503  1.0503    
## bsln_adjusted   -0.0054  0.0172  -0.3106  0.7561  -0.0392  0.0284    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
4.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

4.2.8 Type of HIIE

4.2.8.1 Forest plot

4.2.8.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) HIIE
## Bækkerud 2016           0.0834 [-0.8694; 1.0362]       1.5        1.5 HIIT
## Beetham 2019           -0.7117 [-1.8362; 0.4129]       1.1        1.1 HIIT
## Burgomaster 2008        0.1633 [-0.7147; 1.0413]       1.8        1.8  SIT
## Cocks 2013              0.0000 [-0.9800; 0.9800]       1.4        1.4  SIT
## Conraads 2015          -0.1303 [-0.4278; 0.1673]      15.7       15.7 HIIT
## Currie 2015            -0.0220 [-0.9226; 0.8785]       1.7        1.7 HIIT
## Earnest 2013            0.1102 [-0.5407; 0.7611]       3.3        3.3 HIIT
## Eguchi 2012             0.0711 [-0.8057; 0.9479]       1.8        1.8 HIIT
## Fisher 2015            -0.1077 [-0.9327; 0.7173]       2.0        2.0  SIT
## Gillen 2016            -0.0859 [-0.9869; 0.8150]       1.7        1.7  SIT
## Gorostiaga 1991         0.2781 [-0.8590; 1.4151]       1.1        1.1  SIT
## Granata 2015           -0.0737 [-0.9549; 0.8075]       1.8        1.8 HIIT
## Granata 2015           -0.1189 [-1.0437; 0.8058]       1.6        1.6  SIT
## Grieco 2013            -0.1097 [-0.9495; 0.7302]       2.0        2.0 HIIT
## Helgerud 2007          -0.1347 [-1.0122; 0.7428]       1.8        1.8 HIIT
## Helgerud 2007          -0.2857 [-1.1667; 0.5953]       1.8        1.8  SIT
## Honkala 2017 (Healthy)  0.0925 [-0.6487; 0.8337]       2.5        2.5  SIT
## Honkala 2017 (T2D)      0.0204 [-0.9674; 1.0081]       1.4        1.4  SIT
## Jo 2020                -0.0456 [-0.7179; 0.6268]       3.1        3.1 HIIT
## Keating 2014           -0.2150 [-1.0532; 0.6231]       2.0        2.0 HIIT
## Klonizakis 2014         0.0572 [-0.8906; 1.0050]       1.5        1.5 HIIT
## Macpherson 2011         0.1198 [-0.7576; 0.9971]       1.8        1.8  SIT
## Maillard 2016           0.1024 [-0.8783; 1.0830]       1.4        1.4 HIIT
## Martins 2016            0.1874 [-0.5314; 0.9063]       2.7        2.7  SIT
## Matsuo 2014            -0.3160 [-1.0895; 0.4575]       2.3        2.3 HIIT
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       2.2        2.2 HIIT
## Mitranun 2014           0.2274 [-0.5158; 0.9705]       2.5        2.5 HIIT
## Moreira 2008           -0.0139 [-0.9939; 0.9661]       1.4        1.4 HIIT
## Nalcakan 2014          -0.0992 [-1.1142; 0.9158]       1.4        1.4  SIT
## Nie 2017               -0.0519 [-0.7693; 0.6655]       2.7        2.7 HIIT
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       3.9        3.9 HIIT
## Ramos 2016b             0.3042 [-0.3941; 1.0025]       2.9        2.9 HIIT
## Robinson 2015          -0.0622 [-0.6903; 0.5658]       3.5        3.5 HIIT
## Rognmo 2004            -0.1673 [-1.1213; 0.7867]       1.5        1.5 HIIT
## Sandvei 2012           -0.0307 [-0.8489; 0.7875]       2.1        2.1  SIT
## Sawyer 2016             0.0090 [-0.9149; 0.9330]       1.6        1.6 HIIT
## Scribbans 2014          0.0000 [-0.9005; 0.9005]       1.7        1.7  SIT
## Shepherd 2013          -0.0169 [-0.9969; 0.9631]       1.4        1.4  SIT
## Sjöros 2018             0.0162 [-0.8401; 0.8726]       1.9        1.9  SIT
## Tjønna 2008            -0.1325 [-1.0442; 0.7792]       1.7        1.7 HIIT
## Trapp 2008              0.2121 [-0.5056; 0.9298]       2.7        2.7  SIT
## Winn 2018              -0.5813 [-1.5818; 0.4191]       1.4        1.4 HIIT
## Zapata-Lamana 2018     -0.7184 [-1.4827; 0.0459]       2.4        2.4 HIIT
## 
## Number of studies combined: k = 43
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## Random effects model -0.0494 [-0.1674; 0.0686] -0.82  0.4117
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  10.66   42  1.0000
## 
## Results for subgroups (fixed effect model):
##        k     SMD            95%-CI    Q  I^2
## HIIT  26 -0.0817 [-0.2239; 0.0605] 7.62 0.0%
## SIT   17  0.0276 [-0.1838; 0.2389] 1.50 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.71    1  0.4004
## Within groups  9.12   41  1.0000
## 
## Results for subgroups (random effects model):
##        k     SMD            95%-CI tau^2 tau
## HIIT  26 -0.0817 [-0.2239; 0.0605]     0   0
## SIT   17  0.0276 [-0.1838; 0.2389]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.71    1  0.4004
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
4.2.8.3 Meta-regression
## 
## Mixed-Effects Model (k = 43; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0354)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 41) = 9.9026, p-val = 1.0000
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.7576, p-val = 0.3841
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.0847  0.0725  -1.1670  0.2432  -0.2268  0.0575    
## HIIESIT    0.1131  0.1299   0.8704  0.3841  -0.1416  0.3678    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
4.2.8.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

5. Body Fat

5.1 Overall

5.1.1 Forest plot

5.1.2 R output

##                            SMD            95%-CI %W(fixed) %W(random)
## Beetham 2019            0.0493 [-1.0441; 1.1427]       2.0        2.0
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       3.0        3.0
## Earnest 2013            0.2324 [-0.4202; 0.8849]       5.5        5.5
## Eguchi 2012             0.0631 [-0.8137; 0.9398]       3.0        3.0
## Fisher 2015            -0.2232 [-1.0502; 0.6037]       3.4        3.4
## Gillen 2016             0.0000 [-0.9005; 0.9005]       2.9        2.9
## Grieco 2013            -0.3567 [-1.2024; 0.4891]       3.3        3.3
## Honkala 2017 (Healthy)  0.0861 [-0.6551; 0.8272]       4.3        4.3
## Honkala 2017 (T2D)      0.2716 [-0.7206; 1.2638]       2.4        2.4
## Keating 2014           -0.1327 [-0.9694; 0.7039]       3.3        3.3
## Lunt 2014               0.0000 [-0.8374; 0.8374]       3.3        3.3
## Lunt 2014               0.2120 [-0.6277; 1.0516]       3.3        3.3
## Macpherson 2011         0.1534 [-0.7244; 1.0312]       3.0        3.0
## Maillard 2016          -0.0873 [-1.0678; 0.8931]       2.4        2.4
## Matsuo 2014            -0.2568 [-1.0287; 0.5152]       3.9        3.9
## Matsuo 2015             0.0783 [-0.7222; 0.8788]       3.6        3.6
## Mitranun 2014          -0.0931 [-0.8343; 0.6481]       4.3        4.3
## Moreira 2008           -0.0912 [-1.0717; 0.8893]       2.4        2.4
## Motiani 2017            0.0919 [-0.6772; 0.8611]       3.9        3.9
## Nalcakan 2014           0.1726 [-0.8437; 1.1888]       2.3        2.3
## Nie 2017                0.1287 [-0.5893; 0.8467]       4.5        4.5
## Ramos 2016a             0.0230 [-0.5749; 0.6210]       6.5        6.5
## Ramos 2016b             0.1307 [-0.5643; 0.8258]       4.8        4.8
## Sandvei 2012           -0.0990 [-0.9176; 0.7197]       3.5        3.5
## Sawyer 2016             0.1401 [-0.7849; 1.0652]       2.7        2.7
## Sjöros 2018             0.0381 [-0.8184; 0.8945]       3.2        3.2
## Skleryk 2013            0.4302 [-0.5610; 1.4215]       2.4        2.4
## Trapp 2008              0.5592 [-0.1703; 1.2888]       4.4        4.4
## Winn 2018              -0.3125 [-1.2984; 0.6734]       2.4        2.4
## 
## Number of studies combined: k = 29
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0395 [-0.1133; 0.1923] 0.51  0.6121
## Random effects model 0.0395 [-0.1133; 0.1923] 0.51  0.6121
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.97   28  1.0000
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

5.1.3 Sensitivity analysis

5.1.3.1 Forest plot

5.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                                      SMD            95%-CI p-value   tau^2     tau   I^2
## Omitting Beetham 2019             0.0382 [-0.1162; 0.1926]  0.6275  0.0000  0.0000  0.0%
## Omitting Burgomaster 2008         0.0481 [-0.1071; 0.2033]  0.5434  0.0000  0.0000  0.0%
## Omitting Earnest 2013             0.0274 [-0.1298; 0.1846]  0.7326  0.0000  0.0000  0.0%
## Omitting Eguchi 2012              0.0377 [-0.1176; 0.1929]  0.6342  0.0000  0.0000  0.0%
## Omitting Fisher 2015              0.0473 [-0.1082; 0.2029]  0.5509  0.0000  0.0000  0.0%
## Omitting Gillen 2016              0.0395 [-0.1156; 0.1946]  0.6176  0.0000  0.0000  0.0%
## Omitting Grieco 2013              0.0512 [-0.1042; 0.2066]  0.5183  0.0000  0.0000  0.0%
## Omitting Honkala 2017 (Healthy)   0.0364 [-0.1198; 0.1926]  0.6482  0.0000  0.0000  0.0%
## Omitting Honkala 2017 (T2D)       0.0331 [-0.1216; 0.1878]  0.6752  0.0000  0.0000  0.0%
## Omitting Keating 2014             0.0441 [-0.1114; 0.1996]  0.5782  0.0000  0.0000  0.0%
## Omitting Lunt 2014                0.0397 [-0.1158; 0.1952]  0.6168  0.0000  0.0000  0.0%
## Omitting Lunt 2014                0.0327 [-0.1228; 0.1881]  0.6802  0.0000  0.0000  0.0%
## Omitting Macpherson 2011          0.0350 [-0.1202; 0.1902]  0.6587  0.0000  0.0000  0.0%
## Omitting Maillard 2016            0.0414 [-0.1134; 0.1961]  0.6002  0.0000  0.0000  0.0%
## Omitting Matsuo 2014              0.0501 [-0.1059; 0.2060]  0.5291  0.0000  0.0000  0.0%
## Omitting Matsuo 2015              0.0370 [-0.1188; 0.1927]  0.6418  0.0000  0.0000  0.0%
## Omitting Mitranun 2014            0.0441 [-0.1121; 0.2003]  0.5801  0.0000  0.0000  0.0%
## Omitting Moreira 2008             0.0415 [-0.1133; 0.1962]  0.5994  0.0000  0.0000  0.0%
## Omitting Motiani 2017             0.0363 [-0.1197; 0.1923]  0.6484  0.0000  0.0000  0.0%
## Omitting Nalcakan 2014            0.0355 [-0.1191; 0.1901]  0.6527  0.0000  0.0000  0.0%
## Omitting Nie 2017                 0.0342 [-0.1222; 0.1907]  0.6678  0.0000  0.0000  0.0%
## Omitting Ramos 2016a              0.0395 [-0.1186; 0.1976]  0.6246  0.0000  0.0000  0.0%
## Omitting Ramos 2016b              0.0338 [-0.1228; 0.1905]  0.6720  0.0000  0.0000  0.0%
## Omitting Sandvei 2012             0.0432 [-0.1124; 0.1988]  0.5863  0.0000  0.0000  0.0%
## Omitting Sawyer 2016              0.0357 [-0.1193; 0.1907]  0.6516  0.0000  0.0000  0.0%
## Omitting Sjöros 2018              0.0384 [-0.1169; 0.1938]  0.6278  0.0000  0.0000  0.0%
## Omitting Skleryk 2013             0.0294 [-0.1253; 0.1841]  0.7092  0.0000  0.0000  0.0%
## Omitting Trapp 2008               0.0152 [-0.1411; 0.1715]  0.8485  0.0000  0.0000  0.0%
## Omitting Winn 2018                0.0466 [-0.1081; 0.2013]  0.5552  0.0000  0.0000  0.0%
##                                                                                         
## Pooled estimate                   0.0395 [-0.1133; 0.1923]  0.6121  0.0000  0.0000  0.0%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

5.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

5.2 Subgroups

5.2.1 Overall

5.2.1.1 Forest plot

5.2.1.2 R output
##                        SMD            95%-CI     meta-analysis
##                     0.0395 [-0.1133; 0.1923]           Overall
## Healthy             0.0332 [-0.2027; 0.2691]        Population
## Overweight/obese    0.0410 [-0.2346; 0.3166]        Population
## Metabolic Syndrome  0.0692 [-0.3252; 0.4637]        Population
## T2D                 0.0114 [-0.4255; 0.4484]        Population
## < 30 y             -0.0067 [-0.2683; 0.2548]               Age
## 30 - 50 y           0.0845 [-0.1492; 0.3181]               Age
## > 50 y              0.0197 [-0.2987; 0.3381]               Age
## < 5 weeks           0.0261 [-0.3014; 0.3536] Training Duration
## 5 - 10 weeks       -0.0158 [-0.2733; 0.2417] Training Duration
## > 10 weeks          0.0890 [-0.1441; 0.3221] Training Duration
## < 0.5              -0.0052 [-0.2386; 0.2283]         Men Ratio
## > 0.5               0.0711 [-0.1312; 0.2733]         Men Ratio
## Running             0.0380 [-0.2182; 0.2942]  Type of Exercise
## Cycling             0.0386 [-0.1519; 0.2290]  Type of Exercise
## BMI < 25 kg/m²      0.0526 [-0.2886; 0.3937]   Baseline Values
## BMI 25 - 30 kg/m²  -0.0292 [-0.2740; 0.2156]   Baseline Values
## BMI > 30 kg/m²      0.0958 [-0.1431; 0.3347]   Baseline Values
## HIIT               -0.0091 [-0.2096; 0.1914]      Type of HIIE
## SIT                 0.1042 [-0.1319; 0.3404]      Type of HIIE
## 
## Number of studies combined: k = 29
## 
##                         SMD            95%-CI    z p-value
## Random effects model 0.0395 [-0.1133; 0.1923] 0.51  0.6121
## 
## Quantifying heterogeneity:
##  tau^2 = 0; tau = 0; I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.97   28  1.0000
## 
## Results for meta-analyses (random effects model):
##                     k    SMD            95%-CI tau^2 tau    Q  I^2
## Overall            29 0.0395 [-0.1133; 0.1923]     0   0 6.97 0.0%
## Population         29 0.0395 [-0.1133; 0.1923]     0   0 6.97 0.0%
## Age                29 0.0395 [-0.1133; 0.1923]     0   0 6.97 0.0%
## Training Duration  29 0.0395 [-0.1133; 0.1923]     0   0 6.97 0.0%
## Men Ratio          29 0.0395 [-0.1133; 0.1923]     0   0 6.97 0.0%
## Type of Exercise   29 0.0395 [-0.1133; 0.1923]     0   0 6.97 0.0%
## Baseline Values    29 0.0395 [-0.1133; 0.1923]     0   0 6.97 0.0%
## Type of HIIE       29 0.0395 [-0.1133; 0.1923]     0   0 6.97 0.0%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

5.2.2 Population

5.2.2.1 Forest plot

5.2.2.2 R output
##                            SMD            95%-CI %W(fixed) %W(random)         population
## Beetham 2019            0.0493 [-1.0441; 1.1427]       2.0        2.0   Overweight/obese
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       3.0        3.0            Healthy
## Earnest 2013            0.2324 [-0.4202; 0.8849]       5.5        5.5   Overweight/obese
## Eguchi 2012             0.0631 [-0.8137; 0.9398]       3.0        3.0            Healthy
## Fisher 2015            -0.2232 [-1.0502; 0.6037]       3.4        3.4   Overweight/obese
## Gillen 2016             0.0000 [-0.9005; 0.9005]       2.9        2.9            Healthy
## Grieco 2013            -0.3567 [-1.2024; 0.4891]       3.3        3.3            Healthy
## Honkala 2017 (Healthy)  0.0861 [-0.6551; 0.8272]       4.3        4.3            Healthy
## Honkala 2017 (T2D)      0.2716 [-0.7206; 1.2638]       2.4        2.4                T2D
## Keating 2014           -0.1327 [-0.9694; 0.7039]       3.3        3.3   Overweight/obese
## Lunt 2014               0.0000 [-0.8374; 0.8374]       3.3        3.3   Overweight/obese
## Lunt 2014               0.2120 [-0.6277; 1.0516]       3.3        3.3   Overweight/obese
## Macpherson 2011         0.1534 [-0.7244; 1.0312]       3.0        3.0            Healthy
## Maillard 2016          -0.0873 [-1.0678; 0.8931]       2.4        2.4                T2D
## Matsuo 2014            -0.2568 [-1.0287; 0.5152]       3.9        3.9            Healthy
## Matsuo 2015             0.0783 [-0.7222; 0.8788]       3.6        3.6 Metabolic Syndrome
## Mitranun 2014          -0.0931 [-0.8343; 0.6481]       4.3        4.3                T2D
## Moreira 2008           -0.0912 [-1.0717; 0.8893]       2.4        2.4   Overweight/obese
## Motiani 2017            0.0919 [-0.6772; 0.8611]       3.9        3.9            Healthy
## Nalcakan 2014           0.1726 [-0.8437; 1.1888]       2.3        2.3            Healthy
## Nie 2017                0.1287 [-0.5893; 0.8467]       4.5        4.5            Healthy
## Ramos 2016a             0.0230 [-0.5749; 0.6210]       6.5        6.5 Metabolic Syndrome
## Ramos 2016b             0.1307 [-0.5643; 0.8258]       4.8        4.8 Metabolic Syndrome
## Sandvei 2012           -0.0990 [-0.9176; 0.7197]       3.5        3.5            Healthy
## Sawyer 2016             0.1401 [-0.7849; 1.0652]       2.7        2.7   Overweight/obese
## Sjöros 2018             0.0381 [-0.8184; 0.8945]       3.2        3.2                T2D
## Skleryk 2013            0.4302 [-0.5610; 1.4215]       2.4        2.4   Overweight/obese
## Trapp 2008              0.5592 [-0.1703; 1.2888]       4.4        4.4            Healthy
## Winn 2018              -0.3125 [-1.2984; 0.6734]       2.4        2.4   Overweight/obese
## 
## Number of studies combined: k = 29
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0395 [-0.1133; 0.1923] 0.51  0.6121
## Random effects model 0.0395 [-0.1133; 0.1923] 0.51  0.6121
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.97   28  1.0000
## 
## Results for subgroups (fixed effect model):
##                      k    SMD            95%-CI    Q  I^2
## Healthy             12 0.0332 [-0.2027; 0.2691] 3.95 0.0%
## Overweight/obese    10 0.0410 [-0.2346; 0.3166] 2.06 0.0%
## Metabolic Syndrome   3 0.0692 [-0.3252; 0.4637] 0.05 0.0%
## T2D                  4 0.0114 [-0.4255; 0.4484] 0.35 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.04    3  0.9979
## Within groups  6.41   25  0.9999
## 
## Results for subgroups (random effects model):
##                      k    SMD            95%-CI tau^2 tau
## Healthy             12 0.0332 [-0.2027; 0.2691]     0   0
## Overweight/obese    10 0.0410 [-0.2346; 0.3166]     0   0
## Metabolic Syndrome   3 0.0692 [-0.3252; 0.4637]     0   0
## T2D                  4 0.0114 [-0.4255; 0.4484]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.04    3  0.9979
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
5.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 29; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0509)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 25) = 6.9254, p-val = 0.9999
## 
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 0.0414, p-val = 0.9978
## 
## Model Results:
## 
##                           estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                     0.0340  0.1203   0.2827  0.7774  -0.2018  0.2698    
## .byvarOverweight/obese      0.0082  0.1850   0.0446  0.9644  -0.3544  0.3709    
## .byvarMetabolic Syndrome    0.0371  0.2345   0.1584  0.8742  -0.4224  0.4967    
## .byvarT2D                  -0.0211  0.2533  -0.0834  0.9336  -0.5176  0.4753    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
5.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

5.2.3 Age

5.2.3.1 Forest plot

5.2.3.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_age
## Beetham 2019            0.0493 [-1.0441; 1.1427]       2.0        2.0       > 50 y
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       3.0        3.0       < 30 y
## Earnest 2013            0.2324 [-0.4202; 0.8849]       5.5        5.5    30 - 50 y
## Eguchi 2012             0.0631 [-0.8137; 0.9398]       3.0        3.0       > 50 y
## Fisher 2015            -0.2232 [-1.0502; 0.6037]       3.4        3.4       < 30 y
## Gillen 2016             0.0000 [-0.9005; 0.9005]       2.9        2.9       < 30 y
## Grieco 2013            -0.3567 [-1.2024; 0.4891]       3.3        3.3       < 30 y
## Honkala 2017 (Healthy)  0.0861 [-0.6551; 0.8272]       4.3        4.3    30 - 50 y
## Honkala 2017 (T2D)      0.2716 [-0.7206; 1.2638]       2.4        2.4    30 - 50 y
## Keating 2014           -0.1327 [-0.9694; 0.7039]       3.3        3.3    30 - 50 y
## Lunt 2014               0.0000 [-0.8374; 0.8374]       3.3        3.3    30 - 50 y
## Lunt 2014               0.2120 [-0.6277; 1.0516]       3.3        3.3    30 - 50 y
## Macpherson 2011         0.1534 [-0.7244; 1.0312]       3.0        3.0       < 30 y
## Maillard 2016          -0.0873 [-1.0678; 0.8931]       2.4        2.4       > 50 y
## Matsuo 2014            -0.2568 [-1.0287; 0.5152]       3.9        3.9       < 30 y
## Matsuo 2015             0.0783 [-0.7222; 0.8788]       3.6        3.6    30 - 50 y
## Mitranun 2014          -0.0931 [-0.8343; 0.6481]       4.3        4.3       > 50 y
## Moreira 2008           -0.0912 [-1.0717; 0.8893]       2.4        2.4    30 - 50 y
## Motiani 2017            0.0919 [-0.6772; 0.8611]       3.9        3.9    30 - 50 y
## Nalcakan 2014           0.1726 [-0.8437; 1.1888]       2.3        2.3       < 30 y
## Nie 2017                0.1287 [-0.5893; 0.8467]       4.5        4.5       < 30 y
## Ramos 2016a             0.0230 [-0.5749; 0.6210]       6.5        6.5       > 50 y
## Ramos 2016b             0.1307 [-0.5643; 0.8258]       4.8        4.8       > 50 y
## Sandvei 2012           -0.0990 [-0.9176; 0.7197]       3.5        3.5       < 30 y
## Sawyer 2016             0.1401 [-0.7849; 1.0652]       2.7        2.7    30 - 50 y
## Sjöros 2018             0.0381 [-0.8184; 0.8945]       3.2        3.2    30 - 50 y
## Skleryk 2013            0.4302 [-0.5610; 1.4215]       2.4        2.4    30 - 50 y
## Trapp 2008              0.5592 [-0.1703; 1.2888]       4.4        4.4       < 30 y
## Winn 2018              -0.3125 [-1.2984; 0.6734]       2.4        2.4    30 - 50 y
## 
## Number of studies combined: k = 29
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0395 [-0.1133; 0.1923] 0.51  0.6121
## Random effects model 0.0395 [-0.1133; 0.1923] 0.51  0.6121
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.97   28  1.0000
## 
## Results for subgroups (fixed effect model):
##             k     SMD            95%-CI    Q  I^2
## < 30 y     10 -0.0067 [-0.2683; 0.2548] 4.17 0.0%
## 30 - 50 y  13  0.0845 [-0.1492; 0.3181] 1.78 0.0%
## > 50 y      6  0.0197 [-0.2987; 0.3381] 0.23 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.28    2  0.8707
## Within groups  6.18   26  1.0000
## 
## Results for subgroups (random effects model):
##             k     SMD            95%-CI tau^2 tau
## < 30 y     10 -0.0067 [-0.2683; 0.2548]     0   0
## 30 - 50 y  13  0.0845 [-0.1492; 0.3181]     0   0
## > 50 y      6  0.0197 [-0.2987; 0.3381]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.28    2  0.8707
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
5.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 29; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0483)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 27) = 6.9425, p-val = 1.0000
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0244, p-val = 0.8759
## 
## Model Results:
## 
##          estimate      se    zval    pval    ci.lb   ci.ub 
## intrcpt    0.0042  0.2395  0.0174  0.9861  -0.4652  0.4736    
## age        0.0009  0.0055  0.1562  0.8759  -0.0100  0.0117    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
5.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

5.2.4 Training Duration

5.2.4.1 Forest plot

5.2.4.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_duration
## Beetham 2019            0.0493 [-1.0441; 1.1427]       2.0        2.0        > 10 weeks
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       3.0        3.0      5 - 10 weeks
## Earnest 2013            0.2324 [-0.4202; 0.8849]       5.5        5.5      5 - 10 weeks
## Eguchi 2012             0.0631 [-0.8137; 0.9398]       3.0        3.0        > 10 weeks
## Fisher 2015            -0.2232 [-1.0502; 0.6037]       3.4        3.4      5 - 10 weeks
## Gillen 2016             0.0000 [-0.9005; 0.9005]       2.9        2.9        > 10 weeks
## Grieco 2013            -0.3567 [-1.2024; 0.4891]       3.3        3.3         < 5 weeks
## Honkala 2017 (Healthy)  0.0861 [-0.6551; 0.8272]       4.3        4.3         < 5 weeks
## Honkala 2017 (T2D)      0.2716 [-0.7206; 1.2638]       2.4        2.4         < 5 weeks
## Keating 2014           -0.1327 [-0.9694; 0.7039]       3.3        3.3        > 10 weeks
## Lunt 2014               0.0000 [-0.8374; 0.8374]       3.3        3.3        > 10 weeks
## Lunt 2014               0.2120 [-0.6277; 1.0516]       3.3        3.3        > 10 weeks
## Macpherson 2011         0.1534 [-0.7244; 1.0312]       3.0        3.0      5 - 10 weeks
## Maillard 2016          -0.0873 [-1.0678; 0.8931]       2.4        2.4        > 10 weeks
## Matsuo 2014            -0.2568 [-1.0287; 0.5152]       3.9        3.9      5 - 10 weeks
## Matsuo 2015             0.0783 [-0.7222; 0.8788]       3.6        3.6      5 - 10 weeks
## Mitranun 2014          -0.0931 [-0.8343; 0.6481]       4.3        4.3      5 - 10 weeks
## Moreira 2008           -0.0912 [-1.0717; 0.8893]       2.4        2.4        > 10 weeks
## Motiani 2017            0.0919 [-0.6772; 0.8611]       3.9        3.9         < 5 weeks
## Nalcakan 2014           0.1726 [-0.8437; 1.1888]       2.3        2.3      5 - 10 weeks
## Nie 2017                0.1287 [-0.5893; 0.8467]       4.5        4.5        > 10 weeks
## Ramos 2016a             0.0230 [-0.5749; 0.6210]       6.5        6.5        > 10 weeks
## Ramos 2016b             0.1307 [-0.5643; 0.8258]       4.8        4.8        > 10 weeks
## Sandvei 2012           -0.0990 [-0.9176; 0.7197]       3.5        3.5      5 - 10 weeks
## Sawyer 2016             0.1401 [-0.7849; 1.0652]       2.7        2.7      5 - 10 weeks
## Sjöros 2018             0.0381 [-0.8184; 0.8945]       3.2        3.2         < 5 weeks
## Skleryk 2013            0.4302 [-0.5610; 1.4215]       2.4        2.4         < 5 weeks
## Trapp 2008              0.5592 [-0.1703; 1.2888]       4.4        4.4        > 10 weeks
## Winn 2018              -0.3125 [-1.2984; 0.6734]       2.4        2.4         < 5 weeks
## 
## Number of studies combined: k = 29
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0395 [-0.1133; 0.1923] 0.51  0.6121
## Random effects model 0.0395 [-0.1133; 0.1923] 0.51  0.6121
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.97   28  1.0000
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI    Q  I^2
## < 5 weeks      7  0.0261 [-0.3014; 0.3536] 1.96 0.0%
## 5 - 10 weeks  10 -0.0158 [-0.2733; 0.2417] 1.92 0.0%
## > 10 weeks    12  0.0890 [-0.1441; 0.3221] 2.22 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.36    2  0.8367
## Within groups  6.10   26  1.0000
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI tau^2 tau
## < 5 weeks      7  0.0261 [-0.3014; 0.3536]     0   0
## 5 - 10 weeks  10 -0.0158 [-0.2733; 0.2417]     0   0
## > 10 weeks    12  0.0890 [-0.1441; 0.3221]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.36    2  0.8367
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
5.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 29; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0484)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 27) = 6.8437, p-val = 1.0000
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.1232, p-val = 0.7256
## 
## Model Results:
## 
##           estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    -0.0134  0.1698  -0.0789  0.9371  -0.3462  0.3194    
## duration    0.0059  0.0168   0.3510  0.7256  -0.0271  0.0389    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
5.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

5.2.5 Men Ratio

5.2.5.1 Forest plot

5.2.5.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_men_ratio
## Beetham 2019            0.0493 [-1.0441; 1.1427]       2.0        2.0              > 0.5
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       3.0        3.0              < 0.5
## Earnest 2013            0.2324 [-0.4202; 0.8849]       5.5        5.5              > 0.5
## Eguchi 2012             0.0631 [-0.8137; 0.9398]       3.0        3.0              > 0.5
## Fisher 2015            -0.2232 [-1.0502; 0.6037]       3.4        3.4              > 0.5
## Gillen 2016             0.0000 [-0.9005; 0.9005]       2.9        2.9              > 0.5
## Grieco 2013            -0.3567 [-1.2024; 0.4891]       3.3        3.3              < 0.5
## Honkala 2017 (Healthy)  0.0861 [-0.6551; 0.8272]       4.3        4.3              > 0.5
## Honkala 2017 (T2D)      0.2716 [-0.7206; 1.2638]       2.4        2.4              > 0.5
## Keating 2014           -0.1327 [-0.9694; 0.7039]       3.3        3.3              < 0.5
## Lunt 2014               0.0000 [-0.8374; 0.8374]       3.3        3.3              < 0.5
## Lunt 2014               0.2120 [-0.6277; 1.0516]       3.3        3.3              < 0.5
## Macpherson 2011         0.1534 [-0.7244; 1.0312]       3.0        3.0              > 0.5
## Maillard 2016          -0.0873 [-1.0678; 0.8931]       2.4        2.4              < 0.5
## Matsuo 2014            -0.2568 [-1.0287; 0.5152]       3.9        3.9              > 0.5
## Matsuo 2015             0.0783 [-0.7222; 0.8788]       3.6        3.6              > 0.5
## Mitranun 2014          -0.0931 [-0.8343; 0.6481]       4.3        4.3              < 0.5
## Moreira 2008           -0.0912 [-1.0717; 0.8893]       2.4        2.4              < 0.5
## Motiani 2017            0.0919 [-0.6772; 0.8611]       3.9        3.9              > 0.5
## Nalcakan 2014           0.1726 [-0.8437; 1.1888]       2.3        2.3              > 0.5
## Nie 2017                0.1287 [-0.5893; 0.8467]       4.5        4.5              < 0.5
## Ramos 2016a             0.0230 [-0.5749; 0.6210]       6.5        6.5              > 0.5
## Ramos 2016b             0.1307 [-0.5643; 0.8258]       4.8        4.8              > 0.5
## Sandvei 2012           -0.0990 [-0.9176; 0.7197]       3.5        3.5              < 0.5
## Sawyer 2016             0.1401 [-0.7849; 1.0652]       2.7        2.7              < 0.5
## Sjöros 2018             0.0381 [-0.8184; 0.8945]       3.2        3.2              > 0.5
## Skleryk 2013            0.4302 [-0.5610; 1.4215]       2.4        2.4              > 0.5
## Trapp 2008              0.5592 [-0.1703; 1.2888]       4.4        4.4              < 0.5
## Winn 2018              -0.3125 [-1.2984; 0.6734]       2.4        2.4              < 0.5
## 
## Number of studies combined: k = 29
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0395 [-0.1133; 0.1923] 0.51  0.6121
## Random effects model 0.0395 [-0.1133; 0.1923] 0.51  0.6121
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.97   28  1.0000
## 
## Results for subgroups (fixed effect model):
##         k     SMD            95%-CI    Q  I^2
## < 0.5  13 -0.0052 [-0.2386; 0.2283] 4.16 0.0%
## > 0.5  16  0.0711 [-0.1312; 0.2733] 2.06 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.23    1  0.6285
## Within groups  6.22   27  1.0000
## 
## Results for subgroups (random effects model):
##         k     SMD            95%-CI tau^2 tau
## < 0.5  13 -0.0052 [-0.2386; 0.2283]     0   0
## > 0.5  16  0.0711 [-0.1312; 0.2733]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.23    1  0.6285
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
5.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 29; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0483)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 27) = 6.9652, p-val = 1.0000
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0017, p-val = 0.9676
## 
## Model Results:
## 
##            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt      0.0455  0.1653   0.2750  0.7833  -0.2786  0.3695    
## men_ratio   -0.0094  0.2307  -0.0407  0.9676  -0.4616  0.4429    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
5.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

5.2.6 Type of Exercise

5.2.6.1 Forest plot

5.2.6.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) type_exercise
## Beetham 2019            0.0493 [-1.0441; 1.1427]       2.0        2.0       Running
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       3.0        3.0       Cycling
## Earnest 2013            0.2324 [-0.4202; 0.8849]       5.5        5.5       Running
## Eguchi 2012             0.0631 [-0.8137; 0.9398]       3.0        3.0       Cycling
## Fisher 2015            -0.2232 [-1.0502; 0.6037]       3.4        3.4       Cycling
## Gillen 2016             0.0000 [-0.9005; 0.9005]       2.9        2.9       Cycling
## Grieco 2013            -0.3567 [-1.2024; 0.4891]       3.3        3.3       Cycling
## Honkala 2017 (Healthy)  0.0861 [-0.6551; 0.8272]       4.3        4.3       Cycling
## Honkala 2017 (T2D)      0.2716 [-0.7206; 1.2638]       2.4        2.4       Cycling
## Keating 2014           -0.1327 [-0.9694; 0.7039]       3.3        3.3       Cycling
## Lunt 2014               0.0000 [-0.8374; 0.8374]       3.3        3.3       Running
## Lunt 2014               0.2120 [-0.6277; 1.0516]       3.3        3.3       Running
## Macpherson 2011         0.1534 [-0.7244; 1.0312]       3.0        3.0       Cycling
## Maillard 2016          -0.0873 [-1.0678; 0.8931]       2.4        2.4       Cycling
## Matsuo 2014            -0.2568 [-1.0287; 0.5152]       3.9        3.9       Cycling
## Matsuo 2015             0.0783 [-0.7222; 0.8788]       3.6        3.6       Cycling
## Mitranun 2014          -0.0931 [-0.8343; 0.6481]       4.3        4.3       Running
## Moreira 2008           -0.0912 [-1.0717; 0.8893]       2.4        2.4       Cycling
## Motiani 2017            0.0919 [-0.6772; 0.8611]       3.9        3.9       Cycling
## Nalcakan 2014           0.1726 [-0.8437; 1.1888]       2.3        2.3       Cycling
## Nie 2017                0.1287 [-0.5893; 0.8467]       4.5        4.5       Cycling
## Ramos 2016a             0.0230 [-0.5749; 0.6210]       6.5        6.5       Running
## Ramos 2016b             0.1307 [-0.5643; 0.8258]       4.8        4.8       Running
## Sandvei 2012           -0.0990 [-0.9176; 0.7197]       3.5        3.5       Running
## Sawyer 2016             0.1401 [-0.7849; 1.0652]       2.7        2.7       Cycling
## Sjöros 2018             0.0381 [-0.8184; 0.8945]       3.2        3.2       Cycling
## Skleryk 2013            0.4302 [-0.5610; 1.4215]       2.4        2.4       Cycling
## Trapp 2008              0.5592 [-0.1703; 1.2888]       4.4        4.4       Cycling
## Winn 2018              -0.3125 [-1.2984; 0.6734]       2.4        2.4       Running
## 
## Number of studies combined: k = 29
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0395 [-0.1133; 0.1923] 0.51  0.6121
## Random effects model 0.0395 [-0.1133; 0.1923] 0.51  0.6121
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.97   28  1.0000
## 
## Results for subgroups (fixed effect model):
##           k    SMD            95%-CI    Q  I^2
## Running   9 0.0380 [-0.2182; 0.2942] 1.20 0.0%
## Cycling  20 0.0386 [-0.1519; 0.2290] 5.25 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.00    1  0.9972
## Within groups  6.45   27  1.0000
## 
## Results for subgroups (random effects model):
##           k    SMD            95%-CI tau^2 tau
## Running   9 0.0380 [-0.2182; 0.2942]     0   0
## Cycling  20 0.0386 [-0.1519; 0.2290]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.00    1  0.9972
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
5.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 29; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0484)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 27) = 6.9667, p-val = 1.0000
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0001, p-val = 0.9909
## 
## Model Results:
## 
##                       estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                 0.0402  0.0971   0.4139  0.6790  -0.1502  0.2306    
## type_exerciseRunning   -0.0019  0.1628  -0.0114  0.9909  -0.3210  0.3173    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
5.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

5.2.7 Baseline Values

5.2.7.1 Forest plot

5.2.7.2 R output
##                            SMD            95%-CI %W(fixed) %W(random)     category_bsln
## Beetham 2019            0.0493 [-1.0441; 1.1427]       2.0        2.0    BMI > 30 kg/m²
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       3.0        3.0    BMI < 25 kg/m²
## Earnest 2013            0.2324 [-0.4202; 0.8849]       5.5        5.5    BMI > 30 kg/m²
## Eguchi 2012             0.0631 [-0.8137; 0.9398]       3.0        3.0 BMI 25 - 30 kg/m²
## Fisher 2015            -0.2232 [-1.0502; 0.6037]       3.4        3.4 BMI 25 - 30 kg/m²
## Gillen 2016             0.0000 [-0.9005; 0.9005]       2.9        2.9 BMI 25 - 30 kg/m²
## Grieco 2013            -0.3567 [-1.2024; 0.4891]       3.3        3.3 BMI 25 - 30 kg/m²
## Honkala 2017 (Healthy)  0.0861 [-0.6551; 0.8272]       4.3        4.3 BMI 25 - 30 kg/m²
## Honkala 2017 (T2D)      0.2716 [-0.7206; 1.2638]       2.4        2.4    BMI > 30 kg/m²
## Keating 2014           -0.1327 [-0.9694; 0.7039]       3.3        3.3 BMI 25 - 30 kg/m²
## Lunt 2014               0.0000 [-0.8374; 0.8374]       3.3        3.3    BMI > 30 kg/m²
## Lunt 2014               0.2120 [-0.6277; 1.0516]       3.3        3.3    BMI > 30 kg/m²
## Macpherson 2011         0.1534 [-0.7244; 1.0312]       3.0        3.0    BMI < 25 kg/m²
## Maillard 2016          -0.0873 [-1.0678; 0.8931]       2.4        2.4    BMI > 30 kg/m²
## Matsuo 2014            -0.2568 [-1.0287; 0.5152]       3.9        3.9    BMI < 25 kg/m²
## Matsuo 2015             0.0783 [-0.7222; 0.8788]       3.6        3.6 BMI 25 - 30 kg/m²
## Mitranun 2014          -0.0931 [-0.8343; 0.6481]       4.3        4.3 BMI 25 - 30 kg/m²
## Moreira 2008           -0.0912 [-1.0717; 0.8893]       2.4        2.4 BMI 25 - 30 kg/m²
## Motiani 2017            0.0919 [-0.6772; 0.8611]       3.9        3.9 BMI 25 - 30 kg/m²
## Nalcakan 2014           0.1726 [-0.8437; 1.1888]       2.3        2.3    BMI < 25 kg/m²
## Nie 2017                0.1287 [-0.5893; 0.8467]       4.5        4.5 BMI 25 - 30 kg/m²
## Ramos 2016a             0.0230 [-0.5749; 0.6210]       6.5        6.5    BMI > 30 kg/m²
## Ramos 2016b             0.1307 [-0.5643; 0.8258]       4.8        4.8    BMI > 30 kg/m²
## Sandvei 2012           -0.0990 [-0.9176; 0.7197]       3.5        3.5    BMI < 25 kg/m²
## Sawyer 2016             0.1401 [-0.7849; 1.0652]       2.7        2.7    BMI > 30 kg/m²
## Sjöros 2018             0.0381 [-0.8184; 0.8945]       3.2        3.2    BMI > 30 kg/m²
## Skleryk 2013            0.4302 [-0.5610; 1.4215]       2.4        2.4    BMI > 30 kg/m²
## Trapp 2008              0.5592 [-0.1703; 1.2888]       4.4        4.4    BMI < 25 kg/m²
## Winn 2018              -0.3125 [-1.2984; 0.6734]       2.4        2.4    BMI > 30 kg/m²
## 
## Number of studies combined: k = 29
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0395 [-0.1133; 0.1923] 0.51  0.6121
## Random effects model 0.0395 [-0.1133; 0.1923] 0.51  0.6121
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.97   28  1.0000
## 
## Results for subgroups (fixed effect model):
##                     k     SMD            95%-CI    Q  I^2
## BMI < 25 kg/m²      6  0.0526 [-0.2886; 0.3937] 3.07 0.0%
## BMI 25 - 30 kg/m²  11 -0.0292 [-0.2740; 0.2156] 1.28 0.0%
## BMI > 30 kg/m²     12  0.0958 [-0.1431; 0.3347] 1.58 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.52    2  0.7705
## Within groups  5.93   26  1.0000
## 
## Results for subgroups (random effects model):
##                     k     SMD            95%-CI tau^2 tau
## BMI < 25 kg/m²      6  0.0526 [-0.2886; 0.3937]     0   0
## BMI 25 - 30 kg/m²  11 -0.0292 [-0.2740; 0.2156]     0   0
## BMI > 30 kg/m²     12  0.0958 [-0.1431; 0.3347]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.52    2  0.7705
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
5.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 29; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0482)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 27) = 6.9094, p-val = 1.0000
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0574, p-val = 0.8106
## 
## Model Results:
## 
##                estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt         -0.1011  0.5920  -0.1708  0.8644  -1.2613  1.0591    
## bsln_adjusted    0.0049  0.0205   0.2397  0.8106  -0.0353  0.0452    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
5.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

5.2.8 Type of HIIE

5.2.8.1 Forest plot

5.2.8.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) HIIE
## Beetham 2019            0.0493 [-1.0441; 1.1427]       2.0        2.0 HIIT
## Burgomaster 2008       -0.2887 [-1.1698; 0.5924]       3.0        3.0  SIT
## Earnest 2013            0.2324 [-0.4202; 0.8849]       5.5        5.5 HIIT
## Eguchi 2012             0.0631 [-0.8137; 0.9398]       3.0        3.0 HIIT
## Fisher 2015            -0.2232 [-1.0502; 0.6037]       3.4        3.4  SIT
## Gillen 2016             0.0000 [-0.9005; 0.9005]       2.9        2.9  SIT
## Grieco 2013            -0.3567 [-1.2024; 0.4891]       3.3        3.3 HIIT
## Honkala 2017 (Healthy)  0.0861 [-0.6551; 0.8272]       4.3        4.3  SIT
## Honkala 2017 (T2D)      0.2716 [-0.7206; 1.2638]       2.4        2.4  SIT
## Keating 2014           -0.1327 [-0.9694; 0.7039]       3.3        3.3 HIIT
## Lunt 2014               0.0000 [-0.8374; 0.8374]       3.3        3.3 HIIT
## Lunt 2014               0.2120 [-0.6277; 1.0516]       3.3        3.3  SIT
## Macpherson 2011         0.1534 [-0.7244; 1.0312]       3.0        3.0  SIT
## Maillard 2016          -0.0873 [-1.0678; 0.8931]       2.4        2.4 HIIT
## Matsuo 2014            -0.2568 [-1.0287; 0.5152]       3.9        3.9 HIIT
## Matsuo 2015             0.0783 [-0.7222; 0.8788]       3.6        3.6 HIIT
## Mitranun 2014          -0.0931 [-0.8343; 0.6481]       4.3        4.3 HIIT
## Moreira 2008           -0.0912 [-1.0717; 0.8893]       2.4        2.4 HIIT
## Motiani 2017            0.0919 [-0.6772; 0.8611]       3.9        3.9  SIT
## Nalcakan 2014           0.1726 [-0.8437; 1.1888]       2.3        2.3  SIT
## Nie 2017                0.1287 [-0.5893; 0.8467]       4.5        4.5 HIIT
## Ramos 2016a             0.0230 [-0.5749; 0.6210]       6.5        6.5 HIIT
## Ramos 2016b             0.1307 [-0.5643; 0.8258]       4.8        4.8 HIIT
## Sandvei 2012           -0.0990 [-0.9176; 0.7197]       3.5        3.5  SIT
## Sawyer 2016             0.1401 [-0.7849; 1.0652]       2.7        2.7 HIIT
## Sjöros 2018             0.0381 [-0.8184; 0.8945]       3.2        3.2  SIT
## Skleryk 2013            0.4302 [-0.5610; 1.4215]       2.4        2.4  SIT
## Trapp 2008              0.5592 [-0.1703; 1.2888]       4.4        4.4  SIT
## Winn 2018              -0.3125 [-1.2984; 0.6734]       2.4        2.4 HIIT
## 
## Number of studies combined: k = 29
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0395 [-0.1133; 0.1923] 0.51  0.6121
## Random effects model 0.0395 [-0.1133; 0.1923] 0.51  0.6121
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [<0.0000; <0.0000]; tau = 0 [<0.0000; <0.0000];
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 0.0%]; H = 1.00 [1.00; 1.00]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.97   28  1.0000
## 
## Results for subgroups (fixed effect model):
##        k     SMD            95%-CI    Q  I^2
## HIIT  16 -0.0091 [-0.2096; 0.1914] 2.43 0.0%
## SIT   13  0.1042 [-0.1319; 0.3404] 3.51 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.51    1  0.4733
## Within groups  5.94   27  1.0000
## 
## Results for subgroups (random effects model):
##        k     SMD            95%-CI tau^2 tau
## HIIT  16 -0.0091 [-0.2096; 0.1914]     0   0
## SIT   13  0.1042 [-0.1319; 0.3404]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.51    1  0.4733
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
5.2.8.3 Meta-regression
## 
## Mixed-Effects Model (k = 29; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0482)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 27) = 6.4006, p-val = 1.0000
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.5663, p-val = 0.4517
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.0103  0.1023  -0.1006  0.9199  -0.2108  0.1902    
## HIIESIT    0.1189  0.1580   0.7525  0.4517  -0.1908  0.4286    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
5.2.8.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

6. Systolic Blood Pressure

6.1 Overall

6.1.1 Forest plot

6.1.2 R output

##                            SMD             95%-CI %W(fixed) %W(random)
## Beetham 2019           -0.7906 [-1.9223;  0.3412]       1.6        3.0
## Ciolac 2010            -0.2950 [-0.9917;  0.4017]       4.2        4.4
## Cocks 2013              0.1605 [-0.8210;  1.1421]       2.1        3.4
## Conraads 2015          -0.7030 [-1.0092; -0.3967]      21.7        5.7
## Currie 2015            -0.5007 [-1.4152;  0.4138]       2.4        3.6
## Eguchi 2012            -0.0434 [-0.9201;  0.8332]       2.6        3.8
## Fisher 2015            -0.3829 [-1.2147;  0.4489]       2.9        3.9
## Honkala 2017 (Healthy)  0.5052 [-0.2473;  1.2577]       3.6        4.2
## Honkala 2017 (T2D)     -1.9928 [-3.1979; -0.7876]       1.4        2.8
## Jo 2020                -0.1714 [-0.8449;  0.5021]       4.5        4.5
## Keating 2014            0.1702 [-0.6671;  1.0074]       2.9        3.9
## Keteyian 2014           0.9259 [ 0.1446;  1.7072]       3.3        4.1
## Klonizakis 2014         0.6056 [-0.3625;  1.5736]       2.2        3.5
## Lunt 2014              -0.2297 [-1.0697;  0.6103]       2.9        3.9
## Lunt 2014               0.1292 [-0.7090;  0.9674]       2.9        3.9
## Matsuo 2014             0.1719 [-0.5983;  0.9421]       3.4        4.1
## Matsuo 2015             0.1389 [-0.6622;  0.9401]       3.2        4.0
## Mitranun 2014           0.6948 [-0.0680;  1.4576]       3.5        4.2
## Molmen-Hansen 2011      1.1876 [ 0.6324;  1.7428]       6.6        4.9
## Ramos 2016a            -0.7593 [-1.3784; -0.1402]       5.3        4.7
## Ramos 2016b            -1.0655 [-1.8073; -0.3237]       3.7        4.2
## Rognmo 2004             0.1380 [-0.8155;  1.0915]       2.2        3.5
## Skleryk 2013            0.1021 [-0.8786;  1.0827]       2.1        3.4
## Tjønna 2008            -0.1108 [-1.0222;  0.8006]       2.4        3.6
## Wegmann 2018            0.3909 [-0.1755;  0.9572]       6.3        4.9
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1092 [-0.2518; 0.0333] -1.50  0.1331
## Random effects model -0.0441 [-0.3174; 0.2291] -0.32  0.7515
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3170 [0.1073; 0.7008]; tau = 0.5630 [0.3275; 0.8372];
##  I^2 = 69.7% [54.5%; 79.9%]; H = 1.82 [1.48; 2.23]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  79.32   24 < 0.0001
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

6.1.3 Sensitivity analysis

6.1.3.1 Forest plot

6.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                                       SMD            95%-CI p-value   tau^2     tau    I^2
## Omitting Beetham 2019             -0.0180 [-0.2896; 0.2537]  0.8969  0.2967  0.5447  68.8%
## Omitting Ciolac 2010              -0.0282 [-0.3063; 0.2500]  0.8426  0.3114  0.5580  69.2%
## Omitting Cocks 2013               -0.0461 [-0.3207; 0.2284]  0.7419  0.3041  0.5514  69.2%
## Omitting Conraads 2015             0.0061 [-0.2519; 0.2641]  0.9632  0.2387  0.4886  59.3%
## Omitting Currie 2015              -0.0228 [-0.2974; 0.2517]  0.8705  0.3030  0.5504  69.0%
## Omitting Eguchi 2012              -0.0394 [-0.3154; 0.2367]  0.7797  0.3074  0.5545  69.3%
## Omitting Fisher 2015              -0.0260 [-0.3019; 0.2499]  0.8534  0.3063  0.5534  69.1%
## Omitting Honkala 2017 (Healthy)   -0.0623 [-0.3356; 0.2109]  0.6547  0.2959  0.5439  68.2%
## Omitting Honkala 2017 (T2D)        0.0109 [-0.2479; 0.2698]  0.9340  0.2563  0.5063  65.6%
## Omitting Jo 2020                  -0.0337 [-0.3127; 0.2453]  0.8131  0.3139  0.5603  69.3%
## Omitting Keating 2014             -0.0477 [-0.3235; 0.2282]  0.7348  0.3060  0.5532  69.1%
## Omitting Keteyian 2014            -0.0783 [-0.3437; 0.1871]  0.5631  0.2709  0.5205  66.4%
## Omitting Klonizakis 2014          -0.0610 [-0.3327; 0.2107]  0.6598  0.2944  0.5426  68.5%
## Omitting Lunt 2014                -0.0321 [-0.3085; 0.2443]  0.8198  0.3080  0.5549  69.3%
## Omitting Lunt 2014                -0.0461 [-0.3221; 0.2299]  0.7434  0.3067  0.5538  69.2%
## Omitting Matsuo 2014              -0.0483 [-0.3249; 0.2282]  0.7319  0.3073  0.5544  69.1%
## Omitting Matsuo 2015              -0.0467 [-0.3231; 0.2296]  0.7403  0.3073  0.5543  69.1%
## Omitting Mitranun 2014            -0.0698 [-0.3398; 0.2003]  0.6125  0.2856  0.5345  67.5%
## Omitting Molmen-Hansen 2011       -0.1035 [-0.3427; 0.1358]  0.3967  0.1871  0.4326  56.8%
## Omitting Ramos 2016a              -0.0041 [-0.2765; 0.2683]  0.9764  0.2907  0.5392  67.4%
## Omitting Ramos 2016b               0.0057 [-0.2609; 0.2722]  0.9668  0.2738  0.5233  66.5%
## Omitting Rognmo 2004              -0.0456 [-0.3204; 0.2293]  0.7452  0.3047  0.5520  69.2%
## Omitting Skleryk 2013             -0.0442 [-0.3190; 0.2305]  0.7525  0.3047  0.5520  69.2%
## Omitting Tjønna 2008              -0.0369 [-0.3127; 0.2388]  0.7928  0.3069  0.5540  69.3%
## Omitting Wegmann 2018             -0.0611 [-0.3371; 0.2148]  0.6642  0.3016  0.5492  68.0%
##                                                                                           
## Pooled estimate                   -0.0441 [-0.3174; 0.2291]  0.7515  0.3170  0.5630  69.7%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

6.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

6.2 Subgroups

6.2.1 Overall

6.2.1.1 Forest plot

6.2.1.2 R output
##                            SMD            95%-CI     meta-analysis
##                        -0.0441 [-0.3174; 0.2291]           Overall
## Healthy                 0.2051 [-0.0834; 0.4936]        Population
## Overweight/obese        0.0992 [-0.4309; 0.6293]        Population
## Cardiac Rehabilitation -0.0747 [-0.8639; 0.7146]        Population
## Metabolic Syndrome     -0.4178 [-0.8400; 0.0044]        Population
## T2D                    -0.5561 [-3.0617; 1.9495]        Population
## < 30 y                 -0.1103 [-0.5111; 0.2906]               Age
## 30 - 50 y               0.0355 [-0.3531; 0.4241]               Age
## > 50 y                 -0.0435 [-0.4766; 0.3897]               Age
## < 5 weeks              -0.1006 [-1.0594; 0.8581] Training Duration
## 5 - 10 weeks            0.2030 [-0.0924; 0.4985] Training Duration
## > 10 weeks             -0.1791 [-0.5602; 0.2019] Training Duration
## < 0.5                   0.1745 [-0.0986; 0.4476]         Men Ratio
## > 0.5                  -0.1396 [-0.4987; 0.2195]         Men Ratio
## Running                 0.0300 [-0.3581; 0.4180]  Type of Exercise
## Cycling                -0.1261 [-0.4753; 0.2231]  Type of Exercise
## < 120 mmHg              0.0128 [-0.3886; 0.4142]   Baseline Values
## 120 - 140 mmHg         -0.1525 [-0.4535; 0.1484]   Baseline Values
## > 140 mmHg              0.5533 [-0.2326; 1.3392]   Baseline Values
## HIIT                   -0.0103 [-0.3188; 0.2982]      Type of HIIE
## SIT                    -0.1406 [-0.7051; 0.4240]      Type of HIIE
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Random effects model -0.0441 [-0.3174; 0.2291] -0.32  0.7515
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3170; tau = 0.5630; I^2 = 69.7% [54.5%; 79.9%]; H = 1.82 [1.48; 2.23]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  79.32   24 < 0.0001
## 
## Results for meta-analyses (random effects model):
##                     k     SMD            95%-CI  tau^2    tau     Q   I^2
## Overall            25 -0.0441 [-0.3174; 0.2291] 0.3170 0.5630 79.32 69.7%
## Population         25 -0.0441 [-0.3174; 0.2291] 0.3170 0.5630 79.32 69.7%
## Age                25 -0.0441 [-0.3174; 0.2291] 0.3170 0.5630 79.32 69.7%
## Training Duration  25 -0.0441 [-0.3174; 0.2291] 0.3170 0.5630 79.32 69.7%
## Men Ratio          25 -0.0441 [-0.3174; 0.2291] 0.3170 0.5630 79.32 69.7%
## Type of Exercise   25 -0.0441 [-0.3174; 0.2291] 0.3170 0.5630 79.32 69.7%
## Baseline Values    25 -0.0441 [-0.3174; 0.2291] 0.3170 0.5630 79.32 69.7%
## Type of HIIE       25 -0.0441 [-0.3174; 0.2291] 0.3170 0.5630 79.32 69.7%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

6.2.2 Population

6.2.2.1 Forest plot

6.2.2.2 R output
##                            SMD             95%-CI %W(fixed) %W(random)             population
## Beetham 2019           -0.7906 [-1.9223;  0.3412]       1.6        3.0       Overweight/obese
## Ciolac 2010            -0.2950 [-0.9917;  0.4017]       4.2        4.4                Healthy
## Cocks 2013              0.1605 [-0.8210;  1.1421]       2.1        3.4                Healthy
## Conraads 2015          -0.7030 [-1.0092; -0.3967]      21.7        5.7 Cardiac Rehabilitation
## Currie 2015            -0.5007 [-1.4152;  0.4138]       2.4        3.6 Cardiac Rehabilitation
## Eguchi 2012            -0.0434 [-0.9201;  0.8332]       2.6        3.8                Healthy
## Fisher 2015            -0.3829 [-1.2147;  0.4489]       2.9        3.9       Overweight/obese
## Honkala 2017 (Healthy)  0.5052 [-0.2473;  1.2577]       3.6        4.2                Healthy
## Honkala 2017 (T2D)     -1.9928 [-3.1979; -0.7876]       1.4        2.8                    T2D
## Jo 2020                -0.1714 [-0.8449;  0.5021]       4.5        4.5     Metabolic Syndrome
## Keating 2014            0.1702 [-0.6671;  1.0074]       2.9        3.9       Overweight/obese
## Keteyian 2014           0.9259 [ 0.1446;  1.7072]       3.3        4.1 Cardiac Rehabilitation
## Klonizakis 2014         0.6056 [-0.3625;  1.5736]       2.2        3.5                Healthy
## Lunt 2014              -0.2297 [-1.0697;  0.6103]       2.9        3.9       Overweight/obese
## Lunt 2014               0.1292 [-0.7090;  0.9674]       2.9        3.9       Overweight/obese
## Matsuo 2014             0.1719 [-0.5983;  0.9421]       3.4        4.1                Healthy
## Matsuo 2015             0.1389 [-0.6622;  0.9401]       3.2        4.0     Metabolic Syndrome
## Mitranun 2014           0.6948 [-0.0680;  1.4576]       3.5        4.2                    T2D
## Molmen-Hansen 2011      1.1876 [ 0.6324;  1.7428]       6.6        4.9       Overweight/obese
## Ramos 2016a            -0.7593 [-1.3784; -0.1402]       5.3        4.7     Metabolic Syndrome
## Ramos 2016b            -1.0655 [-1.8073; -0.3237]       3.7        4.2     Metabolic Syndrome
## Rognmo 2004             0.1380 [-0.8155;  1.0915]       2.2        3.5 Cardiac Rehabilitation
## Skleryk 2013            0.1021 [-0.8786;  1.0827]       2.1        3.4       Overweight/obese
## Tjønna 2008            -0.1108 [-1.0222;  0.8006]       2.4        3.6     Metabolic Syndrome
## Wegmann 2018            0.3909 [-0.1755;  0.9572]       6.3        4.9                Healthy
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1092 [-0.2518; 0.0333] -1.50  0.1331
## Random effects model -0.0441 [-0.3174; 0.2291] -0.32  0.7515
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3170 [0.1073; 0.7008]; tau = 0.5630 [0.3275; 0.8372];
##  I^2 = 69.7% [54.5%; 79.9%]; H = 1.82 [1.48; 2.23]
## 
## Quantifying residual heterogeneity:
##  I^2 = 63.2% [41.2%; 77.0%]; H = 1.65 [1.30; 2.08]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  79.32   24 < 0.0001
## 
## Results for subgroups (fixed effect model):
##                          k     SMD             95%-CI     Q   I^2
## Healthy                  7  0.2051 [-0.0834;  0.4936]  3.74  0.0%
## Overweight/obese         7  0.2675 [-0.0376;  0.5727] 16.96 64.6%
## Cardiac Rehabilitation   4 -0.4409 [-0.7029; -0.1788] 15.32 80.4%
## Metabolic Syndrome       5 -0.4371 [-0.7639; -0.1103]  6.52 38.7%
## T2D                      2 -0.0296 [-0.6808;  0.6216] 11.83 91.5%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f.  p-value
## Between groups 20.49    4   0.0004
## Within groups  54.38   20 < 0.0001
## 
## Results for subgroups (random effects model):
##                          k     SMD            95%-CI  tau^2    tau
## Healthy                  7  0.2051 [-0.0834; 0.4936]      0      0
## Overweight/obese         7  0.0992 [-0.4309; 0.6293] 0.3222 0.5676
## Cardiac Rehabilitation   4 -0.0747 [-0.8639; 0.7146] 0.5003 0.7073
## Metabolic Syndrome       5 -0.4178 [-0.8400; 0.0044] 0.0893 0.2988
## T2D                      2 -0.5561 [-3.0617; 1.9495] 2.9966 1.7311
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   6.05    4  0.1954
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
6.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 25; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.2959 (SE = 0.1480)
## tau (square root of estimated tau^2 value):             0.5439
## I^2 (residual heterogeneity / unaccounted variability): 65.73%
## H^2 (unaccounted variability / sampling variability):   2.92
## R^2 (amount of heterogeneity accounted for):            6.66%
## 
## Test for Residual Heterogeneity:
## QE(df = 20) = 58.3624, p-val < .0001
## 
## Test of Moderators (coefficients 2:5):
## QM(df = 4) = 3.2489, p-val = 0.5171
## 
## Model Results:
## 
##                               estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                         0.2096  0.2567   0.8164  0.4143  -0.2935  0.7126    
## .byvarOverweight/obese         -0.1105  0.3676  -0.3007  0.7637  -0.8309  0.6099    
## .byvarCardiac Rehabilitation   -0.3052  0.4186  -0.7290  0.4660  -1.1256  0.5152    
## .byvarMetabolic Syndrome       -0.6221  0.3926  -1.5848  0.1130  -1.3915  0.1473    
## .byvarT2D                      -0.5870  0.5786  -1.0145  0.3103  -1.7209  0.5470    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
6.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

6.2.3 Age

6.2.3.1 Forest plot

6.2.3.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_age
## Beetham 2019           -0.7906 [-1.9223;  0.3412]       1.6        3.0       > 50 y
## Ciolac 2010            -0.2950 [-0.9917;  0.4017]       4.2        4.4       < 30 y
## Cocks 2013              0.1605 [-0.8210;  1.1421]       2.1        3.4       < 30 y
## Conraads 2015          -0.7030 [-1.0092; -0.3967]      21.7        5.7       > 50 y
## Currie 2015            -0.5007 [-1.4152;  0.4138]       2.4        3.6       > 50 y
## Eguchi 2012            -0.0434 [-0.9201;  0.8332]       2.6        3.8       > 50 y
## Fisher 2015            -0.3829 [-1.2147;  0.4489]       2.9        3.9       < 30 y
## Honkala 2017 (Healthy)  0.5052 [-0.2473;  1.2577]       3.6        4.2    30 - 50 y
## Honkala 2017 (T2D)     -1.9928 [-3.1979; -0.7876]       1.4        2.8    30 - 50 y
## Jo 2020                -0.1714 [-0.8449;  0.5021]       4.5        4.5       > 50 y
## Keating 2014            0.1702 [-0.6671;  1.0074]       2.9        3.9    30 - 50 y
## Keteyian 2014           0.9259 [ 0.1446;  1.7072]       3.3        4.1       > 50 y
## Klonizakis 2014         0.6056 [-0.3625;  1.5736]       2.2        3.5       > 50 y
## Lunt 2014              -0.2297 [-1.0697;  0.6103]       2.9        3.9    30 - 50 y
## Lunt 2014               0.1292 [-0.7090;  0.9674]       2.9        3.9    30 - 50 y
## Matsuo 2014             0.1719 [-0.5983;  0.9421]       3.4        4.1       < 30 y
## Matsuo 2015             0.1389 [-0.6622;  0.9401]       3.2        4.0    30 - 50 y
## Mitranun 2014           0.6948 [-0.0680;  1.4576]       3.5        4.2       > 50 y
## Molmen-Hansen 2011      1.1876 [ 0.6324;  1.7428]       6.6        4.9       > 50 y
## Ramos 2016a            -0.7593 [-1.3784; -0.1402]       5.3        4.7       > 50 y
## Ramos 2016b            -1.0655 [-1.8073; -0.3237]       3.7        4.2       > 50 y
## Rognmo 2004             0.1380 [-0.8155;  1.0915]       2.2        3.5       > 50 y
## Skleryk 2013            0.1021 [-0.8786;  1.0827]       2.1        3.4    30 - 50 y
## Tjønna 2008            -0.1108 [-1.0222;  0.8006]       2.4        3.6       > 50 y
## Wegmann 2018            0.3909 [-0.1755;  0.9572]       6.3        4.9    30 - 50 y
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1092 [-0.2518; 0.0333] -1.50  0.1331
## Random effects model -0.0441 [-0.3174; 0.2291] -0.32  0.7515
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3170 [0.1073; 0.7008]; tau = 0.5630 [0.3275; 0.8372];
##  I^2 = 69.7% [54.5%; 79.9%]; H = 1.82 [1.48; 2.23]
## 
## Quantifying residual heterogeneity:
##  I^2 = 69.4% [53.2%; 80.1%]; H = 1.81 [1.46; 2.24]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  79.32   24 < 0.0001
## 
## Results for subgroups (fixed effect model):
##             k     SMD             95%-CI     Q   I^2
## < 30 y      4 -0.1103 [-0.5111;  0.2906]  1.39  0.0%
## 30 - 50 y   8  0.1005 [-0.1836;  0.3847] 12.41 43.6%
## > 50 y     13 -0.1907 [-0.3720; -0.0094] 58.21 79.4%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f.  p-value
## Between groups  2.87    2   0.2383
## Within groups  72.01   22 < 0.0001
## 
## Results for subgroups (random effects model):
##             k     SMD            95%-CI  tau^2    tau
## < 30 y      4 -0.1103 [-0.5111; 0.2906]      0      0
## 30 - 50 y   8  0.0355 [-0.3531; 0.4241] 0.1335 0.3654
## > 50 y     13 -0.0435 [-0.4766; 0.3897] 0.4727 0.6876
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.26    2  0.8769
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
6.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 25; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.3372 (SE = 0.1593)
## tau (square root of estimated tau^2 value):             0.5807
## I^2 (residual heterogeneity / unaccounted variability): 70.74%
## H^2 (unaccounted variability / sampling variability):   3.42
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 23) = 78.6179, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0130, p-val = 0.9092
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.1085  0.5773  -0.1880  0.8509  -1.2400  1.0229    
## age        0.0013  0.0115   0.1141  0.9092  -0.0211  0.0237    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
6.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

6.2.4 Training Duration

6.2.4.1 Forest plot

6.2.4.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_duration
## Beetham 2019           -0.7906 [-1.9223;  0.3412]       1.6        3.0        > 10 weeks
## Ciolac 2010            -0.2950 [-0.9917;  0.4017]       4.2        4.4        > 10 weeks
## Cocks 2013              0.1605 [-0.8210;  1.1421]       2.1        3.4      5 - 10 weeks
## Conraads 2015          -0.7030 [-1.0092; -0.3967]      21.7        5.7        > 10 weeks
## Currie 2015            -0.5007 [-1.4152;  0.4138]       2.4        3.6        > 10 weeks
## Eguchi 2012            -0.0434 [-0.9201;  0.8332]       2.6        3.8        > 10 weeks
## Fisher 2015            -0.3829 [-1.2147;  0.4489]       2.9        3.9      5 - 10 weeks
## Honkala 2017 (Healthy)  0.5052 [-0.2473;  1.2577]       3.6        4.2         < 5 weeks
## Honkala 2017 (T2D)     -1.9928 [-3.1979; -0.7876]       1.4        2.8         < 5 weeks
## Jo 2020                -0.1714 [-0.8449;  0.5021]       4.5        4.5      5 - 10 weeks
## Keating 2014            0.1702 [-0.6671;  1.0074]       2.9        3.9        > 10 weeks
## Keteyian 2014           0.9259 [ 0.1446;  1.7072]       3.3        4.1      5 - 10 weeks
## Klonizakis 2014         0.6056 [-0.3625;  1.5736]       2.2        3.5         < 5 weeks
## Lunt 2014              -0.2297 [-1.0697;  0.6103]       2.9        3.9        > 10 weeks
## Lunt 2014               0.1292 [-0.7090;  0.9674]       2.9        3.9        > 10 weeks
## Matsuo 2014             0.1719 [-0.5983;  0.9421]       3.4        4.1      5 - 10 weeks
## Matsuo 2015             0.1389 [-0.6622;  0.9401]       3.2        4.0      5 - 10 weeks
## Mitranun 2014           0.6948 [-0.0680;  1.4576]       3.5        4.2      5 - 10 weeks
## Molmen-Hansen 2011      1.1876 [ 0.6324;  1.7428]       6.6        4.9        > 10 weeks
## Ramos 2016a            -0.7593 [-1.3784; -0.1402]       5.3        4.7        > 10 weeks
## Ramos 2016b            -1.0655 [-1.8073; -0.3237]       3.7        4.2        > 10 weeks
## Rognmo 2004             0.1380 [-0.8155;  1.0915]       2.2        3.5      5 - 10 weeks
## Skleryk 2013            0.1021 [-0.8786;  1.0827]       2.1        3.4         < 5 weeks
## Tjønna 2008            -0.1108 [-1.0222;  0.8006]       2.4        3.6        > 10 weeks
## Wegmann 2018            0.3909 [-0.1755;  0.9572]       6.3        4.9        > 10 weeks
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1092 [-0.2518; 0.0333] -1.50  0.1331
## Random effects model -0.0441 [-0.3174; 0.2291] -0.32  0.7515
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3170 [0.1073; 0.7008]; tau = 0.5630 [0.3275; 0.8372];
##  I^2 = 69.7% [54.5%; 79.9%]; H = 1.82 [1.48; 2.23]
## 
## Quantifying residual heterogeneity:
##  I^2 = 67.2% [49.4%; 78.8%]; H = 1.75 [1.41; 2.17]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  79.32   24 < 0.0001
## 
## Results for subgroups (fixed effect model):
##                k     SMD             95%-CI     Q   I^2
## < 5 weeks      4  0.0782 [-0.3928;  0.5491] 11.76 74.5%
## 5 - 10 weeks   8  0.2024 [-0.0820;  0.4869]  7.52  7.0%
## > 10 weeks    13 -0.2517 [-0.4280; -0.0754] 47.86 74.9%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f.  p-value
## Between groups  7.73    2   0.0210
## Within groups  67.15   22 < 0.0001
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI  tau^2    tau
## < 5 weeks      4 -0.1006 [-1.0594; 0.8581] 0.7041 0.8391
## 5 - 10 weeks   8  0.2030 [-0.0924; 0.4985] 0.0127 0.1127
## > 10 weeks    13 -0.1791 [-0.5602; 0.2019] 0.3428 0.5855
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   2.51    2  0.2845
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
6.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 25; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.3355 (SE = 0.1623)
## tau (square root of estimated tau^2 value):             0.5792
## I^2 (residual heterogeneity / unaccounted variability): 70.80%
## H^2 (unaccounted variability / sampling variability):   3.42
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 23) = 78.7742, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0899, p-val = 0.7643
## 
## Model Results:
## 
##           estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt     0.0429  0.3249   0.1321  0.8949  -0.5939  0.6797    
## duration   -0.0082  0.0274  -0.2999  0.7643  -0.0618  0.0454    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
6.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

6.2.5 Men Ratio

6.2.5.1 Forest plot

6.2.5.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_men_ratio
## Beetham 2019           -0.7906 [-1.9223;  0.3412]       1.6        3.0              > 0.5
## Ciolac 2010            -0.2950 [-0.9917;  0.4017]       4.2        4.4              < 0.5
## Cocks 2013              0.1605 [-0.8210;  1.1421]       2.1        3.4              > 0.5
## Conraads 2015          -0.7030 [-1.0092; -0.3967]      21.7        5.7              > 0.5
## Currie 2015            -0.5007 [-1.4152;  0.4138]       2.4        3.6              > 0.5
## Eguchi 2012            -0.0434 [-0.9201;  0.8332]       2.6        3.8              > 0.5
## Fisher 2015            -0.3829 [-1.2147;  0.4489]       2.9        3.9              > 0.5
## Honkala 2017 (Healthy)  0.5052 [-0.2473;  1.2577]       3.6        4.2              > 0.5
## Honkala 2017 (T2D)     -1.9928 [-3.1979; -0.7876]       1.4        2.8              > 0.5
## Jo 2020                -0.1714 [-0.8449;  0.5021]       4.5        4.5              > 0.5
## Keating 2014            0.1702 [-0.6671;  1.0074]       2.9        3.9              < 0.5
## Keteyian 2014           0.9259 [ 0.1446;  1.7072]       3.3        4.1              > 0.5
## Klonizakis 2014         0.6056 [-0.3625;  1.5736]       2.2        3.5              < 0.5
## Lunt 2014              -0.2297 [-1.0697;  0.6103]       2.9        3.9              < 0.5
## Lunt 2014               0.1292 [-0.7090;  0.9674]       2.9        3.9              < 0.5
## Matsuo 2014             0.1719 [-0.5983;  0.9421]       3.4        4.1              > 0.5
## Matsuo 2015             0.1389 [-0.6622;  0.9401]       3.2        4.0              > 0.5
## Mitranun 2014           0.6948 [-0.0680;  1.4576]       3.5        4.2              < 0.5
## Molmen-Hansen 2011      1.1876 [ 0.6324;  1.7428]       6.6        4.9              > 0.5
## Ramos 2016a            -0.7593 [-1.3784; -0.1402]       5.3        4.7              > 0.5
## Ramos 2016b            -1.0655 [-1.8073; -0.3237]       3.7        4.2              > 0.5
## Rognmo 2004             0.1380 [-0.8155;  1.0915]       2.2        3.5              > 0.5
## Skleryk 2013            0.1021 [-0.8786;  1.0827]       2.1        3.4              > 0.5
## Tjønna 2008            -0.1108 [-1.0222;  0.8006]       2.4        3.6              < 0.5
## Wegmann 2018            0.3909 [-0.1755;  0.9572]       6.3        4.9              < 0.5
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1092 [-0.2518; 0.0333] -1.50  0.1331
## Random effects model -0.0441 [-0.3174; 0.2291] -0.32  0.7515
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3170 [0.1073; 0.7008]; tau = 0.5630 [0.3275; 0.8372];
##  I^2 = 69.7% [54.5%; 79.9%]; H = 1.82 [1.48; 2.23]
## 
## Quantifying residual heterogeneity:
##  I^2 = 66.8% [49.1%; 78.3%]; H = 1.74 [1.40; 2.15]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  79.32   24 < 0.0001
## 
## Results for subgroups (fixed effect model):
##         k     SMD             95%-CI     Q   I^2
## < 0.5   8  0.1745 [-0.0986;  0.4476]  5.75  0.0%
## > 0.5  17 -0.2129 [-0.3804; -0.0453] 63.51 74.8%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f.  p-value
## Between groups  5.62    1   0.0178
## Within groups  69.26   23 < 0.0001
## 
## Results for subgroups (random effects model):
##         k     SMD            95%-CI  tau^2    tau
## < 0.5   8  0.1745 [-0.0986; 0.4476]      0      0
## > 0.5  17 -0.1396 [-0.4987; 0.2195] 0.3969 0.6300
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.86    1  0.1724
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
6.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 25; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.2990 (SE = 0.1450)
## tau (square root of estimated tau^2 value):             0.5468
## I^2 (residual heterogeneity / unaccounted variability): 68.08%
## H^2 (unaccounted variability / sampling variability):   3.13
## R^2 (amount of heterogeneity accounted for):            5.67%
## 
## Test for Residual Heterogeneity:
## QE(df = 23) = 72.0471, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.0888, p-val = 0.2967
## 
## Model Results:
## 
##            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt      0.2588  0.3205   0.8076  0.4193  -0.3694  0.8870    
## men_ratio   -0.4537  0.4349  -1.0434  0.2967  -1.3060  0.3986    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
6.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

6.2.6 Type of Exercise

6.2.6.1 Forest plot

6.2.6.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) type_exercise
## Beetham 2019           -0.7906 [-1.9223;  0.3412]       1.6        3.0       Running
## Ciolac 2010            -0.2950 [-0.9917;  0.4017]       4.2        4.4       Running
## Cocks 2013              0.1605 [-0.8210;  1.1421]       2.1        3.4       Cycling
## Conraads 2015          -0.7030 [-1.0092; -0.3967]      21.7        5.7       Cycling
## Currie 2015            -0.5007 [-1.4152;  0.4138]       2.4        3.6       Cycling
## Eguchi 2012            -0.0434 [-0.9201;  0.8332]       2.6        3.8       Cycling
## Fisher 2015            -0.3829 [-1.2147;  0.4489]       2.9        3.9       Cycling
## Honkala 2017 (Healthy)  0.5052 [-0.2473;  1.2577]       3.6        4.2       Cycling
## Honkala 2017 (T2D)     -1.9928 [-3.1979; -0.7876]       1.4        2.8       Cycling
## Jo 2020                -0.1714 [-0.8449;  0.5021]       4.5        4.5       Running
## Keating 2014            0.1702 [-0.6671;  1.0074]       2.9        3.9       Cycling
## Keteyian 2014           0.9259 [ 0.1446;  1.7072]       3.3        4.1       Running
## Klonizakis 2014         0.6056 [-0.3625;  1.5736]       2.2        3.5       Cycling
## Lunt 2014              -0.2297 [-1.0697;  0.6103]       2.9        3.9       Running
## Lunt 2014               0.1292 [-0.7090;  0.9674]       2.9        3.9       Running
## Matsuo 2014             0.1719 [-0.5983;  0.9421]       3.4        4.1       Cycling
## Matsuo 2015             0.1389 [-0.6622;  0.9401]       3.2        4.0       Cycling
## Mitranun 2014           0.6948 [-0.0680;  1.4576]       3.5        4.2       Running
## Molmen-Hansen 2011      1.1876 [ 0.6324;  1.7428]       6.6        4.9       Running
## Ramos 2016a            -0.7593 [-1.3784; -0.1402]       5.3        4.7       Running
## Ramos 2016b            -1.0655 [-1.8073; -0.3237]       3.7        4.2       Running
## Rognmo 2004             0.1380 [-0.8155;  1.0915]       2.2        3.5       Running
## Skleryk 2013            0.1021 [-0.8786;  1.0827]       2.1        3.4       Cycling
## Tjønna 2008            -0.1108 [-1.0222;  0.8006]       2.4        3.6       Running
## Wegmann 2018            0.3909 [-0.1755;  0.9572]       6.3        4.9       Running
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1092 [-0.2518; 0.0333] -1.50  0.1331
## Random effects model -0.0441 [-0.3174; 0.2291] -0.32  0.7515
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3170 [0.1073; 0.7008]; tau = 0.5630 [0.3275; 0.8372];
##  I^2 = 69.7% [54.5%; 79.9%]; H = 1.82 [1.48; 2.23]
## 
## Quantifying residual heterogeneity:
##  I^2 = 66.1% [48.0%; 77.9%]; H = 1.72 [1.39; 2.13]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  79.32   24 < 0.0001
## 
## Results for subgroups (fixed effect model):
##           k     SMD             95%-CI     Q   I^2
## Running  13  0.0880 [-0.1151;  0.2911] 42.07 71.5%
## Cycling  12 -0.2975 [-0.4983; -0.0967] 25.81 57.4%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f.  p-value
## Between groups  7.00    1   0.0082
## Within groups  67.88   23 < 0.0001
## 
## Results for subgroups (random effects model):
##           k     SMD            95%-CI  tau^2    tau
## Running  13  0.0300 [-0.3581; 0.4180] 0.3542 0.5952
## Cycling  12 -0.1261 [-0.4753; 0.2231] 0.1983 0.4453
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.34    1  0.5578
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
6.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 25; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.3067 (SE = 0.1456)
## tau (square root of estimated tau^2 value):             0.5538
## I^2 (residual heterogeneity / unaccounted variability): 68.12%
## H^2 (unaccounted variability / sampling variability):   3.14
## R^2 (amount of heterogeneity accounted for):            3.23%
## 
## Test for Residual Heterogeneity:
## QE(df = 23) = 72.1374, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.3359, p-val = 0.5622
## 
## Model Results:
## 
##                       estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                -0.1299  0.2027  -0.6410  0.5215  -0.5271  0.2673    
## type_exerciseRunning    0.1603  0.2766   0.5795  0.5622  -0.3818  0.7025    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
6.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

6.2.7 Baseline Values

6.2.7.1 Forest plot

6.2.7.2 R output
##                            SMD             95%-CI %W(fixed) %W(random)  category_bsln
## Beetham 2019           -0.7906 [-1.9223;  0.3412]       1.6        3.0 120 - 140 mmHg
## Ciolac 2010            -0.2950 [-0.9917;  0.4017]       4.2        4.4     < 120 mmHg
## Cocks 2013              0.1605 [-0.8210;  1.1421]       2.1        3.4     < 120 mmHg
## Conraads 2015          -0.7030 [-1.0092; -0.3967]      21.7        5.7 120 - 140 mmHg
## Currie 2015            -0.5007 [-1.4152;  0.4138]       2.4        3.6 120 - 140 mmHg
## Eguchi 2012            -0.0434 [-0.9201;  0.8332]       2.6        3.8 120 - 140 mmHg
## Fisher 2015            -0.3829 [-1.2147;  0.4489]       2.9        3.9 120 - 140 mmHg
## Honkala 2017 (Healthy)  0.5052 [-0.2473;  1.2577]       3.6        4.2 120 - 140 mmHg
## Honkala 2017 (T2D)     -1.9928 [-3.1979; -0.7876]       1.4        2.8 120 - 140 mmHg
## Jo 2020                -0.1714 [-0.8449;  0.5021]       4.5        4.5 120 - 140 mmHg
## Keating 2014            0.1702 [-0.6671;  1.0074]       2.9        3.9     < 120 mmHg
## Keteyian 2014           0.9259 [ 0.1446;  1.7072]       3.3        4.1 120 - 140 mmHg
## Klonizakis 2014         0.6056 [-0.3625;  1.5736]       2.2        3.5 120 - 140 mmHg
## Lunt 2014              -0.2297 [-1.0697;  0.6103]       2.9        3.9 120 - 140 mmHg
## Lunt 2014               0.1292 [-0.7090;  0.9674]       2.9        3.9 120 - 140 mmHg
## Matsuo 2014             0.1719 [-0.5983;  0.9421]       3.4        4.1     < 120 mmHg
## Matsuo 2015             0.1389 [-0.6622;  0.9401]       3.2        4.0 120 - 140 mmHg
## Mitranun 2014           0.6948 [-0.0680;  1.4576]       3.5        4.2 120 - 140 mmHg
## Molmen-Hansen 2011      1.1876 [ 0.6324;  1.7428]       6.6        4.9     > 140 mmHg
## Ramos 2016a            -0.7593 [-1.3784; -0.1402]       5.3        4.7 120 - 140 mmHg
## Ramos 2016b            -1.0655 [-1.8073; -0.3237]       3.7        4.2 120 - 140 mmHg
## Rognmo 2004             0.1380 [-0.8155;  1.0915]       2.2        3.5     > 140 mmHg
## Skleryk 2013            0.1021 [-0.8786;  1.0827]       2.1        3.4     > 140 mmHg
## Tjønna 2008            -0.1108 [-1.0222;  0.8006]       2.4        3.6 120 - 140 mmHg
## Wegmann 2018            0.3909 [-0.1755;  0.9572]       6.3        4.9 120 - 140 mmHg
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1092 [-0.2518; 0.0333] -1.50  0.1331
## Random effects model -0.0441 [-0.3174; 0.2291] -0.32  0.7515
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3170 [0.1073; 0.7008]; tau = 0.5630 [0.3275; 0.8372];
##  I^2 = 69.7% [54.5%; 79.9%]; H = 1.82 [1.48; 2.23]
## 
## Quantifying residual heterogeneity:
##  I^2 = 61.1% [38.7%; 75.3%]; H = 1.60 [1.28; 2.01]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  79.32   24 < 0.0001
## 
## Results for subgroups (fixed effect model):
##                  k     SMD             95%-CI     Q   I^2
## < 120 mmHg       4  0.0128 [-0.3886;  0.4142]  1.07  0.0%
## 120 - 140 mmHg  18 -0.2494 [-0.4128; -0.0861] 49.89 65.9%
## > 140 mmHg       3  0.7500 [ 0.3182;  1.1818]  5.53 63.8%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f.  p-value
## Between groups 18.39    2   0.0001
## Within groups  56.48   22 < 0.0001
## 
## Results for subgroups (random effects model):
##                  k     SMD            95%-CI  tau^2    tau
## < 120 mmHg       4  0.0128 [-0.3886; 0.4142]      0      0
## 120 - 140 mmHg  18 -0.1525 [-0.4535; 0.1484] 0.2582 0.5082
## > 140 mmHg       3  0.5533 [-0.2326; 1.3392] 0.3064 0.5535
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   2.79    2  0.2475
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
6.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 25; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.3001 (SE = 0.1486)
## tau (square root of estimated tau^2 value):             0.5478
## I^2 (residual heterogeneity / unaccounted variability): 68.21%
## H^2 (unaccounted variability / sampling variability):   3.15
## R^2 (amount of heterogeneity accounted for):            5.33%
## 
## Test for Residual Heterogeneity:
## QE(df = 23) = 72.3477, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.1857, p-val = 0.6665
## 
## Model Results:
## 
##                estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt         -0.8264  1.8215  -0.4537  0.6500  -4.3965  2.7436    
## bsln_adjusted    0.0061  0.0140   0.4310  0.6665  -0.0215  0.0336    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
6.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

6.2.8 Type of HIIE

6.2.8.1 Forest plot

6.2.8.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) HIIE
## Beetham 2019           -0.7906 [-1.9223;  0.3412]       1.6        3.0 HIIT
## Ciolac 2010            -0.2950 [-0.9917;  0.4017]       4.2        4.4 HIIT
## Cocks 2013              0.1605 [-0.8210;  1.1421]       2.1        3.4  SIT
## Conraads 2015          -0.7030 [-1.0092; -0.3967]      21.7        5.7 HIIT
## Currie 2015            -0.5007 [-1.4152;  0.4138]       2.4        3.6 HIIT
## Eguchi 2012            -0.0434 [-0.9201;  0.8332]       2.6        3.8 HIIT
## Fisher 2015            -0.3829 [-1.2147;  0.4489]       2.9        3.9  SIT
## Honkala 2017 (Healthy)  0.5052 [-0.2473;  1.2577]       3.6        4.2  SIT
## Honkala 2017 (T2D)     -1.9928 [-3.1979; -0.7876]       1.4        2.8  SIT
## Jo 2020                -0.1714 [-0.8449;  0.5021]       4.5        4.5 HIIT
## Keating 2014            0.1702 [-0.6671;  1.0074]       2.9        3.9 HIIT
## Keteyian 2014           0.9259 [ 0.1446;  1.7072]       3.3        4.1 HIIT
## Klonizakis 2014         0.6056 [-0.3625;  1.5736]       2.2        3.5 HIIT
## Lunt 2014              -0.2297 [-1.0697;  0.6103]       2.9        3.9 HIIT
## Lunt 2014               0.1292 [-0.7090;  0.9674]       2.9        3.9  SIT
## Matsuo 2014             0.1719 [-0.5983;  0.9421]       3.4        4.1 HIIT
## Matsuo 2015             0.1389 [-0.6622;  0.9401]       3.2        4.0 HIIT
## Mitranun 2014           0.6948 [-0.0680;  1.4576]       3.5        4.2 HIIT
## Molmen-Hansen 2011      1.1876 [ 0.6324;  1.7428]       6.6        4.9 HIIT
## Ramos 2016a            -0.7593 [-1.3784; -0.1402]       5.3        4.7 HIIT
## Ramos 2016b            -1.0655 [-1.8073; -0.3237]       3.7        4.2 HIIT
## Rognmo 2004             0.1380 [-0.8155;  1.0915]       2.2        3.5 HIIT
## Skleryk 2013            0.1021 [-0.8786;  1.0827]       2.1        3.4  SIT
## Tjønna 2008            -0.1108 [-1.0222;  0.8006]       2.4        3.6 HIIT
## Wegmann 2018            0.3909 [-0.1755;  0.9572]       6.3        4.9 HIIT
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1092 [-0.2518; 0.0333] -1.50  0.1331
## Random effects model -0.0441 [-0.3174; 0.2291] -0.32  0.7515
## 
## Quantifying heterogeneity:
##  tau^2 = 0.3170 [0.1073; 0.7008]; tau = 0.5630 [0.3275; 0.8372];
##  I^2 = 69.7% [54.5%; 79.9%]; H = 1.82 [1.48; 2.23]
## 
## Quantifying residual heterogeneity:
##  I^2 = 69.3% [53.3%; 79.8%]; H = 1.80 [1.46; 2.22]
## 
## Test of heterogeneity:
##      Q d.f.  p-value
##  79.32   24 < 0.0001
## 
## Results for subgroups (fixed effect model):
##        k     SMD            95%-CI     Q   I^2
## HIIT  19 -0.1149 [-0.2697; 0.0400] 63.47 71.6%
## SIT    6 -0.0621 [-0.4309; 0.3067] 11.34 55.9%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f.  p-value
## Between groups  0.07    1   0.7960
## Within groups  74.81   23 < 0.0001
## 
## Results for subgroups (random effects model):
##        k     SMD            95%-CI  tau^2    tau
## HIIT  19 -0.0103 [-0.3188; 0.2982] 0.3157 0.5619
## SIT    6 -0.1406 [-0.7051; 0.4240] 0.2739 0.5234
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.16    1  0.6915
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
6.2.8.3 Meta-regression
## 
## Mixed-Effects Model (k = 25; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.3355 (SE = 0.1604)
## tau (square root of estimated tau^2 value):             0.5792
## I^2 (residual heterogeneity / unaccounted variability): 70.99%
## H^2 (unaccounted variability / sampling variability):   3.45
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 23) = 79.2913, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.2043, p-val = 0.6513
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.0108  0.1607  -0.0674  0.9462  -0.3257  0.3040    
## HIIESIT   -0.1558  0.3447  -0.4520  0.6513  -0.8314  0.5198    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
6.2.8.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

7. Diastolic Blood Pressure

7.1 Overall

7.1.1 Forest plot

7.1.2 R output

##                            SMD             95%-CI %W(fixed) %W(random)
## Beetham 2019           -1.2992 [-2.4937; -0.1048]       1.4        2.3
## Ciolac 2010             0.0721 [-0.6211;  0.7653]       4.1        4.5
## Cocks 2013             -0.2152 [-1.1980;  0.7676]       2.1        3.1
## Conraads 2015          -0.3593 [-0.6589; -0.0597]      22.1        7.4
## Currie 2015            -1.4264 [-2.4347; -0.4181]       2.0        3.0
## Eguchi 2012             0.0000 [-0.8765;  0.8765]       2.6        3.5
## Fisher 2015            -0.8645 [-1.7259; -0.0031]       2.7        3.6
## Honkala 2017 (Healthy)  0.3306 [-0.4152;  1.0764]       3.6        4.2
## Honkala 2017 (T2D)      0.3339 [-0.6606;  1.3283]       2.0        3.0
## Jo 2020                -0.1592 [-0.8325;  0.5141]       4.4        4.7
## Keating 2014           -0.1303 [-0.9669;  0.7063]       2.8        3.7
## Keteyian 2014           1.1469 [ 0.3457;  1.9480]       3.1        3.9
## Klonizakis 2014        -0.2983 [-1.2509;  0.6543]       2.2        3.2
## Lunt 2014               0.2262 [-0.6137;  1.0662]       2.8        3.7
## Lunt 2014               0.4890 [-0.3602;  1.3382]       2.8        3.7
## Matsuo 2014             0.3972 [-0.3791;  1.1735]       3.3        4.0
## Matsuo 2015             0.4857 [-0.3262;  1.2976]       3.0        3.9
## Mitranun 2014          -0.3780 [-1.1253;  0.3694]       3.6        4.2
## Molmen-Hansen 2011      0.8739 [ 0.3380;  1.4098]       6.9        5.6
## Ramos 2016a            -0.1587 [-0.7576;  0.4402]       5.5        5.2
## Ramos 2016b             0.2636 [-0.4337;  0.9609]       4.1        4.5
## Rognmo 2004            -0.1462 [-1.0998;  0.8075]       2.2        3.2
## Skleryk 2013            0.2806 [-0.7042;  1.2654]       2.1        3.1
## Tjønna 2008             0.0000 [-0.9107;  0.9107]       2.4        3.4
## Wegmann 2018            0.1864 [-0.3759;  0.7486]       6.3        5.4
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0024 [-0.1434; 0.1386] -0.03  0.9732
## Random effects model  0.0254 [-0.1884; 0.2391]  0.23  0.8161
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1369 [0.0236; 0.4283]; tau = 0.3700 [0.1536; 0.6544];
##  I^2 = 50.4% [21.3%; 68.7%]; H = 1.42 [1.13; 1.79]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  48.37   24  0.0023
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

7.1.3 Sensitivity analysis

7.1.3.1 Forest plot

7.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                                       SMD            95%-CI p-value   tau^2     tau    I^2
## Omitting Beetham 2019              0.0570 [-0.1457; 0.2598]  0.5814  0.1036  0.3218  44.0%
## Omitting Ciolac 2010               0.0262 [-0.1900; 0.2423]  0.8126  0.1297  0.3601  48.8%
## Omitting Cocks 2013                0.0356 [-0.1773; 0.2486]  0.7428  0.1259  0.3548  48.7%
## Omitting Conraads 2015             0.0642 [-0.1459; 0.2743]  0.5490  0.1052  0.3244  39.5%
## Omitting Currie 2015               0.0693 [-0.1267; 0.2653]  0.4886  0.0867  0.2945  39.6%
## Omitting Eguchi 2012               0.0293 [-0.1849; 0.2435]  0.7884  0.1277  0.3573  48.9%
## Omitting Fisher 2015               0.0610 [-0.1445; 0.2665]  0.5607  0.1066  0.3265  44.4%
## Omitting Honkala 2017 (Healthy)    0.0156 [-0.1982; 0.2294]  0.8865  0.1247  0.3532  48.0%
## Omitting Honkala 2017 (T2D)        0.0197 [-0.1926; 0.2320]  0.8558  0.1245  0.3528  48.4%
## Omitting Jo 2020                   0.0372 [-0.1788; 0.2533]  0.7355  0.1291  0.3593  48.6%
## Omitting Keating 2014              0.0342 [-0.1801; 0.2486]  0.7543  0.1275  0.3571  48.8%
## Omitting Keteyian 2014            -0.0121 [-0.2080; 0.1838]  0.9038  0.0841  0.2900  38.5%
## Omitting Klonizakis 2014           0.0386 [-0.1741; 0.2514]  0.7219  0.1251  0.3537  48.5%
## Omitting Lunt 2014                 0.0211 [-0.1928; 0.2350]  0.8468  0.1264  0.3556  48.6%
## Omitting Lunt 2014                 0.0120 [-0.1995; 0.2236]  0.9111  0.1208  0.3475  47.4%
## Omitting Matsuo 2014               0.0136 [-0.1993; 0.2265]  0.9004  0.1230  0.3506  47.7%
## Omitting Matsuo 2015               0.0112 [-0.2004; 0.2228]  0.9174  0.1205  0.3472  47.3%
## Omitting Mitranun 2014             0.0460 [-0.1673; 0.2592]  0.6728  0.1235  0.3514  47.8%
## Omitting Molmen-Hansen 2011       -0.0240 [-0.2152; 0.1671]  0.8053  0.0689  0.2625  32.9%
## Omitting Ramos 2016a               0.0382 [-0.1790; 0.2554]  0.7302  0.1304  0.3611  48.6%
## Omitting Ramos 2016b               0.0175 [-0.1974; 0.2324]  0.8731  0.1266  0.3558  48.2%
## Omitting Rognmo 2004               0.0338 [-0.1795; 0.2472]  0.7561  0.1266  0.3558  48.8%
## Omitting Skleryk 2013              0.0211 [-0.1916; 0.2337]  0.8460  0.1252  0.3538  48.5%
## Omitting Tjønna 2008               0.0293 [-0.1846; 0.2432]  0.7884  0.1274  0.3569  48.9%
## Omitting Wegmann 2018              0.0191 [-0.1984; 0.2366]  0.8632  0.1304  0.3612  48.4%
##                                                                                           
## Pooled estimate                    0.0254 [-0.1884; 0.2391]  0.8161  0.1369  0.3700  50.4%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

7.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

7.2 Subgroups

7.2.1 Overall

7.2.1.1 Forest plot

7.2.1.2 R output
##                            SMD            95%-CI     meta-analysis
##                         0.0254 [-0.1884; 0.2391]           Overall
## Healthy                 0.1168 [-0.1703; 0.4039]        Population
## Overweight/obese        0.0270 [-0.5167; 0.5707]        Population
## Cardiac Rehabilitation -0.1620 [-0.9985; 0.6745]        Population
## Metabolic Syndrome      0.0471 [-0.2729; 0.3671]        Population
## T2D                    -0.1080 [-0.7572; 0.5412]        Population
## < 30 y                 -0.1078 [-0.6101; 0.3945]               Age
## 30 - 50 y               0.2543 [-0.0260; 0.5345]               Age
## > 50 y                 -0.0763 [-0.4131; 0.2605]               Age
## < 5 weeks               0.1733 [-0.2769; 0.6236] Training Duration
## 5 - 10 weeks            0.0438 [-0.3705; 0.4582] Training Duration
## > 10 weeks             -0.0169 [-0.3126; 0.2789] Training Duration
## < 0.5                   0.0397 [-0.2318; 0.3113]         Men Ratio
## > 0.5                   0.0211 [-0.2702; 0.3124]         Men Ratio
## Running                 0.1490 [-0.1282; 0.4262]  Type of Exercise
## Cycling                -0.1186 [-0.3988; 0.1617]  Type of Exercise
## < 80 mmHg              -0.1640 [-0.4587; 0.1306]   Baseline Values
## 80 - 90 mmHg            0.1423 [-0.1203; 0.4049]   Baseline Values
## > 90 mmHg               0.4612 [-0.0590; 0.9814]   Baseline Values
## HIIT                    0.0190 [-0.2230; 0.2611]      Type of HIIE
## SIT                     0.0624 [-0.3422; 0.4671]      Type of HIIE
## 
## Number of studies combined: k = 25
## 
##                         SMD            95%-CI    z p-value
## Random effects model 0.0254 [-0.1884; 0.2391] 0.23  0.8161
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1369; tau = 0.3700; I^2 = 50.4% [21.3%; 68.7%]; H = 1.42 [1.13; 1.79]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  48.37   24  0.0023
## 
## Results for meta-analyses (random effects model):
##                     k    SMD            95%-CI  tau^2    tau     Q   I^2
## Overall            25 0.0254 [-0.1884; 0.2391] 0.1369 0.3700 48.37 50.4%
## Population         25 0.0254 [-0.1884; 0.2391] 0.1369 0.3700 48.37 50.4%
## Age                25 0.0254 [-0.1884; 0.2391] 0.1369 0.3700 48.37 50.4%
## Training Duration  25 0.0254 [-0.1884; 0.2391] 0.1369 0.3700 48.37 50.4%
## Men Ratio          25 0.0254 [-0.1884; 0.2391] 0.1369 0.3700 48.37 50.4%
## Type of Exercise   25 0.0254 [-0.1884; 0.2391] 0.1369 0.3700 48.37 50.4%
## Baseline Values    25 0.0254 [-0.1884; 0.2391] 0.1369 0.3700 48.37 50.4%
## Type of HIIE       25 0.0254 [-0.1884; 0.2391] 0.1369 0.3700 48.37 50.4%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

7.2.2 Population

7.2.2.1 Forest plot

7.2.2.2 R output
##                            SMD             95%-CI %W(fixed) %W(random)             population
## Beetham 2019           -1.2992 [-2.4937; -0.1048]       1.4        2.3       Overweight/obese
## Ciolac 2010             0.0721 [-0.6211;  0.7653]       4.1        4.5                Healthy
## Cocks 2013             -0.2152 [-1.1980;  0.7676]       2.1        3.1                Healthy
## Conraads 2015          -0.3593 [-0.6589; -0.0597]      22.1        7.4 Cardiac Rehabilitation
## Currie 2015            -1.4264 [-2.4347; -0.4181]       2.0        3.0 Cardiac Rehabilitation
## Eguchi 2012             0.0000 [-0.8765;  0.8765]       2.6        3.5                Healthy
## Fisher 2015            -0.8645 [-1.7259; -0.0031]       2.7        3.6       Overweight/obese
## Honkala 2017 (Healthy)  0.3306 [-0.4152;  1.0764]       3.6        4.2                Healthy
## Honkala 2017 (T2D)      0.3339 [-0.6606;  1.3283]       2.0        3.0                    T2D
## Jo 2020                -0.1592 [-0.8325;  0.5141]       4.4        4.7     Metabolic Syndrome
## Keating 2014           -0.1303 [-0.9669;  0.7063]       2.8        3.7       Overweight/obese
## Keteyian 2014           1.1469 [ 0.3457;  1.9480]       3.1        3.9 Cardiac Rehabilitation
## Klonizakis 2014        -0.2983 [-1.2509;  0.6543]       2.2        3.2                Healthy
## Lunt 2014               0.2262 [-0.6137;  1.0662]       2.8        3.7       Overweight/obese
## Lunt 2014               0.4890 [-0.3602;  1.3382]       2.8        3.7       Overweight/obese
## Matsuo 2014             0.3972 [-0.3791;  1.1735]       3.3        4.0                Healthy
## Matsuo 2015             0.4857 [-0.3262;  1.2976]       3.0        3.9     Metabolic Syndrome
## Mitranun 2014          -0.3780 [-1.1253;  0.3694]       3.6        4.2                    T2D
## Molmen-Hansen 2011      0.8739 [ 0.3380;  1.4098]       6.9        5.6       Overweight/obese
## Ramos 2016a            -0.1587 [-0.7576;  0.4402]       5.5        5.2     Metabolic Syndrome
## Ramos 2016b             0.2636 [-0.4337;  0.9609]       4.1        4.5     Metabolic Syndrome
## Rognmo 2004            -0.1462 [-1.0998;  0.8075]       2.2        3.2 Cardiac Rehabilitation
## Skleryk 2013            0.2806 [-0.7042;  1.2654]       2.1        3.1       Overweight/obese
## Tjønna 2008             0.0000 [-0.9107;  0.9107]       2.4        3.4     Metabolic Syndrome
## Wegmann 2018            0.1864 [-0.3759;  0.7486]       6.3        5.4                Healthy
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0024 [-0.1434; 0.1386] -0.03  0.9732
## Random effects model  0.0254 [-0.1884; 0.2391]  0.23  0.8161
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1369 [0.0236; 0.4283]; tau = 0.3700 [0.1536; 0.6544];
##  I^2 = 50.4% [21.3%; 68.7%]; H = 1.42 [1.13; 1.79]
## 
## Quantifying residual heterogeneity:
##  I^2 = 48.6% [14.7%; 69.0%]; H = 1.39 [1.08; 1.80]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  48.37   24  0.0023
## 
## Results for subgroups (fixed effect model):
##                          k     SMD            95%-CI     Q   I^2
## Healthy                  7  0.1168 [-0.1703; 0.4039]  1.97  0.0%
## Overweight/obese         7  0.1966 [-0.1086; 0.5018] 17.52 65.8%
## Cardiac Rehabilitation   4 -0.2529 [-0.5135; 0.0076] 16.06 81.3%
## Metabolic Syndrome       5  0.0471 [-0.2729; 0.3671]  2.18  0.0%
## T2D                      2 -0.1207 [-0.7188; 0.4774]  1.15 13.3%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  6.09    4  0.1926
## Within groups  38.89   20  0.0069
## 
## Results for subgroups (random effects model):
##                          k     SMD            95%-CI  tau^2    tau
## Healthy                  7  0.1168 [-0.1703; 0.4039]      0      0
## Overweight/obese         7  0.0270 [-0.5167; 0.5707] 0.3427 0.5854
## Cardiac Rehabilitation   4 -0.1620 [-0.9985; 0.6745] 0.5666 0.7527
## Metabolic Syndrome       5  0.0471 [-0.2729; 0.3671]      0      0
## T2D                      2 -0.1080 [-0.7572; 0.5412] 0.0311 0.1763
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.69    4  0.9527
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
7.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 25; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.1703 (SE = 0.1049)
## tau (square root of estimated tau^2 value):             0.4127
## I^2 (residual heterogeneity / unaccounted variability): 52.64%
## H^2 (unaccounted variability / sampling variability):   2.11
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 20) = 42.2257, p-val = 0.0026
## 
## Test of Moderators (coefficients 2:5):
## QM(df = 4) = 0.6651, p-val = 0.9556
## 
## Model Results:
## 
##                               estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                         0.0948  0.2177   0.4356  0.6631  -0.3318  0.5214    
## .byvarOverweight/obese         -0.0390  0.3141  -0.1242  0.9012  -0.6547  0.5766    
## .byvarCardiac Rehabilitation   -0.2574  0.3536  -0.7280  0.4666  -0.9503  0.4356    
## .byvarMetabolic Syndrome       -0.0233  0.3306  -0.0704  0.9438  -0.6713  0.6248    
## .byvarT2D                      -0.1705  0.4786  -0.3563  0.7216  -1.1085  0.7675    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
7.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

7.2.3 Age

7.2.3.1 Forest plot

7.2.3.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_age
## Beetham 2019           -1.2992 [-2.4937; -0.1048]       1.4        2.3       > 50 y
## Ciolac 2010             0.0721 [-0.6211;  0.7653]       4.1        4.5       < 30 y
## Cocks 2013             -0.2152 [-1.1980;  0.7676]       2.1        3.1       < 30 y
## Conraads 2015          -0.3593 [-0.6589; -0.0597]      22.1        7.4       > 50 y
## Currie 2015            -1.4264 [-2.4347; -0.4181]       2.0        3.0       > 50 y
## Eguchi 2012             0.0000 [-0.8765;  0.8765]       2.6        3.5       > 50 y
## Fisher 2015            -0.8645 [-1.7259; -0.0031]       2.7        3.6       < 30 y
## Honkala 2017 (Healthy)  0.3306 [-0.4152;  1.0764]       3.6        4.2    30 - 50 y
## Honkala 2017 (T2D)      0.3339 [-0.6606;  1.3283]       2.0        3.0    30 - 50 y
## Jo 2020                -0.1592 [-0.8325;  0.5141]       4.4        4.7       > 50 y
## Keating 2014           -0.1303 [-0.9669;  0.7063]       2.8        3.7    30 - 50 y
## Keteyian 2014           1.1469 [ 0.3457;  1.9480]       3.1        3.9       > 50 y
## Klonizakis 2014        -0.2983 [-1.2509;  0.6543]       2.2        3.2       > 50 y
## Lunt 2014               0.2262 [-0.6137;  1.0662]       2.8        3.7    30 - 50 y
## Lunt 2014               0.4890 [-0.3602;  1.3382]       2.8        3.7    30 - 50 y
## Matsuo 2014             0.3972 [-0.3791;  1.1735]       3.3        4.0       < 30 y
## Matsuo 2015             0.4857 [-0.3262;  1.2976]       3.0        3.9    30 - 50 y
## Mitranun 2014          -0.3780 [-1.1253;  0.3694]       3.6        4.2       > 50 y
## Molmen-Hansen 2011      0.8739 [ 0.3380;  1.4098]       6.9        5.6       > 50 y
## Ramos 2016a            -0.1587 [-0.7576;  0.4402]       5.5        5.2       > 50 y
## Ramos 2016b             0.2636 [-0.4337;  0.9609]       4.1        4.5       > 50 y
## Rognmo 2004            -0.1462 [-1.0998;  0.8075]       2.2        3.2       > 50 y
## Skleryk 2013            0.2806 [-0.7042;  1.2654]       2.1        3.1    30 - 50 y
## Tjønna 2008             0.0000 [-0.9107;  0.9107]       2.4        3.4       > 50 y
## Wegmann 2018            0.1864 [-0.3759;  0.7486]       6.3        5.4    30 - 50 y
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0024 [-0.1434; 0.1386] -0.03  0.9732
## Random effects model  0.0254 [-0.1884; 0.2391]  0.23  0.8161
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1369 [0.0236; 0.4283]; tau = 0.3700 [0.1536; 0.6544];
##  I^2 = 50.4% [21.3%; 68.7%]; H = 1.42 [1.13; 1.79]
## 
## Quantifying residual heterogeneity:
##  I^2 = 45.9% [11.8%; 66.8%]; H = 1.36 [1.06; 1.74]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  48.37   24  0.0023
## 
## Results for subgroups (fixed effect model):
##             k     SMD            95%-CI     Q   I^2
## < 30 y      4 -0.0882 [-0.4930; 0.3165]  4.52 33.6%
## 30 - 50 y   8  0.2543 [-0.0260; 0.5345]  1.42  0.0%
## > 50 y     13 -0.0886 [-0.2673; 0.0901] 34.75 65.5%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  4.29    2  0.1171
## Within groups  40.69   22  0.0090
## 
## Results for subgroups (random effects model):
##             k     SMD            95%-CI  tau^2    tau
## < 30 y      4 -0.1078 [-0.6101; 0.3945] 0.0884 0.2972
## 30 - 50 y   8  0.2543 [-0.0260; 0.5345]      0      0
## > 50 y     13 -0.0763 [-0.4131; 0.2605] 0.2274 0.4769
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   2.85    2  0.2406
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
7.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 25; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.1457 (SE = 0.0897)
## tau (square root of estimated tau^2 value):             0.3817
## I^2 (residual heterogeneity / unaccounted variability): 51.57%
## H^2 (unaccounted variability / sampling variability):   2.06
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 23) = 47.4925, p-val = 0.0019
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.1318, p-val = 0.7166
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    0.1870  0.4603   0.4061  0.6846  -0.7153  1.0892    
## age       -0.0033  0.0091  -0.3630  0.7166  -0.0212  0.0146    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
7.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

7.2.4 Training Duration

7.2.4.1 Forest plot

7.2.4.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_duration
## Beetham 2019           -1.2992 [-2.4937; -0.1048]       1.4        2.3        > 10 weeks
## Ciolac 2010             0.0721 [-0.6211;  0.7653]       4.1        4.5        > 10 weeks
## Cocks 2013             -0.2152 [-1.1980;  0.7676]       2.1        3.1      5 - 10 weeks
## Conraads 2015          -0.3593 [-0.6589; -0.0597]      22.1        7.4        > 10 weeks
## Currie 2015            -1.4264 [-2.4347; -0.4181]       2.0        3.0        > 10 weeks
## Eguchi 2012             0.0000 [-0.8765;  0.8765]       2.6        3.5        > 10 weeks
## Fisher 2015            -0.8645 [-1.7259; -0.0031]       2.7        3.6      5 - 10 weeks
## Honkala 2017 (Healthy)  0.3306 [-0.4152;  1.0764]       3.6        4.2         < 5 weeks
## Honkala 2017 (T2D)      0.3339 [-0.6606;  1.3283]       2.0        3.0         < 5 weeks
## Jo 2020                -0.1592 [-0.8325;  0.5141]       4.4        4.7      5 - 10 weeks
## Keating 2014           -0.1303 [-0.9669;  0.7063]       2.8        3.7        > 10 weeks
## Keteyian 2014           1.1469 [ 0.3457;  1.9480]       3.1        3.9      5 - 10 weeks
## Klonizakis 2014        -0.2983 [-1.2509;  0.6543]       2.2        3.2         < 5 weeks
## Lunt 2014               0.2262 [-0.6137;  1.0662]       2.8        3.7        > 10 weeks
## Lunt 2014               0.4890 [-0.3602;  1.3382]       2.8        3.7        > 10 weeks
## Matsuo 2014             0.3972 [-0.3791;  1.1735]       3.3        4.0      5 - 10 weeks
## Matsuo 2015             0.4857 [-0.3262;  1.2976]       3.0        3.9      5 - 10 weeks
## Mitranun 2014          -0.3780 [-1.1253;  0.3694]       3.6        4.2      5 - 10 weeks
## Molmen-Hansen 2011      0.8739 [ 0.3380;  1.4098]       6.9        5.6        > 10 weeks
## Ramos 2016a            -0.1587 [-0.7576;  0.4402]       5.5        5.2        > 10 weeks
## Ramos 2016b             0.2636 [-0.4337;  0.9609]       4.1        4.5        > 10 weeks
## Rognmo 2004            -0.1462 [-1.0998;  0.8075]       2.2        3.2      5 - 10 weeks
## Skleryk 2013            0.2806 [-0.7042;  1.2654]       2.1        3.1         < 5 weeks
## Tjønna 2008             0.0000 [-0.9107;  0.9107]       2.4        3.4        > 10 weeks
## Wegmann 2018            0.1864 [-0.3759;  0.7486]       6.3        5.4        > 10 weeks
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0024 [-0.1434; 0.1386] -0.03  0.9732
## Random effects model  0.0254 [-0.1884; 0.2391]  0.23  0.8161
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1369 [0.0236; 0.4283]; tau = 0.3700 [0.1536; 0.6544];
##  I^2 = 50.4% [21.3%; 68.7%]; H = 1.42 [1.13; 1.79]
## 
## Quantifying residual heterogeneity:
##  I^2 = 50.0% [19.2%; 69.1%]; H = 1.41 [1.11; 1.80]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  48.37   24  0.0023
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI     Q   I^2
## < 5 weeks      4  0.1733 [-0.2769; 0.6236]  1.15  0.0%
## 5 - 10 weeks   8  0.0481 [-0.2387; 0.3348] 14.36 51.3%
## > 10 weeks    13 -0.0458 [-0.2197; 0.1281] 28.52 57.9%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.94    2  0.6239
## Within groups  44.03   22  0.0035
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI  tau^2    tau
## < 5 weeks      4  0.1733 [-0.2769; 0.6236]      0      0
## 5 - 10 weeks   8  0.0438 [-0.3705; 0.4582] 0.1815 0.4260
## > 10 weeks    13 -0.0169 [-0.3126; 0.2789] 0.1546 0.3931
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.48    2  0.7869
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
7.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 25; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.1496 (SE = 0.0923)
## tau (square root of estimated tau^2 value):             0.3868
## I^2 (residual heterogeneity / unaccounted variability): 52.44%
## H^2 (unaccounted variability / sampling variability):   2.10
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 23) = 48.3620, p-val = 0.0015
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0005, p-val = 0.9822
## 
## Model Results:
## 
##           estimate      se    zval    pval    ci.lb   ci.ub 
## intrcpt     0.0193  0.2556  0.0757  0.9397  -0.4816  0.5202    
## duration    0.0005  0.0212  0.0223  0.9822  -0.0411  0.0420    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
7.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

7.2.5 Men Ratio

7.2.5.1 Forest plot

7.2.5.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_men_ratio
## Beetham 2019           -1.2992 [-2.4937; -0.1048]       1.4        2.3              > 0.5
## Ciolac 2010             0.0721 [-0.6211;  0.7653]       4.1        4.5              < 0.5
## Cocks 2013             -0.2152 [-1.1980;  0.7676]       2.1        3.1              > 0.5
## Conraads 2015          -0.3593 [-0.6589; -0.0597]      22.1        7.4              > 0.5
## Currie 2015            -1.4264 [-2.4347; -0.4181]       2.0        3.0              > 0.5
## Eguchi 2012             0.0000 [-0.8765;  0.8765]       2.6        3.5              > 0.5
## Fisher 2015            -0.8645 [-1.7259; -0.0031]       2.7        3.6              > 0.5
## Honkala 2017 (Healthy)  0.3306 [-0.4152;  1.0764]       3.6        4.2              > 0.5
## Honkala 2017 (T2D)      0.3339 [-0.6606;  1.3283]       2.0        3.0              > 0.5
## Jo 2020                -0.1592 [-0.8325;  0.5141]       4.4        4.7              > 0.5
## Keating 2014           -0.1303 [-0.9669;  0.7063]       2.8        3.7              < 0.5
## Keteyian 2014           1.1469 [ 0.3457;  1.9480]       3.1        3.9              > 0.5
## Klonizakis 2014        -0.2983 [-1.2509;  0.6543]       2.2        3.2              < 0.5
## Lunt 2014               0.2262 [-0.6137;  1.0662]       2.8        3.7              < 0.5
## Lunt 2014               0.4890 [-0.3602;  1.3382]       2.8        3.7              < 0.5
## Matsuo 2014             0.3972 [-0.3791;  1.1735]       3.3        4.0              > 0.5
## Matsuo 2015             0.4857 [-0.3262;  1.2976]       3.0        3.9              > 0.5
## Mitranun 2014          -0.3780 [-1.1253;  0.3694]       3.6        4.2              < 0.5
## Molmen-Hansen 2011      0.8739 [ 0.3380;  1.4098]       6.9        5.6              > 0.5
## Ramos 2016a            -0.1587 [-0.7576;  0.4402]       5.5        5.2              > 0.5
## Ramos 2016b             0.2636 [-0.4337;  0.9609]       4.1        4.5              > 0.5
## Rognmo 2004            -0.1462 [-1.0998;  0.8075]       2.2        3.2              > 0.5
## Skleryk 2013            0.2806 [-0.7042;  1.2654]       2.1        3.1              > 0.5
## Tjønna 2008             0.0000 [-0.9107;  0.9107]       2.4        3.4              < 0.5
## Wegmann 2018            0.1864 [-0.3759;  0.7486]       6.3        5.4              < 0.5
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0024 [-0.1434; 0.1386] -0.03  0.9732
## Random effects model  0.0254 [-0.1884; 0.2391]  0.23  0.8161
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1369 [0.0236; 0.4283]; tau = 0.3700 [0.1536; 0.6544];
##  I^2 = 50.4% [21.3%; 68.7%]; H = 1.42 [1.13; 1.79]
## 
## Quantifying residual heterogeneity:
##  I^2 = 48.7% [17.7%; 68.1%]; H = 1.40 [1.10; 1.77]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  48.37   24  0.0023
## 
## Results for subgroups (fixed effect model):
##         k     SMD            95%-CI     Q   I^2
## < 0.5   8  0.0397 [-0.2318; 0.3113]  3.16  0.0%
## > 0.5  17 -0.0168 [-0.1821; 0.1485] 41.70 61.6%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.12    1  0.7275
## Within groups  44.86   23  0.0041
## 
## Results for subgroups (random effects model):
##         k    SMD            95%-CI  tau^2    tau
## < 0.5   8 0.0397 [-0.2318; 0.3113]      0      0
## > 0.5  17 0.0211 [-0.2702; 0.3124] 0.2099 0.4582
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.01    1  0.9268
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
7.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 25; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.1470 (SE = 0.0901)
## tau (square root of estimated tau^2 value):             0.3834
## I^2 (residual heterogeneity / unaccounted variability): 51.63%
## H^2 (unaccounted variability / sampling variability):   2.07
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 23) = 47.5541, p-val = 0.0019
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0167, p-val = 0.8972
## 
## Model Results:
## 
##            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt      0.0551  0.2607   0.2115  0.8325  -0.4559  0.5662    
## men_ratio   -0.0459  0.3554  -0.1292  0.8972  -0.7425  0.6506    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
7.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

7.2.6 Type of Exercise

7.2.6.1 Forest plot

7.2.6.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) type_exercise
## Beetham 2019           -1.2992 [-2.4937; -0.1048]       1.4        2.3       Running
## Ciolac 2010             0.0721 [-0.6211;  0.7653]       4.1        4.5       Running
## Cocks 2013             -0.2152 [-1.1980;  0.7676]       2.1        3.1       Cycling
## Conraads 2015          -0.3593 [-0.6589; -0.0597]      22.1        7.4       Cycling
## Currie 2015            -1.4264 [-2.4347; -0.4181]       2.0        3.0       Cycling
## Eguchi 2012             0.0000 [-0.8765;  0.8765]       2.6        3.5       Cycling
## Fisher 2015            -0.8645 [-1.7259; -0.0031]       2.7        3.6       Cycling
## Honkala 2017 (Healthy)  0.3306 [-0.4152;  1.0764]       3.6        4.2       Cycling
## Honkala 2017 (T2D)      0.3339 [-0.6606;  1.3283]       2.0        3.0       Cycling
## Jo 2020                -0.1592 [-0.8325;  0.5141]       4.4        4.7       Running
## Keating 2014           -0.1303 [-0.9669;  0.7063]       2.8        3.7       Cycling
## Keteyian 2014           1.1469 [ 0.3457;  1.9480]       3.1        3.9       Running
## Klonizakis 2014        -0.2983 [-1.2509;  0.6543]       2.2        3.2       Cycling
## Lunt 2014               0.2262 [-0.6137;  1.0662]       2.8        3.7       Running
## Lunt 2014               0.4890 [-0.3602;  1.3382]       2.8        3.7       Running
## Matsuo 2014             0.3972 [-0.3791;  1.1735]       3.3        4.0       Cycling
## Matsuo 2015             0.4857 [-0.3262;  1.2976]       3.0        3.9       Cycling
## Mitranun 2014          -0.3780 [-1.1253;  0.3694]       3.6        4.2       Running
## Molmen-Hansen 2011      0.8739 [ 0.3380;  1.4098]       6.9        5.6       Running
## Ramos 2016a            -0.1587 [-0.7576;  0.4402]       5.5        5.2       Running
## Ramos 2016b             0.2636 [-0.4337;  0.9609]       4.1        4.5       Running
## Rognmo 2004            -0.1462 [-1.0998;  0.8075]       2.2        3.2       Running
## Skleryk 2013            0.2806 [-0.7042;  1.2654]       2.1        3.1       Cycling
## Tjønna 2008             0.0000 [-0.9107;  0.9107]       2.4        3.4       Running
## Wegmann 2018            0.1864 [-0.3759;  0.7486]       6.3        5.4       Running
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0024 [-0.1434; 0.1386] -0.03  0.9732
## Random effects model  0.0254 [-0.1884; 0.2391]  0.23  0.8161
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1369 [0.0236; 0.4283]; tau = 0.3700 [0.1536; 0.6544];
##  I^2 = 50.4% [21.3%; 68.7%]; H = 1.42 [1.13; 1.79]
## 
## Quantifying residual heterogeneity:
##  I^2 = 40.5% [3.0%; 63.5%]; H = 1.30 [1.02; 1.65]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  48.37   24  0.0023
## 
## Results for subgroups (fixed effect model):
##           k     SMD            95%-CI     Q   I^2
## Running  13  0.1813 [-0.0192; 0.3819] 21.81 45.0%
## Cycling  12 -0.1812 [-0.3801; 0.0176] 16.83 34.7%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  6.33    1  0.0119
## Within groups  38.65   23  0.0217
## 
## Results for subgroups (random effects model):
##           k     SMD            95%-CI  tau^2    tau
## Running  13  0.1490 [-0.1282; 0.4262] 0.1129 0.3360
## Cycling  12 -0.1186 [-0.3988; 0.1617] 0.0773 0.2781
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.77    1  0.1833
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
7.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 25; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.1156 (SE = 0.0791)
## tau (square root of estimated tau^2 value):             0.3400
## I^2 (residual heterogeneity / unaccounted variability): 45.00%
## H^2 (unaccounted variability / sampling variability):   1.82
## R^2 (amount of heterogeneity accounted for):            15.60%
## 
## Test for Residual Heterogeneity:
## QE(df = 23) = 41.8198, p-val = 0.0095
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.6173, p-val = 0.2035
## 
## Model Results:
## 
##                       estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                -0.1187  0.1551  -0.7652  0.4441  -0.4228  0.1853    
## type_exerciseRunning    0.2675  0.2103   1.2717  0.2035  -0.1448  0.6797    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
7.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

7.2.7 Baseline Values

7.2.7.1 Forest plot

7.2.7.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_bsln
## Beetham 2019           -1.2992 [-2.4937; -0.1048]       1.4        2.3     < 80 mmHg
## Ciolac 2010             0.0721 [-0.6211;  0.7653]       4.1        4.5     < 80 mmHg
## Cocks 2013             -0.2152 [-1.1980;  0.7676]       2.1        3.1     < 80 mmHg
## Conraads 2015          -0.3593 [-0.6589; -0.0597]      22.1        7.4     < 80 mmHg
## Currie 2015            -1.4264 [-2.4347; -0.4181]       2.0        3.0     < 80 mmHg
## Eguchi 2012             0.0000 [-0.8765;  0.8765]       2.6        3.5     < 80 mmHg
## Fisher 2015            -0.8645 [-1.7259; -0.0031]       2.7        3.6     < 80 mmHg
## Honkala 2017 (Healthy)  0.3306 [-0.4152;  1.0764]       3.6        4.2     < 80 mmHg
## Honkala 2017 (T2D)      0.3339 [-0.6606;  1.3283]       2.0        3.0  80 - 90 mmHg
## Jo 2020                -0.1592 [-0.8325;  0.5141]       4.4        4.7  80 - 90 mmHg
## Keating 2014           -0.1303 [-0.9669;  0.7063]       2.8        3.7     < 80 mmHg
## Keteyian 2014           1.1469 [ 0.3457;  1.9480]       3.1        3.9     < 80 mmHg
## Klonizakis 2014        -0.2983 [-1.2509;  0.6543]       2.2        3.2     < 80 mmHg
## Lunt 2014               0.2262 [-0.6137;  1.0662]       2.8        3.7  80 - 90 mmHg
## Lunt 2014               0.4890 [-0.3602;  1.3382]       2.8        3.7  80 - 90 mmHg
## Matsuo 2014             0.3972 [-0.3791;  1.1735]       3.3        4.0     < 80 mmHg
## Matsuo 2015             0.4857 [-0.3262;  1.2976]       3.0        3.9  80 - 90 mmHg
## Mitranun 2014          -0.3780 [-1.1253;  0.3694]       3.6        4.2     < 80 mmHg
## Molmen-Hansen 2011      0.8739 [ 0.3380;  1.4098]       6.9        5.6     > 90 mmHg
## Ramos 2016a            -0.1587 [-0.7576;  0.4402]       5.5        5.2  80 - 90 mmHg
## Ramos 2016b             0.2636 [-0.4337;  0.9609]       4.1        4.5     > 90 mmHg
## Rognmo 2004            -0.1462 [-1.0998;  0.8075]       2.2        3.2     < 80 mmHg
## Skleryk 2013            0.2806 [-0.7042;  1.2654]       2.1        3.1  80 - 90 mmHg
## Tjønna 2008             0.0000 [-0.9107;  0.9107]       2.4        3.4     > 90 mmHg
## Wegmann 2018            0.1864 [-0.3759;  0.7486]       6.3        5.4  80 - 90 mmHg
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0024 [-0.1434; 0.1386] -0.03  0.9732
## Random effects model  0.0254 [-0.1884; 0.2391]  0.23  0.8161
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1369 [0.0236; 0.4283]; tau = 0.3700 [0.1536; 0.6544];
##  I^2 = 50.4% [21.3%; 68.7%]; H = 1.42 [1.13; 1.79]
## 
## Quantifying residual heterogeneity:
##  I^2 = 32.4% [0.0%; 59.3%]; H = 1.22 [1.00; 1.57]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  48.37   24  0.0023
## 
## Results for subgroups (fixed effect model):
##                k     SMD             95%-CI     Q   I^2
## < 80 mmHg     14 -0.1956 [-0.3816; -0.0097] 26.04 50.1%
## 80 - 90 mmHg   8  0.1423 [-0.1203;  0.4049]  3.12  0.0%
## > 90 mmHg      3  0.5230 [ 0.1375;  0.9084]  3.36 40.5%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups 12.45    2  0.0020
## Within groups  32.53   22  0.0688
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI  tau^2    tau
## < 80 mmHg     14 -0.1640 [-0.4587; 0.1306] 0.1431 0.3783
## 80 - 90 mmHg   8  0.1423 [-0.1203; 0.4049]      0      0
## > 90 mmHg      3  0.4612 [-0.0590; 0.9814] 0.0864 0.2939
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   4.88    2  0.0871
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
7.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 25; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0746 (SE = 0.0648)
## tau (square root of estimated tau^2 value):             0.2732
## I^2 (residual heterogeneity / unaccounted variability): 35.30%
## H^2 (unaccounted variability / sampling variability):   1.55
## R^2 (amount of heterogeneity accounted for):            45.50%
## 
## Test for Residual Heterogeneity:
## QE(df = 23) = 35.5466, p-val = 0.0459
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 6.8330, p-val = 0.0089
## 
## Model Results:
## 
##                estimate      se     zval    pval    ci.lb    ci.ub 
## intrcpt         -1.9416  0.7599  -2.5552  0.0106  -3.4310  -0.4523   * 
## bsln_adjusted    0.0242  0.0093   2.6140  0.0089   0.0061   0.0423  ** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
7.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

7.2.8 Type of HIIE

7.2.8.1 Forest plot

7.2.8.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) HIIE
## Beetham 2019           -1.2992 [-2.4937; -0.1048]       1.4        2.3 HIIT
## Ciolac 2010             0.0721 [-0.6211;  0.7653]       4.1        4.5 HIIT
## Cocks 2013             -0.2152 [-1.1980;  0.7676]       2.1        3.1  SIT
## Conraads 2015          -0.3593 [-0.6589; -0.0597]      22.1        7.4 HIIT
## Currie 2015            -1.4264 [-2.4347; -0.4181]       2.0        3.0 HIIT
## Eguchi 2012             0.0000 [-0.8765;  0.8765]       2.6        3.5 HIIT
## Fisher 2015            -0.8645 [-1.7259; -0.0031]       2.7        3.6  SIT
## Honkala 2017 (Healthy)  0.3306 [-0.4152;  1.0764]       3.6        4.2  SIT
## Honkala 2017 (T2D)      0.3339 [-0.6606;  1.3283]       2.0        3.0  SIT
## Jo 2020                -0.1592 [-0.8325;  0.5141]       4.4        4.7 HIIT
## Keating 2014           -0.1303 [-0.9669;  0.7063]       2.8        3.7 HIIT
## Keteyian 2014           1.1469 [ 0.3457;  1.9480]       3.1        3.9 HIIT
## Klonizakis 2014        -0.2983 [-1.2509;  0.6543]       2.2        3.2 HIIT
## Lunt 2014               0.2262 [-0.6137;  1.0662]       2.8        3.7 HIIT
## Lunt 2014               0.4890 [-0.3602;  1.3382]       2.8        3.7  SIT
## Matsuo 2014             0.3972 [-0.3791;  1.1735]       3.3        4.0 HIIT
## Matsuo 2015             0.4857 [-0.3262;  1.2976]       3.0        3.9 HIIT
## Mitranun 2014          -0.3780 [-1.1253;  0.3694]       3.6        4.2 HIIT
## Molmen-Hansen 2011      0.8739 [ 0.3380;  1.4098]       6.9        5.6 HIIT
## Ramos 2016a            -0.1587 [-0.7576;  0.4402]       5.5        5.2 HIIT
## Ramos 2016b             0.2636 [-0.4337;  0.9609]       4.1        4.5 HIIT
## Rognmo 2004            -0.1462 [-1.0998;  0.8075]       2.2        3.2 HIIT
## Skleryk 2013            0.2806 [-0.7042;  1.2654]       2.1        3.1  SIT
## Tjønna 2008             0.0000 [-0.9107;  0.9107]       2.4        3.4 HIIT
## Wegmann 2018            0.1864 [-0.3759;  0.7486]       6.3        5.4 HIIT
## 
## Number of studies combined: k = 25
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0024 [-0.1434; 0.1386] -0.03  0.9732
## Random effects model  0.0254 [-0.1884; 0.2391]  0.23  0.8161
## 
## Quantifying heterogeneity:
##  tau^2 = 0.1369 [0.0236; 0.4283]; tau = 0.3700 [0.1536; 0.6544];
##  I^2 = 50.4% [21.3%; 68.7%]; H = 1.42 [1.13; 1.79]
## 
## Quantifying residual heterogeneity:
##  I^2 = 48.7% [17.6%; 68.1%]; H = 1.40 [1.10; 1.77]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  48.37   24  0.0023
## 
## Results for subgroups (fixed effect model):
##        k     SMD            95%-CI     Q   I^2
## HIIT  19 -0.0134 [-0.1667; 0.1398] 38.67 53.5%
## SIT    6  0.0655 [-0.2976; 0.4287]  6.15 18.7%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.15    1  0.6945
## Within groups  44.82   23  0.0042
## 
## Results for subgroups (random effects model):
##        k    SMD            95%-CI  tau^2    tau
## HIIT  19 0.0190 [-0.2230; 0.2611] 0.1411 0.3756
## SIT    6 0.0624 [-0.3422; 0.4671] 0.0479 0.2189
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.03    1  0.8568
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
7.2.8.3 Meta-regression
## 
## Mixed-Effects Model (k = 25; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.1473 (SE = 0.0905)
## tau (square root of estimated tau^2 value):             0.3838
## I^2 (residual heterogeneity / unaccounted variability): 52.28%
## H^2 (unaccounted variability / sampling variability):   2.10
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 23) = 48.2004, p-val = 0.0016
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0293, p-val = 0.8641
## 
## Model Results:
## 
##          estimate      se    zval    pval    ci.lb   ci.ub 
## intrcpt    0.0149  0.1248  0.1195  0.9049  -0.2296  0.2595    
## HIIESIT    0.0469  0.2739  0.1711  0.8641  -0.4899  0.5837    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
7.2.8.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

8. HDL

8.1 Overall

8.1.1 Forest plot

8.1.2 R output

##                            SMD            95%-CI %W(fixed) %W(random)
## Abdelbasset 2020       -0.1018 [-0.8067; 0.6031]       3.8        3.8
## Ciolac 2010            -0.1776 [-1.0150; 0.6597]       2.7        2.7
## Conraads 2015          -0.0644 [-0.3617; 0.2329]      21.4       21.4
## Currie 2015             0.9662 [ 0.0147; 1.9177]       2.1        2.1
## Eguchi 2012             0.0917 [-0.7852; 0.9687]       2.5        2.5
## Fisher 2015             0.0877 [-0.7371; 0.9125]       2.8        2.8
## Grieco 2013            -0.4309 [-1.2797; 0.4179]       2.6        2.6
## Helgerud 2007          -0.6997 [-1.6026; 0.2032]       2.3        2.3
## Honkala 2017 (Healthy) -0.5327 [-1.2865; 0.2212]       3.3        3.3
## Honkala 2017 (T2D)     -0.6981 [-1.7151; 0.3188]       1.8        1.8
## Jo 2020                -0.2853 [-0.9610; 0.3904]       4.1        4.1
## Keating 2014            0.0000 [-0.8357; 0.8357]       2.7        2.7
## Kim 2015                0.0174 [-0.7234; 0.7582]       3.5        3.5
## Lunt 2014               0.1810 [-0.6580; 1.0201]       2.7        2.7
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.7        2.7
## Madssen 2014           -0.2247 [-0.8893; 0.4399]       4.3        4.3
## Maillard 2016           0.0000 [-0.9800; 0.9800]       2.0        2.0
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       3.0        3.0
## Mitranun 2014           0.0232 [-0.7176; 0.7641]       3.5        3.5
## Motiani 2017           -0.5635 [-1.3474; 0.2204]       3.1        3.1
## Nalcakan 2014           0.4637 [-0.5642; 1.4915]       1.8        1.8
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       5.3        5.3
## Ramos 2016b            -0.1396 [-0.8347; 0.5556]       3.9        3.9
## Sandvei 2012            0.5378 [-0.2949; 1.3706]       2.7        2.7
## Sawyer 2016             0.2947 [-0.6343; 1.2236]       2.2        2.2
## Tjønna 2008             0.5492 [-0.3781; 1.4765]       2.2        2.2
## Winn 2018               1.0842 [ 0.0347; 2.1337]       1.7        1.7
## Zapata-Lamana 2018     -0.5861 [-1.3427; 0.1704]       3.3        3.3
## 
## Number of studies combined: k = 28
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## Random effects model -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1473]; tau = 0 [0.0000; 0.3838];
##  I^2 = 0.0% [0.0%; 37.1%]; H = 1.00 [1.00; 1.26]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  24.91   27  0.5795
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

8.1.3 Sensitivity analysis

8.1.3.1 Forest plot

8.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                                       SMD            95%-CI p-value   tau^2     tau   I^2
## Omitting Abdelbasset 2020         -0.0513 [-0.1918; 0.0891]  0.4737  0.0000  0.0000  0.0%
## Omitting Ciolac 2010              -0.0499 [-0.1895; 0.0898]  0.4838  0.0000  0.0000  0.0%
## Omitting Conraads 2015            -0.0502 [-0.2056; 0.1053]  0.5270  0.0000  0.0000  0.0%
## Omitting Currie 2015              -0.0737 [-0.2129; 0.0655]  0.2992  0.0000  0.0000  0.0%
## Omitting Eguchi 2012              -0.0567 [-0.1962; 0.0827]  0.4253  0.0000  0.0000  0.0%
## Omitting Fisher 2015              -0.0571 [-0.1968; 0.0826]  0.4230  0.0000  0.0000  0.0%
## Omitting Grieco 2013              -0.0434 [-0.1830; 0.0962]  0.5421  0.0000  0.0000  0.0%
## Omitting Helgerud 2007            -0.0386 [-0.1780; 0.1008]  0.5874  0.0000  0.0000  0.0%
## Omitting Honkala 2017 (Healthy)   -0.0372 [-0.1773; 0.1029]  0.6029  0.0000  0.0000  0.0%
## Omitting Honkala 2017 (T2D)       -0.0419 [-0.1810; 0.0971]  0.5543  0.0000  0.0000  0.0%
## Omitting Jo 2020                  -0.0434 [-0.1841; 0.0973]  0.5455  0.0000  0.0000  0.0%
## Omitting Keating 2014             -0.0547 [-0.1943; 0.0850]  0.4431  0.0000  0.0000  0.0%
## Omitting Kim 2015                 -0.0557 [-0.1959; 0.0845]  0.4364  0.0000  0.0000  0.0%
## Omitting Lunt 2014                -0.0595 [-0.1991; 0.0802]  0.4039  0.0000  0.0000  0.0%
## Omitting Lunt 2014                -0.0546 [-0.1943; 0.0850]  0.4431  0.0000  0.0000  0.0%
## Omitting Madssen 2014             -0.0457 [-0.1865; 0.0951]  0.5248  0.0000  0.0000  0.0%
## Omitting Maillard 2016            -0.0542 [-0.1934; 0.0849]  0.4448  0.0000  0.0000  0.0%
## Omitting Matsuo 2015              -0.0548 [-0.1946; 0.0850]  0.4425  0.0000  0.0000  0.0%
## Omitting Mitranun 2014            -0.0559 [-0.1961; 0.0843]  0.4347  0.0000  0.0000  0.0%
## Omitting Motiani 2017             -0.0375 [-0.1775; 0.1024]  0.5990  0.0000  0.0000  0.0%
## Omitting Nalcakan 2014            -0.0621 [-0.2011; 0.0769]  0.3814  0.0000  0.0000  0.0%
## Omitting Ramos 2016a              -0.0561 [-0.1977; 0.0854]  0.4369  0.0000  0.0000  0.0%
## Omitting Ramos 2016b              -0.0498 [-0.1903; 0.0908]  0.4875  0.0000  0.0000  0.0%
## Omitting Sandvei 2012             -0.0692 [-0.2088; 0.0705]  0.3317  0.0000  0.0000  0.0%
## Omitting Sawyer 2016              -0.0607 [-0.1999; 0.0786]  0.3934  0.0000  0.0000  0.0%
## Omitting Tjønna 2008              -0.0661 [-0.2054; 0.0732]  0.3521  0.0000  0.0000  0.0%
## Omitting Winn 2018                -0.0716 [-0.2105; 0.0673]  0.3123  0.0000  0.0000  0.0%
## Omitting Zapata-Lamana 2018       -0.0355 [-0.1756; 0.1045]  0.6189  0.0000  0.0000  0.0%
##                                                                                          
## Pooled estimate                   -0.0529 [-0.1905; 0.0847]  0.4512  0.0000  0.0000  0.0%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

8.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

8.2 Subgroups

8.2.1 Overall

8.2.1.1 Forest plot

8.2.1.2 R output
##                            SMD            95%-CI     meta-analysis
##                        -0.0529 [-0.1905; 0.0847]           Overall
## Healthy                -0.1927 [-0.5024; 0.1171]        Population
## Overweight/obese        0.0662 [-0.2618; 0.3942]        Population
## Cardiac Rehabilitation  0.0282 [-0.3156; 0.3721]        Population
## Metabolic Syndrome     -0.0291 [-0.3490; 0.2908]        Population
## T2D                    -0.1355 [-0.5495; 0.2786]        Population
## < 30 y                 -0.1368 [-0.4814; 0.2079]               Age
## 30 - 50 y              -0.0716 [-0.3771; 0.2339]               Age
## > 50 y                 -0.0167 [-0.1967; 0.1632]               Age
## < 5 weeks              -0.2789 [-0.8082; 0.2505] Training Duration
## 5 - 10 weeks            0.0033 [-0.2497; 0.2563] Training Duration
## > 10 weeks             -0.0251 [-0.2063; 0.1560] Training Duration
## < 0.5                   0.0653 [-0.1822; 0.3128]         Men Ratio
## > 0.5                  -0.1063 [-0.2721; 0.0595]         Men Ratio
## Cycling                -0.0954 [-0.2756; 0.0849]  Type of Exercise
## Running                 0.0060 [-0.2075; 0.2196]  Type of Exercise
## < 1.3 mmol/L           -0.0015 [-0.1605; 0.1574]   Baseline Values
## > 1.3 mmol/L           -0.2087 [-0.4846; 0.0673]   Baseline Values
## HIIT                   -0.0385 [-0.1908; 0.1139]      Type of HIIE
## SIT                    -0.1155 [-0.4638; 0.2327]      Type of HIIE
## 
## Number of studies combined: k = 28
## 
##                          SMD            95%-CI     z p-value
## Random effects model -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## 
## Quantifying heterogeneity:
##  tau^2 = 0; tau = 0; I^2 = 0.0% [0.0%; 37.1%]; H = 1.00 [1.00; 1.26]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  24.91   27  0.5795
## 
## Results for meta-analyses (random effects model):
##                     k     SMD            95%-CI tau^2 tau     Q  I^2
## Overall            28 -0.0529 [-0.1905; 0.0847]     0   0 24.91 0.0%
## Population         28 -0.0529 [-0.1905; 0.0847]     0   0 24.91 0.0%
## Age                28 -0.0529 [-0.1905; 0.0847]     0   0 24.91 0.0%
## Training Duration  28 -0.0529 [-0.1905; 0.0847]     0   0 24.91 0.0%
## Men Ratio          28 -0.0529 [-0.1905; 0.0847]     0   0 24.91 0.0%
## Type of Exercise   28 -0.0529 [-0.1905; 0.0847]     0   0 24.91 0.0%
## Baseline Values    28 -0.0529 [-0.1905; 0.0847]     0   0 24.91 0.0%
## Type of HIIE       28 -0.0529 [-0.1905; 0.0847]     0   0 24.91 0.0%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

8.2.2 Population

8.2.2.1 Forest plot

8.2.2.2 R output
##                            SMD            95%-CI %W(fixed) %W(random)             population
## Abdelbasset 2020       -0.1018 [-0.8067; 0.6031]       3.8        3.8                    T2D
## Ciolac 2010            -0.1776 [-1.0150; 0.6597]       2.7        2.7                Healthy
## Conraads 2015          -0.0644 [-0.3617; 0.2329]      21.4       21.4 Cardiac Rehabilitation
## Currie 2015             0.9662 [ 0.0147; 1.9177]       2.1        2.1 Cardiac Rehabilitation
## Eguchi 2012             0.0917 [-0.7852; 0.9687]       2.5        2.5                Healthy
## Fisher 2015             0.0877 [-0.7371; 0.9125]       2.8        2.8       Overweight/obese
## Grieco 2013            -0.4309 [-1.2797; 0.4179]       2.6        2.6                Healthy
## Helgerud 2007          -0.6997 [-1.6026; 0.2032]       2.3        2.3                Healthy
## Honkala 2017 (Healthy) -0.5327 [-1.2865; 0.2212]       3.3        3.3                Healthy
## Honkala 2017 (T2D)     -0.6981 [-1.7151; 0.3188]       1.8        1.8                    T2D
## Jo 2020                -0.2853 [-0.9610; 0.3904]       4.1        4.1     Metabolic Syndrome
## Keating 2014            0.0000 [-0.8357; 0.8357]       2.7        2.7       Overweight/obese
## Kim 2015                0.0174 [-0.7234; 0.7582]       3.5        3.5 Cardiac Rehabilitation
## Lunt 2014               0.1810 [-0.6580; 1.0201]       2.7        2.7       Overweight/obese
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.7        2.7       Overweight/obese
## Madssen 2014           -0.2247 [-0.8893; 0.4399]       4.3        4.3 Cardiac Rehabilitation
## Maillard 2016           0.0000 [-0.9800; 0.9800]       2.0        2.0                    T2D
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       3.0        3.0     Metabolic Syndrome
## Mitranun 2014           0.0232 [-0.7176; 0.7641]       3.5        3.5                    T2D
## Motiani 2017           -0.5635 [-1.3474; 0.2204]       3.1        3.1                Healthy
## Nalcakan 2014           0.4637 [-0.5642; 1.4915]       1.8        1.8                Healthy
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       5.3        5.3     Metabolic Syndrome
## Ramos 2016b            -0.1396 [-0.8347; 0.5556]       3.9        3.9     Metabolic Syndrome
## Sandvei 2012            0.5378 [-0.2949; 1.3706]       2.7        2.7                Healthy
## Sawyer 2016             0.2947 [-0.6343; 1.2236]       2.2        2.2       Overweight/obese
## Tjønna 2008             0.5492 [-0.3781; 1.4765]       2.2        2.2     Metabolic Syndrome
## Winn 2018               1.0842 [ 0.0347; 2.1337]       1.7        1.7       Overweight/obese
## Zapata-Lamana 2018     -0.5861 [-1.3427; 0.1704]       3.3        3.3       Overweight/obese
## 
## Number of studies combined: k = 28
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## Random effects model -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1473]; tau = 0 [0.0000; 0.3838];
##  I^2 = 0.0% [0.0%; 37.1%]; H = 1.00 [1.00; 1.26]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 39.3%]; H = 1.00 [1.00; 1.28]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  24.91   27  0.5795
## 
## Results for subgroups (fixed effect model):
##                          k     SMD            95%-CI    Q   I^2
## Healthy                  8 -0.1945 [-0.4949; 0.1059] 7.43  5.8%
## Overweight/obese         7  0.0646 [-0.2593; 0.3885] 6.15  2.4%
## Cardiac Rehabilitation   4 -0.0114 [-0.2577; 0.2349] 4.15 27.7%
## Metabolic Syndrome       5 -0.0291 [-0.3490; 0.2908] 1.99  0.0%
## T2D                      4 -0.1355 [-0.5495; 0.2786] 1.27  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  1.64    4  0.8013
## Within groups  20.99   23  0.5819
## 
## Results for subgroups (random effects model):
##                          k     SMD            95%-CI  tau^2    tau
## Healthy                  8 -0.1927 [-0.5024; 0.1171] 0.0115 0.1075
## Overweight/obese         7  0.0662 [-0.2618; 0.3942] 0.0047 0.0689
## Cardiac Rehabilitation   4  0.0282 [-0.3156; 0.3721] 0.0368 0.1918
## Metabolic Syndrome       5 -0.0291 [-0.3490; 0.2908]      0      0
## T2D                      4 -0.1355 [-0.5495; 0.2786]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.68    4  0.7952
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
8.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 28; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0005 (SE = 0.0467)
## tau (square root of estimated tau^2 value):             0.0224
## I^2 (residual heterogeneity / unaccounted variability): 0.32%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 23) = 23.0729, p-val = 0.4565
## 
## Test of Moderators (coefficients 2:5):
## QM(df = 4) = 1.8360, p-val = 0.7659
## 
## Model Results:
## 
##                               estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                        -0.2006  0.1532  -1.3088  0.1906  -0.5009  0.0998    
## .byvarOverweight/obese          0.2727  0.2253   1.2100  0.2263  -0.1690  0.7143    
## .byvarCardiac Rehabilitation    0.1930  0.1988   0.9709  0.3316  -0.1966  0.5826    
## .byvarMetabolic Syndrome        0.1726  0.2241   0.7701  0.4412  -0.2666  0.6117    
## .byvarT2D                       0.0571  0.2611   0.2186  0.8270  -0.4547  0.5688    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
8.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

8.2.3 Age

8.2.3.1 Forest plot

8.2.3.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_age
## Abdelbasset 2020       -0.1018 [-0.8067; 0.6031]       3.8        3.8       > 50 y
## Ciolac 2010            -0.1776 [-1.0150; 0.6597]       2.7        2.7       < 30 y
## Conraads 2015          -0.0644 [-0.3617; 0.2329]      21.4       21.4       > 50 y
## Currie 2015             0.9662 [ 0.0147; 1.9177]       2.1        2.1       > 50 y
## Eguchi 2012             0.0917 [-0.7852; 0.9687]       2.5        2.5       > 50 y
## Fisher 2015             0.0877 [-0.7371; 0.9125]       2.8        2.8       < 30 y
## Grieco 2013            -0.4309 [-1.2797; 0.4179]       2.6        2.6       < 30 y
## Helgerud 2007          -0.6997 [-1.6026; 0.2032]       2.3        2.3       < 30 y
## Honkala 2017 (Healthy) -0.5327 [-1.2865; 0.2212]       3.3        3.3    30 - 50 y
## Honkala 2017 (T2D)     -0.6981 [-1.7151; 0.3188]       1.8        1.8    30 - 50 y
## Jo 2020                -0.2853 [-0.9610; 0.3904]       4.1        4.1       > 50 y
## Keating 2014            0.0000 [-0.8357; 0.8357]       2.7        2.7    30 - 50 y
## Kim 2015                0.0174 [-0.7234; 0.7582]       3.5        3.5       > 50 y
## Lunt 2014               0.1810 [-0.6580; 1.0201]       2.7        2.7    30 - 50 y
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.7        2.7    30 - 50 y
## Madssen 2014           -0.2247 [-0.8893; 0.4399]       4.3        4.3       > 50 y
## Maillard 2016           0.0000 [-0.9800; 0.9800]       2.0        2.0       > 50 y
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       3.0        3.0    30 - 50 y
## Mitranun 2014           0.0232 [-0.7176; 0.7641]       3.5        3.5       > 50 y
## Motiani 2017           -0.5635 [-1.3474; 0.2204]       3.1        3.1    30 - 50 y
## Nalcakan 2014           0.4637 [-0.5642; 1.4915]       1.8        1.8       < 30 y
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       5.3        5.3       > 50 y
## Ramos 2016b            -0.1396 [-0.8347; 0.5556]       3.9        3.9       > 50 y
## Sandvei 2012            0.5378 [-0.2949; 1.3706]       2.7        2.7       < 30 y
## Sawyer 2016             0.2947 [-0.6343; 1.2236]       2.2        2.2    30 - 50 y
## Tjønna 2008             0.5492 [-0.3781; 1.4765]       2.2        2.2       > 50 y
## Winn 2018               1.0842 [ 0.0347; 2.1337]       1.7        1.7    30 - 50 y
## Zapata-Lamana 2018     -0.5861 [-1.3427; 0.1704]       3.3        3.3       < 30 y
## 
## Number of studies combined: k = 28
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## Random effects model -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1473]; tau = 0 [0.0000; 0.3838];
##  I^2 = 0.0% [0.0%; 37.1%]; H = 1.00 [1.00; 1.26]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 35.9%]; H = 1.00 [1.00; 1.25]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  24.91   27  0.5795
## 
## Results for subgroups (fixed effect model):
##             k     SMD            95%-CI    Q   I^2
## < 30 y      7 -0.1397 [-0.4623; 0.1828] 6.83 12.1%
## 30 - 50 y   9 -0.0771 [-0.3632; 0.2090] 9.07 11.8%
## > 50 y     12 -0.0167 [-0.1967; 0.1632] 6.27  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.46    2  0.7941
## Within groups  22.17   25  0.6260
## 
## Results for subgroups (random effects model):
##             k     SMD            95%-CI  tau^2    tau
## < 30 y      7 -0.1368 [-0.4814; 0.2079] 0.0263 0.1620
## 30 - 50 y   9 -0.0716 [-0.3771; 0.2339] 0.0259 0.1609
## > 50 y     12 -0.0167 [-0.1967; 0.1632]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.39    2  0.8214
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
8.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 28; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0404)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 26) = 24.4829, p-val = 0.5484
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.4269, p-val = 0.5135
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.2195  0.2645  -0.8300  0.4066  -0.7379  0.2989    
## age        0.0034  0.0052   0.6534  0.5135  -0.0069  0.0137    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
8.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

8.2.4 Training Duration

8.2.4.1 Forest plot

8.2.4.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_duration
## Abdelbasset 2020       -0.1018 [-0.8067; 0.6031]       3.8        3.8      5 - 10 weeks
## Ciolac 2010            -0.1776 [-1.0150; 0.6597]       2.7        2.7        > 10 weeks
## Conraads 2015          -0.0644 [-0.3617; 0.2329]      21.4       21.4        > 10 weeks
## Currie 2015             0.9662 [ 0.0147; 1.9177]       2.1        2.1        > 10 weeks
## Eguchi 2012             0.0917 [-0.7852; 0.9687]       2.5        2.5        > 10 weeks
## Fisher 2015             0.0877 [-0.7371; 0.9125]       2.8        2.8      5 - 10 weeks
## Grieco 2013            -0.4309 [-1.2797; 0.4179]       2.6        2.6         < 5 weeks
## Helgerud 2007          -0.6997 [-1.6026; 0.2032]       2.3        2.3      5 - 10 weeks
## Honkala 2017 (Healthy) -0.5327 [-1.2865; 0.2212]       3.3        3.3         < 5 weeks
## Honkala 2017 (T2D)     -0.6981 [-1.7151; 0.3188]       1.8        1.8         < 5 weeks
## Jo 2020                -0.2853 [-0.9610; 0.3904]       4.1        4.1      5 - 10 weeks
## Keating 2014            0.0000 [-0.8357; 0.8357]       2.7        2.7        > 10 weeks
## Kim 2015                0.0174 [-0.7234; 0.7582]       3.5        3.5      5 - 10 weeks
## Lunt 2014               0.1810 [-0.6580; 1.0201]       2.7        2.7        > 10 weeks
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.7        2.7        > 10 weeks
## Madssen 2014           -0.2247 [-0.8893; 0.4399]       4.3        4.3        > 10 weeks
## Maillard 2016           0.0000 [-0.9800; 0.9800]       2.0        2.0        > 10 weeks
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       3.0        3.0      5 - 10 weeks
## Mitranun 2014           0.0232 [-0.7176; 0.7641]       3.5        3.5      5 - 10 weeks
## Motiani 2017           -0.5635 [-1.3474; 0.2204]       3.1        3.1         < 5 weeks
## Nalcakan 2014           0.4637 [-0.5642; 1.4915]       1.8        1.8      5 - 10 weeks
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       5.3        5.3        > 10 weeks
## Ramos 2016b            -0.1396 [-0.8347; 0.5556]       3.9        3.9        > 10 weeks
## Sandvei 2012            0.5378 [-0.2949; 1.3706]       2.7        2.7      5 - 10 weeks
## Sawyer 2016             0.2947 [-0.6343; 1.2236]       2.2        2.2      5 - 10 weeks
## Tjønna 2008             0.5492 [-0.3781; 1.4765]       2.2        2.2        > 10 weeks
## Winn 2018               1.0842 [ 0.0347; 2.1337]       1.7        1.7         < 5 weeks
## Zapata-Lamana 2018     -0.5861 [-1.3427; 0.1704]       3.3        3.3        > 10 weeks
## 
## Number of studies combined: k = 28
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## Random effects model -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1473]; tau = 0 [0.0000; 0.3838];
##  I^2 = 0.0% [0.0%; 37.1%]; H = 1.00 [1.00; 1.26]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 31.0%]; H = 1.00 [1.00; 1.20]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  24.91   27  0.5795
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI    Q   I^2
## < 5 weeks      5 -0.3162 [-0.7054; 0.0729] 7.21 44.6%
## 5 - 10 weeks  10  0.0033 [-0.2497; 0.2563] 5.39  0.0%
## > 10 weeks    13 -0.0251 [-0.2063; 0.1560] 7.98  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  2.04    2  0.3608
## Within groups  20.59   25  0.7152
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI  tau^2    tau
## < 5 weeks      5 -0.2789 [-0.8082; 0.2505] 0.1612 0.4015
## 5 - 10 weeks  10  0.0033 [-0.2497; 0.2563]      0      0
## > 10 weeks    13 -0.0251 [-0.2063; 0.1560]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.91    2  0.6330
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
8.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 28; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0399)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 26) = 23.6623, p-val = 0.5953
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.2475, p-val = 0.2640
## 
## Model Results:
## 
##           estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    -0.2522  0.1918  -1.3152  0.1884  -0.6281  0.1236    
## duration    0.0195  0.0175   1.1169  0.2640  -0.0147  0.0538    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
8.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

8.2.5 Men Ratio

8.2.5.1 Forest plot

8.2.5.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_men_ratio
## Abdelbasset 2020       -0.1018 [-0.8067; 0.6031]       3.8        3.8              > 0.5
## Ciolac 2010            -0.1776 [-1.0150; 0.6597]       2.7        2.7              < 0.5
## Conraads 2015          -0.0644 [-0.3617; 0.2329]      21.4       21.4              > 0.5
## Currie 2015             0.9662 [ 0.0147; 1.9177]       2.1        2.1              > 0.5
## Eguchi 2012             0.0917 [-0.7852; 0.9687]       2.5        2.5              > 0.5
## Fisher 2015             0.0877 [-0.7371; 0.9125]       2.8        2.8              > 0.5
## Grieco 2013            -0.4309 [-1.2797; 0.4179]       2.6        2.6              < 0.5
## Helgerud 2007          -0.6997 [-1.6026; 0.2032]       2.3        2.3              > 0.5
## Honkala 2017 (Healthy) -0.5327 [-1.2865; 0.2212]       3.3        3.3              > 0.5
## Honkala 2017 (T2D)     -0.6981 [-1.7151; 0.3188]       1.8        1.8              > 0.5
## Jo 2020                -0.2853 [-0.9610; 0.3904]       4.1        4.1              > 0.5
## Keating 2014            0.0000 [-0.8357; 0.8357]       2.7        2.7              < 0.5
## Kim 2015                0.0174 [-0.7234; 0.7582]       3.5        3.5              > 0.5
## Lunt 2014               0.1810 [-0.6580; 1.0201]       2.7        2.7              < 0.5
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.7        2.7              < 0.5
## Madssen 2014           -0.2247 [-0.8893; 0.4399]       4.3        4.3              > 0.5
## Maillard 2016           0.0000 [-0.9800; 0.9800]       2.0        2.0              < 0.5
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       3.0        3.0              > 0.5
## Mitranun 2014           0.0232 [-0.7176; 0.7641]       3.5        3.5              < 0.5
## Motiani 2017           -0.5635 [-1.3474; 0.2204]       3.1        3.1              > 0.5
## Nalcakan 2014           0.4637 [-0.5642; 1.4915]       1.8        1.8              > 0.5
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       5.3        5.3              > 0.5
## Ramos 2016b            -0.1396 [-0.8347; 0.5556]       3.9        3.9              > 0.5
## Sandvei 2012            0.5378 [-0.2949; 1.3706]       2.7        2.7              < 0.5
## Sawyer 2016             0.2947 [-0.6343; 1.2236]       2.2        2.2              < 0.5
## Tjønna 2008             0.5492 [-0.3781; 1.4765]       2.2        2.2              < 0.5
## Winn 2018               1.0842 [ 0.0347; 2.1337]       1.7        1.7              < 0.5
## Zapata-Lamana 2018     -0.5861 [-1.3427; 0.1704]       3.3        3.3              < 0.5
## 
## Number of studies combined: k = 28
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## Random effects model -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1473]; tau = 0 [0.0000; 0.3838];
##  I^2 = 0.0% [0.0%; 37.1%]; H = 1.00 [1.00; 1.26]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 30.0%]; H = 1.00 [1.00; 1.20]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  24.91   27  0.5795
## 
## Results for subgroups (fixed effect model):
##         k     SMD            95%-CI     Q  I^2
## < 0.5  12  0.0653 [-0.1822; 0.3128]  9.78 0.0%
## > 0.5  16 -0.1063 [-0.2721; 0.0595] 11.58 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  1.28    1  0.2588
## Within groups  21.35   26  0.7235
## 
## Results for subgroups (random effects model):
##         k     SMD            95%-CI tau^2 tau
## < 0.5  12  0.0653 [-0.1822; 0.3128]     0   0
## > 0.5  16 -0.1063 [-0.2721; 0.0595]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.28    1  0.2588
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
8.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 28; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0404)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 26) = 24.7473, p-val = 0.5333
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.1625, p-val = 0.6868
## 
## Model Results:
## 
##            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt      0.0107  0.1728   0.0622  0.9504  -0.3279  0.3494    
## men_ratio   -0.0937  0.2324  -0.4032  0.6868  -0.5492  0.3618    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
8.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

8.2.6 Type of Exercise

8.2.6.1 Forest plot

8.2.6.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) type_exercise
## Abdelbasset 2020       -0.1018 [-0.8067; 0.6031]       3.8        3.8       Cycling
## Ciolac 2010            -0.1776 [-1.0150; 0.6597]       2.7        2.7       Running
## Conraads 2015          -0.0644 [-0.3617; 0.2329]      21.4       21.4       Cycling
## Currie 2015             0.9662 [ 0.0147; 1.9177]       2.1        2.1       Cycling
## Eguchi 2012             0.0917 [-0.7852; 0.9687]       2.5        2.5       Cycling
## Fisher 2015             0.0877 [-0.7371; 0.9125]       2.8        2.8       Cycling
## Grieco 2013            -0.4309 [-1.2797; 0.4179]       2.6        2.6       Cycling
## Helgerud 2007          -0.6997 [-1.6026; 0.2032]       2.3        2.3       Running
## Honkala 2017 (Healthy) -0.5327 [-1.2865; 0.2212]       3.3        3.3       Cycling
## Honkala 2017 (T2D)     -0.6981 [-1.7151; 0.3188]       1.8        1.8       Cycling
## Jo 2020                -0.2853 [-0.9610; 0.3904]       4.1        4.1       Running
## Keating 2014            0.0000 [-0.8357; 0.8357]       2.7        2.7       Cycling
## Kim 2015                0.0174 [-0.7234; 0.7582]       3.5        3.5       Running
## Lunt 2014               0.1810 [-0.6580; 1.0201]       2.7        2.7       Running
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.7        2.7       Running
## Madssen 2014           -0.2247 [-0.8893; 0.4399]       4.3        4.3       Running
## Maillard 2016           0.0000 [-0.9800; 0.9800]       2.0        2.0       Cycling
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       3.0        3.0       Cycling
## Mitranun 2014           0.0232 [-0.7176; 0.7641]       3.5        3.5       Running
## Motiani 2017           -0.5635 [-1.3474; 0.2204]       3.1        3.1       Cycling
## Nalcakan 2014           0.4637 [-0.5642; 1.4915]       1.8        1.8       Cycling
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       5.3        5.3       Running
## Ramos 2016b            -0.1396 [-0.8347; 0.5556]       3.9        3.9       Running
## Sandvei 2012            0.5378 [-0.2949; 1.3706]       2.7        2.7       Running
## Sawyer 2016             0.2947 [-0.6343; 1.2236]       2.2        2.2       Cycling
## Tjønna 2008             0.5492 [-0.3781; 1.4765]       2.2        2.2       Running
## Winn 2018               1.0842 [ 0.0347; 2.1337]       1.7        1.7       Running
## Zapata-Lamana 2018     -0.5861 [-1.3427; 0.1704]       3.3        3.3       Cycling
## 
## Number of studies combined: k = 28
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## Random effects model -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1473]; tau = 0 [0.0000; 0.3838];
##  I^2 = 0.0% [0.0%; 37.1%]; H = 1.00 [1.00; 1.26]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 32.5%]; H = 1.00 [1.00; 1.22]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  24.91   27  0.5795
## 
## Results for subgroups (fixed effect model):
##           k     SMD            95%-CI     Q  I^2
## Cycling  15 -0.0954 [-0.2756; 0.0849] 12.19 0.0%
## Running  13  0.0060 [-0.2075; 0.2196]  9.93 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.51    1  0.4770
## Within groups  22.12   26  0.6820
## 
## Results for subgroups (random effects model):
##           k     SMD            95%-CI tau^2 tau
## Cycling  15 -0.0954 [-0.2756; 0.0849]     0   0
## Running  13  0.0060 [-0.2075; 0.2196]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.51    1  0.4770
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
8.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 28; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0409)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 26) = 24.3574, p-val = 0.5555
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.5524, p-val = 0.4573
## 
## Model Results:
## 
##                       estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                -0.0970  0.0919  -1.0552  0.2913  -0.2771  0.0831    
## type_exerciseRunning    0.1059  0.1424   0.7433  0.4573  -0.1733  0.3850    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
8.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

8.2.7 Baseline Values

8.2.7.1 Forest plot

8.2.7.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_bsln
## Abdelbasset 2020       -0.1018 [-0.8067; 0.6031]       3.8        3.8  < 1.3 mmol/L
## Ciolac 2010            -0.1776 [-1.0150; 0.6597]       2.7        2.7  > 1.3 mmol/L
## Conraads 2015          -0.0644 [-0.3617; 0.2329]      21.4       21.4  < 1.3 mmol/L
## Currie 2015             0.9662 [ 0.0147; 1.9177]       2.1        2.1  < 1.3 mmol/L
## Eguchi 2012             0.0917 [-0.7852; 0.9687]       2.5        2.5  > 1.3 mmol/L
## Fisher 2015             0.0877 [-0.7371; 0.9125]       2.8        2.8  < 1.3 mmol/L
## Grieco 2013            -0.4309 [-1.2797; 0.4179]       2.6        2.6  > 1.3 mmol/L
## Helgerud 2007          -0.6997 [-1.6026; 0.2032]       2.3        2.3  < 1.3 mmol/L
## Honkala 2017 (Healthy) -0.5327 [-1.2865; 0.2212]       3.3        3.3  > 1.3 mmol/L
## Honkala 2017 (T2D)     -0.6981 [-1.7151; 0.3188]       1.8        1.8  < 1.3 mmol/L
## Jo 2020                -0.2853 [-0.9610; 0.3904]       4.1        4.1  < 1.3 mmol/L
## Keating 2014            0.0000 [-0.8357; 0.8357]       2.7        2.7  > 1.3 mmol/L
## Kim 2015                0.0174 [-0.7234; 0.7582]       3.5        3.5  < 1.3 mmol/L
## Lunt 2014               0.1810 [-0.6580; 1.0201]       2.7        2.7  < 1.3 mmol/L
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.7        2.7  < 1.3 mmol/L
## Madssen 2014           -0.2247 [-0.8893; 0.4399]       4.3        4.3  < 1.3 mmol/L
## Maillard 2016           0.0000 [-0.9800; 0.9800]       2.0        2.0  > 1.3 mmol/L
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       3.0        3.0  < 1.3 mmol/L
## Mitranun 2014           0.0232 [-0.7176; 0.7641]       3.5        3.5  < 1.3 mmol/L
## Motiani 2017           -0.5635 [-1.3474; 0.2204]       3.1        3.1  > 1.3 mmol/L
## Nalcakan 2014           0.4637 [-0.5642; 1.4915]       1.8        1.8  < 1.3 mmol/L
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       5.3        5.3  < 1.3 mmol/L
## Ramos 2016b            -0.1396 [-0.8347; 0.5556]       3.9        3.9  < 1.3 mmol/L
## Sandvei 2012            0.5378 [-0.2949; 1.3706]       2.7        2.7  > 1.3 mmol/L
## Sawyer 2016             0.2947 [-0.6343; 1.2236]       2.2        2.2  < 1.3 mmol/L
## Tjønna 2008             0.5492 [-0.3781; 1.4765]       2.2        2.2  < 1.3 mmol/L
## Winn 2018               1.0842 [ 0.0347; 2.1337]       1.7        1.7  < 1.3 mmol/L
## Zapata-Lamana 2018     -0.5861 [-1.3427; 0.1704]       3.3        3.3  > 1.3 mmol/L
## 
## Number of studies combined: k = 28
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## Random effects model -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1473]; tau = 0 [0.0000; 0.3838];
##  I^2 = 0.0% [0.0%; 37.1%]; H = 1.00 [1.00; 1.26]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 28.9%]; H = 1.00 [1.00; 1.19]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  24.91   27  0.5795
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI     Q  I^2
## < 1.3 mmol/L  19 -0.0015 [-0.1605; 0.1574] 14.79 0.0%
## > 1.3 mmol/L   9 -0.2087 [-0.4846; 0.0673]  6.22 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  1.63    1  0.2024
## Within groups  21.00   26  0.7418
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI tau^2 tau
## < 1.3 mmol/L  19 -0.0015 [-0.1605; 0.1574]     0   0
## > 1.3 mmol/L   9 -0.2087 [-0.4846; 0.0673]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.63    1  0.2024
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
8.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 28; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0400)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 26) = 22.0064, p-val = 0.6883
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 2.9034, p-val = 0.0884
## 
## Model Results:
## 
##                estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt          0.8125  0.5127   1.5847  0.1130  -0.1924  1.8173    
## bsln_adjusted   -0.7112  0.4174  -1.7039  0.0884  -1.5293  0.1069  . 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
8.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

8.2.8 Type of HIIE

8.2.8.1 Forest plot

8.2.8.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) HIIE
## Abdelbasset 2020       -0.1018 [-0.8067; 0.6031]       3.8        3.8 HIIT
## Ciolac 2010            -0.1776 [-1.0150; 0.6597]       2.7        2.7 HIIT
## Conraads 2015          -0.0644 [-0.3617; 0.2329]      21.4       21.4 HIIT
## Currie 2015             0.9662 [ 0.0147; 1.9177]       2.1        2.1 HIIT
## Eguchi 2012             0.0917 [-0.7852; 0.9687]       2.5        2.5 HIIT
## Fisher 2015             0.0877 [-0.7371; 0.9125]       2.8        2.8  SIT
## Grieco 2013            -0.4309 [-1.2797; 0.4179]       2.6        2.6 HIIT
## Helgerud 2007          -0.6997 [-1.6026; 0.2032]       2.3        2.3 HIIT
## Honkala 2017 (Healthy) -0.5327 [-1.2865; 0.2212]       3.3        3.3  SIT
## Honkala 2017 (T2D)     -0.6981 [-1.7151; 0.3188]       1.8        1.8  SIT
## Jo 2020                -0.2853 [-0.9610; 0.3904]       4.1        4.1 HIIT
## Keating 2014            0.0000 [-0.8357; 0.8357]       2.7        2.7 HIIT
## Kim 2015                0.0174 [-0.7234; 0.7582]       3.5        3.5 HIIT
## Lunt 2014               0.1810 [-0.6580; 1.0201]       2.7        2.7 HIIT
## Lunt 2014               0.0000 [-0.8374; 0.8374]       2.7        2.7  SIT
## Madssen 2014           -0.2247 [-0.8893; 0.4399]       4.3        4.3 HIIT
## Maillard 2016           0.0000 [-0.9800; 0.9800]       2.0        2.0 HIIT
## Matsuo 2015             0.0000 [-0.8002; 0.8002]       3.0        3.0 HIIT
## Mitranun 2014           0.0232 [-0.7176; 0.7641]       3.5        3.5 HIIT
## Motiani 2017           -0.5635 [-1.3474; 0.2204]       3.1        3.1  SIT
## Nalcakan 2014           0.4637 [-0.5642; 1.4915]       1.8        1.8  SIT
## Ramos 2016a             0.0000 [-0.5979; 0.5979]       5.3        5.3 HIIT
## Ramos 2016b            -0.1396 [-0.8347; 0.5556]       3.9        3.9 HIIT
## Sandvei 2012            0.5378 [-0.2949; 1.3706]       2.7        2.7  SIT
## Sawyer 2016             0.2947 [-0.6343; 1.2236]       2.2        2.2 HIIT
## Tjønna 2008             0.5492 [-0.3781; 1.4765]       2.2        2.2 HIIT
## Winn 2018               1.0842 [ 0.0347; 2.1337]       1.7        1.7 HIIT
## Zapata-Lamana 2018     -0.5861 [-1.3427; 0.1704]       3.3        3.3 HIIT
## 
## Number of studies combined: k = 28
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## Random effects model -0.0529 [-0.1905; 0.0847] -0.75  0.4512
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1473]; tau = 0 [0.0000; 0.3838];
##  I^2 = 0.0% [0.0%; 37.1%]; H = 1.00 [1.00; 1.26]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 33.4%]; H = 1.00 [1.00; 1.23]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  24.91   27  0.5795
## 
## Results for subgroups (fixed effect model):
##        k     SMD            95%-CI     Q   I^2
## HIIT  21 -0.0385 [-0.1908; 0.1139] 15.49  0.0%
## SIT    7 -0.1191 [-0.4417; 0.2036]  6.95 13.6%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.20    1  0.6579
## Within groups  22.43   26  0.6648
## 
## Results for subgroups (random effects model):
##        k     SMD            95%-CI  tau^2    tau
## HIIT  21 -0.0385 [-0.1908; 0.1139]      0      0
## SIT    7 -0.1155 [-0.4638; 0.2327] 0.0301 0.1736
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.16    1  0.6912
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
8.2.8.3 Meta-regression
## 
## Mixed-Effects Model (k = 28; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0398)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 26) = 24.6868, p-val = 0.5368
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.2230, p-val = 0.6368
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.0372  0.0777  -0.4795  0.6316  -0.1894  0.1150    
## HIIESIT   -0.0858  0.1818  -0.4722  0.6368  -0.4421  0.2704    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
8.2.8.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

9. LDL

9.1 Overall

9.1.1 Forest plot

9.1.2 R output

##                            SMD            95%-CI %W(fixed) %W(random)
## Abdelbasset 2020       -0.4098 [-1.1216; 0.3019]       4.5        5.0
## Ciolac 2010             0.4300 [-0.4153; 1.2754]       3.2        3.9
## Conraads 2015          -0.0612 [-0.3586; 0.2361]      25.6       10.5
## Currie 2015            -0.8003 [-1.7361; 0.1355]       2.6        3.4
## Eguchi 2012             0.2575 [-0.6227; 1.1376]       2.9        3.7
## Fisher 2015            -0.1711 [-0.9970; 0.6548]       3.3        4.1
## Helgerud 2007          -0.2220 [-1.1012; 0.6573]       2.9        3.7
## Honkala 2017 (Healthy)  1.0104 [ 0.2237; 1.7970]       3.7        4.3
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.3        3.1
## Jo 2020                 0.1476 [-0.5256; 0.8208]       5.0        5.3
## Keating 2014            0.6396 [-0.2172; 1.4964]       3.1        3.9
## Kim 2015                0.7581 [-0.0089; 1.5250]       3.8        4.5
## Madssen 2014            0.1632 [-0.5005; 0.8268]       5.1        5.4
## Maillard 2016          -0.6396 [-1.6443; 0.3651]       2.2        3.0
## Matsuo 2015            -0.3539 [-1.1603; 0.4525]       3.5        4.2
## Mitranun 2014           0.0053 [-0.7355; 0.7461]       4.1        4.7
## Motiani 2017            1.0575 [ 0.2367; 1.8782]       3.4        4.1
## Nalcakan 2014          -0.4452 [-1.4721; 0.5816]       2.1        2.9
## Ramos 2016b            -0.5073 [-1.2127; 0.1980]       4.5        5.0
## Sandvei 2012            0.6442 [-0.1948; 1.4833]       3.2        4.0
## Sawyer 2016            -0.1191 [-1.0439; 0.8056]       2.6        3.5
## Winn 2018               0.6875 [-0.3211; 1.6960]       2.2        3.0
## Zapata-Lamana 2018     -0.1122 [-0.8536; 0.6292]       4.1        4.7
## 
## Number of studies combined: k = 23
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0605 [-0.0898; 0.2108] 0.79  0.4303
## Random effects model 0.0846 [-0.1141; 0.2832] 0.83  0.4041
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0751 [0.0000; 0.3441]; tau = 0.2741 [0.0000; 0.5866];
##  I^2 = 34.5% [0.0%; 60.5%]; H = 1.24 [1.00; 1.59]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  33.59   22  0.0540
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

9.1.3 Sensitivity analysis

9.1.3.1 Forest plot

9.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                                      SMD            95%-CI p-value   tau^2     tau    I^2
## Omitting Abdelbasset 2020         0.1049 [-0.0903; 0.3001]  0.2920  0.0574  0.2396  28.5%
## Omitting Ciolac 2010              0.0670 [-0.1299; 0.2639]  0.5047  0.0630  0.2509  30.8%
## Omitting Conraads 2015            0.0983 [-0.1126; 0.3093]  0.3609  0.0768  0.2770  30.5%
## Omitting Currie 2015              0.1063 [-0.0816; 0.2941]  0.2675  0.0470  0.2169  25.0%
## Omitting Eguchi 2012              0.0745 [-0.1239; 0.2729]  0.4616  0.0663  0.2575  31.9%
## Omitting Fisher 2015              0.0912 [-0.1074; 0.2897]  0.3682  0.0659  0.2567  31.7%
## Omitting Helgerud 2007            0.0919 [-0.1058; 0.2896]  0.3622  0.0650  0.2550  31.5%
## Omitting Honkala 2017 (Healthy)   0.0359 [-0.1435; 0.2154]  0.6946  0.0312  0.1767  18.0%
## Omitting Honkala 2017 (T2D)       0.0835 [-0.1146; 0.2816]  0.4088  0.0669  0.2587  32.3%
## Omitting Jo 2020                  0.0775 [-0.1238; 0.2788]  0.4504  0.0687  0.2621  32.2%
## Omitting Keating 2014             0.0588 [-0.1347; 0.2522]  0.5516  0.0564  0.2374  28.5%
## Omitting Kim 2015                 0.0480 [-0.1411; 0.2371]  0.6189  0.0471  0.2171  24.8%
## Omitting Madssen 2014             0.0766 [-0.1248; 0.2779]  0.4562  0.0686  0.2620  32.1%
## Omitting Maillard 2016            0.1001 [-0.0920; 0.2923]  0.3071  0.0557  0.2359  28.4%
## Omitting Matsuo 2015              0.0987 [-0.0977; 0.2951]  0.3247  0.0614  0.2477  30.1%
## Omitting Mitranun 2014            0.0848 [-0.1157; 0.2853]  0.4070  0.0683  0.2614  32.3%
## Omitting Motiani 2017             0.0371 [-0.1420; 0.2161]  0.6849  0.0310  0.1762  17.9%
## Omitting Nalcakan 2014            0.0950 [-0.1000; 0.2900]  0.3398  0.0613  0.2475  30.4%
## Omitting Ramos 2016b              0.1095 [-0.0828; 0.3018]  0.2644  0.0519  0.2279  26.5%
## Omitting Sandvei 2012             0.0577 [-0.1355; 0.2509]  0.5580  0.0557  0.2361  28.2%
## Omitting Sawyer 2016              0.0877 [-0.1104; 0.2859]  0.3855  0.0664  0.2577  32.0%
## Omitting Winn 2018                0.0629 [-0.1305; 0.2563]  0.5239  0.0580  0.2408  29.3%
## Omitting Zapata-Lamana 2018       0.0903 [-0.1095; 0.2902]  0.3757  0.0671  0.2590  31.9%
##                                                                                          
## Pooled estimate                   0.0846 [-0.1141; 0.2832]  0.4041  0.0751  0.2741  34.5%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

9.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

9.2 Subgroups

9.2.1 Overall

9.2.1.1 Forest plot

9.2.1.2 R output
##                            SMD            95%-CI     meta-analysis
##                         0.0846 [-0.1141; 0.2832]           Overall
## Healthy                 0.4247 [ 0.0269; 0.8225]        Population
## Overweight/obese        0.1317 [-0.2523; 0.5157]        Population
## Cardiac Rehabilitation  0.0425 [-0.4124; 0.4973]        Population
## Metabolic Syndrome     -0.2085 [-0.6257; 0.2087]        Population
## T2D                    -0.2364 [-0.6516; 0.1789]        Population
## < 30 y                  0.0413 [-0.3052; 0.3878]               Age
## 30 - 50 y               0.4161 [-0.0086; 0.8408]               Age
## > 50 y                 -0.0661 [-0.2982; 0.1660]               Age
## < 5 weeks               0.7312 [ 0.2867; 1.1756] Training Duration
## 5 - 10 weeks            0.0059 [-0.2480; 0.2599] Training Duration
## > 10 weeks             -0.0517 [-0.2919; 0.1885] Training Duration
## < 0.5                   0.1835 [-0.1191; 0.4860]         Men Ratio
## > 0.5                   0.0360 [-0.2102; 0.2822]         Men Ratio
## Cycling                 0.0073 [-0.2563; 0.2708]  Type of Exercise
## Running                 0.1857 [-0.0833; 0.4547]  Type of Exercise
## < 3 mmol/L             -0.0719 [-0.2451; 0.1012]   Baseline Values
## > 3 mmol/L              0.4630 [ 0.0777; 0.8483]   Baseline Values
## HIIT                   -0.0184 [-0.1936; 0.1569]      Type of HIIE
## SIT                     0.3846 [-0.1053; 0.8746]      Type of HIIE
## 
## Number of studies combined: k = 23
## 
##                         SMD            95%-CI    z p-value
## Random effects model 0.0846 [-0.1141; 0.2832] 0.83  0.4041
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0751; tau = 0.2741; I^2 = 34.5% [0.0%; 60.5%]; H = 1.24 [1.00; 1.59]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  33.59   22  0.0540
## 
## Results for meta-analyses (random effects model):
##                     k    SMD            95%-CI  tau^2    tau     Q   I^2
## Overall            23 0.0846 [-0.1141; 0.2832] 0.0751 0.2741 33.59 34.5%
## Population         23 0.0846 [-0.1141; 0.2832] 0.0751 0.2741 33.59 34.5%
## Age                23 0.0846 [-0.1141; 0.2832] 0.0751 0.2741 33.59 34.5%
## Training Duration  23 0.0846 [-0.1141; 0.2832] 0.0751 0.2741 33.59 34.5%
## Men Ratio          23 0.0846 [-0.1141; 0.2832] 0.0751 0.2741 33.59 34.5%
## Type of Exercise   23 0.0846 [-0.1141; 0.2832] 0.0751 0.2741 33.59 34.5%
## Baseline Values    23 0.0846 [-0.1141; 0.2832] 0.0751 0.2741 33.59 34.5%
## Type of HIIE       23 0.0846 [-0.1141; 0.2832] 0.0751 0.2741 33.59 34.5%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

9.2.2 Population

9.2.2.1 Forest plot

9.2.2.2 R output
##                            SMD            95%-CI %W(fixed) %W(random)             population
## Abdelbasset 2020       -0.4098 [-1.1216; 0.3019]       4.5        5.0                    T2D
## Ciolac 2010             0.4300 [-0.4153; 1.2754]       3.2        3.9                Healthy
## Conraads 2015          -0.0612 [-0.3586; 0.2361]      25.6       10.5 Cardiac Rehabilitation
## Currie 2015            -0.8003 [-1.7361; 0.1355]       2.6        3.4 Cardiac Rehabilitation
## Eguchi 2012             0.2575 [-0.6227; 1.1376]       2.9        3.7                Healthy
## Fisher 2015            -0.1711 [-0.9970; 0.6548]       3.3        4.1       Overweight/obese
## Helgerud 2007          -0.2220 [-1.1012; 0.6573]       2.9        3.7                Healthy
## Honkala 2017 (Healthy)  1.0104 [ 0.2237; 1.7970]       3.7        4.3                Healthy
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.3        3.1                    T2D
## Jo 2020                 0.1476 [-0.5256; 0.8208]       5.0        5.3     Metabolic Syndrome
## Keating 2014            0.6396 [-0.2172; 1.4964]       3.1        3.9       Overweight/obese
## Kim 2015                0.7581 [-0.0089; 1.5250]       3.8        4.5 Cardiac Rehabilitation
## Madssen 2014            0.1632 [-0.5005; 0.8268]       5.1        5.4 Cardiac Rehabilitation
## Maillard 2016          -0.6396 [-1.6443; 0.3651]       2.2        3.0                    T2D
## Matsuo 2015            -0.3539 [-1.1603; 0.4525]       3.5        4.2     Metabolic Syndrome
## Mitranun 2014           0.0053 [-0.7355; 0.7461]       4.1        4.7                    T2D
## Motiani 2017            1.0575 [ 0.2367; 1.8782]       3.4        4.1                Healthy
## Nalcakan 2014          -0.4452 [-1.4721; 0.5816]       2.1        2.9                Healthy
## Ramos 2016b            -0.5073 [-1.2127; 0.1980]       4.5        5.0     Metabolic Syndrome
## Sandvei 2012            0.6442 [-0.1948; 1.4833]       3.2        4.0                Healthy
## Sawyer 2016            -0.1191 [-1.0439; 0.8056]       2.6        3.5       Overweight/obese
## Winn 2018               0.6875 [-0.3211; 1.6960]       2.2        3.0       Overweight/obese
## Zapata-Lamana 2018     -0.1122 [-0.8536; 0.6292]       4.1        4.7       Overweight/obese
## 
## Number of studies combined: k = 23
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0605 [-0.0898; 0.2108] 0.79  0.4303
## Random effects model 0.0846 [-0.1141; 0.2832] 0.83  0.4041
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0751 [0.0000; 0.3441]; tau = 0.2741 [0.0000; 0.5866];
##  I^2 = 34.5% [0.0%; 60.5%]; H = 1.24 [1.00; 1.59]
## 
## Quantifying residual heterogeneity:
##  I^2 = 17.5% [0.0%; 52.3%]; H = 1.10 [1.00; 1.45]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  33.59   22  0.0540
## 
## Results for subgroups (fixed effect model):
##                          k     SMD            95%-CI    Q   I^2
## Healthy                  7  0.4438 [ 0.1176; 0.7700] 8.87 32.4%
## Overweight/obese         5  0.1317 [-0.2523; 0.5157] 3.39  0.0%
## Cardiac Rehabilitation   4  0.0032 [-0.2437; 0.2501] 6.43 53.4%
## Metabolic Syndrome       3 -0.2085 [-0.6257; 0.2087] 1.79  0.0%
## T2D                      4 -0.2364 [-0.6516; 0.1789] 1.34  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  9.20    4  0.0562
## Within groups  21.83   18  0.2398
## 
## Results for subgroups (random effects model):
##                          k     SMD            95%-CI  tau^2    tau
## Healthy                  7  0.4247 [ 0.0269; 0.8225] 0.0932 0.3053
## Overweight/obese         5  0.1317 [-0.2523; 0.5157]      0      0
## Cardiac Rehabilitation   4  0.0425 [-0.4124; 0.4973] 0.1113 0.3336
## Metabolic Syndrome       3 -0.2085 [-0.6257; 0.2087]      0      0
## T2D                      4 -0.2364 [-0.6516; 0.1789]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   6.88    4  0.1426
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
9.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 23; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0498 (SE = 0.0700)
## tau (square root of estimated tau^2 value):             0.2232
## I^2 (residual heterogeneity / unaccounted variability): 23.91%
## H^2 (unaccounted variability / sampling variability):   1.31
## R^2 (amount of heterogeneity accounted for):            33.65%
## 
## Test for Residual Heterogeneity:
## QE(df = 18) = 23.6548, p-val = 0.1667
## 
## Test of Moderators (coefficients 2:5):
## QM(df = 4) = 7.3362, p-val = 0.1192
## 
## Model Results:
## 
##                               estimate      se     zval    pval    ci.lb    ci.ub 
## intrcpt                         0.4465  0.1865   2.3945  0.0166   0.0810   0.8120  * 
## .byvarOverweight/obese         -0.2974  0.2887  -1.0302  0.3029  -0.8634   0.2685    
## .byvarCardiac Rehabilitation   -0.4110  0.2647  -1.5528  0.1205  -0.9299   0.1078    
## .byvarMetabolic Syndrome       -0.6679  0.3113  -2.1458  0.0319  -1.2780  -0.0578  * 
## .byvarT2D                      -0.6952  0.3049  -2.2801  0.0226  -1.2929  -0.0976  * 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
9.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

9.2.3 Age

9.2.3.1 Forest plot

9.2.3.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_age
## Abdelbasset 2020       -0.4098 [-1.1216; 0.3019]       4.5        5.0       > 50 y
## Ciolac 2010             0.4300 [-0.4153; 1.2754]       3.2        3.9       < 30 y
## Conraads 2015          -0.0612 [-0.3586; 0.2361]      25.6       10.5       > 50 y
## Currie 2015            -0.8003 [-1.7361; 0.1355]       2.6        3.4       > 50 y
## Eguchi 2012             0.2575 [-0.6227; 1.1376]       2.9        3.7       > 50 y
## Fisher 2015            -0.1711 [-0.9970; 0.6548]       3.3        4.1       < 30 y
## Helgerud 2007          -0.2220 [-1.1012; 0.6573]       2.9        3.7       < 30 y
## Honkala 2017 (Healthy)  1.0104 [ 0.2237; 1.7970]       3.7        4.3    30 - 50 y
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.3        3.1    30 - 50 y
## Jo 2020                 0.1476 [-0.5256; 0.8208]       5.0        5.3       > 50 y
## Keating 2014            0.6396 [-0.2172; 1.4964]       3.1        3.9    30 - 50 y
## Kim 2015                0.7581 [-0.0089; 1.5250]       3.8        4.5       > 50 y
## Madssen 2014            0.1632 [-0.5005; 0.8268]       5.1        5.4       > 50 y
## Maillard 2016          -0.6396 [-1.6443; 0.3651]       2.2        3.0       > 50 y
## Matsuo 2015            -0.3539 [-1.1603; 0.4525]       3.5        4.2    30 - 50 y
## Mitranun 2014           0.0053 [-0.7355; 0.7461]       4.1        4.7       > 50 y
## Motiani 2017            1.0575 [ 0.2367; 1.8782]       3.4        4.1    30 - 50 y
## Nalcakan 2014          -0.4452 [-1.4721; 0.5816]       2.1        2.9       < 30 y
## Ramos 2016b            -0.5073 [-1.2127; 0.1980]       4.5        5.0       > 50 y
## Sandvei 2012            0.6442 [-0.1948; 1.4833]       3.2        4.0       < 30 y
## Sawyer 2016            -0.1191 [-1.0439; 0.8056]       2.6        3.5    30 - 50 y
## Winn 2018               0.6875 [-0.3211; 1.6960]       2.2        3.0    30 - 50 y
## Zapata-Lamana 2018     -0.1122 [-0.8536; 0.6292]       4.1        4.7       < 30 y
## 
## Number of studies combined: k = 23
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0605 [-0.0898; 0.2108] 0.79  0.4303
## Random effects model 0.0846 [-0.1141; 0.2832] 0.83  0.4041
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0751 [0.0000; 0.3441]; tau = 0.2741 [0.0000; 0.5866];
##  I^2 = 34.5% [0.0%; 60.5%]; H = 1.24 [1.00; 1.59]
## 
## Quantifying residual heterogeneity:
##  I^2 = 19.4% [0.0%; 52.5%]; H = 1.11 [1.00; 1.45]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  33.59   22  0.0540
## 
## Results for subgroups (fixed effect model):
##             k     SMD            95%-CI     Q   I^2
## < 30 y      6  0.0413 [-0.3052; 0.3878]  4.05  0.0%
## 30 - 50 y   7  0.4251 [ 0.0940; 0.7563]  9.78 38.6%
## > 50 y     10 -0.0628 [-0.2564; 0.1309] 10.98 18.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  6.23    2  0.0445
## Within groups  24.80   20  0.2091
## 
## Results for subgroups (random effects model):
##             k     SMD            95%-CI  tau^2    tau
## < 30 y      6  0.0413 [-0.3052; 0.3878]      0      0
## 30 - 50 y   7  0.4161 [-0.0086; 0.8408] 0.1264 0.3555
## > 50 y     10 -0.0661 [-0.2982; 0.1660] 0.0247 0.1572
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   3.81    2  0.1486
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
9.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 23; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0818 (SE = 0.0729)
## tau (square root of estimated tau^2 value):             0.2860
## I^2 (residual heterogeneity / unaccounted variability): 35.90%
## H^2 (unaccounted variability / sampling variability):   1.56
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 21) = 32.7590, p-val = 0.0490
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.3736, p-val = 0.5411
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    0.2877  0.3472   0.8287  0.4073  -0.3928  0.9683    
## age       -0.0044  0.0072  -0.6112  0.5411  -0.0184  0.0097    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
9.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

9.2.4 Training Duration

9.2.4.1 Forest plot

9.2.4.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_duration
## Abdelbasset 2020       -0.4098 [-1.1216; 0.3019]       4.5        5.0      5 - 10 weeks
## Ciolac 2010             0.4300 [-0.4153; 1.2754]       3.2        3.9        > 10 weeks
## Conraads 2015          -0.0612 [-0.3586; 0.2361]      25.6       10.5        > 10 weeks
## Currie 2015            -0.8003 [-1.7361; 0.1355]       2.6        3.4        > 10 weeks
## Eguchi 2012             0.2575 [-0.6227; 1.1376]       2.9        3.7        > 10 weeks
## Fisher 2015            -0.1711 [-0.9970; 0.6548]       3.3        4.1      5 - 10 weeks
## Helgerud 2007          -0.2220 [-1.1012; 0.6573]       2.9        3.7      5 - 10 weeks
## Honkala 2017 (Healthy)  1.0104 [ 0.2237; 1.7970]       3.7        4.3         < 5 weeks
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.3        3.1         < 5 weeks
## Jo 2020                 0.1476 [-0.5256; 0.8208]       5.0        5.3      5 - 10 weeks
## Keating 2014            0.6396 [-0.2172; 1.4964]       3.1        3.9        > 10 weeks
## Kim 2015                0.7581 [-0.0089; 1.5250]       3.8        4.5      5 - 10 weeks
## Madssen 2014            0.1632 [-0.5005; 0.8268]       5.1        5.4        > 10 weeks
## Maillard 2016          -0.6396 [-1.6443; 0.3651]       2.2        3.0        > 10 weeks
## Matsuo 2015            -0.3539 [-1.1603; 0.4525]       3.5        4.2      5 - 10 weeks
## Mitranun 2014           0.0053 [-0.7355; 0.7461]       4.1        4.7      5 - 10 weeks
## Motiani 2017            1.0575 [ 0.2367; 1.8782]       3.4        4.1         < 5 weeks
## Nalcakan 2014          -0.4452 [-1.4721; 0.5816]       2.1        2.9      5 - 10 weeks
## Ramos 2016b            -0.5073 [-1.2127; 0.1980]       4.5        5.0        > 10 weeks
## Sandvei 2012            0.6442 [-0.1948; 1.4833]       3.2        4.0      5 - 10 weeks
## Sawyer 2016            -0.1191 [-1.0439; 0.8056]       2.6        3.5      5 - 10 weeks
## Winn 2018               0.6875 [-0.3211; 1.6960]       2.2        3.0         < 5 weeks
## Zapata-Lamana 2018     -0.1122 [-0.8536; 0.6292]       4.1        4.7        > 10 weeks
## 
## Number of studies combined: k = 23
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0605 [-0.0898; 0.2108] 0.79  0.4303
## Random effects model 0.0846 [-0.1141; 0.2832] 0.83  0.4041
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0751 [0.0000; 0.3441]; tau = 0.2741 [0.0000; 0.5866];
##  I^2 = 34.5% [0.0%; 60.5%]; H = 1.24 [1.00; 1.59]
## 
## Quantifying residual heterogeneity:
##  I^2 = 4.5% [0.0%; 49.4%]; H = 1.02 [1.00; 1.41]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  33.59   22  0.0540
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI    Q   I^2
## < 5 weeks      4  0.7312 [ 0.2867; 1.1756] 3.00  0.0%
## 5 - 10 weeks  10  0.0059 [-0.2480; 0.2599] 8.75  0.0%
## > 10 weeks     9 -0.0529 [-0.2589; 0.1531] 9.20 13.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups 10.09    2  0.0065
## Within groups  20.94   20  0.4006
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI  tau^2    tau
## < 5 weeks      4  0.7312 [ 0.2867; 1.1756]      0      0
## 5 - 10 weeks  10  0.0059 [-0.2480; 0.2599]      0      0
## > 10 weeks     9 -0.0517 [-0.2919; 0.1885] 0.0181 0.1344
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   9.73    2  0.0077
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
9.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 23; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0269 (SE = 0.0533)
## tau (square root of estimated tau^2 value):             0.1641
## I^2 (residual heterogeneity / unaccounted variability): 15.71%
## H^2 (unaccounted variability / sampling variability):   1.19
## R^2 (amount of heterogeneity accounted for):            64.14%
## 
## Test for Residual Heterogeneity:
## QE(df = 21) = 24.9145, p-val = 0.2509
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 7.1406, p-val = 0.0075
## 
## Model Results:
## 
##           estimate      se     zval    pval    ci.lb    ci.ub 
## intrcpt     0.6429  0.2295   2.8020  0.0051   0.1932   1.0926  ** 
## duration   -0.0597  0.0223  -2.6722  0.0075  -0.1035  -0.0159  ** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
9.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

9.2.5 Men Ratio

9.2.5.1 Forest plot

9.2.5.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_men_ratio
## Abdelbasset 2020       -0.4098 [-1.1216; 0.3019]       4.5        5.0              > 0.5
## Ciolac 2010             0.4300 [-0.4153; 1.2754]       3.2        3.9              < 0.5
## Conraads 2015          -0.0612 [-0.3586; 0.2361]      25.6       10.5              > 0.5
## Currie 2015            -0.8003 [-1.7361; 0.1355]       2.6        3.4              > 0.5
## Eguchi 2012             0.2575 [-0.6227; 1.1376]       2.9        3.7              > 0.5
## Fisher 2015            -0.1711 [-0.9970; 0.6548]       3.3        4.1              > 0.5
## Helgerud 2007          -0.2220 [-1.1012; 0.6573]       2.9        3.7              > 0.5
## Honkala 2017 (Healthy)  1.0104 [ 0.2237; 1.7970]       3.7        4.3              > 0.5
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.3        3.1              > 0.5
## Jo 2020                 0.1476 [-0.5256; 0.8208]       5.0        5.3              > 0.5
## Keating 2014            0.6396 [-0.2172; 1.4964]       3.1        3.9              < 0.5
## Kim 2015                0.7581 [-0.0089; 1.5250]       3.8        4.5              > 0.5
## Madssen 2014            0.1632 [-0.5005; 0.8268]       5.1        5.4              > 0.5
## Maillard 2016          -0.6396 [-1.6443; 0.3651]       2.2        3.0              < 0.5
## Matsuo 2015            -0.3539 [-1.1603; 0.4525]       3.5        4.2              > 0.5
## Mitranun 2014           0.0053 [-0.7355; 0.7461]       4.1        4.7              < 0.5
## Motiani 2017            1.0575 [ 0.2367; 1.8782]       3.4        4.1              > 0.5
## Nalcakan 2014          -0.4452 [-1.4721; 0.5816]       2.1        2.9              > 0.5
## Ramos 2016b            -0.5073 [-1.2127; 0.1980]       4.5        5.0              > 0.5
## Sandvei 2012            0.6442 [-0.1948; 1.4833]       3.2        4.0              < 0.5
## Sawyer 2016            -0.1191 [-1.0439; 0.8056]       2.6        3.5              < 0.5
## Winn 2018               0.6875 [-0.3211; 1.6960]       2.2        3.0              < 0.5
## Zapata-Lamana 2018     -0.1122 [-0.8536; 0.6292]       4.1        4.7              < 0.5
## 
## Number of studies combined: k = 23
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0605 [-0.0898; 0.2108] 0.79  0.4303
## Random effects model 0.0846 [-0.1141; 0.2832] 0.83  0.4041
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0751 [0.0000; 0.3441]; tau = 0.2741 [0.0000; 0.5866];
##  I^2 = 34.5% [0.0%; 60.5%]; H = 1.24 [1.00; 1.59]
## 
## Quantifying residual heterogeneity:
##  I^2 = 30.3% [0.0%; 58.6%]; H = 1.20 [1.00; 1.55]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  33.59   22  0.0540
## 
## Results for subgroups (fixed effect model):
##         k    SMD            95%-CI     Q   I^2
## < 0.5   8 0.1835 [-0.1191; 0.4860]  6.66  0.0%
## > 0.5  15 0.0163 [-0.1572; 0.1899] 23.48 40.4%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.88    1  0.3477
## Within groups  30.15   21  0.0891
## 
## Results for subgroups (random effects model):
##         k    SMD            95%-CI  tau^2    tau
## < 0.5   8 0.1835 [-0.1191; 0.4860]      0      0
## > 0.5  15 0.0360 [-0.2102; 0.2822] 0.0874 0.2956
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.55    1  0.4588
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
9.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 23; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0864 (SE = 0.0743)
## tau (square root of estimated tau^2 value):             0.2939
## I^2 (residual heterogeneity / unaccounted variability): 37.38%
## H^2 (unaccounted variability / sampling variability):   1.60
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 21) = 33.5333, p-val = 0.0406
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0066, p-val = 0.9354
## 
## Model Results:
## 
##            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt      0.1020  0.2298   0.4438  0.6572  -0.3484  0.5524    
## men_ratio   -0.0248  0.3056  -0.0811  0.9354  -0.6238  0.5742    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
9.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

9.2.6 Type of Exercise

9.2.6.1 Forest plot

9.2.6.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) type_exercise
## Abdelbasset 2020       -0.4098 [-1.1216; 0.3019]       4.5        5.0       Cycling
## Ciolac 2010             0.4300 [-0.4153; 1.2754]       3.2        3.9       Running
## Conraads 2015          -0.0612 [-0.3586; 0.2361]      25.6       10.5       Cycling
## Currie 2015            -0.8003 [-1.7361; 0.1355]       2.6        3.4       Cycling
## Eguchi 2012             0.2575 [-0.6227; 1.1376]       2.9        3.7       Cycling
## Fisher 2015            -0.1711 [-0.9970; 0.6548]       3.3        4.1       Cycling
## Helgerud 2007          -0.2220 [-1.1012; 0.6573]       2.9        3.7       Running
## Honkala 2017 (Healthy)  1.0104 [ 0.2237; 1.7970]       3.7        4.3       Cycling
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.3        3.1       Cycling
## Jo 2020                 0.1476 [-0.5256; 0.8208]       5.0        5.3       Running
## Keating 2014            0.6396 [-0.2172; 1.4964]       3.1        3.9       Cycling
## Kim 2015                0.7581 [-0.0089; 1.5250]       3.8        4.5       Running
## Madssen 2014            0.1632 [-0.5005; 0.8268]       5.1        5.4       Running
## Maillard 2016          -0.6396 [-1.6443; 0.3651]       2.2        3.0       Cycling
## Matsuo 2015            -0.3539 [-1.1603; 0.4525]       3.5        4.2       Cycling
## Mitranun 2014           0.0053 [-0.7355; 0.7461]       4.1        4.7       Running
## Motiani 2017            1.0575 [ 0.2367; 1.8782]       3.4        4.1       Cycling
## Nalcakan 2014          -0.4452 [-1.4721; 0.5816]       2.1        2.9       Cycling
## Ramos 2016b            -0.5073 [-1.2127; 0.1980]       4.5        5.0       Running
## Sandvei 2012            0.6442 [-0.1948; 1.4833]       3.2        4.0       Running
## Sawyer 2016            -0.1191 [-1.0439; 0.8056]       2.6        3.5       Cycling
## Winn 2018               0.6875 [-0.3211; 1.6960]       2.2        3.0       Running
## Zapata-Lamana 2018     -0.1122 [-0.8536; 0.6292]       4.1        4.7       Cycling
## 
## Number of studies combined: k = 23
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0605 [-0.0898; 0.2108] 0.79  0.4303
## Random effects model 0.0846 [-0.1141; 0.2832] 0.83  0.4041
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0751 [0.0000; 0.3441]; tau = 0.2741 [0.0000; 0.5866];
##  I^2 = 34.5% [0.0%; 60.5%]; H = 1.24 [1.00; 1.59]
## 
## Quantifying residual heterogeneity:
##  I^2 = 29.2% [0.0%; 57.9%]; H = 1.19 [1.00; 1.54]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  33.59   22  0.0540
## 
## Results for subgroups (fixed effect model):
##           k     SMD            95%-CI     Q   I^2
## Cycling  14 -0.0071 [-0.1926; 0.1784] 20.98 38.0%
## Running   9  0.1827 [-0.0749; 0.4404]  8.68  7.8%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  1.37    1  0.2412
## Within groups  29.65   21  0.0991
## 
## Results for subgroups (random effects model):
##           k    SMD            95%-CI  tau^2    tau
## Cycling  14 0.0073 [-0.2563; 0.2708] 0.0873 0.2955
## Running   9 0.1857 [-0.0833; 0.4547] 0.0132 0.1151
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.86    1  0.3531
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
9.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 23; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0777 (SE = 0.0716)
## tau (square root of estimated tau^2 value):             0.2787
## I^2 (residual heterogeneity / unaccounted variability): 34.57%
## H^2 (unaccounted variability / sampling variability):   1.53
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 21) = 32.0959, p-val = 0.0573
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.8963, p-val = 0.3438
## 
## Model Results:
## 
##                       estimate      se    zval    pval    ci.lb   ci.ub 
## intrcpt                 0.0065  0.1313  0.0496  0.9604  -0.2508  0.2638    
## type_exerciseRunning    0.1973  0.2084  0.9467  0.3438  -0.2112  0.6059    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
9.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

9.2.7 Baseline Values

9.2.7.1 Forest plot

9.2.7.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_bsln
## Abdelbasset 2020       -0.4098 [-1.1216; 0.3019]       4.5        5.0    < 3 mmol/L
## Ciolac 2010             0.4300 [-0.4153; 1.2754]       3.2        3.9    < 3 mmol/L
## Conraads 2015          -0.0612 [-0.3586; 0.2361]      25.6       10.5    < 3 mmol/L
## Currie 2015            -0.8003 [-1.7361; 0.1355]       2.6        3.4    < 3 mmol/L
## Eguchi 2012             0.2575 [-0.6227; 1.1376]       2.9        3.7    > 3 mmol/L
## Fisher 2015            -0.1711 [-0.9970; 0.6548]       3.3        4.1    < 3 mmol/L
## Helgerud 2007          -0.2220 [-1.1012; 0.6573]       2.9        3.7    < 3 mmol/L
## Honkala 2017 (Healthy)  1.0104 [ 0.2237; 1.7970]       3.7        4.3    > 3 mmol/L
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.3        3.1    < 3 mmol/L
## Jo 2020                 0.1476 [-0.5256; 0.8208]       5.0        5.3    < 3 mmol/L
## Keating 2014            0.6396 [-0.2172; 1.4964]       3.1        3.9    > 3 mmol/L
## Kim 2015                0.7581 [-0.0089; 1.5250]       3.8        4.5    > 3 mmol/L
## Madssen 2014            0.1632 [-0.5005; 0.8268]       5.1        5.4    < 3 mmol/L
## Maillard 2016          -0.6396 [-1.6443; 0.3651]       2.2        3.0    < 3 mmol/L
## Matsuo 2015            -0.3539 [-1.1603; 0.4525]       3.5        4.2    > 3 mmol/L
## Mitranun 2014           0.0053 [-0.7355; 0.7461]       4.1        4.7    > 3 mmol/L
## Motiani 2017            1.0575 [ 0.2367; 1.8782]       3.4        4.1    > 3 mmol/L
## Nalcakan 2014          -0.4452 [-1.4721; 0.5816]       2.1        2.9    < 3 mmol/L
## Ramos 2016b            -0.5073 [-1.2127; 0.1980]       4.5        5.0    < 3 mmol/L
## Sandvei 2012            0.6442 [-0.1948; 1.4833]       3.2        4.0    < 3 mmol/L
## Sawyer 2016            -0.1191 [-1.0439; 0.8056]       2.6        3.5    < 3 mmol/L
## Winn 2018               0.6875 [-0.3211; 1.6960]       2.2        3.0    < 3 mmol/L
## Zapata-Lamana 2018     -0.1122 [-0.8536; 0.6292]       4.1        4.7    < 3 mmol/L
## 
## Number of studies combined: k = 23
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0605 [-0.0898; 0.2108] 0.79  0.4303
## Random effects model 0.0846 [-0.1141; 0.2832] 0.83  0.4041
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0751 [0.0000; 0.3441]; tau = 0.2741 [0.0000; 0.5866];
##  I^2 = 34.5% [0.0%; 60.5%]; H = 1.24 [1.00; 1.59]
## 
## Quantifying residual heterogeneity:
##  I^2 = 5.3% [0.0%; 37.6%]; H = 1.03 [1.00; 1.27]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  33.59   22  0.0540
## 
## Results for subgroups (fixed effect model):
##              k     SMD            95%-CI     Q   I^2
## < 3 mmol/L  16 -0.0719 [-0.2451; 0.1012] 12.64  0.0%
## > 3 mmol/L   7  0.4601 [ 0.1552; 0.7651]  9.54 37.1%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  8.84    1  0.0029
## Within groups  22.18   21  0.3890
## 
## Results for subgroups (random effects model):
##              k     SMD            95%-CI  tau^2    tau
## < 3 mmol/L  16 -0.0719 [-0.2451; 0.1012]      0      0
## > 3 mmol/L   7  0.4630 [ 0.0777; 0.8483] 0.1002 0.3166
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   6.16    1  0.0131
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
9.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 23; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0351 (SE = 0.0601)
## tau (square root of estimated tau^2 value):             0.1874
## I^2 (residual heterogeneity / unaccounted variability): 18.11%
## H^2 (unaccounted variability / sampling variability):   1.22
## R^2 (amount of heterogeneity accounted for):            53.24%
## 
## Test for Residual Heterogeneity:
## QE(df = 21) = 25.6426, p-val = 0.2204
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 7.1423, p-val = 0.0075
## 
## Model Results:
## 
##                estimate      se     zval    pval    ci.lb    ci.ub 
## intrcpt         -1.3340  0.5365  -2.4867  0.0129  -2.3854  -0.2826   * 
## bsln_adjusted    0.5284  0.1977   2.6725  0.0075   0.1409   0.9160  ** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
9.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

9.2.8 Type of HIIE

9.2.8.1 Forest plot

9.2.8.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) HIIE
## Abdelbasset 2020       -0.4098 [-1.1216; 0.3019]       4.5        5.0 HIIT
## Ciolac 2010             0.4300 [-0.4153; 1.2754]       3.2        3.9 HIIT
## Conraads 2015          -0.0612 [-0.3586; 0.2361]      25.6       10.5 HIIT
## Currie 2015            -0.8003 [-1.7361; 0.1355]       2.6        3.4 HIIT
## Eguchi 2012             0.2575 [-0.6227; 1.1376]       2.9        3.7 HIIT
## Fisher 2015            -0.1711 [-0.9970; 0.6548]       3.3        4.1  SIT
## Helgerud 2007          -0.2220 [-1.1012; 0.6573]       2.9        3.7 HIIT
## Honkala 2017 (Healthy)  1.0104 [ 0.2237; 1.7970]       3.7        4.3  SIT
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.3        3.1  SIT
## Jo 2020                 0.1476 [-0.5256; 0.8208]       5.0        5.3 HIIT
## Keating 2014            0.6396 [-0.2172; 1.4964]       3.1        3.9 HIIT
## Kim 2015                0.7581 [-0.0089; 1.5250]       3.8        4.5 HIIT
## Madssen 2014            0.1632 [-0.5005; 0.8268]       5.1        5.4 HIIT
## Maillard 2016          -0.6396 [-1.6443; 0.3651]       2.2        3.0 HIIT
## Matsuo 2015            -0.3539 [-1.1603; 0.4525]       3.5        4.2 HIIT
## Mitranun 2014           0.0053 [-0.7355; 0.7461]       4.1        4.7 HIIT
## Motiani 2017            1.0575 [ 0.2367; 1.8782]       3.4        4.1  SIT
## Nalcakan 2014          -0.4452 [-1.4721; 0.5816]       2.1        2.9  SIT
## Ramos 2016b            -0.5073 [-1.2127; 0.1980]       4.5        5.0 HIIT
## Sandvei 2012            0.6442 [-0.1948; 1.4833]       3.2        4.0  SIT
## Sawyer 2016            -0.1191 [-1.0439; 0.8056]       2.6        3.5 HIIT
## Winn 2018               0.6875 [-0.3211; 1.6960]       2.2        3.0 HIIT
## Zapata-Lamana 2018     -0.1122 [-0.8536; 0.6292]       4.1        4.7 HIIT
## 
## Number of studies combined: k = 23
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0605 [-0.0898; 0.2108] 0.79  0.4303
## Random effects model 0.0846 [-0.1141; 0.2832] 0.83  0.4041
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0751 [0.0000; 0.3441]; tau = 0.2741 [0.0000; 0.5866];
##  I^2 = 34.5% [0.0%; 60.5%]; H = 1.24 [1.00; 1.59]
## 
## Quantifying residual heterogeneity:
##  I^2 = 19.9% [0.0%; 52.3%]; H = 1.12 [1.00; 1.45]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  33.59   22  0.0540
## 
## Results for subgroups (fixed effect model):
##        k     SMD            95%-CI     Q   I^2
## HIIT  17 -0.0210 [-0.1872; 0.1451] 16.84  5.0%
## SIT    6  0.4182 [ 0.0627; 0.7738]  9.38 46.7%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  4.81    1  0.0283
## Within groups  26.22   21  0.1983
## 
## Results for subgroups (random effects model):
##        k     SMD            95%-CI  tau^2    tau
## HIIT  17 -0.0184 [-0.1936; 0.1569] 0.0069 0.0831
## SIT    6  0.3846 [-0.1053; 0.8746] 0.1741 0.4173
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   2.30    1  0.1291
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
9.2.8.3 Meta-regression
## 
## Mixed-Effects Model (k = 23; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0506 (SE = 0.0611)
## tau (square root of estimated tau^2 value):             0.2249
## I^2 (residual heterogeneity / unaccounted variability): 26.12%
## H^2 (unaccounted variability / sampling variability):   1.35
## R^2 (amount of heterogeneity accounted for):            32.65%
## 
## Test for Residual Heterogeneity:
## QE(df = 21) = 28.4236, p-val = 0.1285
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 3.4823, p-val = 0.0620
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.0120  0.1076  -0.1119  0.9109  -0.2230  0.1989    
## HIIESIT    0.4295  0.2301   1.8661  0.0620  -0.0216  0.8805  . 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
9.2.8.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

10. Triglycerides

10.1 Overall

10.1.1 Forest plot

10.1.2 R output

##                            SMD            95%-CI %W(fixed) %W(random)
## Abdelbasset 2020        0.2323 [-0.4745; 0.9391]       3.9        3.9
## Ciolac 2010             0.1845 [-0.6530; 1.0220]       2.8        2.8
## Conraads 2015          -0.0242 [-0.3215; 0.2730]      22.0       22.0
## Currie 2015             0.3256 [-0.5809; 1.2321]       2.4        2.4
## Eguchi 2012            -0.6465 [-1.5457; 0.2526]       2.4        2.4
## Fisher 2015            -0.0086 [-0.8330; 0.8159]       2.9        2.9
## Helgerud 2007          -0.1255 [-1.0029; 0.7519]       2.5        2.5
## Honkala 2017 (Healthy) -0.4795 [-1.2309; 0.2719]       3.4        3.4
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.0        2.0
## Jo 2020                 0.4589 [-0.2222; 1.1399]       4.2        4.2
## Keating 2014           -0.2384 [-1.0771; 0.6003]       2.8        2.8
## Kim 2015                0.4689 [-0.2820; 1.2198]       3.5        3.5
## Lunt 2014              -0.2087 [-1.0482; 0.6309]       2.8        2.8
## Lunt 2014              -0.7385 [-1.6026; 0.1257]       2.6        2.6
## Madssen 2014            0.4002 [-0.2688; 1.0692]       4.3        4.3
## Maillard 2016          -0.1685 [-1.1503; 0.8132]       2.0        2.0
## Matsuo 2015             0.6403 [-0.1801; 1.4607]       2.9        2.9
## Mitranun 2014           0.0035 [-0.7373; 0.7443]       3.5        3.5
## Motiani 2017           -0.6864 [-1.4775; 0.1046]       3.1        3.1
## Nalcakan 2014           0.4655 [-0.5624; 1.4935]       1.8        1.8
## Ramos 2016a             0.0347 [-0.5633; 0.6327]       5.4        5.4
## Ramos 2016b             0.2851 [-0.4127; 0.9829]       4.0        4.0
## Sandvei 2012            0.6305 [-0.2077; 1.4686]       2.8        2.8
## Sawyer 2016             0.6031 [-0.3416; 1.5478]       2.2        2.2
## Tjønna 2008             0.2944 [-0.6211; 1.2100]       2.3        2.3
## Winn 2018              -0.4553 [-1.4479; 0.5373]       2.0        2.0
## Zapata-Lamana 2018     -0.4783 [-1.2297; 0.2730]       3.4        3.4
## 
## Number of studies combined: k = 27
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0300 [-0.1095; 0.1695] 0.42  0.6736
## Random effects model 0.0300 [-0.1095; 0.1695] 0.42  0.6736
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1436]; tau = 0 [0.0000; 0.3789];
##  I^2 = 0.0% [0.0%; 42.2%]; H = 1.00 [1.00; 1.32]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  25.83   26  0.4723
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

10.1.3 Sensitivity analysis

10.1.3.1 Forest plot

10.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                                      SMD            95%-CI p-value   tau^2     tau   I^2
## Omitting Abdelbasset 2020         0.0212 [-0.1212; 0.1637]  0.7701  0.0000  0.0000  0.0%
## Omitting Ciolac 2010              0.0250 [-0.1166; 0.1666]  0.7294  0.0000  0.0000  0.0%
## Omitting Conraads 2015            0.0444 [-0.1138; 0.2025]  0.5826  0.0000  0.0000  0.0%
## Omitting Currie 2015              0.0224 [-0.1189; 0.1637]  0.7561  0.0000  0.0000  0.0%
## Omitting Eguchi 2012              0.0452 [-0.0962; 0.1865]  0.5313  0.0000  0.0000  0.0%
## Omitting Fisher 2015              0.0303 [-0.1114; 0.1720]  0.6747  0.0000  0.0000  0.0%
## Omitting Helgerud 2007            0.0331 [-0.1083; 0.1746]  0.6463  0.0000  0.0000  0.0%
## Omitting Honkala 2017 (Healthy)   0.0469 [-0.0952; 0.1890]  0.5178  0.0000  0.0000  0.0%
## Omitting Honkala 2017 (T2D)       0.0298 [-0.1112; 0.1709]  0.6785  0.0000  0.0000  0.0%
## Omitting Jo 2020                  0.0109 [-0.1318; 0.1536]  0.8810  0.0000  0.0000  0.0%
## Omitting Keating 2014             0.0366 [-0.1050; 0.1782]  0.6125  0.0000  0.0000  0.0%
## Omitting Kim 2015                 0.0140 [-0.1281; 0.1561]  0.8467  0.0000  0.0000  0.0%
## Omitting Lunt 2014                0.0358 [-0.1058; 0.1774]  0.6204  0.0000  0.0000  0.0%
## Omitting Lunt 2014                0.0490 [-0.0925; 0.1905]  0.4977  0.0000  0.0000  0.0%
## Omitting Madssen 2014             0.0128 [-0.1300; 0.1556]  0.8609  0.0000  0.0000  0.0%
## Omitting Maillard 2016            0.0331 [-0.1080; 0.1742]  0.6453  0.0000  0.0000  0.0%
## Omitting Matsuo 2015              0.0118 [-0.1299; 0.1535]  0.8707  0.0000  0.0000  0.0%
## Omitting Mitranun 2014            0.0302 [-0.1120; 0.1724]  0.6773  0.0000  0.0000  0.0%
## Omitting Motiani 2017             0.0514 [-0.0904; 0.1933]  0.4774  0.0000  0.0000  0.0%
## Omitting Nalcakan 2014            0.0216 [-0.1194; 0.1625]  0.7640  0.0000  0.0000  0.0%
## Omitting Ramos 2016a              0.0290 [-0.1147; 0.1726]  0.6927  0.0000  0.0000  0.0%
## Omitting Ramos 2016b              0.0189 [-0.1237; 0.1614]  0.7952  0.0000  0.0000  0.0%
## Omitting Sandvei 2012             0.0128 [-0.1288; 0.1544]  0.8592  0.0000  0.0000  0.0%
## Omitting Sawyer 2016              0.0171 [-0.1240; 0.1583]  0.8119  0.0000  0.0000  0.0%
## Omitting Tjønna 2008              0.0232 [-0.1181; 0.1645]  0.7471  0.0000  0.0000  0.0%
## Omitting Winn 2018                0.0385 [-0.1026; 0.1795]  0.5929  0.0000  0.0000  0.0%
## Omitting Zapata-Lamana 2018       0.0469 [-0.0953; 0.1890]  0.5182  0.0000  0.0000  0.0%
##                                                                                         
## Pooled estimate                   0.0300 [-0.1095; 0.1695]  0.6736  0.0000  0.0000  0.0%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

10.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

10.2 Subgroups

10.2.1 Overall

10.2.1.1 Forest plot

10.2.1.2 R output
##                            SMD            95%-CI     meta-analysis
##                         0.0300 [-0.1095; 0.1695]           Overall
## Healthy                -0.1211 [-0.5006; 0.2583]        Population
## Overweight/obese       -0.2294 [-0.5533; 0.0946]        Population
## Cardiac Rehabilitation  0.1078 [-0.1381; 0.3538]        Population
## Metabolic Syndrome      0.2977 [-0.0239; 0.6193]        Population
## T2D                     0.0499 [-0.3623; 0.4621]        Population
## < 30 y                  0.0645 [-0.2822; 0.4112]               Age
## 30 - 50 y              -0.1862 [-0.5057; 0.1332]               Age
## > 50 y                  0.1065 [-0.0737; 0.2866]               Age
## < 5 weeks              -0.4292 [-0.8599; 0.0016] Training Duration
## 5 - 10 weeks            0.3160 [ 0.0616; 0.5703] Training Duration
## > 10 weeks             -0.0352 [-0.2165; 0.1460] Training Duration
## < 0.5                  -0.0592 [-0.3178; 0.1995]         Men Ratio
## > 0.5                   0.0656 [-0.1003; 0.2315]         Men Ratio
## Cycling                -0.0491 [-0.2336; 0.1355]  Type of Exercise
## Running                 0.1342 [-0.0795; 0.3478]  Type of Exercise
## < 1.7 mmol/L           -0.0191 [-0.1802; 0.1420]   Baseline Values
## > 1.7 mmol/L            0.1755 [-0.1047; 0.4558]   Baseline Values
## HIIT                    0.0734 [-0.0814; 0.2282]      Type of HIIE
## SIT                    -0.1506 [-0.5374; 0.2361]      Type of HIIE
## 
## Number of studies combined: k = 27
## 
##                         SMD            95%-CI    z p-value
## Random effects model 0.0300 [-0.1095; 0.1695] 0.42  0.6736
## 
## Quantifying heterogeneity:
##  tau^2 = 0; tau = 0; I^2 = 0.0% [0.0%; 42.2%]; H = 1.00 [1.00; 1.32]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  25.83   26  0.4723
## 
## Results for meta-analyses (random effects model):
##                     k    SMD            95%-CI tau^2 tau     Q  I^2
## Overall            27 0.0300 [-0.1095; 0.1695]     0   0 25.83 0.0%
## Population         27 0.0300 [-0.1095; 0.1695]     0   0 25.83 0.0%
## Age                27 0.0300 [-0.1095; 0.1695]     0   0 25.83 0.0%
## Training Duration  27 0.0300 [-0.1095; 0.1695]     0   0 25.83 0.0%
## Men Ratio          27 0.0300 [-0.1095; 0.1695]     0   0 25.83 0.0%
## Type of Exercise   27 0.0300 [-0.1095; 0.1695]     0   0 25.83 0.0%
## Baseline Values    27 0.0300 [-0.1095; 0.1695]     0   0 25.83 0.0%
## Type of HIIE       27 0.0300 [-0.1095; 0.1695]     0   0 25.83 0.0%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

10.2.2 Population

10.2.2.1 Forest plot

10.2.2.2 R output
##                            SMD            95%-CI %W(fixed) %W(random)             population
## Abdelbasset 2020        0.2323 [-0.4745; 0.9391]       3.9        3.9                    T2D
## Ciolac 2010             0.1845 [-0.6530; 1.0220]       2.8        2.8                Healthy
## Conraads 2015          -0.0242 [-0.3215; 0.2730]      22.0       22.0 Cardiac Rehabilitation
## Currie 2015             0.3256 [-0.5809; 1.2321]       2.4        2.4 Cardiac Rehabilitation
## Eguchi 2012            -0.6465 [-1.5457; 0.2526]       2.4        2.4                Healthy
## Fisher 2015            -0.0086 [-0.8330; 0.8159]       2.9        2.9       Overweight/obese
## Helgerud 2007          -0.1255 [-1.0029; 0.7519]       2.5        2.5                Healthy
## Honkala 2017 (Healthy) -0.4795 [-1.2309; 0.2719]       3.4        3.4                Healthy
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.0        2.0                    T2D
## Jo 2020                 0.4589 [-0.2222; 1.1399]       4.2        4.2     Metabolic Syndrome
## Keating 2014           -0.2384 [-1.0771; 0.6003]       2.8        2.8       Overweight/obese
## Kim 2015                0.4689 [-0.2820; 1.2198]       3.5        3.5 Cardiac Rehabilitation
## Lunt 2014              -0.2087 [-1.0482; 0.6309]       2.8        2.8       Overweight/obese
## Lunt 2014              -0.7385 [-1.6026; 0.1257]       2.6        2.6       Overweight/obese
## Madssen 2014            0.4002 [-0.2688; 1.0692]       4.3        4.3 Cardiac Rehabilitation
## Maillard 2016          -0.1685 [-1.1503; 0.8132]       2.0        2.0                    T2D
## Matsuo 2015             0.6403 [-0.1801; 1.4607]       2.9        2.9     Metabolic Syndrome
## Mitranun 2014           0.0035 [-0.7373; 0.7443]       3.5        3.5                    T2D
## Motiani 2017           -0.6864 [-1.4775; 0.1046]       3.1        3.1                Healthy
## Nalcakan 2014           0.4655 [-0.5624; 1.4935]       1.8        1.8                Healthy
## Ramos 2016a             0.0347 [-0.5633; 0.6327]       5.4        5.4     Metabolic Syndrome
## Ramos 2016b             0.2851 [-0.4127; 0.9829]       4.0        4.0     Metabolic Syndrome
## Sandvei 2012            0.6305 [-0.2077; 1.4686]       2.8        2.8                Healthy
## Sawyer 2016             0.6031 [-0.3416; 1.5478]       2.2        2.2       Overweight/obese
## Tjønna 2008             0.2944 [-0.6211; 1.2100]       2.3        2.3     Metabolic Syndrome
## Winn 2018              -0.4553 [-1.4479; 0.5373]       2.0        2.0       Overweight/obese
## Zapata-Lamana 2018     -0.4783 [-1.2297; 0.2730]       3.4        3.4       Overweight/obese
## 
## Number of studies combined: k = 27
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0300 [-0.1095; 0.1695] 0.42  0.6736
## Random effects model 0.0300 [-0.1095; 0.1695] 0.42  0.6736
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1436]; tau = 0 [0.0000; 0.3789];
##  I^2 = 0.0% [0.0%; 42.2%]; H = 1.00 [1.00; 1.32]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 31.2%]; H = 1.00 [1.00; 1.21]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  25.83   26  0.4723
## 
## Results for subgroups (fixed effect model):
##                          k     SMD            95%-CI    Q   I^2
## Healthy                  7 -0.1314 [-0.4531; 0.1903] 8.28 27.6%
## Overweight/obese         7 -0.2294 [-0.5533; 0.0946] 4.76  0.0%
## Cardiac Rehabilitation   4  0.1078 [-0.1381; 0.3538] 2.46  0.0%
## Metabolic Syndrome       5  0.2977 [-0.0239; 0.6193] 1.52  0.0%
## T2D                      4  0.0499 [-0.3623; 0.4621] 0.44  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  6.48    4  0.1658
## Within groups  17.46   22  0.7373
## 
## Results for subgroups (random effects model):
##                          k     SMD            95%-CI  tau^2    tau
## Healthy                  7 -0.1211 [-0.5006; 0.2583] 0.0721 0.2685
## Overweight/obese         7 -0.2294 [-0.5533; 0.0946]      0      0
## Cardiac Rehabilitation   4  0.1078 [-0.1381; 0.3538]      0      0
## Metabolic Syndrome       5  0.2977 [-0.0239; 0.6193]      0      0
## T2D                      4  0.0499 [-0.3623; 0.4621]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   6.11    4  0.1912
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
10.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 27; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0473)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 22) = 18.9030, p-val = 0.6513
## 
## Test of Moderators (coefficients 2:5):
## QM(df = 4) = 6.9308, p-val = 0.1396
## 
## Model Results:
## 
##                               estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                        -0.1348  0.1638  -0.8231  0.4105  -0.4559  0.1862    
## .byvarOverweight/obese         -0.1027  0.2325  -0.4417  0.6587  -0.5585  0.3531    
## .byvarCardiac Rehabilitation    0.2465  0.2063   1.1948  0.2322  -0.1579  0.6509    
## .byvarMetabolic Syndrome        0.4419  0.2318   1.9069  0.0565  -0.0123  0.8962  . 
## .byvarT2D                       0.1852  0.2666   0.6948  0.4872  -0.3373  0.7077    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
10.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

10.2.3 Age

10.2.3.1 Forest plot

10.2.3.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_age
## Abdelbasset 2020        0.2323 [-0.4745; 0.9391]       3.9        3.9       > 50 y
## Ciolac 2010             0.1845 [-0.6530; 1.0220]       2.8        2.8       < 30 y
## Conraads 2015          -0.0242 [-0.3215; 0.2730]      22.0       22.0       > 50 y
## Currie 2015             0.3256 [-0.5809; 1.2321]       2.4        2.4       > 50 y
## Eguchi 2012            -0.6465 [-1.5457; 0.2526]       2.4        2.4       > 50 y
## Fisher 2015            -0.0086 [-0.8330; 0.8159]       2.9        2.9       < 30 y
## Helgerud 2007          -0.1255 [-1.0029; 0.7519]       2.5        2.5       < 30 y
## Honkala 2017 (Healthy) -0.4795 [-1.2309; 0.2719]       3.4        3.4    30 - 50 y
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.0        2.0    30 - 50 y
## Jo 2020                 0.4589 [-0.2222; 1.1399]       4.2        4.2       > 50 y
## Keating 2014           -0.2384 [-1.0771; 0.6003]       2.8        2.8    30 - 50 y
## Kim 2015                0.4689 [-0.2820; 1.2198]       3.5        3.5       > 50 y
## Lunt 2014              -0.2087 [-1.0482; 0.6309]       2.8        2.8    30 - 50 y
## Lunt 2014              -0.7385 [-1.6026; 0.1257]       2.6        2.6    30 - 50 y
## Madssen 2014            0.4002 [-0.2688; 1.0692]       4.3        4.3       > 50 y
## Maillard 2016          -0.1685 [-1.1503; 0.8132]       2.0        2.0       > 50 y
## Matsuo 2015             0.6403 [-0.1801; 1.4607]       2.9        2.9    30 - 50 y
## Mitranun 2014           0.0035 [-0.7373; 0.7443]       3.5        3.5       > 50 y
## Motiani 2017           -0.6864 [-1.4775; 0.1046]       3.1        3.1    30 - 50 y
## Nalcakan 2014           0.4655 [-0.5624; 1.4935]       1.8        1.8       < 30 y
## Ramos 2016a             0.0347 [-0.5633; 0.6327]       5.4        5.4       > 50 y
## Ramos 2016b             0.2851 [-0.4127; 0.9829]       4.0        4.0       > 50 y
## Sandvei 2012            0.6305 [-0.2077; 1.4686]       2.8        2.8       < 30 y
## Sawyer 2016             0.6031 [-0.3416; 1.5478]       2.2        2.2    30 - 50 y
## Tjønna 2008             0.2944 [-0.6211; 1.2100]       2.3        2.3       > 50 y
## Winn 2018              -0.4553 [-1.4479; 0.5373]       2.0        2.0    30 - 50 y
## Zapata-Lamana 2018     -0.4783 [-1.2297; 0.2730]       3.4        3.4       < 30 y
## 
## Number of studies combined: k = 27
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0300 [-0.1095; 0.1695] 0.42  0.6736
## Random effects model 0.0300 [-0.1095; 0.1695] 0.42  0.6736
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1436]; tau = 0 [0.0000; 0.3789];
##  I^2 = 0.0% [0.0%; 42.2%]; H = 1.00 [1.00; 1.32]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 35.7%]; H = 1.00 [1.00; 1.25]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  25.83   26  0.4723
## 
## Results for subgroups (fixed effect model):
##             k     SMD            95%-CI    Q   I^2
## < 30 y      6  0.0645 [-0.2822; 0.4112] 4.28  0.0%
## 30 - 50 y   9 -0.1908 [-0.4777; 0.0961] 9.86 18.9%
## > 50 y     12  0.1065 [-0.0737; 0.2866] 6.80  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  3.01    2  0.2225
## Within groups  20.94   24  0.6422
## 
## Results for subgroups (random effects model):
##             k     SMD            95%-CI  tau^2    tau
## < 30 y      6  0.0645 [-0.2822; 0.4112]      0      0
## 30 - 50 y   9 -0.1862 [-0.5057; 0.1332] 0.0451 0.2123
## > 50 y     12  0.1065 [-0.0737; 0.2866]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   2.47    2  0.2911
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
10.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 27; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0047 (SE = 0.0422)
## tau (square root of estimated tau^2 value):             0.0683
## I^2 (residual heterogeneity / unaccounted variability): 3.13%
## H^2 (unaccounted variability / sampling variability):   1.03
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 25) = 25.8076, p-val = 0.4179
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0285, p-val = 0.8659
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.0154  0.2863  -0.0538  0.9571  -0.5765  0.5457    
## age        0.0010  0.0057   0.1689  0.8659  -0.0101  0.0120    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
10.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

10.2.4 Training Duration

10.2.4.1 Forest plot

10.2.4.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_duration
## Abdelbasset 2020        0.2323 [-0.4745; 0.9391]       3.9        3.9      5 - 10 weeks
## Ciolac 2010             0.1845 [-0.6530; 1.0220]       2.8        2.8        > 10 weeks
## Conraads 2015          -0.0242 [-0.3215; 0.2730]      22.0       22.0        > 10 weeks
## Currie 2015             0.3256 [-0.5809; 1.2321]       2.4        2.4        > 10 weeks
## Eguchi 2012            -0.6465 [-1.5457; 0.2526]       2.4        2.4        > 10 weeks
## Fisher 2015            -0.0086 [-0.8330; 0.8159]       2.9        2.9      5 - 10 weeks
## Helgerud 2007          -0.1255 [-1.0029; 0.7519]       2.5        2.5      5 - 10 weeks
## Honkala 2017 (Healthy) -0.4795 [-1.2309; 0.2719]       3.4        3.4         < 5 weeks
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.0        2.0         < 5 weeks
## Jo 2020                 0.4589 [-0.2222; 1.1399]       4.2        4.2      5 - 10 weeks
## Keating 2014           -0.2384 [-1.0771; 0.6003]       2.8        2.8        > 10 weeks
## Kim 2015                0.4689 [-0.2820; 1.2198]       3.5        3.5      5 - 10 weeks
## Lunt 2014              -0.2087 [-1.0482; 0.6309]       2.8        2.8        > 10 weeks
## Lunt 2014              -0.7385 [-1.6026; 0.1257]       2.6        2.6        > 10 weeks
## Madssen 2014            0.4002 [-0.2688; 1.0692]       4.3        4.3        > 10 weeks
## Maillard 2016          -0.1685 [-1.1503; 0.8132]       2.0        2.0        > 10 weeks
## Matsuo 2015             0.6403 [-0.1801; 1.4607]       2.9        2.9      5 - 10 weeks
## Mitranun 2014           0.0035 [-0.7373; 0.7443]       3.5        3.5      5 - 10 weeks
## Motiani 2017           -0.6864 [-1.4775; 0.1046]       3.1        3.1         < 5 weeks
## Nalcakan 2014           0.4655 [-0.5624; 1.4935]       1.8        1.8      5 - 10 weeks
## Ramos 2016a             0.0347 [-0.5633; 0.6327]       5.4        5.4        > 10 weeks
## Ramos 2016b             0.2851 [-0.4127; 0.9829]       4.0        4.0        > 10 weeks
## Sandvei 2012            0.6305 [-0.2077; 1.4686]       2.8        2.8      5 - 10 weeks
## Sawyer 2016             0.6031 [-0.3416; 1.5478]       2.2        2.2      5 - 10 weeks
## Tjønna 2008             0.2944 [-0.6211; 1.2100]       2.3        2.3        > 10 weeks
## Winn 2018              -0.4553 [-1.4479; 0.5373]       2.0        2.0         < 5 weeks
## Zapata-Lamana 2018     -0.4783 [-1.2297; 0.2730]       3.4        3.4        > 10 weeks
## 
## Number of studies combined: k = 27
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0300 [-0.1095; 0.1695] 0.42  0.6736
## Random effects model 0.0300 [-0.1095; 0.1695] 0.42  0.6736
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1436]; tau = 0 [0.0000; 0.3789];
##  I^2 = 0.0% [0.0%; 42.2%]; H = 1.00 [1.00; 1.32]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 5.3%]; H = 1.00 [1.00; 1.03]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  25.83   26  0.4723
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI    Q  I^2
## < 5 weeks      4 -0.4292 [-0.8599; 0.0016] 1.07 0.0%
## 5 - 10 weeks  10  0.3160 [ 0.0616; 0.5703] 3.89 0.0%
## > 10 weeks    13 -0.0352 [-0.2165; 0.1460] 9.27 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  9.72    2  0.0078
## Within groups  14.23   24  0.9414
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI tau^2 tau
## < 5 weeks      4 -0.4292 [-0.8599; 0.0016]     0   0
## 5 - 10 weeks  10  0.3160 [ 0.0616; 0.5703]     0   0
## > 10 weeks    13 -0.0352 [-0.2165; 0.1460]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   9.72    2  0.0078
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
10.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 27; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0031 (SE = 0.0413)
## tau (square root of estimated tau^2 value):             0.0558
## I^2 (residual heterogeneity / unaccounted variability): 2.13%
## H^2 (unaccounted variability / sampling variability):   1.02
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 25) = 25.5452, p-val = 0.4322
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.2906, p-val = 0.5898
## 
## Model Results:
## 
##           estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    -0.0705  0.2017  -0.3495  0.7267  -0.4657  0.3248    
## duration    0.0098  0.0182   0.5391  0.5898  -0.0259  0.0455    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
10.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

10.2.5 Men Ratio

10.2.5.1 Forest plot

10.2.5.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_men_ratio
## Abdelbasset 2020        0.2323 [-0.4745; 0.9391]       3.9        3.9              > 0.5
## Ciolac 2010             0.1845 [-0.6530; 1.0220]       2.8        2.8              < 0.5
## Conraads 2015          -0.0242 [-0.3215; 0.2730]      22.0       22.0              > 0.5
## Currie 2015             0.3256 [-0.5809; 1.2321]       2.4        2.4              > 0.5
## Eguchi 2012            -0.6465 [-1.5457; 0.2526]       2.4        2.4              > 0.5
## Fisher 2015            -0.0086 [-0.8330; 0.8159]       2.9        2.9              > 0.5
## Helgerud 2007          -0.1255 [-1.0029; 0.7519]       2.5        2.5              > 0.5
## Honkala 2017 (Healthy) -0.4795 [-1.2309; 0.2719]       3.4        3.4              > 0.5
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.0        2.0              > 0.5
## Jo 2020                 0.4589 [-0.2222; 1.1399]       4.2        4.2              > 0.5
## Keating 2014           -0.2384 [-1.0771; 0.6003]       2.8        2.8              < 0.5
## Kim 2015                0.4689 [-0.2820; 1.2198]       3.5        3.5              > 0.5
## Lunt 2014              -0.2087 [-1.0482; 0.6309]       2.8        2.8              < 0.5
## Lunt 2014              -0.7385 [-1.6026; 0.1257]       2.6        2.6              < 0.5
## Madssen 2014            0.4002 [-0.2688; 1.0692]       4.3        4.3              > 0.5
## Maillard 2016          -0.1685 [-1.1503; 0.8132]       2.0        2.0              < 0.5
## Matsuo 2015             0.6403 [-0.1801; 1.4607]       2.9        2.9              > 0.5
## Mitranun 2014           0.0035 [-0.7373; 0.7443]       3.5        3.5              < 0.5
## Motiani 2017           -0.6864 [-1.4775; 0.1046]       3.1        3.1              > 0.5
## Nalcakan 2014           0.4655 [-0.5624; 1.4935]       1.8        1.8              > 0.5
## Ramos 2016a             0.0347 [-0.5633; 0.6327]       5.4        5.4              > 0.5
## Ramos 2016b             0.2851 [-0.4127; 0.9829]       4.0        4.0              > 0.5
## Sandvei 2012            0.6305 [-0.2077; 1.4686]       2.8        2.8              < 0.5
## Sawyer 2016             0.6031 [-0.3416; 1.5478]       2.2        2.2              < 0.5
## Tjønna 2008             0.2944 [-0.6211; 1.2100]       2.3        2.3              < 0.5
## Winn 2018              -0.4553 [-1.4479; 0.5373]       2.0        2.0              < 0.5
## Zapata-Lamana 2018     -0.4783 [-1.2297; 0.2730]       3.4        3.4              < 0.5
## 
## Number of studies combined: k = 27
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0300 [-0.1095; 0.1695] 0.42  0.6736
## Random effects model 0.0300 [-0.1095; 0.1695] 0.42  0.6736
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1436]; tau = 0 [0.0000; 0.3789];
##  I^2 = 0.0% [0.0%; 42.2%]; H = 1.00 [1.00; 1.32]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 39.1%]; H = 1.00 [1.00; 1.28]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  25.83   26  0.4723
## 
## Results for subgroups (fixed effect model):
##         k     SMD            95%-CI     Q  I^2
## < 0.5  11 -0.0592 [-0.3178; 0.1995]  9.13 0.0%
## > 0.5  16  0.0656 [-0.1003; 0.2315] 14.18 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.63    1  0.4261
## Within groups  23.31   25  0.5593
## 
## Results for subgroups (random effects model):
##         k     SMD            95%-CI tau^2 tau
## < 0.5  11 -0.0592 [-0.3178; 0.1995]     0   0
## > 0.5  16  0.0656 [-0.1003; 0.2315]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.63    1  0.4261
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
10.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 27; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0048 (SE = 0.0422)
## tau (square root of estimated tau^2 value):             0.0690
## I^2 (residual heterogeneity / unaccounted variability): 3.19%
## H^2 (unaccounted variability / sampling variability):   1.03
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 25) = 25.8228, p-val = 0.4171
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0150, p-val = 0.9024
## 
## Model Results:
## 
##            estimate      se    zval    pval    ci.lb   ci.ub 
## intrcpt      0.0115  0.1774  0.0650  0.9482  -0.3362  0.3593    
## men_ratio    0.0292  0.2385  0.1226  0.9024  -0.4381  0.4966    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
10.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

10.2.6 Type of Exercise

10.2.6.1 Forest plot

10.2.6.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) type_exercise
## Abdelbasset 2020        0.2323 [-0.4745; 0.9391]       3.9        3.9       Cycling
## Ciolac 2010             0.1845 [-0.6530; 1.0220]       2.8        2.8       Running
## Conraads 2015          -0.0242 [-0.3215; 0.2730]      22.0       22.0       Cycling
## Currie 2015             0.3256 [-0.5809; 1.2321]       2.4        2.4       Cycling
## Eguchi 2012            -0.6465 [-1.5457; 0.2526]       2.4        2.4       Cycling
## Fisher 2015            -0.0086 [-0.8330; 0.8159]       2.9        2.9       Cycling
## Helgerud 2007          -0.1255 [-1.0029; 0.7519]       2.5        2.5       Running
## Honkala 2017 (Healthy) -0.4795 [-1.2309; 0.2719]       3.4        3.4       Cycling
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.0        2.0       Cycling
## Jo 2020                 0.4589 [-0.2222; 1.1399]       4.2        4.2       Running
## Keating 2014           -0.2384 [-1.0771; 0.6003]       2.8        2.8       Cycling
## Kim 2015                0.4689 [-0.2820; 1.2198]       3.5        3.5       Running
## Lunt 2014              -0.2087 [-1.0482; 0.6309]       2.8        2.8       Running
## Lunt 2014              -0.7385 [-1.6026; 0.1257]       2.6        2.6       Running
## Madssen 2014            0.4002 [-0.2688; 1.0692]       4.3        4.3       Running
## Maillard 2016          -0.1685 [-1.1503; 0.8132]       2.0        2.0       Cycling
## Matsuo 2015             0.6403 [-0.1801; 1.4607]       2.9        2.9       Cycling
## Mitranun 2014           0.0035 [-0.7373; 0.7443]       3.5        3.5       Running
## Motiani 2017           -0.6864 [-1.4775; 0.1046]       3.1        3.1       Cycling
## Nalcakan 2014           0.4655 [-0.5624; 1.4935]       1.8        1.8       Cycling
## Ramos 2016a             0.0347 [-0.5633; 0.6327]       5.4        5.4       Running
## Ramos 2016b             0.2851 [-0.4127; 0.9829]       4.0        4.0       Running
## Sandvei 2012            0.6305 [-0.2077; 1.4686]       2.8        2.8       Running
## Sawyer 2016             0.6031 [-0.3416; 1.5478]       2.2        2.2       Cycling
## Tjønna 2008             0.2944 [-0.6211; 1.2100]       2.3        2.3       Running
## Winn 2018              -0.4553 [-1.4479; 0.5373]       2.0        2.0       Running
## Zapata-Lamana 2018     -0.4783 [-1.2297; 0.2730]       3.4        3.4       Cycling
## 
## Number of studies combined: k = 27
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0300 [-0.1095; 0.1695] 0.42  0.6736
## Random effects model 0.0300 [-0.1095; 0.1695] 0.42  0.6736
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1436]; tau = 0 [0.0000; 0.3789];
##  I^2 = 0.0% [0.0%; 42.2%]; H = 1.00 [1.00; 1.32]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 36.4%]; H = 1.00 [1.00; 1.25]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  25.83   26  0.4723
## 
## Results for subgroups (fixed effect model):
##           k     SMD            95%-CI     Q  I^2
## Cycling  14 -0.0491 [-0.2336; 0.1355] 12.70 0.0%
## Running  13  0.1342 [-0.0795; 0.3478]  9.63 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  1.62    1  0.2033
## Within groups  22.33   25  0.6168
## 
## Results for subgroups (random effects model):
##           k     SMD            95%-CI tau^2 tau
## Cycling  14 -0.0491 [-0.2336; 0.1355]     0   0
## Running  13  0.1342 [-0.0795; 0.3478]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.62    1  0.2033
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
10.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 27; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0416)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 25) = 24.1499, p-val = 0.5107
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.6839, p-val = 0.1944
## 
## Model Results:
## 
##                       estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                -0.0498  0.0941  -0.5294  0.5965  -0.2342  0.1346    
## type_exerciseRunning    0.1867  0.1439   1.2976  0.1944  -0.0953  0.4688    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
10.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

10.2.7 Baseline Values

10.2.7.1 Forest plot

10.2.7.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) category_bsln
## Abdelbasset 2020        0.2323 [-0.4745; 0.9391]       3.9        3.9  > 1.7 mmol/L
## Ciolac 2010             0.1845 [-0.6530; 1.0220]       2.8        2.8  < 1.7 mmol/L
## Conraads 2015          -0.0242 [-0.3215; 0.2730]      22.0       22.0  < 1.7 mmol/L
## Currie 2015             0.3256 [-0.5809; 1.2321]       2.4        2.4  < 1.7 mmol/L
## Eguchi 2012            -0.6465 [-1.5457; 0.2526]       2.4        2.4  > 1.7 mmol/L
## Fisher 2015            -0.0086 [-0.8330; 0.8159]       2.9        2.9  < 1.7 mmol/L
## Helgerud 2007          -0.1255 [-1.0029; 0.7519]       2.5        2.5  < 1.7 mmol/L
## Honkala 2017 (Healthy) -0.4795 [-1.2309; 0.2719]       3.4        3.4  < 1.7 mmol/L
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.0        2.0  > 1.7 mmol/L
## Jo 2020                 0.4589 [-0.2222; 1.1399]       4.2        4.2  > 1.7 mmol/L
## Keating 2014           -0.2384 [-1.0771; 0.6003]       2.8        2.8  < 1.7 mmol/L
## Kim 2015                0.4689 [-0.2820; 1.2198]       3.5        3.5  < 1.7 mmol/L
## Lunt 2014              -0.2087 [-1.0482; 0.6309]       2.8        2.8  < 1.7 mmol/L
## Lunt 2014              -0.7385 [-1.6026; 0.1257]       2.6        2.6  < 1.7 mmol/L
## Madssen 2014            0.4002 [-0.2688; 1.0692]       4.3        4.3  < 1.7 mmol/L
## Maillard 2016          -0.1685 [-1.1503; 0.8132]       2.0        2.0  < 1.7 mmol/L
## Matsuo 2015             0.6403 [-0.1801; 1.4607]       2.9        2.9  > 1.7 mmol/L
## Mitranun 2014           0.0035 [-0.7373; 0.7443]       3.5        3.5  < 1.7 mmol/L
## Motiani 2017           -0.6864 [-1.4775; 0.1046]       3.1        3.1  < 1.7 mmol/L
## Nalcakan 2014           0.4655 [-0.5624; 1.4935]       1.8        1.8  < 1.7 mmol/L
## Ramos 2016a             0.0347 [-0.5633; 0.6327]       5.4        5.4  > 1.7 mmol/L
## Ramos 2016b             0.2851 [-0.4127; 0.9829]       4.0        4.0  > 1.7 mmol/L
## Sandvei 2012            0.6305 [-0.2077; 1.4686]       2.8        2.8  < 1.7 mmol/L
## Sawyer 2016             0.6031 [-0.3416; 1.5478]       2.2        2.2  < 1.7 mmol/L
## Tjønna 2008             0.2944 [-0.6211; 1.2100]       2.3        2.3  < 1.7 mmol/L
## Winn 2018              -0.4553 [-1.4479; 0.5373]       2.0        2.0  < 1.7 mmol/L
## Zapata-Lamana 2018     -0.4783 [-1.2297; 0.2730]       3.4        3.4  < 1.7 mmol/L
## 
## Number of studies combined: k = 27
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0300 [-0.1095; 0.1695] 0.42  0.6736
## Random effects model 0.0300 [-0.1095; 0.1695] 0.42  0.6736
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1436]; tau = 0 [0.0000; 0.3789];
##  I^2 = 0.0% [0.0%; 42.2%]; H = 1.00 [1.00; 1.32]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 37.0%]; H = 1.00 [1.00; 1.26]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  25.83   26  0.4723
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI     Q  I^2
## < 1.7 mmol/L  20 -0.0191 [-0.1802; 0.1420] 17.41 0.0%
## > 1.7 mmol/L   7  0.1755 [-0.1047; 0.4558]  5.14 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  1.39    1  0.2380
## Within groups  22.55   25  0.6036
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI tau^2 tau
## < 1.7 mmol/L  20 -0.0191 [-0.1802; 0.1420]     0   0
## > 1.7 mmol/L   7  0.1755 [-0.1047; 0.4558]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.39    1  0.2380
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
10.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 27; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0402)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 25) = 24.2709, p-val = 0.5038
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.5629, p-val = 0.2112
## 
## Model Results:
## 
##                estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt         -0.2945  0.2692  -1.0943  0.2738  -0.8221  0.2330    
## bsln_adjusted    0.2311  0.1849   1.2502  0.2112  -0.1312  0.5934    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
10.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

10.2.8 Type of HIIE

10.2.8.1 Forest plot

10.2.8.2 R output
##                            SMD            95%-CI %W(fixed) %W(random) HIIE
## Abdelbasset 2020        0.2323 [-0.4745; 0.9391]       3.9        3.9 HIIT
## Ciolac 2010             0.1845 [-0.6530; 1.0220]       2.8        2.8 HIIT
## Conraads 2015          -0.0242 [-0.3215; 0.2730]      22.0       22.0 HIIT
## Currie 2015             0.3256 [-0.5809; 1.2321]       2.4        2.4 HIIT
## Eguchi 2012            -0.6465 [-1.5457; 0.2526]       2.4        2.4 HIIT
## Fisher 2015            -0.0086 [-0.8330; 0.8159]       2.9        2.9  SIT
## Helgerud 2007          -0.1255 [-1.0029; 0.7519]       2.5        2.5 HIIT
## Honkala 2017 (Healthy) -0.4795 [-1.2309; 0.2719]       3.4        3.4  SIT
## Honkala 2017 (T2D)      0.0000 [-0.9877; 0.9877]       2.0        2.0  SIT
## Jo 2020                 0.4589 [-0.2222; 1.1399]       4.2        4.2 HIIT
## Keating 2014           -0.2384 [-1.0771; 0.6003]       2.8        2.8 HIIT
## Kim 2015                0.4689 [-0.2820; 1.2198]       3.5        3.5 HIIT
## Lunt 2014              -0.2087 [-1.0482; 0.6309]       2.8        2.8 HIIT
## Lunt 2014              -0.7385 [-1.6026; 0.1257]       2.6        2.6  SIT
## Madssen 2014            0.4002 [-0.2688; 1.0692]       4.3        4.3 HIIT
## Maillard 2016          -0.1685 [-1.1503; 0.8132]       2.0        2.0 HIIT
## Matsuo 2015             0.6403 [-0.1801; 1.4607]       2.9        2.9 HIIT
## Mitranun 2014           0.0035 [-0.7373; 0.7443]       3.5        3.5 HIIT
## Motiani 2017           -0.6864 [-1.4775; 0.1046]       3.1        3.1  SIT
## Nalcakan 2014           0.4655 [-0.5624; 1.4935]       1.8        1.8  SIT
## Ramos 2016a             0.0347 [-0.5633; 0.6327]       5.4        5.4 HIIT
## Ramos 2016b             0.2851 [-0.4127; 0.9829]       4.0        4.0 HIIT
## Sandvei 2012            0.6305 [-0.2077; 1.4686]       2.8        2.8  SIT
## Sawyer 2016             0.6031 [-0.3416; 1.5478]       2.2        2.2 HIIT
## Tjønna 2008             0.2944 [-0.6211; 1.2100]       2.3        2.3 HIIT
## Winn 2018              -0.4553 [-1.4479; 0.5373]       2.0        2.0 HIIT
## Zapata-Lamana 2018     -0.4783 [-1.2297; 0.2730]       3.4        3.4 HIIT
## 
## Number of studies combined: k = 27
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0300 [-0.1095; 0.1695] 0.42  0.6736
## Random effects model 0.0300 [-0.1095; 0.1695] 0.42  0.6736
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1436]; tau = 0 [0.0000; 0.3789];
##  I^2 = 0.0% [0.0%; 42.2%]; H = 1.00 [1.00; 1.32]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 36.2%]; H = 1.00 [1.00; 1.25]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  25.83   26  0.4723
## 
## Results for subgroups (fixed effect model):
##        k     SMD            95%-CI     Q   I^2
## HIIT  20  0.0734 [-0.0814; 0.2282] 13.79  0.0%
## SIT    7 -0.1640 [-0.4878; 0.1598]  8.47 29.2%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  1.68    1  0.1948
## Within groups  22.27   25  0.6204
## 
## Results for subgroups (random effects model):
##        k     SMD            95%-CI  tau^2    tau
## HIIT  20  0.0734 [-0.0814; 0.2282]      0      0
## SIT    7 -0.1506 [-0.5374; 0.2361] 0.0792 0.2815
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.11    1  0.2919
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
10.2.8.3 Meta-regression
## 
## Mixed-Effects Model (k = 27; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0404)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 25) = 24.0595, p-val = 0.5159
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.7743, p-val = 0.1829
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    0.0754  0.0789   0.9550  0.3396  -0.0793  0.2300    
## HIIESIT   -0.2435  0.1828  -1.3320  0.1829  -0.6018  0.1148    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
10.2.8.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

11. Total Cholesterol

11.1 Overall

11.1.1 Forest plot

11.1.2 R output

##                            SMD             95%-CI %W(fixed) %W(random)
## Abdelbasset 2020        0.4282 [-0.2843;  1.1406]       4.1        4.4
## Ciolac 2010             0.3701 [-0.4728;  1.2130]       2.9        3.6
## Conraads 2015          -0.0242 [-0.3215;  0.2731]      23.4        9.1
## Currie 2015            -1.0152 [-1.9719; -0.0586]       2.3        3.0
## Eguchi 2012            -0.5729 [-1.4672;  0.3214]       2.6        3.3
## Fisher 2015            -0.1696 [-0.9955;  0.6562]       3.0        3.7
## Grieco 2013             0.9216 [ 0.0393;  1.8038]       2.7        3.3
## Helgerud 2007          -0.0341 [-0.9107;  0.8424]       2.7        3.4
## Honkala 2017 (Healthy)  0.8577 [ 0.0836;  1.6318]       3.5        4.0
## Honkala 2017 (T2D)      0.7046 [-0.3128;  1.7221]       2.0        2.7
## Jo 2020                 0.8928 [ 0.1878;  1.5978]       4.2        4.5
## Keating 2014            0.3604 [-0.4821;  1.2029]       2.9        3.6
## Kim 2015                0.6605 [-0.1002;  1.4213]       3.6        4.1
## Lunt 2014              -0.7221 [-1.5851;  0.1409]       2.8        3.4
## Lunt 2014              -0.3736 [-1.2179;  0.4707]       2.9        3.5
## Madssen 2014            0.1319 [-0.5314;  0.7952]       4.7        4.8
## Maillard 2016          -0.1494 [-1.1308;  0.8319]       2.2        2.8
## Matsuo 2015             0.1370 [-0.6641;  0.9380]       3.2        3.8
## Mitranun 2014           0.0115 [-0.7293;  0.7523]       3.8        4.2
## Motiani 2017            0.9454 [ 0.1348;  1.7559]       3.2        3.8
## Nalcakan 2014           0.2218 [-0.7957;  1.2393]       2.0        2.7
## Ramos 2016b            -0.5535 [-1.2609;  0.1539]       4.1        4.5
## Sandvei 2012            0.4295 [-0.3980;  1.2570]       3.0        3.6
## Sawyer 2016            -0.0089 [-0.9328;  0.9150]       2.4        3.1
## Winn 2018               0.0419 [-0.9382;  1.0220]       2.2        2.9
## Zapata-Lamana 2018     -0.0156 [-0.7564;  0.7252]       3.8        4.2
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.1196 [-0.0243; 0.2635] 1.63  0.1032
## Random effects model 0.1400 [-0.0502; 0.3303] 1.44  0.1491
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0801 [0.0000; 0.3233]; tau = 0.2830 [0.0000; 0.5686];
##  I^2 = 35.4% [0.0%; 59.9%]; H = 1.24 [1.00; 1.58]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  38.68   25  0.0397
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

11.1.3 Sensitivity analysis

11.1.3.1 Forest plot

11.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                                      SMD            95%-CI p-value   tau^2     tau    I^2
## Omitting Abdelbasset 2020         0.1220 [-0.0676; 0.3116]  0.2071  0.0683  0.2614  31.6%
## Omitting Ciolac 2010              0.1272 [-0.0622; 0.3165]  0.1881  0.0698  0.2643  32.4%
## Omitting Conraads 2015            0.1518 [-0.0476; 0.3512]  0.1356  0.0790  0.2810  30.8%
## Omitting Currie 2015              0.1639 [-0.0109; 0.3386]  0.0661  0.0412  0.2031  22.2%
## Omitting Eguchi 2012              0.1570 [-0.0267; 0.3408]  0.0939  0.0584  0.2417  28.7%
## Omitting Fisher 2015              0.1465 [-0.0427; 0.3356]  0.1290  0.0691  0.2629  32.1%
## Omitting Grieco 2013              0.1093 [-0.0721; 0.2907]  0.2375  0.0535  0.2313  26.9%
## Omitting Helgerud 2007            0.1411 [-0.0486; 0.3308]  0.1448  0.0709  0.2663  32.8%
## Omitting Honkala 2017 (Healthy)   0.1055 [-0.0756; 0.2866]  0.2535  0.0515  0.2269  26.0%
## Omitting Honkala 2017 (T2D)       0.1208 [-0.0651; 0.3067]  0.2029  0.0642  0.2533  30.8%
## Omitting Jo 2020                  0.0985 [-0.0797; 0.2766]  0.2785  0.0446  0.2111  23.2%
## Omitting Keating 2014             0.1275 [-0.0619; 0.3169]  0.1871  0.0700  0.2645  32.4%
## Omitting Kim 2015                 0.1133 [-0.0723; 0.2990]  0.2316  0.0607  0.2464  29.2%
## Omitting Lunt 2014                0.1625 [-0.0176; 0.3426]  0.0770  0.0506  0.2250  25.8%
## Omitting Lunt 2014                0.1529 [-0.0338; 0.3397]  0.1085  0.0642  0.2533  30.5%
## Omitting Madssen 2014             0.1357 [-0.0565; 0.3280]  0.1665  0.0733  0.2707  33.0%
## Omitting Maillard 2016            0.1432 [-0.0453; 0.3317]  0.1365  0.0695  0.2637  32.5%
## Omitting Matsuo 2015              0.1355 [-0.0552; 0.3261]  0.1637  0.0720  0.2684  33.0%
## Omitting Mitranun 2014            0.1408 [-0.0502; 0.3319]  0.1485  0.0720  0.2684  32.8%
## Omitting Motiani 2017             0.1042 [-0.0752; 0.2837]  0.2549  0.0487  0.2206  25.0%
## Omitting Nalcakan 2014            0.1333 [-0.0557; 0.3223]  0.1668  0.0708  0.2661  32.9%
## Omitting Ramos 2016b              0.1658 [-0.0159; 0.3476]  0.0737  0.0517  0.2273  25.9%
## Omitting Sandvei 2012             0.1247 [-0.0642; 0.3137]  0.1958  0.0687  0.2622  32.0%
## Omitting Sawyer 2016              0.1399 [-0.0496; 0.3293]  0.1478  0.0710  0.2664  32.8%
## Omitting Winn 2018                0.1381 [-0.0511; 0.3273]  0.1526  0.0710  0.2665  32.9%
## Omitting Zapata-Lamana 2018       0.1420 [-0.0489; 0.3329]  0.1450  0.0717  0.2678  32.7%
##                                                                                          
## Pooled estimate                   0.1400 [-0.0502; 0.3303]  0.1491  0.0801  0.2830  35.4%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

11.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

11.2 Subgroups

11.2.1 Overall

11.2.1.1 Forest plot

11.2.1.2 R output
##                            SMD            95%-CI     meta-analysis
##                         0.1400 [-0.0502; 0.3303]           Overall
## Healthy                 0.4047 [ 0.0539; 0.7554]        Population
## Overweight/obese       -0.1225 [-0.4446; 0.1996]        Population
## Cardiac Rehabilitation  0.0035 [-0.4653; 0.4723]        Population
## Metabolic Syndrome      0.1555 [-0.6771; 0.9881]        Population
## T2D                     0.2301 [-0.1858; 0.6460]        Population
## < 30 y                  0.2189 [-0.1024; 0.5402]               Age
## 30 - 50 y               0.2159 [-0.1462; 0.5781]               Age
## > 50 y                  0.0279 [-0.2774; 0.3332]               Age
## < 5 weeks               0.7095 [ 0.3151; 1.1040] Training Duration
## 5 - 10 weeks            0.2855 [ 0.0308; 0.5401] Training Duration
## > 10 weeks             -0.1411 [-0.3509; 0.0687] Training Duration
## < 0.5                   0.0766 [-0.1802; 0.3334]         Men Ratio
## > 0.5                   0.1744 [-0.0890; 0.4377]         Men Ratio
## Cycling                 0.1667 [-0.0777; 0.4112]  Type of Exercise
## Running                 0.0915 [-0.2020; 0.3850]  Type of Exercise
## < 5.2 mmol/L            0.2066 [ 0.0076; 0.4057]   Baseline Values
## > 5.2 mmol/L           -0.2059 [-0.5974; 0.1855]   Baseline Values
## HIIT                    0.0642 [-0.1394; 0.2677]      Type of HIIE
## SIT                     0.3642 [-0.0135; 0.7419]      Type of HIIE
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Random effects model 0.1400 [-0.0502; 0.3303] 1.44  0.1491
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0801; tau = 0.2830; I^2 = 35.4% [0.0%; 59.9%]; H = 1.24 [1.00; 1.58]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  38.68   25  0.0397
## 
## Results for meta-analyses (random effects model):
##                     k    SMD            95%-CI  tau^2    tau     Q   I^2
## Overall            26 0.1400 [-0.0502; 0.3303] 0.0801 0.2830 38.68 35.4%
## Population         26 0.1400 [-0.0502; 0.3303] 0.0801 0.2830 38.68 35.4%
## Age                26 0.1400 [-0.0502; 0.3303] 0.0801 0.2830 38.68 35.4%
## Training Duration  26 0.1400 [-0.0502; 0.3303] 0.0801 0.2830 38.68 35.4%
## Men Ratio          26 0.1400 [-0.0502; 0.3303] 0.0801 0.2830 38.68 35.4%
## Type of Exercise   26 0.1400 [-0.0502; 0.3303] 0.0801 0.2830 38.68 35.4%
## Baseline Values    26 0.1400 [-0.0502; 0.3303] 0.0801 0.2830 38.68 35.4%
## Type of HIIE       26 0.1400 [-0.0502; 0.3303] 0.0801 0.2830 38.68 35.4%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

11.2.2 Population

11.2.2.1 Forest plot

11.2.2.2 R output
##                            SMD             95%-CI %W(fixed) %W(random)             population
## Abdelbasset 2020        0.4282 [-0.2843;  1.1406]       4.1        4.4                    T2D
## Ciolac 2010             0.3701 [-0.4728;  1.2130]       2.9        3.6                Healthy
## Conraads 2015          -0.0242 [-0.3215;  0.2731]      23.4        9.1 Cardiac Rehabilitation
## Currie 2015            -1.0152 [-1.9719; -0.0586]       2.3        3.0 Cardiac Rehabilitation
## Eguchi 2012            -0.5729 [-1.4672;  0.3214]       2.6        3.3                Healthy
## Fisher 2015            -0.1696 [-0.9955;  0.6562]       3.0        3.7       Overweight/obese
## Grieco 2013             0.9216 [ 0.0393;  1.8038]       2.7        3.3                Healthy
## Helgerud 2007          -0.0341 [-0.9107;  0.8424]       2.7        3.4                Healthy
## Honkala 2017 (Healthy)  0.8577 [ 0.0836;  1.6318]       3.5        4.0                Healthy
## Honkala 2017 (T2D)      0.7046 [-0.3128;  1.7221]       2.0        2.7                    T2D
## Jo 2020                 0.8928 [ 0.1878;  1.5978]       4.2        4.5     Metabolic Syndrome
## Keating 2014            0.3604 [-0.4821;  1.2029]       2.9        3.6       Overweight/obese
## Kim 2015                0.6605 [-0.1002;  1.4213]       3.6        4.1 Cardiac Rehabilitation
## Lunt 2014              -0.7221 [-1.5851;  0.1409]       2.8        3.4       Overweight/obese
## Lunt 2014              -0.3736 [-1.2179;  0.4707]       2.9        3.5       Overweight/obese
## Madssen 2014            0.1319 [-0.5314;  0.7952]       4.7        4.8 Cardiac Rehabilitation
## Maillard 2016          -0.1494 [-1.1308;  0.8319]       2.2        2.8                    T2D
## Matsuo 2015             0.1370 [-0.6641;  0.9380]       3.2        3.8     Metabolic Syndrome
## Mitranun 2014           0.0115 [-0.7293;  0.7523]       3.8        4.2                    T2D
## Motiani 2017            0.9454 [ 0.1348;  1.7559]       3.2        3.8                Healthy
## Nalcakan 2014           0.2218 [-0.7957;  1.2393]       2.0        2.7                Healthy
## Ramos 2016b            -0.5535 [-1.2609;  0.1539]       4.1        4.5     Metabolic Syndrome
## Sandvei 2012            0.4295 [-0.3980;  1.2570]       3.0        3.6                Healthy
## Sawyer 2016            -0.0089 [-0.9328;  0.9150]       2.4        3.1       Overweight/obese
## Winn 2018               0.0419 [-0.9382;  1.0220]       2.2        2.9       Overweight/obese
## Zapata-Lamana 2018     -0.0156 [-0.7564;  0.7252]       3.8        4.2       Overweight/obese
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.1196 [-0.0243; 0.2635] 1.63  0.1032
## Random effects model 0.1400 [-0.0502; 0.3303] 1.44  0.1491
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0801 [0.0000; 0.3233]; tau = 0.2830 [0.0000; 0.5686];
##  I^2 = 35.4% [0.0%; 59.9%]; H = 1.24 [1.00; 1.58]
## 
## Quantifying residual heterogeneity:
##  I^2 = 27.4% [0.0%; 56.9%]; H = 1.17 [1.00; 1.52]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  38.68   25  0.0397
## 
## Results for subgroups (fixed effect model):
##                          k     SMD            95%-CI    Q   I^2
## Healthy                  8  0.4131 [ 0.1088; 0.7174] 9.26 24.4%
## Overweight/obese         7 -0.1225 [-0.4446; 0.1996] 3.43  0.0%
## Cardiac Rehabilitation   4  0.0050 [-0.2421; 0.2520] 6.77 55.7%
## Metabolic Syndrome       3  0.1573 [-0.2672; 0.5819] 7.64 73.8%
## T2D                      4  0.2301 [-0.1858; 0.6460] 1.85  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  6.87    4  0.1429
## Within groups  28.94   21  0.1155
## 
## Results for subgroups (random effects model):
##                          k     SMD            95%-CI  tau^2    tau
## Healthy                  8  0.4047 [ 0.0539; 0.7554] 0.0624 0.2497
## Overweight/obese         7 -0.1225 [-0.4446; 0.1996]      0      0
## Cardiac Rehabilitation   4  0.0035 [-0.4653; 0.4723] 0.1225 0.3501
## Metabolic Syndrome       3  0.1555 [-0.6771; 0.9881] 0.3992 0.6319
## T2D                      4  0.2301 [-0.1858; 0.6460]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   5.22    4  0.2656
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
11.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 26; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0791 (SE = 0.0755)
## tau (square root of estimated tau^2 value):             0.2812
## I^2 (residual heterogeneity / unaccounted variability): 32.74%
## H^2 (unaccounted variability / sampling variability):   1.49
## R^2 (amount of heterogeneity accounted for):            1.30%
## 
## Test for Residual Heterogeneity:
## QE(df = 21) = 31.2204, p-val = 0.0701
## 
## Test of Moderators (coefficients 2:5):
## QM(df = 4) = 4.7048, p-val = 0.3190
## 
## Model Results:
## 
##                               estimate      se     zval    pval    ci.lb    ci.ub 
## intrcpt                         0.4179  0.1844   2.2664  0.0234   0.0565   0.7793  * 
## .byvarOverweight/obese         -0.5458  0.2692  -2.0275  0.0426  -1.0735  -0.0182  * 
## .byvarCardiac Rehabilitation   -0.4126  0.2802  -1.4724  0.1409  -0.9618   0.1366    
## .byvarMetabolic Syndrome       -0.2569  0.3276  -0.7842  0.4329  -0.8990   0.3852    
## .byvarT2D                      -0.1770  0.3163  -0.5597  0.5757  -0.7970   0.4429    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
11.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

11.2.3 Age

11.2.3.1 Forest plot

11.2.3.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_age
## Abdelbasset 2020        0.4282 [-0.2843;  1.1406]       4.1        4.4       > 50 y
## Ciolac 2010             0.3701 [-0.4728;  1.2130]       2.9        3.6       < 30 y
## Conraads 2015          -0.0242 [-0.3215;  0.2731]      23.4        9.1       > 50 y
## Currie 2015            -1.0152 [-1.9719; -0.0586]       2.3        3.0       > 50 y
## Eguchi 2012            -0.5729 [-1.4672;  0.3214]       2.6        3.3       > 50 y
## Fisher 2015            -0.1696 [-0.9955;  0.6562]       3.0        3.7       < 30 y
## Grieco 2013             0.9216 [ 0.0393;  1.8038]       2.7        3.3       < 30 y
## Helgerud 2007          -0.0341 [-0.9107;  0.8424]       2.7        3.4       < 30 y
## Honkala 2017 (Healthy)  0.8577 [ 0.0836;  1.6318]       3.5        4.0    30 - 50 y
## Honkala 2017 (T2D)      0.7046 [-0.3128;  1.7221]       2.0        2.7    30 - 50 y
## Jo 2020                 0.8928 [ 0.1878;  1.5978]       4.2        4.5       > 50 y
## Keating 2014            0.3604 [-0.4821;  1.2029]       2.9        3.6    30 - 50 y
## Kim 2015                0.6605 [-0.1002;  1.4213]       3.6        4.1       > 50 y
## Lunt 2014              -0.7221 [-1.5851;  0.1409]       2.8        3.4    30 - 50 y
## Lunt 2014              -0.3736 [-1.2179;  0.4707]       2.9        3.5    30 - 50 y
## Madssen 2014            0.1319 [-0.5314;  0.7952]       4.7        4.8       > 50 y
## Maillard 2016          -0.1494 [-1.1308;  0.8319]       2.2        2.8       > 50 y
## Matsuo 2015             0.1370 [-0.6641;  0.9380]       3.2        3.8    30 - 50 y
## Mitranun 2014           0.0115 [-0.7293;  0.7523]       3.8        4.2       > 50 y
## Motiani 2017            0.9454 [ 0.1348;  1.7559]       3.2        3.8    30 - 50 y
## Nalcakan 2014           0.2218 [-0.7957;  1.2393]       2.0        2.7       < 30 y
## Ramos 2016b            -0.5535 [-1.2609;  0.1539]       4.1        4.5       > 50 y
## Sandvei 2012            0.4295 [-0.3980;  1.2570]       3.0        3.6       < 30 y
## Sawyer 2016            -0.0089 [-0.9328;  0.9150]       2.4        3.1    30 - 50 y
## Winn 2018               0.0419 [-0.9382;  1.0220]       2.2        2.9    30 - 50 y
## Zapata-Lamana 2018     -0.0156 [-0.7564;  0.7252]       3.8        4.2       < 30 y
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.1196 [-0.0243; 0.2635] 1.63  0.1032
## Random effects model 0.1400 [-0.0502; 0.3303] 1.44  0.1491
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0801 [0.0000; 0.3233]; tau = 0.2830 [0.0000; 0.5686];
##  I^2 = 35.4% [0.0%; 59.9%]; H = 1.24 [1.00; 1.58]
## 
## Quantifying residual heterogeneity:
##  I^2 = 32.6% [0.0%; 59.0%]; H = 1.22 [1.00; 1.56]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  38.68   25  0.0397
## 
## Results for subgroups (fixed effect model):
##             k    SMD            95%-CI     Q   I^2
## < 30 y      7 0.2189 [-0.1024; 0.5402]  4.01  0.0%
## 30 - 50 y   9 0.2234 [-0.0650; 0.5118] 12.52 36.1%
## > 50 y     10 0.0287 [-0.1657; 0.2232] 17.58 48.8%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  1.70    2  0.4274
## Within groups  34.11   23  0.0636
## 
## Results for subgroups (random effects model):
##             k    SMD            95%-CI  tau^2    tau
## < 30 y      7 0.2189 [-0.1024; 0.5402]      0      0
## 30 - 50 y   9 0.2159 [-0.1462; 0.5781] 0.1105 0.3324
## > 50 y     10 0.0279 [-0.2774; 0.3332] 0.1085 0.3294
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.92    2  0.6326
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
11.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 26; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0796 (SE = 0.0685)
## tau (square root of estimated tau^2 value):             0.2821
## I^2 (residual heterogeneity / unaccounted variability): 34.66%
## H^2 (unaccounted variability / sampling variability):   1.53
## R^2 (amount of heterogeneity accounted for):            0.66%
## 
## Test for Residual Heterogeneity:
## QE(df = 24) = 36.7329, p-val = 0.0465
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.1284, p-val = 0.2881
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    0.4679  0.3235   1.4462  0.1481  -0.1662  1.1020    
## age       -0.0072  0.0068  -1.0623  0.2881  -0.0204  0.0061    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
11.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

11.2.4 Training Duration

11.2.4.1 Forest plot

11.2.4.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_duration
## Abdelbasset 2020        0.4282 [-0.2843;  1.1406]       4.1        4.4      5 - 10 weeks
## Ciolac 2010             0.3701 [-0.4728;  1.2130]       2.9        3.6        > 10 weeks
## Conraads 2015          -0.0242 [-0.3215;  0.2731]      23.4        9.1        > 10 weeks
## Currie 2015            -1.0152 [-1.9719; -0.0586]       2.3        3.0        > 10 weeks
## Eguchi 2012            -0.5729 [-1.4672;  0.3214]       2.6        3.3        > 10 weeks
## Fisher 2015            -0.1696 [-0.9955;  0.6562]       3.0        3.7      5 - 10 weeks
## Grieco 2013             0.9216 [ 0.0393;  1.8038]       2.7        3.3         < 5 weeks
## Helgerud 2007          -0.0341 [-0.9107;  0.8424]       2.7        3.4      5 - 10 weeks
## Honkala 2017 (Healthy)  0.8577 [ 0.0836;  1.6318]       3.5        4.0         < 5 weeks
## Honkala 2017 (T2D)      0.7046 [-0.3128;  1.7221]       2.0        2.7         < 5 weeks
## Jo 2020                 0.8928 [ 0.1878;  1.5978]       4.2        4.5      5 - 10 weeks
## Keating 2014            0.3604 [-0.4821;  1.2029]       2.9        3.6        > 10 weeks
## Kim 2015                0.6605 [-0.1002;  1.4213]       3.6        4.1      5 - 10 weeks
## Lunt 2014              -0.7221 [-1.5851;  0.1409]       2.8        3.4        > 10 weeks
## Lunt 2014              -0.3736 [-1.2179;  0.4707]       2.9        3.5        > 10 weeks
## Madssen 2014            0.1319 [-0.5314;  0.7952]       4.7        4.8        > 10 weeks
## Maillard 2016          -0.1494 [-1.1308;  0.8319]       2.2        2.8        > 10 weeks
## Matsuo 2015             0.1370 [-0.6641;  0.9380]       3.2        3.8      5 - 10 weeks
## Mitranun 2014           0.0115 [-0.7293;  0.7523]       3.8        4.2      5 - 10 weeks
## Motiani 2017            0.9454 [ 0.1348;  1.7559]       3.2        3.8         < 5 weeks
## Nalcakan 2014           0.2218 [-0.7957;  1.2393]       2.0        2.7      5 - 10 weeks
## Ramos 2016b            -0.5535 [-1.2609;  0.1539]       4.1        4.5        > 10 weeks
## Sandvei 2012            0.4295 [-0.3980;  1.2570]       3.0        3.6      5 - 10 weeks
## Sawyer 2016            -0.0089 [-0.9328;  0.9150]       2.4        3.1      5 - 10 weeks
## Winn 2018               0.0419 [-0.9382;  1.0220]       2.2        2.9         < 5 weeks
## Zapata-Lamana 2018     -0.0156 [-0.7564;  0.7252]       3.8        4.2        > 10 weeks
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.1196 [-0.0243; 0.2635] 1.63  0.1032
## Random effects model 0.1400 [-0.0502; 0.3303] 1.44  0.1491
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0801 [0.0000; 0.3233]; tau = 0.2830 [0.0000; 0.5686];
##  I^2 = 35.4% [0.0%; 59.9%]; H = 1.24 [1.00; 1.58]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 34.1%]; H = 1.00 [1.00; 1.23]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  38.68   25  0.0397
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI     Q  I^2
## < 5 weeks      5  0.7095 [ 0.3151; 1.1040]  2.30 0.0%
## 5 - 10 weeks  10  0.2855 [ 0.0308; 0.5401]  6.42 0.0%
## > 10 weeks    11 -0.1291 [-0.3241; 0.0658] 10.62 5.9%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups 16.47    2  0.0003
## Within groups  19.34   23  0.6815
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI  tau^2    tau
## < 5 weeks      5  0.7095 [ 0.3151; 1.1040]      0      0
## 5 - 10 weeks  10  0.2855 [ 0.0308; 0.5401]      0      0
## > 10 weeks    11 -0.1411 [-0.3509; 0.0687] 0.0079 0.0890
## 
## Test for subgroup differences (random effects model):
##                      Q d.f. p-value
## Between groups   16.14    2  0.0003
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
11.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 26; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0428)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            100.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 24) = 21.5591, p-val = 0.6056
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 17.1191, p-val < .0001
## 
## Model Results:
## 
##           estimate      se     zval    pval    ci.lb    ci.ub 
## intrcpt     0.9030  0.2031   4.4468  <.0001   0.5050   1.3011  *** 
## duration   -0.0802  0.0194  -4.1375  <.0001  -0.1182  -0.0422  *** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
11.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

11.2.5 Men Ratio

11.2.5.1 Forest plot

11.2.5.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_men_ratio
## Abdelbasset 2020        0.4282 [-0.2843;  1.1406]       4.1        4.4              > 0.5
## Ciolac 2010             0.3701 [-0.4728;  1.2130]       2.9        3.6              < 0.5
## Conraads 2015          -0.0242 [-0.3215;  0.2731]      23.4        9.1              > 0.5
## Currie 2015            -1.0152 [-1.9719; -0.0586]       2.3        3.0              > 0.5
## Eguchi 2012            -0.5729 [-1.4672;  0.3214]       2.6        3.3              > 0.5
## Fisher 2015            -0.1696 [-0.9955;  0.6562]       3.0        3.7              > 0.5
## Grieco 2013             0.9216 [ 0.0393;  1.8038]       2.7        3.3              < 0.5
## Helgerud 2007          -0.0341 [-0.9107;  0.8424]       2.7        3.4              > 0.5
## Honkala 2017 (Healthy)  0.8577 [ 0.0836;  1.6318]       3.5        4.0              > 0.5
## Honkala 2017 (T2D)      0.7046 [-0.3128;  1.7221]       2.0        2.7              > 0.5
## Jo 2020                 0.8928 [ 0.1878;  1.5978]       4.2        4.5              > 0.5
## Keating 2014            0.3604 [-0.4821;  1.2029]       2.9        3.6              < 0.5
## Kim 2015                0.6605 [-0.1002;  1.4213]       3.6        4.1              > 0.5
## Lunt 2014              -0.7221 [-1.5851;  0.1409]       2.8        3.4              < 0.5
## Lunt 2014              -0.3736 [-1.2179;  0.4707]       2.9        3.5              < 0.5
## Madssen 2014            0.1319 [-0.5314;  0.7952]       4.7        4.8              > 0.5
## Maillard 2016          -0.1494 [-1.1308;  0.8319]       2.2        2.8              < 0.5
## Matsuo 2015             0.1370 [-0.6641;  0.9380]       3.2        3.8              > 0.5
## Mitranun 2014           0.0115 [-0.7293;  0.7523]       3.8        4.2              < 0.5
## Motiani 2017            0.9454 [ 0.1348;  1.7559]       3.2        3.8              > 0.5
## Nalcakan 2014           0.2218 [-0.7957;  1.2393]       2.0        2.7              > 0.5
## Ramos 2016b            -0.5535 [-1.2609;  0.1539]       4.1        4.5              > 0.5
## Sandvei 2012            0.4295 [-0.3980;  1.2570]       3.0        3.6              < 0.5
## Sawyer 2016            -0.0089 [-0.9328;  0.9150]       2.4        3.1              < 0.5
## Winn 2018               0.0419 [-0.9382;  1.0220]       2.2        2.9              < 0.5
## Zapata-Lamana 2018     -0.0156 [-0.7564;  0.7252]       3.8        4.2              < 0.5
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.1196 [-0.0243; 0.2635] 1.63  0.1032
## Random effects model 0.1400 [-0.0502; 0.3303] 1.44  0.1491
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0801 [0.0000; 0.3233]; tau = 0.2830 [0.0000; 0.5686];
##  I^2 = 35.4% [0.0%; 59.9%]; H = 1.24 [1.00; 1.58]
## 
## Quantifying residual heterogeneity:
##  I^2 = 32.7% [0.0%; 58.7%]; H = 1.22 [1.00; 1.56]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  38.68   25  0.0397
## 
## Results for subgroups (fixed effect model):
##         k    SMD            95%-CI     Q   I^2
## < 0.5  11 0.0766 [-0.1802; 0.3334]  9.05  0.0%
## > 0.5  15 0.1335 [-0.0406; 0.3076] 26.63 47.4%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.13    1  0.7189
## Within groups  35.68   24  0.0589
## 
## Results for subgroups (random effects model):
##         k    SMD            95%-CI  tau^2    tau
## < 0.5  11 0.0766 [-0.1802; 0.3334]      0      0
## > 0.5  15 0.1744 [-0.0890; 0.4377] 0.1173 0.3425
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.27    1  0.6024
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
11.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 26; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0910 (SE = 0.0722)
## tau (square root of estimated tau^2 value):             0.3016
## I^2 (residual heterogeneity / unaccounted variability): 37.90%
## H^2 (unaccounted variability / sampling variability):   1.61
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 24) = 38.6472, p-val = 0.0297
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0027, p-val = 0.9587
## 
## Model Results:
## 
##            estimate      se    zval    pval    ci.lb   ci.ub 
## intrcpt      0.1302  0.2190  0.5943  0.5523  -0.2991  0.5594    
## men_ratio    0.0156  0.3016  0.0518  0.9587  -0.5755  0.6068    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
11.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

11.2.6 Type of Exercise

11.2.6.1 Forest plot

11.2.6.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) type_exercise
## Abdelbasset 2020        0.4282 [-0.2843;  1.1406]       4.1        4.4       Cycling
## Ciolac 2010             0.3701 [-0.4728;  1.2130]       2.9        3.6       Running
## Conraads 2015          -0.0242 [-0.3215;  0.2731]      23.4        9.1       Cycling
## Currie 2015            -1.0152 [-1.9719; -0.0586]       2.3        3.0       Cycling
## Eguchi 2012            -0.5729 [-1.4672;  0.3214]       2.6        3.3       Cycling
## Fisher 2015            -0.1696 [-0.9955;  0.6562]       3.0        3.7       Cycling
## Grieco 2013             0.9216 [ 0.0393;  1.8038]       2.7        3.3       Cycling
## Helgerud 2007          -0.0341 [-0.9107;  0.8424]       2.7        3.4       Running
## Honkala 2017 (Healthy)  0.8577 [ 0.0836;  1.6318]       3.5        4.0       Cycling
## Honkala 2017 (T2D)      0.7046 [-0.3128;  1.7221]       2.0        2.7       Cycling
## Jo 2020                 0.8928 [ 0.1878;  1.5978]       4.2        4.5       Running
## Keating 2014            0.3604 [-0.4821;  1.2029]       2.9        3.6       Cycling
## Kim 2015                0.6605 [-0.1002;  1.4213]       3.6        4.1       Running
## Lunt 2014              -0.7221 [-1.5851;  0.1409]       2.8        3.4       Running
## Lunt 2014              -0.3736 [-1.2179;  0.4707]       2.9        3.5       Running
## Madssen 2014            0.1319 [-0.5314;  0.7952]       4.7        4.8       Running
## Maillard 2016          -0.1494 [-1.1308;  0.8319]       2.2        2.8       Cycling
## Matsuo 2015             0.1370 [-0.6641;  0.9380]       3.2        3.8       Cycling
## Mitranun 2014           0.0115 [-0.7293;  0.7523]       3.8        4.2       Running
## Motiani 2017            0.9454 [ 0.1348;  1.7559]       3.2        3.8       Cycling
## Nalcakan 2014           0.2218 [-0.7957;  1.2393]       2.0        2.7       Cycling
## Ramos 2016b            -0.5535 [-1.2609;  0.1539]       4.1        4.5       Running
## Sandvei 2012            0.4295 [-0.3980;  1.2570]       3.0        3.6       Running
## Sawyer 2016            -0.0089 [-0.9328;  0.9150]       2.4        3.1       Cycling
## Winn 2018               0.0419 [-0.9382;  1.0220]       2.2        2.9       Running
## Zapata-Lamana 2018     -0.0156 [-0.7564;  0.7252]       3.8        4.2       Cycling
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.1196 [-0.0243; 0.2635] 1.63  0.1032
## Random effects model 0.1400 [-0.0502; 0.3303] 1.44  0.1491
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0801 [0.0000; 0.3233]; tau = 0.2830 [0.0000; 0.5686];
##  I^2 = 35.4% [0.0%; 59.9%]; H = 1.24 [1.00; 1.58]
## 
## Quantifying residual heterogeneity:
##  I^2 = 32.9% [0.0%; 58.8%]; H = 1.22 [1.00; 1.56]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  38.68   25  0.0397
## 
## Results for subgroups (fixed effect model):
##           k    SMD            95%-CI     Q   I^2
## Cycling  15 0.1254 [-0.0559; 0.3067] 20.69 32.3%
## Running  11 0.0988 [-0.1386; 0.3362] 15.09 33.7%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.03    1  0.8614
## Within groups  35.78   24  0.0577
## 
## Results for subgroups (random effects model):
##           k    SMD            95%-CI  tau^2    tau
## Cycling  15 0.1667 [-0.0777; 0.4112] 0.0688 0.2623
## Running  11 0.0915 [-0.2020; 0.3850] 0.0826 0.2873
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.15    1  0.6995
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
11.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 26; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0919 (SE = 0.0727)
## tau (square root of estimated tau^2 value):             0.3031
## I^2 (residual heterogeneity / unaccounted variability): 37.89%
## H^2 (unaccounted variability / sampling variability):   1.61
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 24) = 38.6428, p-val = 0.0298
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.1655, p-val = 0.6842
## 
## Model Results:
## 
##                       estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                 0.1752  0.1315   1.3327  0.1826  -0.0825  0.4329    
## type_exerciseRunning   -0.0819  0.2013  -0.4068  0.6842  -0.4765  0.3127    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
11.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

11.2.7 Baseline Values

11.2.7.1 Forest plot

11.2.7.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_bsln
## Abdelbasset 2020        0.4282 [-0.2843;  1.1406]       4.1        4.4  < 5.2 mmol/L
## Ciolac 2010             0.3701 [-0.4728;  1.2130]       2.9        3.6  < 5.2 mmol/L
## Conraads 2015          -0.0242 [-0.3215;  0.2731]      23.4        9.1  < 5.2 mmol/L
## Currie 2015            -1.0152 [-1.9719; -0.0586]       2.3        3.0  < 5.2 mmol/L
## Eguchi 2012            -0.5729 [-1.4672;  0.3214]       2.6        3.3  > 5.2 mmol/L
## Fisher 2015            -0.1696 [-0.9955;  0.6562]       3.0        3.7  < 5.2 mmol/L
## Grieco 2013             0.9216 [ 0.0393;  1.8038]       2.7        3.3  < 5.2 mmol/L
## Helgerud 2007          -0.0341 [-0.9107;  0.8424]       2.7        3.4  < 5.2 mmol/L
## Honkala 2017 (Healthy)  0.8577 [ 0.0836;  1.6318]       3.5        4.0  < 5.2 mmol/L
## Honkala 2017 (T2D)      0.7046 [-0.3128;  1.7221]       2.0        2.7  < 5.2 mmol/L
## Jo 2020                 0.8928 [ 0.1878;  1.5978]       4.2        4.5  < 5.2 mmol/L
## Keating 2014            0.3604 [-0.4821;  1.2029]       2.9        3.6  > 5.2 mmol/L
## Kim 2015                0.6605 [-0.1002;  1.4213]       3.6        4.1  < 5.2 mmol/L
## Lunt 2014              -0.7221 [-1.5851;  0.1409]       2.8        3.4  > 5.2 mmol/L
## Lunt 2014              -0.3736 [-1.2179;  0.4707]       2.9        3.5  > 5.2 mmol/L
## Madssen 2014            0.1319 [-0.5314;  0.7952]       4.7        4.8  < 5.2 mmol/L
## Maillard 2016          -0.1494 [-1.1308;  0.8319]       2.2        2.8  < 5.2 mmol/L
## Matsuo 2015             0.1370 [-0.6641;  0.9380]       3.2        3.8  > 5.2 mmol/L
## Mitranun 2014           0.0115 [-0.7293;  0.7523]       3.8        4.2  < 5.2 mmol/L
## Motiani 2017            0.9454 [ 0.1348;  1.7559]       3.2        3.8  < 5.2 mmol/L
## Nalcakan 2014           0.2218 [-0.7957;  1.2393]       2.0        2.7  < 5.2 mmol/L
## Ramos 2016b            -0.5535 [-1.2609;  0.1539]       4.1        4.5  < 5.2 mmol/L
## Sandvei 2012            0.4295 [-0.3980;  1.2570]       3.0        3.6  < 5.2 mmol/L
## Sawyer 2016            -0.0089 [-0.9328;  0.9150]       2.4        3.1  < 5.2 mmol/L
## Winn 2018               0.0419 [-0.9382;  1.0220]       2.2        2.9  < 5.2 mmol/L
## Zapata-Lamana 2018     -0.0156 [-0.7564;  0.7252]       3.8        4.2  < 5.2 mmol/L
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.1196 [-0.0243; 0.2635] 1.63  0.1032
## Random effects model 0.1400 [-0.0502; 0.3303] 1.44  0.1491
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0801 [0.0000; 0.3233]; tau = 0.2830 [0.0000; 0.5686];
##  I^2 = 35.4% [0.0%; 59.9%]; H = 1.24 [1.00; 1.58]
## 
## Quantifying residual heterogeneity:
##  I^2 = 26.4% [0.0%; 55.0%]; H = 1.17 [1.00; 1.49]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  38.68   25  0.0397
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI     Q   I^2
## < 5.2 mmol/L  21  0.1695 [ 0.0138; 0.3253] 28.36 29.5%
## > 5.2 mmol/L   5 -0.2047 [-0.5843; 0.1749]  4.25  5.9%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  3.20    1  0.0738
## Within groups  32.61   24  0.1125
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI  tau^2    tau
## < 5.2 mmol/L  21  0.2066 [ 0.0076; 0.4057] 0.0589 0.2427
## > 5.2 mmol/L   5 -0.2059 [-0.5974; 0.1855] 0.0119 0.1089
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   3.39    1  0.0656
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
11.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 26; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0887 (SE = 0.0725)
## tau (square root of estimated tau^2 value):             0.2978
## I^2 (residual heterogeneity / unaccounted variability): 35.96%
## H^2 (unaccounted variability / sampling variability):   1.56
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 24) = 37.4792, p-val = 0.0392
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.3526, p-val = 0.5527
## 
## Model Results:
## 
##                estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt         -0.2894  0.7303  -0.3963  0.6919  -1.7207  1.1419    
## bsln_adjusted    0.0939  0.1582   0.5938  0.5527  -0.2161  0.4040    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
11.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

11.2.8 Type of HIIE

11.2.8.1 Forest plot

11.2.8.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) HIIE
## Abdelbasset 2020        0.4282 [-0.2843;  1.1406]       4.1        4.4 HIIT
## Ciolac 2010             0.3701 [-0.4728;  1.2130]       2.9        3.6 HIIT
## Conraads 2015          -0.0242 [-0.3215;  0.2731]      23.4        9.1 HIIT
## Currie 2015            -1.0152 [-1.9719; -0.0586]       2.3        3.0 HIIT
## Eguchi 2012            -0.5729 [-1.4672;  0.3214]       2.6        3.3 HIIT
## Fisher 2015            -0.1696 [-0.9955;  0.6562]       3.0        3.7  SIT
## Grieco 2013             0.9216 [ 0.0393;  1.8038]       2.7        3.3 HIIT
## Helgerud 2007          -0.0341 [-0.9107;  0.8424]       2.7        3.4 HIIT
## Honkala 2017 (Healthy)  0.8577 [ 0.0836;  1.6318]       3.5        4.0  SIT
## Honkala 2017 (T2D)      0.7046 [-0.3128;  1.7221]       2.0        2.7  SIT
## Jo 2020                 0.8928 [ 0.1878;  1.5978]       4.2        4.5 HIIT
## Keating 2014            0.3604 [-0.4821;  1.2029]       2.9        3.6 HIIT
## Kim 2015                0.6605 [-0.1002;  1.4213]       3.6        4.1 HIIT
## Lunt 2014              -0.7221 [-1.5851;  0.1409]       2.8        3.4 HIIT
## Lunt 2014              -0.3736 [-1.2179;  0.4707]       2.9        3.5  SIT
## Madssen 2014            0.1319 [-0.5314;  0.7952]       4.7        4.8 HIIT
## Maillard 2016          -0.1494 [-1.1308;  0.8319]       2.2        2.8 HIIT
## Matsuo 2015             0.1370 [-0.6641;  0.9380]       3.2        3.8 HIIT
## Mitranun 2014           0.0115 [-0.7293;  0.7523]       3.8        4.2 HIIT
## Motiani 2017            0.9454 [ 0.1348;  1.7559]       3.2        3.8  SIT
## Nalcakan 2014           0.2218 [-0.7957;  1.2393]       2.0        2.7  SIT
## Ramos 2016b            -0.5535 [-1.2609;  0.1539]       4.1        4.5 HIIT
## Sandvei 2012            0.4295 [-0.3980;  1.2570]       3.0        3.6  SIT
## Sawyer 2016            -0.0089 [-0.9328;  0.9150]       2.4        3.1 HIIT
## Winn 2018               0.0419 [-0.9382;  1.0220]       2.2        2.9 HIIT
## Zapata-Lamana 2018     -0.0156 [-0.7564;  0.7252]       3.8        4.2 HIIT
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.1196 [-0.0243; 0.2635] 1.63  0.1032
## Random effects model 0.1400 [-0.0502; 0.3303] 1.44  0.1491
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0801 [0.0000; 0.3233]; tau = 0.2830 [0.0000; 0.5686];
##  I^2 = 35.4% [0.0%; 59.9%]; H = 1.24 [1.00; 1.58]
## 
## Quantifying residual heterogeneity:
##  I^2 = 27.2% [0.0%; 55.5%]; H = 1.17 [1.00; 1.50]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  38.68   25  0.0397
## 
## Results for subgroups (fixed effect model):
##        k    SMD            95%-CI     Q   I^2
## HIIT  19 0.0545 [-0.1061; 0.2151] 24.98 27.9%
## SIT    7 0.3673 [ 0.0413; 0.6934]  7.98 24.8%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  2.85    1  0.0916
## Within groups  32.96   24  0.1048
## 
## Results for subgroups (random effects model):
##        k    SMD            95%-CI  tau^2    tau
## HIIT  19 0.0642 [-0.1394; 0.2677] 0.0530 0.2302
## SIT    7 0.3642 [-0.0135; 0.7419] 0.0644 0.2537
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.88    1  0.1705
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
11.2.8.3 Meta-regression
## 
## Mixed-Effects Model (k = 26; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0708 (SE = 0.0651)
## tau (square root of estimated tau^2 value):             0.2661
## I^2 (residual heterogeneity / unaccounted variability): 32.49%
## H^2 (unaccounted variability / sampling variability):   1.48
## R^2 (amount of heterogeneity accounted for):            11.60%
## 
## Test for Residual Heterogeneity:
## QE(df = 24) = 35.5486, p-val = 0.0607
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.9847, p-val = 0.1589
## 
## Model Results:
## 
##          estimate      se    zval    pval    ci.lb   ci.ub 
## intrcpt    0.0649  0.1088  0.5964  0.5509  -0.1484  0.2782    
## HIIESIT    0.3143  0.2231  1.4088  0.1589  -0.1230  0.7516    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
11.2.8.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

12. C-reactive Protein

12.1 Overall

12.1.1 Forest plot

12.1.2 R output

##                   SMD            95%-CI %W(fixed) %W(random)
## Conraads 2015 -0.1003 [-0.3978; 0.1971]      57.1       50.5
## Hovanloo 2013 -0.4405 [-1.4323; 0.5513]       5.1        6.0
## Keating 2014  -0.4402 [-1.2860; 0.4056]       7.1        8.2
## Kim 2015      -0.2307 [-0.9739; 0.5126]       9.1       10.5
## Madssen 2014   0.5620 [-0.1132; 1.2371]      11.1       12.5
## Nalcakan 2014 -0.7619 [-1.8123; 0.2885]       4.6        5.4
## Sawyer 2016   -0.0607 [-0.9849; 0.8634]       5.9        6.9
## 
## Number of studies combined: k = 7
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1083 [-0.3330; 0.1165] -0.94  0.3450
## Random effects model -0.1122 [-0.3595; 0.1351] -0.89  0.3740
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0085 [0.0000; 0.6243]; tau = 0.0921 [0.0000; 0.7901];
##  I^2 = 6.4% [0.0%; 72.7%]; H = 1.03 [1.00; 1.91]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.41    6  0.3786
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

12.1.3 Sensitivity analysis

12.1.3.1 Forest plot

12.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                              SMD            95%-CI p-value   tau^2     tau    I^2
## Omitting Conraads 2015   -0.1274 [-0.5043; 0.2495]  0.5077  0.0345  0.1856  15.5%
## Omitting Hovanloo 2013   -0.0876 [-0.3511; 0.1759]  0.5147  0.0119  0.1090   9.3%
## Omitting Keating 2014    -0.0789 [-0.3356; 0.1779]  0.5471  0.0079  0.0888   6.2%
## Omitting Kim 2015        -0.0967 [-0.3857; 0.1924]  0.5122  0.0204  0.1427  13.9%
## Omitting Madssen 2014    -0.1851 [-0.4236; 0.0534]  0.1282  0.0000  0.0000   0.0%
## Omitting Nalcakan 2014   -0.0748 [-0.3049; 0.1554]  0.5243  0.0000  0.0000   0.0%
## Omitting Sawyer 2016     -0.1166 [-0.4013; 0.1682]  0.4225  0.0213  0.1458  15.3%
##                                                                                  
## Pooled estimate          -0.1122 [-0.3595; 0.1351]  0.3740  0.0085  0.0921   6.4%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

12.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

12.2 Subgroups

12.2.1 Overall

12.2.1.1 Forest plot

12.2.1.2 R output
##                            SMD            95%-CI     meta-analysis
##                        -0.1122 [-0.3595; 0.1351]           Overall
## Healthy                -0.5575 [-1.2818; 0.1668]        Population
## Overweight/obese       -0.2565 [-0.8810; 0.3679]        Population
## Cardiac Rehabilitation  0.0306 [-0.3678; 0.4290]        Population
## < 30 y                 -0.5575 [-1.2818; 0.1668]               Age
## 30 - 50 y              -0.2565 [-0.8810; 0.3679]               Age
## > 50 y                  0.0306 [-0.3678; 0.4290]               Age
## < 5 weeks              -0.4165 [-1.4105; 0.5775] Training Duration
## 5 - 10 weeks           -0.2876 [-0.7956; 0.2205] Training Duration
## > 10 weeks              0.0115 [-0.4491; 0.4721] Training Duration
## < 0.5                  -0.3018 [-0.8306; 0.2270]         Men Ratio
## > 0.5                  -0.0504 [-0.4433; 0.3425]         Men Ratio
## Cycling                -0.1807 [-0.4324; 0.0711]  Type of Exercise
## Running                 0.1789 [-0.5784; 0.9361]  Type of Exercise
## < 2 mg/L               -0.1385 [-0.4045; 0.1276]   Baseline Values
## > 2 mg/L               -0.0844 [-0.6548; 0.4860]   Baseline Values
## HIIT                   -0.0536 [-0.2976; 0.1903]      Type of HIIE
## SIT                    -0.5575 [-1.2818; 0.1668]      Type of HIIE
## 
## Number of studies combined: k = 7
## 
##                          SMD            95%-CI     z p-value
## Random effects model -0.1122 [-0.3595; 0.1351] -0.89  0.3740
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0085; tau = 0.0921; I^2 = 6.4% [0.0%; 72.7%]; H = 1.03 [1.00; 1.91]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.41    6  0.3786
## 
## Results for meta-analyses (random effects model):
##                     k     SMD            95%-CI  tau^2    tau    Q  I^2
## Overall             7 -0.1122 [-0.3595; 0.1351] 0.0085 0.0921 6.41 6.4%
## Population          7 -0.1122 [-0.3595; 0.1351] 0.0085 0.0921 6.41 6.4%
## Age                 7 -0.1122 [-0.3595; 0.1351] 0.0085 0.0921 6.41 6.4%
## Training Duration   7 -0.1122 [-0.3595; 0.1351] 0.0085 0.0921 6.41 6.4%
## Men Ratio           7 -0.1122 [-0.3595; 0.1351] 0.0085 0.0921 6.41 6.4%
## Type of Exercise    7 -0.1122 [-0.3595; 0.1351] 0.0085 0.0921 6.41 6.4%
## Baseline Values     7 -0.1122 [-0.3595; 0.1351] 0.0085 0.0921 6.41 6.4%
## Type of HIIE        7 -0.1122 [-0.3595; 0.1351] 0.0085 0.0921 6.41 6.4%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

12.2.2 Population

12.2.2.1 Forest plot

12.2.2.2 R output
##                   SMD            95%-CI %W(fixed) %W(random)             population
## Conraads 2015 -0.1003 [-0.3978; 0.1971]      57.1       50.5 Cardiac Rehabilitation
## Hovanloo 2013 -0.4405 [-1.4323; 0.5513]       5.1        6.0                Healthy
## Keating 2014  -0.4402 [-1.2860; 0.4056]       7.1        8.2       Overweight/obese
## Kim 2015      -0.2307 [-0.9739; 0.5126]       9.1       10.5 Cardiac Rehabilitation
## Madssen 2014   0.5620 [-0.1132; 1.2371]      11.1       12.5 Cardiac Rehabilitation
## Nalcakan 2014 -0.7619 [-1.8123; 0.2885]       4.6        5.4                Healthy
## Sawyer 2016   -0.0607 [-0.9849; 0.8634]       5.9        6.9       Overweight/obese
## 
## Number of studies combined: k = 7
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1083 [-0.3330; 0.1165] -0.94  0.3450
## Random effects model -0.1122 [-0.3595; 0.1351] -0.89  0.3740
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0085 [0.0000; 0.6243]; tau = 0.0921 [0.0000; 0.7901];
##  I^2 = 6.4% [0.0%; 72.7%]; H = 1.03 [1.00; 1.91]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 78.0%]; H = 1.00 [1.00; 2.13]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.41    6  0.3786
## 
## Results for subgroups (fixed effect model):
##                          k     SMD            95%-CI    Q   I^2
## Healthy                  2 -0.5575 [-1.2818; 0.1668] 0.16  0.0%
## Overweight/obese         2 -0.2565 [-0.8810; 0.3679] 0.33  0.0%
## Cardiac Rehabilitation   3 -0.0217 [-0.2774; 0.2339] 3.29 39.2%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 2.13    2  0.3442
## Within groups  3.78    4  0.4362
## 
## Results for subgroups (random effects model):
##                          k     SMD            95%-CI  tau^2    tau
## Healthy                  2 -0.5575 [-1.2818; 0.1668]      0      0
## Overweight/obese         2 -0.2565 [-0.8810; 0.3679]      0      0
## Cardiac Rehabilitation   3  0.0306 [-0.3678; 0.4290] 0.0523 0.2288
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   2.11    2  0.3474
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
12.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 7; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0850)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            100.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 4) = 3.9852, p-val = 0.4080
## 
## Test of Moderators (coefficients 2:3):
## QM(df = 2) = 2.4274, p-val = 0.2971
## 
## Model Results:
## 
##                               estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                        -0.5920  0.3679  -1.6090  0.1076  -1.3131  0.1291    
## .byvarOverweight/obese          0.3248  0.4865   0.6676  0.5044  -0.6288  1.2784    
## .byvarCardiac Rehabilitation    0.5712  0.3904   1.4632  0.1434  -0.1939  1.3363    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
12.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

12.2.3 Age

12.2.3.1 Forest plot

12.2.3.2 R output
##                   SMD            95%-CI %W(fixed) %W(random) category_age
## Conraads 2015 -0.1003 [-0.3978; 0.1971]      57.1       50.5       > 50 y
## Hovanloo 2013 -0.4405 [-1.4323; 0.5513]       5.1        6.0       < 30 y
## Keating 2014  -0.4402 [-1.2860; 0.4056]       7.1        8.2    30 - 50 y
## Kim 2015      -0.2307 [-0.9739; 0.5126]       9.1       10.5       > 50 y
## Madssen 2014   0.5620 [-0.1132; 1.2371]      11.1       12.5       > 50 y
## Nalcakan 2014 -0.7619 [-1.8123; 0.2885]       4.6        5.4       < 30 y
## Sawyer 2016   -0.0607 [-0.9849; 0.8634]       5.9        6.9    30 - 50 y
## 
## Number of studies combined: k = 7
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1083 [-0.3330; 0.1165] -0.94  0.3450
## Random effects model -0.1122 [-0.3595; 0.1351] -0.89  0.3740
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0085 [0.0000; 0.6243]; tau = 0.0921 [0.0000; 0.7901];
##  I^2 = 6.4% [0.0%; 72.7%]; H = 1.03 [1.00; 1.91]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 78.0%]; H = 1.00 [1.00; 2.13]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.41    6  0.3786
## 
## Results for subgroups (fixed effect model):
##             k     SMD            95%-CI    Q   I^2
## < 30 y      2 -0.5575 [-1.2818; 0.1668] 0.16  0.0%
## 30 - 50 y   2 -0.2565 [-0.8810; 0.3679] 0.33  0.0%
## > 50 y      3 -0.0217 [-0.2774; 0.2339] 3.29 39.2%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 2.13    2  0.3442
## Within groups  3.78    4  0.4362
## 
## Results for subgroups (random effects model):
##             k     SMD            95%-CI  tau^2    tau
## < 30 y      2 -0.5575 [-1.2818; 0.1668]      0      0
## 30 - 50 y   2 -0.2565 [-0.8810; 0.3679]      0      0
## > 50 y      3  0.0306 [-0.3678; 0.4290] 0.0523 0.2288
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   2.11    2  0.3474
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
12.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 7; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0817)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            100.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 5) = 4.3002, p-val = 0.5071
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 2.1125, p-val = 0.1461
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.8531  0.5251  -1.6245  0.1043  -1.8824  0.1761    
## age        0.0142  0.0098   1.4534  0.1461  -0.0049  0.0333    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
12.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

12.2.4 Training Duration

12.2.4.1 Forest plot

12.2.4.2 R output
##                   SMD            95%-CI %W(fixed) %W(random) category_duration
## Conraads 2015 -0.1003 [-0.3978; 0.1971]      57.1       50.5        > 10 weeks
## Hovanloo 2013 -0.4405 [-1.4323; 0.5513]       5.1        6.0         < 5 weeks
## Keating 2014  -0.4402 [-1.2860; 0.4056]       7.1        8.2        > 10 weeks
## Kim 2015      -0.2307 [-0.9739; 0.5126]       9.1       10.5      5 - 10 weeks
## Madssen 2014   0.5620 [-0.1132; 1.2371]      11.1       12.5        > 10 weeks
## Nalcakan 2014 -0.7619 [-1.8123; 0.2885]       4.6        5.4      5 - 10 weeks
## Sawyer 2016   -0.0607 [-0.9849; 0.8634]       5.9        6.9      5 - 10 weeks
## 
## Number of studies combined: k = 7
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1083 [-0.3330; 0.1165] -0.94  0.3450
## Random effects model -0.1122 [-0.3595; 0.1351] -0.89  0.3740
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0085 [0.0000; 0.6243]; tau = 0.0921 [0.0000; 0.7901];
##  I^2 = 6.4% [0.0%; 72.7%]; H = 1.03 [1.00; 1.91]
## 
## Quantifying residual heterogeneity:
##  I^2 = 16.0% [0.0%; 82.5%]; H = 1.09 [1.00; 2.39]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.41    6  0.3786
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI    Q   I^2
## < 5 weeks      1 -0.4165 [-1.4105; 0.5775] 0.00    --
## 5 - 10 weeks   3 -0.2876 [-0.7956; 0.2205] 0.90  0.0%
## > 10 weeks     3 -0.0347 [-0.2939; 0.2244] 3.86 48.2%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 1.16    2  0.5612
## Within groups  4.76    4  0.3127
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI  tau^2    tau
## < 5 weeks      1 -0.4165 [-1.4105; 0.5775]     --     --
## 5 - 10 weeks   3 -0.2876 [-0.7956; 0.2205]      0      0
## > 10 weeks     3  0.0115 [-0.4491; 0.4721] 0.0828 0.2878
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.03    2  0.5977
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
12.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 7; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0026 (SE = 0.0860)
## tau (square root of estimated tau^2 value):             0.0511
## I^2 (residual heterogeneity / unaccounted variability): 1.92%
## H^2 (unaccounted variability / sampling variability):   1.02
## R^2 (amount of heterogeneity accounted for):            69.18%
## 
## Test for Residual Heterogeneity:
## QE(df = 5) = 5.0980, p-val = 0.4040
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.3101, p-val = 0.2524
## 
## Model Results:
## 
##           estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    -0.5929  0.4387  -1.3515  0.1765  -1.4526  0.2669    
## duration    0.0465  0.0407   1.1446  0.2524  -0.0332  0.1262    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
12.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

12.2.5 Men Ratio

12.2.5.1 Forest plot

12.2.5.2 R output
##                   SMD            95%-CI %W(fixed) %W(random) category_men_ratio
## Conraads 2015 -0.1003 [-0.3978; 0.1971]      57.1       50.5              > 0.5
## Hovanloo 2013 -0.4405 [-1.4323; 0.5513]       5.1        6.0              < 0.5
## Keating 2014  -0.4402 [-1.2860; 0.4056]       7.1        8.2              < 0.5
## Kim 2015      -0.2307 [-0.9739; 0.5126]       9.1       10.5              > 0.5
## Madssen 2014   0.5620 [-0.1132; 1.2371]      11.1       12.5              > 0.5
## Nalcakan 2014 -0.7619 [-1.8123; 0.2885]       4.6        5.4              > 0.5
## Sawyer 2016   -0.0607 [-0.9849; 0.8634]       5.9        6.9              < 0.5
## 
## Number of studies combined: k = 7
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1083 [-0.3330; 0.1165] -0.94  0.3450
## Random effects model -0.1122 [-0.3595; 0.1351] -0.89  0.3740
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0085 [0.0000; 0.6243]; tau = 0.0921 [0.0000; 0.7901];
##  I^2 = 6.4% [0.0%; 72.7%]; H = 1.03 [1.00; 1.91]
## 
## Quantifying residual heterogeneity:
##  I^2 = 4.9% [0.0%; 75.9%]; H = 1.03 [1.00; 2.04]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.41    6  0.3786
## 
## Results for subgroups (fixed effect model):
##         k     SMD            95%-CI    Q   I^2
## < 0.5   3 -0.3018 [-0.8306; 0.2270] 0.40  0.0%
## > 0.5   4 -0.0601 [-0.3086; 0.1884] 4.86 38.3%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.66    1  0.4175
## Within groups  5.26    5  0.3851
## 
## Results for subgroups (random effects model):
##         k     SMD            95%-CI  tau^2    tau
## < 0.5   3 -0.3018 [-0.8306; 0.2270]      0      0
## > 0.5   4 -0.0504 [-0.4433; 0.3425] 0.0624 0.2499
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.56    1  0.4544
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
12.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 7; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0276 (SE = 0.1020)
## tau (square root of estimated tau^2 value):             0.1663
## I^2 (residual heterogeneity / unaccounted variability): 17.24%
## H^2 (unaccounted variability / sampling variability):   1.21
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 5) = 6.0415, p-val = 0.3022
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.3146, p-val = 0.5748
## 
## Model Results:
## 
##            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt     -0.3912  0.5038  -0.7765  0.4374  -1.3786  0.5962    
## men_ratio    0.3595  0.6409   0.5609  0.5748  -0.8966  1.6156    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
12.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

12.2.6 Type of Exercise

12.2.6.1 Forest plot

12.2.6.2 R output
##                   SMD            95%-CI %W(fixed) %W(random) type_exercise
## Conraads 2015 -0.1003 [-0.3978; 0.1971]      57.1       50.5       Cycling
## Hovanloo 2013 -0.4405 [-1.4323; 0.5513]       5.1        6.0       Cycling
## Keating 2014  -0.4402 [-1.2860; 0.4056]       7.1        8.2       Cycling
## Kim 2015      -0.2307 [-0.9739; 0.5126]       9.1       10.5       Running
## Madssen 2014   0.5620 [-0.1132; 1.2371]      11.1       12.5       Running
## Nalcakan 2014 -0.7619 [-1.8123; 0.2885]       4.6        5.4       Cycling
## Sawyer 2016   -0.0607 [-0.9849; 0.8634]       5.9        6.9       Cycling
## 
## Number of studies combined: k = 7
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1083 [-0.3330; 0.1165] -0.94  0.3450
## Random effects model -0.1122 [-0.3595; 0.1351] -0.89  0.3740
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0085 [0.0000; 0.6243]; tau = 0.0921 [0.0000; 0.7901];
##  I^2 = 6.4% [0.0%; 72.7%]; H = 1.03 [1.00; 1.91]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 69.4%]; H = 1.00 [1.00; 1.81]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.41    6  0.3786
## 
## Results for subgroups (fixed effect model):
##           k     SMD            95%-CI    Q   I^2
## Cycling   5 -0.1807 [-0.4324; 0.0711] 1.87  0.0%
## Running   2  0.1994 [-0.3008; 0.6996] 2.28 56.1%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 1.77    1  0.1835
## Within groups  4.15    5  0.5285
## 
## Results for subgroups (random effects model):
##           k     SMD            95%-CI  tau^2    tau
## Cycling   5 -0.1807 [-0.4324; 0.0711]      0      0
## Running   2  0.1789 [-0.5784; 0.9361] 0.1676 0.4094
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.78    1  0.3772
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
12.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 7; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0879)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            100.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 5) = 4.5375, p-val = 0.4749
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.8752, p-val = 0.1709
## 
## Model Results:
## 
##                       estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                -0.1873  0.1284  -1.4592  0.1445  -0.4389  0.0643    
## type_exerciseRunning    0.3909  0.2855   1.3694  0.1709  -0.1686  0.9505    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
12.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

12.2.7 Baseline Values

12.2.7.1 Forest plot

12.2.7.2 R output
##                   SMD            95%-CI %W(fixed) %W(random) category_bsln
## Conraads 2015 -0.1003 [-0.3978; 0.1971]      57.1       50.5      < 2 mg/L
## Hovanloo 2013 -0.4405 [-1.4323; 0.5513]       5.1        6.0      < 2 mg/L
## Keating 2014  -0.4402 [-1.2860; 0.4056]       7.1        8.2      > 2 mg/L
## Kim 2015      -0.2307 [-0.9739; 0.5126]       9.1       10.5      < 2 mg/L
## Madssen 2014   0.5620 [-0.1132; 1.2371]      11.1       12.5      > 2 mg/L
## Nalcakan 2014 -0.7619 [-1.8123; 0.2885]       4.6        5.4      > 2 mg/L
## Sawyer 2016   -0.0607 [-0.9849; 0.8634]       5.9        6.9      > 2 mg/L
## 
## Number of studies combined: k = 7
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1083 [-0.3330; 0.1165] -0.94  0.3450
## Random effects model -0.1122 [-0.3595; 0.1351] -0.89  0.3740
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0085 [0.0000; 0.6243]; tau = 0.0921 [0.0000; 0.7901];
##  I^2 = 6.4% [0.0%; 72.7%]; H = 1.03 [1.00; 1.91]
## 
## Quantifying residual heterogeneity:
##  I^2 = 12.1% [0.0%; 77.7%]; H = 1.07 [1.00; 2.12]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.41    6  0.3786
## 
## Results for subgroups (fixed effect model):
##            k     SMD            95%-CI    Q   I^2
## < 2 mg/L   3 -0.1385 [-0.4045; 0.1276] 0.42  0.0%
## > 2 mg/L   4 -0.0172 [-0.4380; 0.4037] 5.27 43.1%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.23    1  0.6331
## Within groups  5.69    5  0.3377
## 
## Results for subgroups (random effects model):
##            k     SMD            95%-CI  tau^2    tau
## < 2 mg/L   3 -0.1385 [-0.4045; 0.1276]      0      0
## > 2 mg/L   4 -0.0844 [-0.6548; 0.4860] 0.1453 0.3812
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.03    1  0.8663
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
12.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 7; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0379 (SE = 0.1128)
## tau (square root of estimated tau^2 value):             0.1946
## I^2 (residual heterogeneity / unaccounted variability): 21.37%
## H^2 (unaccounted variability / sampling variability):   1.27
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 5) = 6.3588, p-val = 0.2729
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0279, p-val = 0.8674
## 
## Model Results:
## 
##                estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt         -0.1719  0.3218  -0.5340  0.5933  -0.8027  0.4589    
## bsln_adjusted    0.0226  0.1353   0.1669  0.8674  -0.2426  0.2877    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
12.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

12.2.8 Type of HIIE

12.2.8.1 Forest plot

12.2.8.2 R output
##                   SMD            95%-CI %W(fixed) %W(random) HIIE
## Conraads 2015 -0.1003 [-0.3978; 0.1971]      57.1       50.5 HIIT
## Hovanloo 2013 -0.4405 [-1.4323; 0.5513]       5.1        6.0  SIT
## Keating 2014  -0.4402 [-1.2860; 0.4056]       7.1        8.2 HIIT
## Kim 2015      -0.2307 [-0.9739; 0.5126]       9.1       10.5 HIIT
## Madssen 2014   0.5620 [-0.1132; 1.2371]      11.1       12.5 HIIT
## Nalcakan 2014 -0.7619 [-1.8123; 0.2885]       4.6        5.4  SIT
## Sawyer 2016   -0.0607 [-0.9849; 0.8634]       5.9        6.9 HIIT
## 
## Number of studies combined: k = 7
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.1083 [-0.3330; 0.1165] -0.94  0.3450
## Random effects model -0.1122 [-0.3595; 0.1351] -0.89  0.3740
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0085 [0.0000; 0.6243]; tau = 0.0921 [0.0000; 0.7901];
##  I^2 = 6.4% [0.0%; 72.7%]; H = 1.03 [1.00; 1.91]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 70.1%]; H = 1.00 [1.00; 1.83]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.41    6  0.3786
## 
## Results for subgroups (fixed effect model):
##        k     SMD            95%-CI    Q  I^2
## HIIT   5 -0.0554 [-0.2920; 0.1812] 4.08 2.0%
## SIT    2 -0.5575 [-1.2818; 0.1668] 0.16 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 1.67    1  0.1965
## Within groups  4.25    5  0.5143
## 
## Results for subgroups (random effects model):
##        k     SMD            95%-CI  tau^2    tau
## HIIT   5 -0.0536 [-0.2976; 0.1903] 0.0022 0.0464
## SIT    2 -0.5575 [-1.2818; 0.1668]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.67    1  0.1963
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
12.2.8.3 Meta-regression
## 
## Mixed-Effects Model (k = 7; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0745)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            100.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 5) = 4.4981, p-val = 0.4801
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.9145, p-val = 0.1665
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.0562  0.1207  -0.4660  0.6412  -0.2928  0.1803    
## HIIESIT   -0.5358  0.3872  -1.3837  0.1665  -1.2947  0.2231    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
12.2.8.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

13. Fasting Insulin

13.1 Overall

13.1.1 Forest plot

13.1.2 R output

##                            SMD             95%-CI %W(fixed) %W(random)
## Ciolac 2010             0.1742 [-0.6631;  1.0115]       5.2        5.2
## Gillen 2016             0.0000 [-0.9005;  0.9005]       4.5        4.5
## Grieco 2013             0.3461 [-0.4993;  1.1915]       5.1        5.1
## Honkala 2017 (Healthy)  0.1562 [-0.5857;  0.8981]       6.6        6.6
## Honkala 2017 (T2D)     -0.0238 [-1.0115;  0.9640]       3.7        3.7
## Keating 2014           -0.0439 [-0.8797;  0.7919]       5.2        5.2
## Matsuo 2015             0.0148 [-0.7854;  0.8150]       5.7        5.7
## Motiani 2017            0.3542 [-0.4206;  1.1289]       6.0        6.0
## Ramos 2016a            -0.3385 [-0.9407;  0.2637]      10.0       10.0
## Robinson 2015           0.2792 [-0.3518;  0.9101]       9.1        9.1
## Sandvei 2012           -0.2927 [-1.1152;  0.5298]       5.4        5.4
## Sawyer 2016            -0.0258 [-0.9498;  0.8982]       4.2        4.2
## Sjöros 2018             0.0543 [-0.8022;  0.9108]       4.9        4.9
## Skleryk 2013           -0.0164 [-0.9964;  0.9636]       3.8        3.8
## Tjønna 2008            -0.0090 [-0.9197;  0.9018]       4.4        4.4
## Trapp 2008              0.6862 [-0.0503;  1.4226]       6.7        6.7
## Winn 2018              -1.2046 [-2.2698; -0.1395]       3.2        3.2
## Zapata-Lamana 2018     -0.2451 [-0.9887;  0.4985]       6.5        6.5
## 
## Number of studies combined: k = 18
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0237 [-0.1666; 0.2139] 0.24  0.8074
## Random effects model 0.0237 [-0.1666; 0.2139] 0.24  0.8074
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1336]; tau = 0 [0.0000; 0.3655];
##  I^2 = 0.0% [0.0%; 33.9%]; H = 1.00 [1.00; 1.23]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  12.88   17  0.7445
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

13.1.3 Sensitivity analysis

13.1.3.1 Forest plot

13.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                                       SMD            95%-CI p-value   tau^2     tau   I^2
## Omitting Ciolac 2010               0.0168 [-0.1788; 0.2123]  0.8664  0.0000  0.0000  0.0%
## Omitting Gillen 2016               0.0257 [-0.1691; 0.2206]  0.7957  0.0000  0.0000  0.0%
## Omitting Grieco 2013               0.0081 [-0.1873; 0.2036]  0.9350  0.0000  0.0000  0.0%
## Omitting Honkala 2017 (Healthy)    0.0156 [-0.1814; 0.2127]  0.8765  0.0000  0.0000  0.0%
## Omitting Honkala 2017 (T2D)        0.0264 [-0.1677; 0.2205]  0.7898  0.0000  0.0000  0.0%
## Omitting Keating 2014              0.0282 [-0.1673; 0.2238]  0.7771  0.0000  0.0000  0.0%
## Omitting Matsuo 2015               0.0252 [-0.1709; 0.2213]  0.8011  0.0000  0.0000  0.0%
## Omitting Motiani 2017              0.0041 [-0.1923; 0.2006]  0.9670  0.0000  0.0000  0.0%
## Omitting Ramos 2016a               0.0642 [-0.1365; 0.2649]  0.5308  0.0000  0.0000  0.0%
## Omitting Robinson 2015            -0.0003 [-0.2001; 0.1994]  0.9973  0.0000  0.0000  0.0%
## Omitting Sandvei 2012              0.0419 [-0.1538; 0.2377]  0.6746  0.0000  0.0000  0.0%
## Omitting Sawyer 2016               0.0268 [-0.1678; 0.2214]  0.7875  0.0000  0.0000  0.0%
## Omitting Sjöros 2018               0.0232 [-0.1722; 0.2185]  0.8163  0.0000  0.0000  0.0%
## Omitting Skleryk 2013              0.0262 [-0.1680; 0.2203]  0.7917  0.0000  0.0000  0.0%
## Omitting Tjønna 2008               0.0261 [-0.1686; 0.2208]  0.7928  0.0000  0.0000  0.0%
## Omitting Trapp 2008               -0.0212 [-0.2184; 0.1759]  0.8327  0.0000  0.0000  0.0%
## Omitting Winn 2018                 0.0619 [-0.1316; 0.2554]  0.5306  0.0000  0.0000  0.0%
## Omitting Zapata-Lamana 2018        0.0430 [-0.1540; 0.2400]  0.6688  0.0000  0.0000  0.0%
##                                                                                          
## Pooled estimate                    0.0237 [-0.1666; 0.2139]  0.8074  0.0000  0.0000  0.0%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

13.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

13.2 Subgroups

13.2.1 Overall

13.2.1.1 Forest plot

13.2.1.2 R output
##                        SMD            95%-CI     meta-analysis
##                     0.0237 [-0.1666; 0.2139]           Overall
## Healthy             0.2174 [-0.0862; 0.5211]        Population
## Overweight/obese   -0.2396 [-0.6377; 0.1586]        Population
## Metabolic Syndrome -0.0270 [-0.3798; 0.3258]        Population
## T2D                 0.0201 [-0.6270; 0.6672]        Population
## < 30 y              0.1180 [-0.2122; 0.4481]               Age
## 30 - 50 y          -0.0138 [-0.3034; 0.2757]               Age
## > 50 y             -0.0370 [-0.4301; 0.3562]               Age
## < 5 weeks           0.0902 [-0.2025; 0.3829] Training Duration
## 5 - 10 weeks       -0.1005 [-0.5879; 0.3869] Training Duration
## > 10 weeks          0.0041 [-0.2884; 0.2965] Training Duration
## < 0.5               0.0417 [-0.2249; 0.3083]         Men Ratio
## > 0.5              -0.0012 [-0.2845; 0.2821]         Men Ratio
## Running            -0.2694 [-0.6293; 0.0906]  Type of Exercise
## Cycling             0.1389 [-0.0856; 0.3633]  Type of Exercise
## < 40 pmol/L         0.1317 [-0.1912; 0.4546]   Baseline Values
## > 40 pmol/L        -0.0553 [-0.3020; 0.1914]   Baseline Values
## HIIT               -0.0625 [-0.3116; 0.1865]      Type of HIIE
## SIT                 0.1472 [-0.1483; 0.4427]      Type of HIIE
## 
## Number of studies combined: k = 18
## 
##                         SMD            95%-CI    z p-value
## Random effects model 0.0237 [-0.1666; 0.2139] 0.24  0.8074
## 
## Quantifying heterogeneity:
##  tau^2 = 0; tau = 0; I^2 = 0.0% [0.0%; 33.9%]; H = 1.00 [1.00; 1.23]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  12.88   17  0.7445
## 
## Results for meta-analyses (random effects model):
##                     k    SMD            95%-CI tau^2 tau     Q  I^2
## Overall            18 0.0237 [-0.1666; 0.2139]     0   0 12.88 0.0%
## Population         18 0.0237 [-0.1666; 0.2139]     0   0 12.88 0.0%
## Age                18 0.0237 [-0.1666; 0.2139]     0   0 12.88 0.0%
## Training Duration  18 0.0237 [-0.1666; 0.2139]     0   0 12.88 0.0%
## Men Ratio          18 0.0237 [-0.1666; 0.2139]     0   0 12.88 0.0%
## Type of Exercise   18 0.0237 [-0.1666; 0.2139]     0   0 12.88 0.0%
## Baseline Values    18 0.0237 [-0.1666; 0.2139]     0   0 12.88 0.0%
## Type of HIIE       18 0.0237 [-0.1666; 0.2139]     0   0 12.88 0.0%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

13.2.2 Population

13.2.2.1 Forest plot

13.2.2.2 R output
##                            SMD             95%-CI %W(fixed) %W(random)         population
## Ciolac 2010             0.1742 [-0.6631;  1.0115]       5.2        5.2            Healthy
## Gillen 2016             0.0000 [-0.9005;  0.9005]       4.5        4.5            Healthy
## Grieco 2013             0.3461 [-0.4993;  1.1915]       5.1        5.1            Healthy
## Honkala 2017 (Healthy)  0.1562 [-0.5857;  0.8981]       6.6        6.6            Healthy
## Honkala 2017 (T2D)     -0.0238 [-1.0115;  0.9640]       3.7        3.7                T2D
## Keating 2014           -0.0439 [-0.8797;  0.7919]       5.2        5.2   Overweight/obese
## Matsuo 2015             0.0148 [-0.7854;  0.8150]       5.7        5.7 Metabolic Syndrome
## Motiani 2017            0.3542 [-0.4206;  1.1289]       6.0        6.0            Healthy
## Ramos 2016a            -0.3385 [-0.9407;  0.2637]      10.0       10.0 Metabolic Syndrome
## Robinson 2015           0.2792 [-0.3518;  0.9101]       9.1        9.1 Metabolic Syndrome
## Sandvei 2012           -0.2927 [-1.1152;  0.5298]       5.4        5.4            Healthy
## Sawyer 2016            -0.0258 [-0.9498;  0.8982]       4.2        4.2   Overweight/obese
## Sjöros 2018             0.0543 [-0.8022;  0.9108]       4.9        4.9                T2D
## Skleryk 2013           -0.0164 [-0.9964;  0.9636]       3.8        3.8   Overweight/obese
## Tjønna 2008            -0.0090 [-0.9197;  0.9018]       4.4        4.4 Metabolic Syndrome
## Trapp 2008              0.6862 [-0.0503;  1.4226]       6.7        6.7            Healthy
## Winn 2018              -1.2046 [-2.2698; -0.1395]       3.2        3.2   Overweight/obese
## Zapata-Lamana 2018     -0.2451 [-0.9887;  0.4985]       6.5        6.5   Overweight/obese
## 
## Number of studies combined: k = 18
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0237 [-0.1666; 0.2139] 0.24  0.8074
## Random effects model 0.0237 [-0.1666; 0.2139] 0.24  0.8074
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1336]; tau = 0 [0.0000; 0.3655];
##  I^2 = 0.0% [0.0%; 33.9%]; H = 1.00 [1.00; 1.23]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 23.1%]; H = 1.00 [1.00; 1.14]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  12.88   17  0.7445
## 
## Results for subgroups (fixed effect model):
##                      k     SMD            95%-CI    Q  I^2
## Healthy              7  0.2174 [-0.0862; 0.5211] 3.28 0.0%
## Overweight/obese     5 -0.2396 [-0.6377; 0.1586] 3.29 0.0%
## Metabolic Syndrome   4 -0.0270 [-0.3798; 0.3258] 1.87 0.0%
## T2D                  2  0.0201 [-0.6270; 0.6672] 0.01 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 3.32    3  0.3445
## Within groups  8.45   14  0.8646
## 
## Results for subgroups (random effects model):
##                      k     SMD            95%-CI tau^2 tau
## Healthy              7  0.2174 [-0.0862; 0.5211]     0   0
## Overweight/obese     5 -0.2396 [-0.6377; 0.1586]     0   0
## Metabolic Syndrome   4 -0.0270 [-0.3798; 0.3258]     0   0
## T2D                  2  0.0201 [-0.6270; 0.6672]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   3.32    3  0.3445
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
13.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 18; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0648)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 14) = 9.2220, p-val = 0.8166
## 
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 3.6531, p-val = 0.3014
## 
## Model Results:
## 
##                           estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                     0.2245  0.1548   1.4503  0.1470  -0.0789  0.5279    
## .byvarOverweight/obese     -0.4795  0.2551  -1.8799  0.0601  -0.9794  0.0204  . 
## .byvarMetabolic Syndrome   -0.2519  0.2374  -1.0609  0.2887  -0.7171  0.2134    
## .byvarT2D                  -0.2037  0.3647  -0.5587  0.5764  -0.9184  0.5110    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
13.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

13.2.3 Age

13.2.3.1 Forest plot

13.2.3.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_age
## Ciolac 2010             0.1742 [-0.6631;  1.0115]       5.2        5.2       < 30 y
## Gillen 2016             0.0000 [-0.9005;  0.9005]       4.5        4.5       < 30 y
## Grieco 2013             0.3461 [-0.4993;  1.1915]       5.1        5.1       < 30 y
## Honkala 2017 (Healthy)  0.1562 [-0.5857;  0.8981]       6.6        6.6    30 - 50 y
## Honkala 2017 (T2D)     -0.0238 [-1.0115;  0.9640]       3.7        3.7    30 - 50 y
## Keating 2014           -0.0439 [-0.8797;  0.7919]       5.2        5.2    30 - 50 y
## Matsuo 2015             0.0148 [-0.7854;  0.8150]       5.7        5.7    30 - 50 y
## Motiani 2017            0.3542 [-0.4206;  1.1289]       6.0        6.0    30 - 50 y
## Ramos 2016a            -0.3385 [-0.9407;  0.2637]      10.0       10.0       > 50 y
## Robinson 2015           0.2792 [-0.3518;  0.9101]       9.1        9.1       > 50 y
## Sandvei 2012           -0.2927 [-1.1152;  0.5298]       5.4        5.4       < 30 y
## Sawyer 2016            -0.0258 [-0.9498;  0.8982]       4.2        4.2    30 - 50 y
## Sjöros 2018             0.0543 [-0.8022;  0.9108]       4.9        4.9    30 - 50 y
## Skleryk 2013           -0.0164 [-0.9964;  0.9636]       3.8        3.8    30 - 50 y
## Tjønna 2008            -0.0090 [-0.9197;  0.9018]       4.4        4.4       > 50 y
## Trapp 2008              0.6862 [-0.0503;  1.4226]       6.7        6.7       < 30 y
## Winn 2018              -1.2046 [-2.2698; -0.1395]       3.2        3.2    30 - 50 y
## Zapata-Lamana 2018     -0.2451 [-0.9887;  0.4985]       6.5        6.5       < 30 y
## 
## Number of studies combined: k = 18
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0237 [-0.1666; 0.2139] 0.24  0.8074
## Random effects model 0.0237 [-0.1666; 0.2139] 0.24  0.8074
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1336]; tau = 0 [0.0000; 0.3655];
##  I^2 = 0.0% [0.0%; 33.9%]; H = 1.00 [1.00; 1.23]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 36.7%]; H = 1.00 [1.00; 1.26]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  12.88   17  0.7445
## 
## Results for subgroups (fixed effect model):
##             k     SMD            95%-CI    Q  I^2
## < 30 y      6  0.1180 [-0.2122; 0.4481] 4.24 0.0%
## 30 - 50 y   9 -0.0138 [-0.3034; 0.2757] 5.20 0.0%
## > 50 y      3 -0.0370 [-0.4301; 0.3562] 1.86 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.47    2  0.7909
## Within groups  11.30   15  0.7308
## 
## Results for subgroups (random effects model):
##             k     SMD            95%-CI tau^2 tau
## < 30 y      6  0.1180 [-0.2122; 0.4481]     0   0
## 30 - 50 y   9 -0.0138 [-0.3034; 0.2757]     0   0
## > 50 y      3 -0.0370 [-0.4301; 0.3562]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.47    2  0.7909
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
13.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 18; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0611)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 16) = 12.3043, p-val = 0.7228
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.5709, p-val = 0.4499
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    0.2613  0.3291   0.7939  0.4273  -0.3838  0.9063    
## age       -0.0059  0.0078  -0.7556  0.4499  -0.0213  0.0094    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
13.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

13.2.4 Training Duration

13.2.4.1 Forest plot

13.2.4.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_duration
## Ciolac 2010             0.1742 [-0.6631;  1.0115]       5.2        5.2        > 10 weeks
## Gillen 2016             0.0000 [-0.9005;  0.9005]       4.5        4.5        > 10 weeks
## Grieco 2013             0.3461 [-0.4993;  1.1915]       5.1        5.1         < 5 weeks
## Honkala 2017 (Healthy)  0.1562 [-0.5857;  0.8981]       6.6        6.6         < 5 weeks
## Honkala 2017 (T2D)     -0.0238 [-1.0115;  0.9640]       3.7        3.7         < 5 weeks
## Keating 2014           -0.0439 [-0.8797;  0.7919]       5.2        5.2        > 10 weeks
## Matsuo 2015             0.0148 [-0.7854;  0.8150]       5.7        5.7      5 - 10 weeks
## Motiani 2017            0.3542 [-0.4206;  1.1289]       6.0        6.0         < 5 weeks
## Ramos 2016a            -0.3385 [-0.9407;  0.2637]      10.0       10.0        > 10 weeks
## Robinson 2015           0.2792 [-0.3518;  0.9101]       9.1        9.1         < 5 weeks
## Sandvei 2012           -0.2927 [-1.1152;  0.5298]       5.4        5.4      5 - 10 weeks
## Sawyer 2016            -0.0258 [-0.9498;  0.8982]       4.2        4.2      5 - 10 weeks
## Sjöros 2018             0.0543 [-0.8022;  0.9108]       4.9        4.9         < 5 weeks
## Skleryk 2013           -0.0164 [-0.9964;  0.9636]       3.8        3.8         < 5 weeks
## Tjønna 2008            -0.0090 [-0.9197;  0.9018]       4.4        4.4        > 10 weeks
## Trapp 2008              0.6862 [-0.0503;  1.4226]       6.7        6.7        > 10 weeks
## Winn 2018              -1.2046 [-2.2698; -0.1395]       3.2        3.2         < 5 weeks
## Zapata-Lamana 2018     -0.2451 [-0.9887;  0.4985]       6.5        6.5        > 10 weeks
## 
## Number of studies combined: k = 18
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0237 [-0.1666; 0.2139] 0.24  0.8074
## Random effects model 0.0237 [-0.1666; 0.2139] 0.24  0.8074
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1336]; tau = 0 [0.0000; 0.3655];
##  I^2 = 0.0% [0.0%; 33.9%]; H = 1.00 [1.00; 1.23]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 36.7%]; H = 1.00 [1.00; 1.26]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  12.88   17  0.7445
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI    Q  I^2
## < 5 weeks      8  0.0902 [-0.2025; 0.3829] 6.15 0.0%
## 5 - 10 weeks   3 -0.1005 [-0.5879; 0.3869] 0.29 0.0%
## > 10 weeks     7  0.0041 [-0.2884; 0.2965] 4.87 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.47    2  0.7925
## Within groups  11.31   15  0.7305
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI tau^2 tau
## < 5 weeks      8  0.0902 [-0.2025; 0.3829]     0   0
## 5 - 10 weeks   3 -0.1005 [-0.5879; 0.3869]     0   0
## > 10 weeks     7  0.0041 [-0.2884; 0.2965]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.47    2  0.7925
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
13.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 18; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0611)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 16) = 12.7039, p-val = 0.6943
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.1713, p-val = 0.6789
## 
## Model Results:
## 
##           estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt     0.0827  0.1726   0.4794  0.6317  -0.2556  0.4211    
## duration   -0.0071  0.0172  -0.4139  0.6789  -0.0408  0.0266    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
13.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

13.2.5 Men Ratio

13.2.5.1 Forest plot

13.2.5.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_men_ratio
## Ciolac 2010             0.1742 [-0.6631;  1.0115]       5.2        5.2              < 0.5
## Gillen 2016             0.0000 [-0.9005;  0.9005]       4.5        4.5              > 0.5
## Grieco 2013             0.3461 [-0.4993;  1.1915]       5.1        5.1              < 0.5
## Honkala 2017 (Healthy)  0.1562 [-0.5857;  0.8981]       6.6        6.6              > 0.5
## Honkala 2017 (T2D)     -0.0238 [-1.0115;  0.9640]       3.7        3.7              > 0.5
## Keating 2014           -0.0439 [-0.8797;  0.7919]       5.2        5.2              < 0.5
## Matsuo 2015             0.0148 [-0.7854;  0.8150]       5.7        5.7              > 0.5
## Motiani 2017            0.3542 [-0.4206;  1.1289]       6.0        6.0              > 0.5
## Ramos 2016a            -0.3385 [-0.9407;  0.2637]      10.0       10.0              > 0.5
## Robinson 2015           0.2792 [-0.3518;  0.9101]       9.1        9.1              < 0.5
## Sandvei 2012           -0.2927 [-1.1152;  0.5298]       5.4        5.4              < 0.5
## Sawyer 2016            -0.0258 [-0.9498;  0.8982]       4.2        4.2              < 0.5
## Sjöros 2018             0.0543 [-0.8022;  0.9108]       4.9        4.9              > 0.5
## Skleryk 2013           -0.0164 [-0.9964;  0.9636]       3.8        3.8              > 0.5
## Tjønna 2008            -0.0090 [-0.9197;  0.9018]       4.4        4.4              < 0.5
## Trapp 2008              0.6862 [-0.0503;  1.4226]       6.7        6.7              < 0.5
## Winn 2018              -1.2046 [-2.2698; -0.1395]       3.2        3.2              < 0.5
## Zapata-Lamana 2018     -0.2451 [-0.9887;  0.4985]       6.5        6.5              < 0.5
## 
## Number of studies combined: k = 18
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0237 [-0.1666; 0.2139] 0.24  0.8074
## Random effects model 0.0237 [-0.1666; 0.2139] 0.24  0.8074
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1336]; tau = 0 [0.0000; 0.3655];
##  I^2 = 0.0% [0.0%; 33.9%]; H = 1.00 [1.00; 1.23]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 33.2%]; H = 1.00 [1.00; 1.22]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  12.88   17  0.7445
## 
## Results for subgroups (fixed effect model):
##         k     SMD            95%-CI    Q  I^2
## < 0.5  10  0.0459 [-0.2114; 0.3031] 9.62 6.4%
## > 0.5   8 -0.0012 [-0.2845; 0.2821] 2.10 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.06    1  0.8095
## Within groups  11.72   16  0.7634
## 
## Results for subgroups (random effects model):
##         k     SMD            95%-CI  tau^2    tau
## < 0.5  10  0.0417 [-0.2249; 0.3083] 0.0119 0.1091
## > 0.5   8 -0.0012 [-0.2845; 0.2821]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.05    1  0.8288
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
13.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 18; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0606)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 16) = 12.7336, p-val = 0.6921
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.1415, p-val = 0.7067
## 
## Model Results:
## 
##            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt      0.0751  0.1677   0.4479  0.6542  -0.2536  0.4039    
## men_ratio   -0.0983  0.2614  -0.3762  0.7067  -0.6106  0.4139    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
13.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

13.2.6 Type of Exercise

13.2.6.1 Forest plot

13.2.6.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) type_exercise
## Ciolac 2010             0.1742 [-0.6631;  1.0115]       5.2        5.2       Running
## Gillen 2016             0.0000 [-0.9005;  0.9005]       4.5        4.5       Cycling
## Grieco 2013             0.3461 [-0.4993;  1.1915]       5.1        5.1       Cycling
## Honkala 2017 (Healthy)  0.1562 [-0.5857;  0.8981]       6.6        6.6       Cycling
## Honkala 2017 (T2D)     -0.0238 [-1.0115;  0.9640]       3.7        3.7       Cycling
## Keating 2014           -0.0439 [-0.8797;  0.7919]       5.2        5.2       Cycling
## Matsuo 2015             0.0148 [-0.7854;  0.8150]       5.7        5.7       Cycling
## Motiani 2017            0.3542 [-0.4206;  1.1289]       6.0        6.0       Cycling
## Ramos 2016a            -0.3385 [-0.9407;  0.2637]      10.0       10.0       Running
## Robinson 2015           0.2792 [-0.3518;  0.9101]       9.1        9.1       Cycling
## Sandvei 2012           -0.2927 [-1.1152;  0.5298]       5.4        5.4       Running
## Sawyer 2016            -0.0258 [-0.9498;  0.8982]       4.2        4.2       Cycling
## Sjöros 2018             0.0543 [-0.8022;  0.9108]       4.9        4.9       Cycling
## Skleryk 2013           -0.0164 [-0.9964;  0.9636]       3.8        3.8       Cycling
## Tjønna 2008            -0.0090 [-0.9197;  0.9018]       4.4        4.4       Running
## Trapp 2008              0.6862 [-0.0503;  1.4226]       6.7        6.7       Cycling
## Winn 2018              -1.2046 [-2.2698; -0.1395]       3.2        3.2       Running
## Zapata-Lamana 2018     -0.2451 [-0.9887;  0.4985]       6.5        6.5       Cycling
## 
## Number of studies combined: k = 18
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0237 [-0.1666; 0.2139] 0.24  0.8074
## Random effects model 0.0237 [-0.1666; 0.2139] 0.24  0.8074
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1336]; tau = 0 [0.0000; 0.3655];
##  I^2 = 0.0% [0.0%; 33.9%]; H = 1.00 [1.00; 1.23]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 4.7%]; H = 1.00 [1.00; 1.02]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  12.88   17  0.7445
## 
## Results for subgroups (fixed effect model):
##           k     SMD            95%-CI    Q  I^2
## Running   5 -0.2694 [-0.6293; 0.0906] 3.89 0.0%
## Cycling  13  0.1389 [-0.0856; 0.3633] 4.32 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 3.56    1  0.0592
## Within groups  8.21   16  0.9422
## 
## Results for subgroups (random effects model):
##           k     SMD            95%-CI tau^2 tau
## Running   5 -0.2694 [-0.6293; 0.0906]     0   0
## Cycling  13  0.1389 [-0.0856; 0.3633]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   3.56    1  0.0592
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
13.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 18; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0609)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 16) = 8.9933, p-val = 0.9137
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 3.8819, p-val = 0.0488
## 
## Model Results:
## 
##                       estimate      se     zval    pval    ci.lb    ci.ub 
## intrcpt                 0.1431  0.1145   1.2504  0.2112  -0.0812   0.3674    
## type_exerciseRunning   -0.4257  0.2161  -1.9703  0.0488  -0.8493  -0.0022  * 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
13.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

13.2.7 Baseline Values

13.2.7.1 Forest plot

13.2.7.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_bsln
## Ciolac 2010             0.1742 [-0.6631;  1.0115]       5.2        5.2   > 40 pmol/L
## Gillen 2016             0.0000 [-0.9005;  0.9005]       4.5        4.5   > 40 pmol/L
## Grieco 2013             0.3461 [-0.4993;  1.1915]       5.1        5.1   < 40 pmol/L
## Honkala 2017 (Healthy)  0.1562 [-0.5857;  0.8981]       6.6        6.6   < 40 pmol/L
## Honkala 2017 (T2D)     -0.0238 [-1.0115;  0.9640]       3.7        3.7   > 40 pmol/L
## Keating 2014           -0.0439 [-0.8797;  0.7919]       5.2        5.2   > 40 pmol/L
## Matsuo 2015             0.0148 [-0.7854;  0.8150]       5.7        5.7   > 40 pmol/L
## Motiani 2017            0.3542 [-0.4206;  1.1289]       6.0        6.0   < 40 pmol/L
## Ramos 2016a            -0.3385 [-0.9407;  0.2637]      10.0       10.0   > 40 pmol/L
## Robinson 2015           0.2792 [-0.3518;  0.9101]       9.1        9.1   > 40 pmol/L
## Sandvei 2012           -0.2927 [-1.1152;  0.5298]       5.4        5.4   > 40 pmol/L
## Sawyer 2016            -0.0258 [-0.9498;  0.8982]       4.2        4.2   < 40 pmol/L
## Sjöros 2018             0.0543 [-0.8022;  0.9108]       4.9        4.9   < 40 pmol/L
## Skleryk 2013           -0.0164 [-0.9964;  0.9636]       3.8        3.8   < 40 pmol/L
## Tjønna 2008            -0.0090 [-0.9197;  0.9018]       4.4        4.4   > 40 pmol/L
## Trapp 2008              0.6862 [-0.0503;  1.4226]       6.7        6.7   < 40 pmol/L
## Winn 2018              -1.2046 [-2.2698; -0.1395]       3.2        3.2   < 40 pmol/L
## Zapata-Lamana 2018     -0.2451 [-0.9887;  0.4985]       6.5        6.5   > 40 pmol/L
## 
## Number of studies combined: k = 18
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0237 [-0.1666; 0.2139] 0.24  0.8074
## Random effects model 0.0237 [-0.1666; 0.2139] 0.24  0.8074
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1336]; tau = 0 [0.0000; 0.3655];
##  I^2 = 0.0% [0.0%; 33.9%]; H = 1.00 [1.00; 1.23]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 27.4%]; H = 1.00 [1.00; 1.17]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  12.88   17  0.7445
## 
## Results for subgroups (fixed effect model):
##               k     SMD            95%-CI    Q   I^2
## < 40 pmol/L   8  0.1423 [-0.1572; 0.4419] 8.07 13.2%
## > 40 pmol/L  10 -0.0553 [-0.3020; 0.1914] 2.71  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  1.00    1  0.3183
## Within groups  10.78   16  0.8230
## 
## Results for subgroups (random effects model):
##               k     SMD            95%-CI  tau^2    tau
## < 40 pmol/L   8  0.1317 [-0.1912; 0.4546] 0.0288 0.1696
## > 40 pmol/L  10 -0.0553 [-0.3020; 0.1914]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.81    1  0.3673
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
13.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 18; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0608)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 16) = 12.0213, p-val = 0.7425
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.8539, p-val = 0.3555
## 
## Model Results:
## 
##                estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt          0.1647  0.1809   0.9105  0.3625  -0.1898  0.5192    
## bsln_adjusted   -0.0022  0.0024  -0.9240  0.3555  -0.0069  0.0025    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
13.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

13.2.8 Type of HIIE

13.2.8.1 Forest plot

13.2.8.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) HIIE
## Ciolac 2010             0.1742 [-0.6631;  1.0115]       5.2        5.2 HIIT
## Gillen 2016             0.0000 [-0.9005;  0.9005]       4.5        4.5  SIT
## Grieco 2013             0.3461 [-0.4993;  1.1915]       5.1        5.1 HIIT
## Honkala 2017 (Healthy)  0.1562 [-0.5857;  0.8981]       6.6        6.6  SIT
## Honkala 2017 (T2D)     -0.0238 [-1.0115;  0.9640]       3.7        3.7  SIT
## Keating 2014           -0.0439 [-0.8797;  0.7919]       5.2        5.2 HIIT
## Matsuo 2015             0.0148 [-0.7854;  0.8150]       5.7        5.7 HIIT
## Motiani 2017            0.3542 [-0.4206;  1.1289]       6.0        6.0  SIT
## Ramos 2016a            -0.3385 [-0.9407;  0.2637]      10.0       10.0 HIIT
## Robinson 2015           0.2792 [-0.3518;  0.9101]       9.1        9.1 HIIT
## Sandvei 2012           -0.2927 [-1.1152;  0.5298]       5.4        5.4  SIT
## Sawyer 2016            -0.0258 [-0.9498;  0.8982]       4.2        4.2 HIIT
## Sjöros 2018             0.0543 [-0.8022;  0.9108]       4.9        4.9  SIT
## Skleryk 2013           -0.0164 [-0.9964;  0.9636]       3.8        3.8  SIT
## Tjønna 2008            -0.0090 [-0.9197;  0.9018]       4.4        4.4 HIIT
## Trapp 2008              0.6862 [-0.0503;  1.4226]       6.7        6.7  SIT
## Winn 2018              -1.2046 [-2.2698; -0.1395]       3.2        3.2 HIIT
## Zapata-Lamana 2018     -0.2451 [-0.9887;  0.4985]       6.5        6.5 HIIT
## 
## Number of studies combined: k = 18
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0237 [-0.1666; 0.2139] 0.24  0.8074
## Random effects model 0.0237 [-0.1666; 0.2139] 0.24  0.8074
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.1336]; tau = 0 [0.0000; 0.3655];
##  I^2 = 0.0% [0.0%; 33.9%]; H = 1.00 [1.00; 1.23]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 26.5%]; H = 1.00 [1.00; 1.17]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  12.88   17  0.7445
## 
## Results for subgroups (fixed effect model):
##        k     SMD            95%-CI    Q  I^2
## HIIT  10 -0.0625 [-0.3116; 0.1865] 7.07 0.0%
## SIT    8  0.1472 [-0.1483; 0.4427] 3.57 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  1.13    1  0.2874
## Within groups  10.64   16  0.8311
## 
## Results for subgroups (random effects model):
##        k     SMD            95%-CI tau^2 tau
## HIIT  10 -0.0625 [-0.3116; 0.1865]     0   0
## SIT    8  0.1472 [-0.1483; 0.4427]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.13    1  0.2874
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
13.2.8.3 Meta-regression
## 
## Mixed-Effects Model (k = 18; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0606)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 16) = 11.6417, p-val = 0.7683
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.2335, p-val = 0.2667
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.0672  0.1269  -0.5291  0.5967  -0.3160  0.1816    
## HIIESIT    0.2188  0.1970   1.1106  0.2667  -0.1673  0.6050    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
13.2.8.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

14. Fasting Glucose

14.1 Overall

14.1.1 Forest plot

14.1.2 R output

##                            SMD             95%-CI %W(fixed) %W(random)
## Ciolac 2010             0.1301 [-0.7065;  0.9667]       2.8        3.5
## Conraads 2015          -0.0131 [-0.3103;  0.2842]      22.6        9.4
## Eguchi 2012             0.2888 [-0.5923;  1.1699]       2.6        3.2
## Gillen 2016             0.0000 [-0.9005;  0.9005]       2.5        3.1
## Grieco 2013             0.7150 [-0.1504;  1.5804]       2.7        3.3
## Honkala 2017 (Healthy) -0.5370 [-1.2910;  0.2171]       3.5        4.0
## Honkala 2017 (T2D)      1.4863 [ 0.3724;  2.6002]       1.6        2.2
## Jo 2020                -0.0169 [-0.6892;  0.6554]       4.4        4.7
## Keating 2014            0.6187 [-0.2368;  1.4742]       2.7        3.4
## Lira 2017              -0.5603 [-1.4539;  0.3332]       2.5        3.2
## Madssen 2014            0.3177 [-0.3489;  0.9844]       4.5        4.7
## Maillard 2016           0.0663 [-0.9140;  1.0465]       2.1        2.7
## Matsuo 2015            -0.2904 [-1.0947;  0.5140]       3.1        3.7
## Mitranun 2014           0.0012 [-0.7396;  0.7420]       3.6        4.1
## Motiani 2017           -0.0000 [-0.7688;  0.7688]       3.4        3.9
## Ramos 2016a            -0.4043 [-1.0083;  0.1998]       5.5        5.3
## Ramos 2016b            -0.0631 [-0.7575;  0.6314]       4.1        4.5
## Robinson 2015          -0.6719 [-1.3172; -0.0265]       4.8        4.9
## Sandvei 2012            0.0000 [-0.8181;  0.8181]       3.0        3.6
## Sawyer 2016             0.8925 [-0.0763;  1.8613]       2.1        2.8
## Sjöros 2018             0.6632 [-0.2163;  1.5427]       2.6        3.2
## Skleryk 2013           -0.6900 [-1.6987;  0.3187]       2.0        2.6
## Tjønna 2008             0.6903 [-0.2465;  1.6271]       2.3        2.9
## Trapp 2008              0.2981 [-0.4215;  1.0178]       3.8        4.3
## Winn 2018              -1.1390 [-2.1954; -0.0825]       1.8        2.4
## Zapata-Lamana 2018     -0.2544 [-0.9982;  0.4894]       3.6        4.1
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0044 [-0.1368; 0.1456] 0.06  0.9512
## Random effects model 0.0224 [-0.1617; 0.2066] 0.24  0.8114
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0711 [0.0025; 0.3551]; tau = 0.2667 [0.0495; 0.5959];
##  I^2 = 33.6% [0.0%; 58.8%]; H = 1.23 [1.00; 1.56]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  37.65   25  0.0500
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

14.1.3 Sensitivity analysis

14.1.3.1 Forest plot

14.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                                       SMD            95%-CI p-value   tau^2     tau    I^2
## Omitting Ciolac 2010               0.0136 [-0.1681; 0.1954]  0.8833  0.0586  0.2420  29.5%
## Omitting Conraads 2015             0.0229 [-0.1708; 0.2167]  0.8166  0.0712  0.2668  29.7%
## Omitting Eguchi 2012               0.0086 [-0.1720; 0.1892]  0.9255  0.0567  0.2380  28.9%
## Omitting Gillen 2016               0.0181 [-0.1635; 0.1996]  0.8454  0.0588  0.2425  29.7%
## Omitting Grieco 2013              -0.0073 [-0.1818; 0.1672]  0.9345  0.0446  0.2112  24.2%
## Omitting Honkala 2017 (Healthy)    0.0378 [-0.1393; 0.2149]  0.6756  0.0483  0.2197  25.5%
## Omitting Honkala 2017 (T2D)       -0.0163 [-0.1779; 0.1452]  0.8428  0.0240  0.1549  14.8%
## Omitting Jo 2020                   0.0194 [-0.1643; 0.2031]  0.8358  0.0601  0.2452  29.7%
## Omitting Keating 2014             -0.0041 [-0.1804; 0.1722]  0.9637  0.0480  0.2191  25.6%
## Omitting Lira 2017                 0.0334 [-0.1440; 0.2108]  0.7119  0.0505  0.2248  26.6%
## Omitting Madssen 2014              0.0021 [-0.1792; 0.1834]  0.9818  0.0551  0.2347  27.9%
## Omitting Maillard 2016             0.0162 [-0.1648; 0.1972]  0.8608  0.0585  0.2418  29.6%
## Omitting Matsuo 2015               0.0283 [-0.1525; 0.2092]  0.7589  0.0564  0.2374  28.7%
## Omitting Mitranun 2014             0.0184 [-0.1646; 0.2013]  0.8441  0.0596  0.2441  29.7%
## Omitting Motiani 2017              0.0183 [-0.1643; 0.2009]  0.8441  0.0594  0.2438  29.7%
## Omitting Ramos 2016a               0.0397 [-0.1397; 0.2192]  0.6643  0.0502  0.2240  25.8%
## Omitting Ramos 2016b               0.0214 [-0.1619; 0.2047]  0.8190  0.0597  0.2444  29.6%
## Omitting Robinson 2015             0.0482 [-0.1228; 0.2192]  0.5804  0.0353  0.1878  19.8%
## Omitting Sandvei 2012              0.0182 [-0.1639; 0.2004]  0.8446  0.0592  0.2432  29.7%
## Omitting Sawyer 2016              -0.0082 [-0.1804; 0.1641]  0.9259  0.0414  0.2036  23.0%
## Omitting Sjöros 2018              -0.0047 [-0.1804; 0.1710]  0.9583  0.0470  0.2169  25.2%
## Omitting Skleryk 2013              0.0327 [-0.1434; 0.2088]  0.7158  0.0490  0.2214  26.1%
## Omitting Tjønna 2008              -0.0032 [-0.1789; 0.1724]  0.9713  0.0476  0.2181  25.5%
## Omitting Trapp 2008                0.0048 [-0.1765; 0.1860]  0.9590  0.0560  0.2366  28.3%
## Omitting Winn 2018                 0.0370 [-0.1316; 0.2056]  0.6672  0.0354  0.1882  20.4%
## Omitting Zapata-Lamana 2018        0.0284 [-0.1531; 0.2100]  0.7588  0.0569  0.2386  28.7%
##                                                                                           
## Pooled estimate                    0.0224 [-0.1617; 0.2066]  0.8114  0.0711  0.2667  33.6%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

14.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

14.2 Subgroups

14.2.1 Overall

14.2.1.1 Forest plot

14.2.1.2 R output
##                            SMD            95%-CI     meta-analysis
##                         0.0224 [-0.1617; 0.2066]           Overall
## Healthy                 0.0316 [-0.2417; 0.3049]        Population
## Overweight/obese       -0.0805 [-0.7515; 0.5904]        Population
## Cardiac Rehabilitation  0.0406 [-0.2309; 0.3121]        Population
## Metabolic Syndrome     -0.1970 [-0.5145; 0.1205]        Population
## T2D                     0.4465 [-0.1372; 1.0302]        Population
## < 30 y                  0.0512 [-0.2580; 0.3603]               Age
## 30 - 50 y               0.0884 [-0.3905; 0.5673]               Age
## > 50 y                 -0.0391 [-0.2272; 0.1490]               Age
## < 5 weeks              -0.1055 [-0.5849; 0.3739] Training Duration
## 5 - 10 weeks            0.0523 [-0.2985; 0.4032] Training Duration
## > 10 weeks              0.0542 [-0.1296; 0.2381] Training Duration
## < 0.5                   0.0902 [-0.2156; 0.3961]         Men Ratio
## > 0.5                  -0.0327 [-0.2420; 0.1766]         Men Ratio
## Running                -0.0716 [-0.3185; 0.1753]  Type of Exercise
## Cycling                 0.0837 [-0.1607; 0.3282]  Type of Exercise
## < 5.6 mmol/L            0.0276 [-0.1913; 0.2465]   Baseline Values
## > 5.6 mmol/L            0.0139 [-0.2785; 0.3062]   Baseline Values
## HIIT                   -0.0131 [-0.2091; 0.1830]      Type of HIIE
## SIT                     0.1009 [-0.2898; 0.4916]      Type of HIIE
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Random effects model 0.0224 [-0.1617; 0.2066] 0.24  0.8114
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0711; tau = 0.2667; I^2 = 33.6% [0.0%; 58.8%]; H = 1.23 [1.00; 1.56]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  37.65   25  0.0500
## 
## Results for meta-analyses (random effects model):
##                     k    SMD            95%-CI  tau^2    tau     Q   I^2
## Overall            26 0.0224 [-0.1617; 0.2066] 0.0711 0.2667 37.65 33.6%
## Population         26 0.0224 [-0.1617; 0.2066] 0.0711 0.2667 37.65 33.6%
## Age                26 0.0224 [-0.1617; 0.2066] 0.0711 0.2667 37.65 33.6%
## Training Duration  26 0.0224 [-0.1617; 0.2066] 0.0711 0.2667 37.65 33.6%
## Men Ratio          26 0.0224 [-0.1617; 0.2066] 0.0711 0.2667 37.65 33.6%
## Type of Exercise   26 0.0224 [-0.1617; 0.2066] 0.0711 0.2667 37.65 33.6%
## Baseline Values    26 0.0224 [-0.1617; 0.2066] 0.0711 0.2667 37.65 33.6%
## Type of HIIE       26 0.0224 [-0.1617; 0.2066] 0.0711 0.2667 37.65 33.6%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

14.2.2 Population

14.2.2.1 Forest plot

14.2.2.2 R output
##                            SMD             95%-CI %W(fixed) %W(random)             population
## Ciolac 2010             0.1301 [-0.7065;  0.9667]       2.8        3.5                Healthy
## Conraads 2015          -0.0131 [-0.3103;  0.2842]      22.6        9.4 Cardiac Rehabilitation
## Eguchi 2012             0.2888 [-0.5923;  1.1699]       2.6        3.2                Healthy
## Gillen 2016             0.0000 [-0.9005;  0.9005]       2.5        3.1                Healthy
## Grieco 2013             0.7150 [-0.1504;  1.5804]       2.7        3.3                Healthy
## Honkala 2017 (Healthy) -0.5370 [-1.2910;  0.2171]       3.5        4.0                Healthy
## Honkala 2017 (T2D)      1.4863 [ 0.3724;  2.6002]       1.6        2.2                    T2D
## Jo 2020                -0.0169 [-0.6892;  0.6554]       4.4        4.7     Metabolic Syndrome
## Keating 2014            0.6187 [-0.2368;  1.4742]       2.7        3.4       Overweight/obese
## Lira 2017              -0.5603 [-1.4539;  0.3332]       2.5        3.2                Healthy
## Madssen 2014            0.3177 [-0.3489;  0.9844]       4.5        4.7 Cardiac Rehabilitation
## Maillard 2016           0.0663 [-0.9140;  1.0465]       2.1        2.7                    T2D
## Matsuo 2015            -0.2904 [-1.0947;  0.5140]       3.1        3.7     Metabolic Syndrome
## Mitranun 2014           0.0012 [-0.7396;  0.7420]       3.6        4.1                    T2D
## Motiani 2017           -0.0000 [-0.7688;  0.7688]       3.4        3.9                Healthy
## Ramos 2016a            -0.4043 [-1.0083;  0.1998]       5.5        5.3     Metabolic Syndrome
## Ramos 2016b            -0.0631 [-0.7575;  0.6314]       4.1        4.5     Metabolic Syndrome
## Robinson 2015          -0.6719 [-1.3172; -0.0265]       4.8        4.9     Metabolic Syndrome
## Sandvei 2012            0.0000 [-0.8181;  0.8181]       3.0        3.6                Healthy
## Sawyer 2016             0.8925 [-0.0763;  1.8613]       2.1        2.8       Overweight/obese
## Sjöros 2018             0.6632 [-0.2163;  1.5427]       2.6        3.2                    T2D
## Skleryk 2013           -0.6900 [-1.6987;  0.3187]       2.0        2.6       Overweight/obese
## Tjønna 2008             0.6903 [-0.2465;  1.6271]       2.3        2.9     Metabolic Syndrome
## Trapp 2008              0.2981 [-0.4215;  1.0178]       3.8        4.3                Healthy
## Winn 2018              -1.1390 [-2.1954; -0.0825]       1.8        2.4       Overweight/obese
## Zapata-Lamana 2018     -0.2544 [-0.9982;  0.4894]       3.6        4.1       Overweight/obese
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0044 [-0.1368; 0.1456] 0.06  0.9512
## Random effects model 0.0224 [-0.1617; 0.2066] 0.24  0.8114
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0711 [0.0025; 0.3551]; tau = 0.2667 [0.0495; 0.5959];
##  I^2 = 33.6% [0.0%; 58.8%]; H = 1.23 [1.00; 1.56]
## 
## Quantifying residual heterogeneity:
##  I^2 = 27.3% [0.0%; 56.8%]; H = 1.17 [1.00; 1.52]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  37.65   25  0.0500
## 
## Results for subgroups (fixed effect model):
##                          k     SMD            95%-CI     Q   I^2
## Healthy                  9  0.0316 [-0.2417; 0.3049]  6.65  0.0%
## Overweight/obese         5 -0.0528 [-0.4590; 0.3534] 10.60 62.3%
## Cardiac Rehabilitation   2  0.0406 [-0.2309; 0.3121]  0.75  0.0%
## Metabolic Syndrome       6 -0.2078 [-0.4954; 0.0798]  6.02 16.9%
## T2D                      4  0.4010 [-0.0497; 0.8518]  4.85 38.2%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  5.25    4  0.2628
## Within groups  28.88   21  0.1169
## 
## Results for subgroups (random effects model):
##                          k     SMD            95%-CI  tau^2    tau
## Healthy                  9  0.0316 [-0.2417; 0.3049]      0      0
## Overweight/obese         5 -0.0805 [-0.7515; 0.5904] 0.3615 0.6012
## Cardiac Rehabilitation   2  0.0406 [-0.2309; 0.3121]      0      0
## Metabolic Syndrome       6 -0.1970 [-0.5145; 0.1205] 0.0267 0.1633
## T2D                      4  0.4465 [-0.1372; 1.0302] 0.1352 0.3676
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   3.92    4  0.4174
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
14.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 26; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0831 (SE = 0.0760)
## tau (square root of estimated tau^2 value):             0.2883
## I^2 (residual heterogeneity / unaccounted variability): 34.08%
## H^2 (unaccounted variability / sampling variability):   1.52
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 21) = 31.8581, p-val = 0.0605
## 
## Test of Moderators (coefficients 2:5):
## QM(df = 4) = 4.0100, p-val = 0.4046
## 
## Model Results:
## 
##                               estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                         0.0344  0.1697   0.2025  0.8395  -0.2982  0.3670    
## .byvarOverweight/obese         -0.1043  0.2980  -0.3499  0.7264  -0.6884  0.4798    
## .byvarCardiac Rehabilitation    0.0677  0.3130   0.2162  0.8288  -0.5458  0.6812    
## .byvarMetabolic Syndrome       -0.2180  0.2545  -0.8564  0.3918  -0.7168  0.2809    
## .byvarT2D                       0.4278  0.3216   1.3305  0.1834  -0.2024  1.0581    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
14.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

14.2.3 Age

14.2.3.1 Forest plot

14.2.3.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_age
## Ciolac 2010             0.1301 [-0.7065;  0.9667]       2.8        3.5       < 30 y
## Conraads 2015          -0.0131 [-0.3103;  0.2842]      22.6        9.4       > 50 y
## Eguchi 2012             0.2888 [-0.5923;  1.1699]       2.6        3.2       > 50 y
## Gillen 2016             0.0000 [-0.9005;  0.9005]       2.5        3.1       < 30 y
## Grieco 2013             0.7150 [-0.1504;  1.5804]       2.7        3.3       < 30 y
## Honkala 2017 (Healthy) -0.5370 [-1.2910;  0.2171]       3.5        4.0    30 - 50 y
## Honkala 2017 (T2D)      1.4863 [ 0.3724;  2.6002]       1.6        2.2    30 - 50 y
## Jo 2020                -0.0169 [-0.6892;  0.6554]       4.4        4.7       > 50 y
## Keating 2014            0.6187 [-0.2368;  1.4742]       2.7        3.4    30 - 50 y
## Lira 2017              -0.5603 [-1.4539;  0.3332]       2.5        3.2       < 30 y
## Madssen 2014            0.3177 [-0.3489;  0.9844]       4.5        4.7       > 50 y
## Maillard 2016           0.0663 [-0.9140;  1.0465]       2.1        2.7       > 50 y
## Matsuo 2015            -0.2904 [-1.0947;  0.5140]       3.1        3.7    30 - 50 y
## Mitranun 2014           0.0012 [-0.7396;  0.7420]       3.6        4.1       > 50 y
## Motiani 2017           -0.0000 [-0.7688;  0.7688]       3.4        3.9    30 - 50 y
## Ramos 2016a            -0.4043 [-1.0083;  0.1998]       5.5        5.3       > 50 y
## Ramos 2016b            -0.0631 [-0.7575;  0.6314]       4.1        4.5       > 50 y
## Robinson 2015          -0.6719 [-1.3172; -0.0265]       4.8        4.9       > 50 y
## Sandvei 2012            0.0000 [-0.8181;  0.8181]       3.0        3.6       < 30 y
## Sawyer 2016             0.8925 [-0.0763;  1.8613]       2.1        2.8    30 - 50 y
## Sjöros 2018             0.6632 [-0.2163;  1.5427]       2.6        3.2    30 - 50 y
## Skleryk 2013           -0.6900 [-1.6987;  0.3187]       2.0        2.6    30 - 50 y
## Tjønna 2008             0.6903 [-0.2465;  1.6271]       2.3        2.9       > 50 y
## Trapp 2008              0.2981 [-0.4215;  1.0178]       3.8        4.3       < 30 y
## Winn 2018              -1.1390 [-2.1954; -0.0825]       1.8        2.4    30 - 50 y
## Zapata-Lamana 2018     -0.2544 [-0.9982;  0.4894]       3.6        4.1       < 30 y
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0044 [-0.1368; 0.1456] 0.06  0.9512
## Random effects model 0.0224 [-0.1617; 0.2066] 0.24  0.8114
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0711 [0.0025; 0.3551]; tau = 0.2667 [0.0495; 0.5959];
##  I^2 = 33.6% [0.0%; 58.8%]; H = 1.23 [1.00; 1.56]
## 
## Quantifying residual heterogeneity:
##  I^2 = 31.7% [0.0%; 58.5%]; H = 1.21 [1.00; 1.55]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  37.65   25  0.0500
## 
## Results for subgroups (fixed effect model):
##             k     SMD            95%-CI     Q   I^2
## < 30 y      7  0.0512 [-0.2580; 0.3603]  4.81  0.0%
## 30 - 50 y   9  0.0607 [-0.2368; 0.3582] 20.26 60.5%
## > 50 y     10 -0.0391 [-0.2272; 0.1490]  8.63  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.43    2  0.8064
## Within groups  33.70   23  0.0696
## 
## Results for subgroups (random effects model):
##             k     SMD            95%-CI  tau^2    tau
## < 30 y      7  0.0512 [-0.2580; 0.3603]      0      0
## 30 - 50 y   9  0.0884 [-0.3905; 0.5673] 0.3205 0.5662
## > 50 y     10 -0.0391 [-0.2272; 0.1490]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.40    2  0.8191
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
14.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 26; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0804 (SE = 0.0672)
## tau (square root of estimated tau^2 value):             0.2835
## I^2 (residual heterogeneity / unaccounted variability): 35.72%
## H^2 (unaccounted variability / sampling variability):   1.56
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 24) = 37.3371, p-val = 0.0405
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.1380, p-val = 0.7103
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    0.1418  0.3318   0.4275  0.6690  -0.5084  0.7921    
## age       -0.0026  0.0070  -0.3715  0.7103  -0.0164  0.0112    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
14.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

14.2.4 Training Duration

14.2.4.1 Forest plot

14.2.4.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_duration
## Ciolac 2010             0.1301 [-0.7065;  0.9667]       2.8        3.5        > 10 weeks
## Conraads 2015          -0.0131 [-0.3103;  0.2842]      22.6        9.4        > 10 weeks
## Eguchi 2012             0.2888 [-0.5923;  1.1699]       2.6        3.2        > 10 weeks
## Gillen 2016             0.0000 [-0.9005;  0.9005]       2.5        3.1        > 10 weeks
## Grieco 2013             0.7150 [-0.1504;  1.5804]       2.7        3.3         < 5 weeks
## Honkala 2017 (Healthy) -0.5370 [-1.2910;  0.2171]       3.5        4.0         < 5 weeks
## Honkala 2017 (T2D)      1.4863 [ 0.3724;  2.6002]       1.6        2.2         < 5 weeks
## Jo 2020                -0.0169 [-0.6892;  0.6554]       4.4        4.7      5 - 10 weeks
## Keating 2014            0.6187 [-0.2368;  1.4742]       2.7        3.4        > 10 weeks
## Lira 2017              -0.5603 [-1.4539;  0.3332]       2.5        3.2         < 5 weeks
## Madssen 2014            0.3177 [-0.3489;  0.9844]       4.5        4.7        > 10 weeks
## Maillard 2016           0.0663 [-0.9140;  1.0465]       2.1        2.7        > 10 weeks
## Matsuo 2015            -0.2904 [-1.0947;  0.5140]       3.1        3.7      5 - 10 weeks
## Mitranun 2014           0.0012 [-0.7396;  0.7420]       3.6        4.1      5 - 10 weeks
## Motiani 2017           -0.0000 [-0.7688;  0.7688]       3.4        3.9         < 5 weeks
## Ramos 2016a            -0.4043 [-1.0083;  0.1998]       5.5        5.3        > 10 weeks
## Ramos 2016b            -0.0631 [-0.7575;  0.6314]       4.1        4.5        > 10 weeks
## Robinson 2015          -0.6719 [-1.3172; -0.0265]       4.8        4.9         < 5 weeks
## Sandvei 2012            0.0000 [-0.8181;  0.8181]       3.0        3.6      5 - 10 weeks
## Sawyer 2016             0.8925 [-0.0763;  1.8613]       2.1        2.8      5 - 10 weeks
## Sjöros 2018             0.6632 [-0.2163;  1.5427]       2.6        3.2         < 5 weeks
## Skleryk 2013           -0.6900 [-1.6987;  0.3187]       2.0        2.6         < 5 weeks
## Tjønna 2008             0.6903 [-0.2465;  1.6271]       2.3        2.9        > 10 weeks
## Trapp 2008              0.2981 [-0.4215;  1.0178]       3.8        4.3        > 10 weeks
## Winn 2018              -1.1390 [-2.1954; -0.0825]       1.8        2.4         < 5 weeks
## Zapata-Lamana 2018     -0.2544 [-0.9982;  0.4894]       3.6        4.1        > 10 weeks
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0044 [-0.1368; 0.1456] 0.06  0.9512
## Random effects model 0.0224 [-0.1617; 0.2066] 0.24  0.8114
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0711 [0.0025; 0.3551]; tau = 0.2667 [0.0495; 0.5959];
##  I^2 = 33.6% [0.0%; 58.8%]; H = 1.23 [1.00; 1.56]
## 
## Quantifying residual heterogeneity:
##  I^2 = 29.4% [0.0%; 57.2%]; H = 1.19 [1.00; 1.53]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  37.65   25  0.0500
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI     Q   I^2
## < 5 weeks      9 -0.1554 [-0.4404; 0.1295] 21.81 63.3%
## 5 - 10 weeks   5  0.0523 [-0.2985; 0.4032]  3.30  0.0%
## > 10 weeks    12  0.0542 [-0.1296; 0.2381]  7.46  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  1.56    2  0.4581
## Within groups  32.57   23  0.0889
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI  tau^2    tau
## < 5 weeks      9 -0.1055 [-0.5849; 0.3739] 0.3337 0.5777
## 5 - 10 weeks   5  0.0523 [-0.2985; 0.4032]      0      0
## > 10 weeks    12  0.0542 [-0.1296; 0.2381]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.38    2  0.8268
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
14.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 26; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0748 (SE = 0.0649)
## tau (square root of estimated tau^2 value):             0.2736
## I^2 (residual heterogeneity / unaccounted variability): 34.60%
## H^2 (unaccounted variability / sampling variability):   1.53
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 24) = 36.6967, p-val = 0.0469
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.6565, p-val = 0.4178
## 
## Model Results:
## 
##           estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    -0.1200  0.2004  -0.5990  0.5492  -0.5128  0.2727    
## duration    0.0152  0.0187   0.8102  0.4178  -0.0215  0.0519    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
14.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

14.2.5 Men Ratio

14.2.5.1 Forest plot

14.2.5.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_men_ratio
## Ciolac 2010             0.1301 [-0.7065;  0.9667]       2.8        3.5              < 0.5
## Conraads 2015          -0.0131 [-0.3103;  0.2842]      22.6        9.4              > 0.5
## Eguchi 2012             0.2888 [-0.5923;  1.1699]       2.6        3.2              > 0.5
## Gillen 2016             0.0000 [-0.9005;  0.9005]       2.5        3.1              > 0.5
## Grieco 2013             0.7150 [-0.1504;  1.5804]       2.7        3.3              < 0.5
## Honkala 2017 (Healthy) -0.5370 [-1.2910;  0.2171]       3.5        4.0              > 0.5
## Honkala 2017 (T2D)      1.4863 [ 0.3724;  2.6002]       1.6        2.2              > 0.5
## Jo 2020                -0.0169 [-0.6892;  0.6554]       4.4        4.7              > 0.5
## Keating 2014            0.6187 [-0.2368;  1.4742]       2.7        3.4              < 0.5
## Lira 2017              -0.5603 [-1.4539;  0.3332]       2.5        3.2              > 0.5
## Madssen 2014            0.3177 [-0.3489;  0.9844]       4.5        4.7              > 0.5
## Maillard 2016           0.0663 [-0.9140;  1.0465]       2.1        2.7              < 0.5
## Matsuo 2015            -0.2904 [-1.0947;  0.5140]       3.1        3.7              > 0.5
## Mitranun 2014           0.0012 [-0.7396;  0.7420]       3.6        4.1              < 0.5
## Motiani 2017           -0.0000 [-0.7688;  0.7688]       3.4        3.9              > 0.5
## Ramos 2016a            -0.4043 [-1.0083;  0.1998]       5.5        5.3              > 0.5
## Ramos 2016b            -0.0631 [-0.7575;  0.6314]       4.1        4.5              > 0.5
## Robinson 2015          -0.6719 [-1.3172; -0.0265]       4.8        4.9              < 0.5
## Sandvei 2012            0.0000 [-0.8181;  0.8181]       3.0        3.6              < 0.5
## Sawyer 2016             0.8925 [-0.0763;  1.8613]       2.1        2.8              < 0.5
## Sjöros 2018             0.6632 [-0.2163;  1.5427]       2.6        3.2              > 0.5
## Skleryk 2013           -0.6900 [-1.6987;  0.3187]       2.0        2.6              > 0.5
## Tjønna 2008             0.6903 [-0.2465;  1.6271]       2.3        2.9              < 0.5
## Trapp 2008              0.2981 [-0.4215;  1.0178]       3.8        4.3              < 0.5
## Winn 2018              -1.1390 [-2.1954; -0.0825]       1.8        2.4              < 0.5
## Zapata-Lamana 2018     -0.2544 [-0.9982;  0.4894]       3.6        4.1              < 0.5
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0044 [-0.1368; 0.1456] 0.06  0.9512
## Random effects model 0.0224 [-0.1617; 0.2066] 0.24  0.8114
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0711 [0.0025; 0.3551]; tau = 0.2667 [0.0495; 0.5959];
##  I^2 = 33.6% [0.0%; 58.8%]; H = 1.23 [1.00; 1.56]
## 
## Quantifying residual heterogeneity:
##  I^2 = 28.7% [0.0%; 56.4%]; H = 1.18 [1.00; 1.51]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  37.65   25  0.0500
## 
## Results for subgroups (fixed effect model):
##         k     SMD            95%-CI     Q   I^2
## < 0.5  12  0.0681 [-0.1699; 0.3061] 17.76 38.1%
## > 0.5  14 -0.0336 [-0.2093; 0.1422] 15.91 18.3%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.45    1  0.5006
## Within groups  33.67   24  0.0906
## 
## Results for subgroups (random effects model):
##         k     SMD            95%-CI  tau^2    tau
## < 0.5  12  0.0902 [-0.2156; 0.3961] 0.1097 0.3312
## > 0.5  14 -0.0327 [-0.2420; 0.1766] 0.0278 0.1668
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.42    1  0.5156
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
14.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 26; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0811 (SE = 0.0673)
## tau (square root of estimated tau^2 value):             0.2847
## I^2 (residual heterogeneity / unaccounted variability): 36.00%
## H^2 (unaccounted variability / sampling variability):   1.56
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 24) = 37.4985, p-val = 0.0390
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.1297, p-val = 0.7188
## 
## Model Results:
## 
##            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt      0.0817  0.1869   0.4369  0.6622  -0.2847  0.4480    
## men_ratio   -0.0977  0.2712  -0.3601  0.7188  -0.6293  0.4339    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
14.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

14.2.6 Type of Exercise

14.2.6.1 Forest plot

14.2.6.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) type_exercise
## Ciolac 2010             0.1301 [-0.7065;  0.9667]       2.8        3.5       Running
## Conraads 2015          -0.0131 [-0.3103;  0.2842]      22.6        9.4       Cycling
## Eguchi 2012             0.2888 [-0.5923;  1.1699]       2.6        3.2       Cycling
## Gillen 2016             0.0000 [-0.9005;  0.9005]       2.5        3.1       Cycling
## Grieco 2013             0.7150 [-0.1504;  1.5804]       2.7        3.3       Cycling
## Honkala 2017 (Healthy) -0.5370 [-1.2910;  0.2171]       3.5        4.0       Cycling
## Honkala 2017 (T2D)      1.4863 [ 0.3724;  2.6002]       1.6        2.2       Cycling
## Jo 2020                -0.0169 [-0.6892;  0.6554]       4.4        4.7       Running
## Keating 2014            0.6187 [-0.2368;  1.4742]       2.7        3.4       Cycling
## Lira 2017              -0.5603 [-1.4539;  0.3332]       2.5        3.2       Running
## Madssen 2014            0.3177 [-0.3489;  0.9844]       4.5        4.7       Running
## Maillard 2016           0.0663 [-0.9140;  1.0465]       2.1        2.7       Cycling
## Matsuo 2015            -0.2904 [-1.0947;  0.5140]       3.1        3.7       Cycling
## Mitranun 2014           0.0012 [-0.7396;  0.7420]       3.6        4.1       Running
## Motiani 2017           -0.0000 [-0.7688;  0.7688]       3.4        3.9       Cycling
## Ramos 2016a            -0.4043 [-1.0083;  0.1998]       5.5        5.3       Running
## Ramos 2016b            -0.0631 [-0.7575;  0.6314]       4.1        4.5       Running
## Robinson 2015          -0.6719 [-1.3172; -0.0265]       4.8        4.9       Cycling
## Sandvei 2012            0.0000 [-0.8181;  0.8181]       3.0        3.6       Running
## Sawyer 2016             0.8925 [-0.0763;  1.8613]       2.1        2.8       Cycling
## Sjöros 2018             0.6632 [-0.2163;  1.5427]       2.6        3.2       Cycling
## Skleryk 2013           -0.6900 [-1.6987;  0.3187]       2.0        2.6       Cycling
## Tjønna 2008             0.6903 [-0.2465;  1.6271]       2.3        2.9       Running
## Trapp 2008              0.2981 [-0.4215;  1.0178]       3.8        4.3       Cycling
## Winn 2018              -1.1390 [-2.1954; -0.0825]       1.8        2.4       Running
## Zapata-Lamana 2018     -0.2544 [-0.9982;  0.4894]       3.6        4.1       Cycling
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0044 [-0.1368; 0.1456] 0.06  0.9512
## Random effects model 0.0224 [-0.1617; 0.2066] 0.24  0.8114
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0711 [0.0025; 0.3551]; tau = 0.2667 [0.0495; 0.5959];
##  I^2 = 33.6% [0.0%; 58.8%]; H = 1.23 [1.00; 1.56]
## 
## Quantifying residual heterogeneity:
##  I^2 = 28.5% [0.0%; 56.3%]; H = 1.18 [1.00; 1.51]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  37.65   25  0.0500
## 
## Results for subgroups (fixed effect model):
##           k     SMD            95%-CI     Q   I^2
## Running  10 -0.0715 [-0.3122; 0.1691]  9.42  4.5%
## Cycling  16  0.0412 [-0.1335; 0.2159] 24.15 37.9%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.55    1  0.4575
## Within groups  33.58   24  0.0925
## 
## Results for subgroups (random effects model):
##           k     SMD            95%-CI  tau^2    tau
## Running  10 -0.0716 [-0.3185; 0.1753] 0.0072 0.0849
## Cycling  16  0.0837 [-0.1607; 0.3282] 0.0856 0.2925
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.77    1  0.3808
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
14.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 26; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0781 (SE = 0.0664)
## tau (square root of estimated tau^2 value):             0.2794
## I^2 (residual heterogeneity / unaccounted variability): 35.13%
## H^2 (unaccounted variability / sampling variability):   1.54
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 24) = 36.9987, p-val = 0.0438
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.7391, p-val = 0.3900
## 
## Model Results:
## 
##                       estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                 0.0892  0.1224   0.7292  0.4659  -0.1506  0.3290    
## type_exerciseRunning   -0.1684  0.1958  -0.8597  0.3900  -0.5522  0.2155    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
14.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

14.2.7 Baseline Values

14.2.7.1 Forest plot

14.2.7.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_bsln
## Ciolac 2010             0.1301 [-0.7065;  0.9667]       2.8        3.5  < 5.6 mmol/L
## Conraads 2015          -0.0131 [-0.3103;  0.2842]      22.6        9.4  < 5.6 mmol/L
## Eguchi 2012             0.2888 [-0.5923;  1.1699]       2.6        3.2  < 5.6 mmol/L
## Gillen 2016             0.0000 [-0.9005;  0.9005]       2.5        3.1  < 5.6 mmol/L
## Grieco 2013             0.7150 [-0.1504;  1.5804]       2.7        3.3  < 5.6 mmol/L
## Honkala 2017 (Healthy) -0.5370 [-1.2910;  0.2171]       3.5        4.0  < 5.6 mmol/L
## Honkala 2017 (T2D)      1.4863 [ 0.3724;  2.6002]       1.6        2.2  > 5.6 mmol/L
## Jo 2020                -0.0169 [-0.6892;  0.6554]       4.4        4.7  > 5.6 mmol/L
## Keating 2014            0.6187 [-0.2368;  1.4742]       2.7        3.4  < 5.6 mmol/L
## Lira 2017              -0.5603 [-1.4539;  0.3332]       2.5        3.2  < 5.6 mmol/L
## Madssen 2014            0.3177 [-0.3489;  0.9844]       4.5        4.7  > 5.6 mmol/L
## Maillard 2016           0.0663 [-0.9140;  1.0465]       2.1        2.7  > 5.6 mmol/L
## Matsuo 2015            -0.2904 [-1.0947;  0.5140]       3.1        3.7  > 5.6 mmol/L
## Mitranun 2014           0.0012 [-0.7396;  0.7420]       3.6        4.1  > 5.6 mmol/L
## Motiani 2017           -0.0000 [-0.7688;  0.7688]       3.4        3.9  < 5.6 mmol/L
## Ramos 2016a            -0.4043 [-1.0083;  0.1998]       5.5        5.3  > 5.6 mmol/L
## Ramos 2016b            -0.0631 [-0.7575;  0.6314]       4.1        4.5  > 5.6 mmol/L
## Robinson 2015          -0.6719 [-1.3172; -0.0265]       4.8        4.9  > 5.6 mmol/L
## Sandvei 2012            0.0000 [-0.8181;  0.8181]       3.0        3.6  < 5.6 mmol/L
## Sawyer 2016             0.8925 [-0.0763;  1.8613]       2.1        2.8  < 5.6 mmol/L
## Sjöros 2018             0.6632 [-0.2163;  1.5427]       2.6        3.2  > 5.6 mmol/L
## Skleryk 2013           -0.6900 [-1.6987;  0.3187]       2.0        2.6  > 5.6 mmol/L
## Tjønna 2008             0.6903 [-0.2465;  1.6271]       2.3        2.9  > 5.6 mmol/L
## Trapp 2008              0.2981 [-0.4215;  1.0178]       3.8        4.3  < 5.6 mmol/L
## Winn 2018              -1.1390 [-2.1954; -0.0825]       1.8        2.4  < 5.6 mmol/L
## Zapata-Lamana 2018     -0.2544 [-0.9982;  0.4894]       3.6        4.1  < 5.6 mmol/L
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0044 [-0.1368; 0.1456] 0.06  0.9512
## Random effects model 0.0224 [-0.1617; 0.2066] 0.24  0.8114
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0711 [0.0025; 0.3551]; tau = 0.2667 [0.0495; 0.5959];
##  I^2 = 33.6% [0.0%; 58.8%]; H = 1.23 [1.00; 1.56]
## 
## Quantifying residual heterogeneity:
##  I^2 = 29.5% [0.0%; 56.8%]; H = 1.19 [1.00; 1.52]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  37.65   25  0.0500
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI     Q   I^2
## < 5.6 mmol/L  14  0.0198 [-0.1634; 0.2030] 15.74 17.4%
## > 5.6 mmol/L  12 -0.0235 [-0.2458; 0.1989] 18.30 39.9%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.09    1  0.7684
## Within groups  34.04   24  0.0839
## 
## Results for subgroups (random effects model):
##                k    SMD            95%-CI  tau^2    tau
## < 5.6 mmol/L  14 0.0276 [-0.1913; 0.2465] 0.0291 0.1705
## > 5.6 mmol/L  12 0.0139 [-0.2785; 0.3062] 0.1037 0.3221
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.01    1  0.9412
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
14.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 26; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0798 (SE = 0.0663)
## tau (square root of estimated tau^2 value):             0.2825
## I^2 (residual heterogeneity / unaccounted variability): 36.25%
## H^2 (unaccounted variability / sampling variability):   1.57
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 24) = 37.6474, p-val = 0.0377
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0224, p-val = 0.8809
## 
## Model Results:
## 
##                estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt         -0.0568  0.5465  -0.1040  0.9172  -1.1280  1.0144    
## bsln_adjusted    0.0140  0.0936   0.1498  0.8809  -0.1695  0.1975    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
14.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

14.2.8 Type of HIIE

14.2.8.1 Forest plot

14.2.8.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) HIIE
## Ciolac 2010             0.1301 [-0.7065;  0.9667]       2.8        3.5 HIIT
## Conraads 2015          -0.0131 [-0.3103;  0.2842]      22.6        9.4 HIIT
## Eguchi 2012             0.2888 [-0.5923;  1.1699]       2.6        3.2 HIIT
## Gillen 2016             0.0000 [-0.9005;  0.9005]       2.5        3.1  SIT
## Grieco 2013             0.7150 [-0.1504;  1.5804]       2.7        3.3 HIIT
## Honkala 2017 (Healthy) -0.5370 [-1.2910;  0.2171]       3.5        4.0  SIT
## Honkala 2017 (T2D)      1.4863 [ 0.3724;  2.6002]       1.6        2.2  SIT
## Jo 2020                -0.0169 [-0.6892;  0.6554]       4.4        4.7 HIIT
## Keating 2014            0.6187 [-0.2368;  1.4742]       2.7        3.4 HIIT
## Lira 2017              -0.5603 [-1.4539;  0.3332]       2.5        3.2 HIIT
## Madssen 2014            0.3177 [-0.3489;  0.9844]       4.5        4.7 HIIT
## Maillard 2016           0.0663 [-0.9140;  1.0465]       2.1        2.7 HIIT
## Matsuo 2015            -0.2904 [-1.0947;  0.5140]       3.1        3.7 HIIT
## Mitranun 2014           0.0012 [-0.7396;  0.7420]       3.6        4.1 HIIT
## Motiani 2017           -0.0000 [-0.7688;  0.7688]       3.4        3.9  SIT
## Ramos 2016a            -0.4043 [-1.0083;  0.1998]       5.5        5.3 HIIT
## Ramos 2016b            -0.0631 [-0.7575;  0.6314]       4.1        4.5 HIIT
## Robinson 2015          -0.6719 [-1.3172; -0.0265]       4.8        4.9 HIIT
## Sandvei 2012            0.0000 [-0.8181;  0.8181]       3.0        3.6  SIT
## Sawyer 2016             0.8925 [-0.0763;  1.8613]       2.1        2.8 HIIT
## Sjöros 2018             0.6632 [-0.2163;  1.5427]       2.6        3.2  SIT
## Skleryk 2013           -0.6900 [-1.6987;  0.3187]       2.0        2.6  SIT
## Tjønna 2008             0.6903 [-0.2465;  1.6271]       2.3        2.9 HIIT
## Trapp 2008              0.2981 [-0.4215;  1.0178]       3.8        4.3  SIT
## Winn 2018              -1.1390 [-2.1954; -0.0825]       1.8        2.4 HIIT
## Zapata-Lamana 2018     -0.2544 [-0.9982;  0.4894]       3.6        4.1 HIIT
## 
## Number of studies combined: k = 26
## 
##                         SMD            95%-CI    z p-value
## Fixed effect model   0.0044 [-0.1368; 0.1456] 0.06  0.9512
## Random effects model 0.0224 [-0.1617; 0.2066] 0.24  0.8114
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0711 [0.0025; 0.3551]; tau = 0.2667 [0.0495; 0.5959];
##  I^2 = 33.6% [0.0%; 58.8%]; H = 1.23 [1.00; 1.56]
## 
## Quantifying residual heterogeneity:
##  I^2 = 28.9% [0.0%; 56.5%]; H = 1.19 [1.00; 1.52]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  37.65   25  0.0500
## 
## Results for subgroups (fixed effect model):
##        k     SMD            95%-CI     Q   I^2
## HIIT  18 -0.0206 [-0.1810; 0.1397] 22.14 23.2%
## SIT    8  0.0824 [-0.2173; 0.3822] 11.63 39.8%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.35    1  0.5523
## Within groups  33.78   24  0.0887
## 
## Results for subgroups (random effects model):
##        k     SMD            95%-CI  tau^2    tau
## HIIT  18 -0.0131 [-0.2091; 0.1830] 0.0390 0.1975
## SIT    8  0.1009 [-0.2898; 0.4916] 0.1251 0.3536
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.26    1  0.6094
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
14.2.8.3 Meta-regression
## 
## Mixed-Effects Model (k = 26; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0781 (SE = 0.0660)
## tau (square root of estimated tau^2 value):             0.2795
## I^2 (residual heterogeneity / unaccounted variability): 35.56%
## H^2 (unaccounted variability / sampling variability):   1.55
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 24) = 37.2425, p-val = 0.0414
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.2632, p-val = 0.6079
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.0065  0.1120  -0.0579  0.9538  -0.2260  0.2131    
## HIIESIT    0.1101  0.2146   0.5131  0.6079  -0.3105  0.5308    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
14.2.8.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

15. HbA1c

15.1 Overall

15.1.1 Forest plot

15.1.2 R output

##                            SMD             95%-CI %W(fixed) %W(random)
## Abdelbasset 2020       -0.0000 [-0.7044;  0.7044]      12.9       12.9
## Eguchi 2012            -0.0554 [-0.9321;  0.8213]       8.4        8.4
## Honkala 2017 (Healthy) -0.7456 [-1.5117;  0.0205]      10.9       10.9
## Madssen 2014           -0.1727 [-0.8365;  0.4911]      14.6       14.6
## Maillard 2016          -0.2152 [-1.1980;  0.7676]       6.7        6.7
## Matsuo 2015            -0.1827 [-0.9845;  0.6192]      10.0       10.0
## Motiani 2017           -0.9164 [-1.7245; -0.1083]       9.8        9.8
## Ramos 2016a            -0.1349 [-0.7335;  0.4638]      17.9       17.9
## Sjöros 2018            -0.0817 [-0.9384;  0.7750]       8.8        8.8
## 
## Number of studies combined: k = 9
## 
##                          SMD             95%-CI     z p-value
## Fixed effect model   -0.2655 [-0.5190; -0.0120] -2.05  0.0401
## Random effects model -0.2655 [-0.5190; -0.0120] -2.05  0.0401
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.2195]; tau = 0 [0.0000; 0.4685];
##  I^2 = 0.0% [0.0%; 46.4%]; H = 1.00 [1.00; 1.37]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  5.25    8  0.7302
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

15.1.3 Sensitivity analysis

15.1.3.1 Forest plot

15.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                                       SMD             95%-CI p-value   tau^2     tau   I^2
## Omitting Abdelbasset 2020         -0.2589 [-0.5212;  0.0033]  0.0530  0.0000  0.0000  0.0%
## Omitting Eguchi 2012              -0.2423 [-0.4983;  0.0138]  0.0637  0.0000  0.0000  0.0%
## Omitting Honkala 2017 (Healthy)   -0.1708 [-0.4303;  0.0886]  0.1968  0.0000  0.0000  0.0%
## Omitting Honkala 2017 (T2D)       -0.2565 [-0.5102; -0.0027]  0.0476  0.0000  0.0000  0.0%
## Omitting Madssen 2014             -0.2367 [-0.5013;  0.0279]  0.0795  0.0000  0.0000  0.0%
## Omitting Maillard 2016            -0.2290 [-0.4829;  0.0248]  0.0770  0.0000  0.0000  0.0%
## Omitting Matsuo 2015              -0.2327 [-0.4909;  0.0255]  0.0773  0.0000  0.0000  0.0%
## Omitting Motiani 2017             -0.1609 [-0.4187;  0.0970]  0.2215  0.0000  0.0000  0.0%
## Omitting Ramos 2016a              -0.2467 [-0.5162;  0.0229]  0.0729  0.0000  0.0000  0.0%
## Omitting Sjöros 2018              -0.2408 [-0.4973;  0.0158]  0.0659  0.0000  0.0000  0.0%
##                                                                                           
## Pooled estimate                   -0.2352 [-0.4807;  0.0104]  0.0605  0.0000  0.0000  0.0%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

15.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

15.2 Subgroups

15.2.1 Overall

15.2.1.1 Forest plot

15.2.1.2 R output
##                            SMD             95%-CI     meta-analysis
##                        -0.2655 [-0.5190; -0.0120]           Overall
## Healthy                -0.5844 [-1.0638; -0.1049]        Population
## Cardiac Rehabilitation -0.1689 [-0.8328;  0.4950]        Population
## Metabolic Syndrome     -0.1481 [-0.6279;  0.3316]        Population
## T2D                    -0.0180 [-0.4471;  0.4112]        Population
## 30 - 50 y              -0.3785 [-0.7706;  0.0135]               Age
## > 50 y                 -0.1097 [-0.4357;  0.2163]               Age
## < 5 weeks              -0.4228 [-0.9169;  0.0713] Training Duration
## 5 - 10 weeks           -0.0768 [-0.6061;  0.4525] Training Duration
## > 10 weeks             -0.1396 [-0.5073;  0.2282] Training Duration
## < 0.5                  -0.2034 [-1.1868;  0.7799]         Men Ratio
## > 0.5                  -0.2290 [-0.4829;  0.0248]         Men Ratio
## Cycling                -0.2620 [-0.5570;  0.0329]  Type of Exercise
## Running                -0.1487 [-0.5933;  0.2958]  Type of Exercise
## < 40 mmol/mol          -0.3318 [-0.6759;  0.0124]   Baseline Values
## > 40 mmol/mol          -0.1188 [-0.4699;  0.2324]   Baseline Values
## HIIT                   -0.1191 [-0.4211;  0.1829]      Type of HIIE
## SIT                    -0.4228 [-0.9169;  0.0713]      Type of HIIE

15.2.2 Population

15.2.2.1 Forest plot

15.2.2.2 R output
##                            SMD             95%-CI %W(fixed) %W(random)             population
## Abdelbasset 2020       -0.0000 [-0.7044;  0.7044]      12.2       12.2                    T2D
## Eguchi 2012            -0.0554 [-0.9321;  0.8213]       7.8        7.8                Healthy
## Honkala 2017 (Healthy) -0.7456 [-1.5117;  0.0205]      10.3       10.3                Healthy
## Honkala 2017 (T2D)      0.2285 [-0.7624;  1.2194]       6.1        6.1                    T2D
## Madssen 2014           -0.1727 [-0.8365;  0.4911]      13.7       13.7 Cardiac Rehabilitation
## Maillard 2016          -0.2152 [-1.1980;  0.7676]       6.2        6.2                    T2D
## Matsuo 2015            -0.1827 [-0.9845;  0.6192]       9.4        9.4     Metabolic Syndrome
## Motiani 2017           -0.9164 [-1.7245; -0.1083]       9.2        9.2                Healthy
## Ramos 2016a            -0.1349 [-0.7335;  0.4638]      16.8       16.8     Metabolic Syndrome
## Sjöros 2018            -0.0817 [-0.9384;  0.7750]       8.2        8.2                    T2D
## 
## Number of studies combined: k = 10
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.2352 [-0.4807; 0.0104] -1.88  0.0605
## Random effects model -0.2352 [-0.4807; 0.0104] -1.88  0.0605
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.2142]; tau = 0 [0.0000; 0.4628];
##  I^2 = 0.0% [0.0%; 44.9%]; H = 1.00 [1.00; 1.35]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 28.6%]; H = 1.00 [1.00; 1.18]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.15    9  0.7249
## 
## Results for subgroups (fixed effect model):
##                          k     SMD             95%-CI    Q  I^2
## Healthy                  3 -0.5854 [-1.0562; -0.1145] 2.07 3.5%
## Cardiac Rehabilitation   1 -0.1689 [-0.8328;  0.4950] 0.00   --
## Metabolic Syndrome       2 -0.1481 [-0.6279;  0.3316] 0.01 0.0%
## T2D                      4 -0.0180 [-0.4471;  0.4112] 0.37 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 3.27    3  0.3518
## Within groups  2.45    6  0.8738
## 
## Results for subgroups (random effects model):
##                          k     SMD             95%-CI  tau^2    tau
## Healthy                  3 -0.5844 [-1.0638; -0.1049] 0.0063 0.0794
## Cardiac Rehabilitation   1 -0.1689 [-0.8328;  0.4950]     --     --
## Metabolic Syndrome       2 -0.1481 [-0.6279;  0.3316]      0      0
## T2D                      4 -0.0180 [-0.4471;  0.4112]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   3.18    3  0.3649
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
15.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 10; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.1007)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 6) = 2.6343, p-val = 0.8531
## 
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 3.5150, p-val = 0.3188
## 
## Model Results:
## 
##                               estimate      se     zval    pval    ci.lb    ci.ub 
## intrcpt                        -0.6053  0.2396  -2.5268  0.0115  -1.0748  -0.1358  * 
## .byvarCardiac Rehabilitation    0.4326  0.4148   1.0428  0.2970  -0.3805   1.2456    
## .byvarMetabolic Syndrome        0.4533  0.3425   1.3237  0.1856  -0.2179   1.1246    
## .byvarT2D                       0.5866  0.3245   1.8077  0.0706  -0.0494   1.2227  . 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
15.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

15.2.3 Age

15.2.3.1 Forest plot

15.2.3.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_age
## Abdelbasset 2020       -0.0000 [-0.7044;  0.7044]      12.2       12.2       > 50 y
## Eguchi 2012            -0.0554 [-0.9321;  0.8213]       7.8        7.8       > 50 y
## Honkala 2017 (Healthy) -0.7456 [-1.5117;  0.0205]      10.3       10.3    30 - 50 y
## Honkala 2017 (T2D)      0.2285 [-0.7624;  1.2194]       6.1        6.1    30 - 50 y
## Madssen 2014           -0.1727 [-0.8365;  0.4911]      13.7       13.7       > 50 y
## Maillard 2016          -0.2152 [-1.1980;  0.7676]       6.2        6.2       > 50 y
## Matsuo 2015            -0.1827 [-0.9845;  0.6192]       9.4        9.4    30 - 50 y
## Motiani 2017           -0.9164 [-1.7245; -0.1083]       9.2        9.2    30 - 50 y
## Ramos 2016a            -0.1349 [-0.7335;  0.4638]      16.8       16.8       > 50 y
## Sjöros 2018            -0.0817 [-0.9384;  0.7750]       8.2        8.2    30 - 50 y
## 
## Number of studies combined: k = 10
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.2352 [-0.4807; 0.0104] -1.88  0.0605
## Random effects model -0.2352 [-0.4807; 0.0104] -1.88  0.0605
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.2142]; tau = 0 [0.0000; 0.4628];
##  I^2 = 0.0% [0.0%; 44.9%]; H = 1.00 [1.00; 1.35]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 38.3%]; H = 1.00 [1.00; 1.27]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.15    9  0.7249
## 
## Results for subgroups (fixed effect model):
##             k     SMD             95%-CI    Q  I^2
## 30 - 50 y   5 -0.3825 [-0.7567; -0.0084] 4.38 8.7%
## > 50 y      5 -0.1097 [-0.4357;  0.2163] 0.18 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 1.16    1  0.2812
## Within groups  4.56    8  0.8033
## 
## Results for subgroups (random effects model):
##             k     SMD            95%-CI  tau^2    tau
## 30 - 50 y   5 -0.3785 [-0.7706; 0.0135] 0.0175 0.1321
## > 50 y      5 -0.1097 [-0.4357; 0.2163]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.07    1  0.3014
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
15.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 10; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0781)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 8) = 5.6930, p-val = 0.6816
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.4564, p-val = 0.4993
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -1.0026  1.1429  -0.8772  0.3804  -3.2426  1.2374    
## age        0.0145  0.0214   0.6755  0.4993  -0.0275  0.0564    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
15.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

15.2.4 Training Duration

15.2.4.1 Forest plot

15.2.4.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_duration
## Abdelbasset 2020       -0.0000 [-0.7044;  0.7044]      12.2       12.2      5 - 10 weeks
## Eguchi 2012            -0.0554 [-0.9321;  0.8213]       7.8        7.8        > 10 weeks
## Honkala 2017 (Healthy) -0.7456 [-1.5117;  0.0205]      10.3       10.3         < 5 weeks
## Honkala 2017 (T2D)      0.2285 [-0.7624;  1.2194]       6.1        6.1         < 5 weeks
## Madssen 2014           -0.1727 [-0.8365;  0.4911]      13.7       13.7        > 10 weeks
## Maillard 2016          -0.2152 [-1.1980;  0.7676]       6.2        6.2        > 10 weeks
## Matsuo 2015            -0.1827 [-0.9845;  0.6192]       9.4        9.4      5 - 10 weeks
## Motiani 2017           -0.9164 [-1.7245; -0.1083]       9.2        9.2         < 5 weeks
## Ramos 2016a            -0.1349 [-0.7335;  0.4638]      16.8       16.8        > 10 weeks
## Sjöros 2018            -0.0817 [-0.9384;  0.7750]       8.2        8.2         < 5 weeks
## 
## Number of studies combined: k = 10
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.2352 [-0.4807; 0.0104] -1.88  0.0605
## Random effects model -0.2352 [-0.4807; 0.0104] -1.88  0.0605
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.2142]; tau = 0 [0.0000; 0.4628];
##  I^2 = 0.0% [0.0%; 44.9%]; H = 1.00 [1.00; 1.35]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 46.3%]; H = 1.00 [1.00; 1.36]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.15    9  0.7249
## 
## Results for subgroups (fixed effect model):
##                k     SMD             95%-CI    Q   I^2
## < 5 weeks      4 -0.4398 [-0.8628; -0.0169] 4.06 26.0%
## 5 - 10 weeks   2 -0.0768 [-0.6061;  0.4525] 0.10  0.0%
## > 10 weeks     4 -0.1396 [-0.5073;  0.2282] 0.06  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 1.50    2  0.4726
## Within groups  4.22    7  0.7538
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI  tau^2    tau
## < 5 weeks      4 -0.4228 [-0.9169; 0.0713] 0.0663 0.2575
## 5 - 10 weeks   2 -0.0768 [-0.6061; 0.4525]      0      0
## > 10 weeks     4 -0.1396 [-0.5073; 0.2282]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.09    2  0.5790
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
15.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 10; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0807)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 8) = 5.1622, p-val = 0.7401
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.9871, p-val = 0.3204
## 
## Model Results:
## 
##           estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    -0.4325  0.2348  -1.8417  0.0655  -0.8927  0.0278  . 
## duration    0.0227  0.0229   0.9935  0.3204  -0.0221  0.0676    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
15.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

15.2.5 Men Ratio

15.2.5.1 Forest plot

15.2.5.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_men_ratio
## Abdelbasset 2020       -0.0000 [-0.7044;  0.7044]      12.2       12.2              > 0.5
## Eguchi 2012            -0.0554 [-0.9321;  0.8213]       7.8        7.8              > 0.5
## Honkala 2017 (Healthy) -0.7456 [-1.5117;  0.0205]      10.3       10.3              > 0.5
## Honkala 2017 (T2D)      0.2285 [-0.7624;  1.2194]       6.1        6.1              > 0.5
## Madssen 2014           -0.1727 [-0.8365;  0.4911]      13.7       13.7              > 0.5
## Maillard 2016          -0.2152 [-1.1980;  0.7676]       6.2        6.2              < 0.5
## Matsuo 2015            -0.1827 [-0.9845;  0.6192]       9.4        9.4              > 0.5
## Motiani 2017           -0.9164 [-1.7245; -0.1083]       9.2        9.2              > 0.5
## Ramos 2016a            -0.1349 [-0.7335;  0.4638]      16.8       16.8              > 0.5
## Sjöros 2018            -0.0817 [-0.9384;  0.7750]       8.2        8.2              > 0.5
## 
## Number of studies combined: k = 10
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.2352 [-0.4807; 0.0104] -1.88  0.0605
## Random effects model -0.2352 [-0.4807; 0.0104] -1.88  0.0605
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.2142]; tau = 0 [0.0000; 0.4628];
##  I^2 = 0.0% [0.0%; 44.9%]; H = 1.00 [1.00; 1.35]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 50.8%]; H = 1.00 [1.00; 1.43]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.15    9  0.7249
## 
## Results for subgroups (fixed effect model):
##         k     SMD            95%-CI    Q  I^2
## < 0.5   1 -0.2034 [-1.1868; 0.7799] 0.00   --
## > 0.5   9 -0.2290 [-0.4829; 0.0248] 5.72 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.00    1  0.9606
## Within groups  5.72    8  0.6786
## 
## Results for subgroups (random effects model):
##         k     SMD            95%-CI tau^2 tau
## < 0.5   1 -0.2034 [-1.1868; 0.7799]    --  --
## > 0.5   9 -0.2290 [-0.4829; 0.0248]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.00    1  0.9606
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
15.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 10; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0776)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 8) = 5.5401, p-val = 0.6986
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.6093, p-val = 0.4351
## 
## Model Results:
## 
##            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt      0.0497  0.3858   0.1288  0.8975  -0.7065  0.8059    
## men_ratio   -0.3705  0.4747  -0.7806  0.4351  -1.3009  0.5598    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
15.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

15.2.6 Type of Exercise

15.2.6.1 Forest plot

15.2.6.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) type_exercise
## Abdelbasset 2020       -0.0000 [-0.7044;  0.7044]      12.2       12.2       Cycling
## Eguchi 2012            -0.0554 [-0.9321;  0.8213]       7.8        7.8       Cycling
## Honkala 2017 (Healthy) -0.7456 [-1.5117;  0.0205]      10.3       10.3       Cycling
## Honkala 2017 (T2D)      0.2285 [-0.7624;  1.2194]       6.1        6.1       Cycling
## Madssen 2014           -0.1727 [-0.8365;  0.4911]      13.7       13.7       Running
## Maillard 2016          -0.2152 [-1.1980;  0.7676]       6.2        6.2       Cycling
## Matsuo 2015            -0.1827 [-0.9845;  0.6192]       9.4        9.4       Cycling
## Motiani 2017           -0.9164 [-1.7245; -0.1083]       9.2        9.2       Cycling
## Ramos 2016a            -0.1349 [-0.7335;  0.4638]      16.8       16.8       Running
## Sjöros 2018            -0.0817 [-0.9384;  0.7750]       8.2        8.2       Cycling
## 
## Number of studies combined: k = 10
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.2352 [-0.4807; 0.0104] -1.88  0.0605
## Random effects model -0.2352 [-0.4807; 0.0104] -1.88  0.0605
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.2142]; tau = 0 [0.0000; 0.4628];
##  I^2 = 0.0% [0.0%; 44.9%]; H = 1.00 [1.00; 1.35]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 49.2%]; H = 1.00 [1.00; 1.40]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.15    9  0.7249
## 
## Results for subgroups (fixed effect model):
##           k     SMD            95%-CI    Q  I^2
## Cycling   8 -0.2620 [-0.5570; 0.0329] 5.54 0.0%
## Running   2 -0.1487 [-0.5933; 0.2958] 0.01 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.17    1  0.6773
## Within groups  5.55    8  0.6976
## 
## Results for subgroups (random effects model):
##           k     SMD            95%-CI tau^2 tau
## Cycling   8 -0.2620 [-0.5570; 0.0329]     0   0
## Running   2 -0.1487 [-0.5933; 0.2958]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.17    1  0.6773
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
15.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 10; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0832)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 8) = 5.9552, p-val = 0.6523
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.1942, p-val = 0.6595
## 
## Model Results:
## 
##                       estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                -0.2717  0.1503  -1.8079  0.0706  -0.5663  0.0229  . 
## type_exerciseRunning    0.1199  0.2721   0.4406  0.6595  -0.4134  0.6532    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
15.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

15.2.7 Baseline Values

15.2.7.1 Forest plot

15.2.7.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) category_bsln
## Abdelbasset 2020       -0.0000 [-0.7044;  0.7044]      12.2       12.2 > 40 mmol/mol
## Eguchi 2012            -0.0554 [-0.9321;  0.8213]       7.8        7.8 < 40 mmol/mol
## Honkala 2017 (Healthy) -0.7456 [-1.5117;  0.0205]      10.3       10.3 < 40 mmol/mol
## Honkala 2017 (T2D)      0.2285 [-0.7624;  1.2194]       6.1        6.1 < 40 mmol/mol
## Madssen 2014           -0.1727 [-0.8365;  0.4911]      13.7       13.7 > 40 mmol/mol
## Maillard 2016          -0.2152 [-1.1980;  0.7676]       6.2        6.2 > 40 mmol/mol
## Matsuo 2015            -0.1827 [-0.9845;  0.6192]       9.4        9.4 < 40 mmol/mol
## Motiani 2017           -0.9164 [-1.7245; -0.1083]       9.2        9.2 < 40 mmol/mol
## Ramos 2016a            -0.1349 [-0.7335;  0.4638]      16.8       16.8 > 40 mmol/mol
## Sjöros 2018            -0.0817 [-0.9384;  0.7750]       8.2        8.2 < 40 mmol/mol
## 
## Number of studies combined: k = 10
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.2352 [-0.4807; 0.0104] -1.88  0.0605
## Random effects model -0.2352 [-0.4807; 0.0104] -1.88  0.0605
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.2142]; tau = 0 [0.0000; 0.4628];
##  I^2 = 0.0% [0.0%; 44.9%]; H = 1.00 [1.00; 1.35]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 43.7%]; H = 1.00 [1.00; 1.33]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.15    9  0.7249
## 
## Results for subgroups (fixed effect model):
##                 k     SMD            95%-CI    Q  I^2
## < 40 mmol/mol   6 -0.3318 [-0.6759; 0.0124] 4.84 0.0%
## > 40 mmol/mol   4 -0.1188 [-0.4699; 0.2324] 0.16 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 0.72    1  0.3958
## Within groups  5.00    8  0.7575
## 
## Results for subgroups (random effects model):
##                 k     SMD            95%-CI tau^2 tau
## < 40 mmol/mol   6 -0.3318 [-0.6759; 0.0124]     0   0
## > 40 mmol/mol   4 -0.1188 [-0.4699; 0.2324]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.72    1  0.3958
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
15.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 10; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0780)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 8) = 5.4957, p-val = 0.7035
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.6536, p-val = 0.4188
## 
## Model Results:
## 
##                estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt         -0.9685  0.9157  -1.0577  0.2902  -2.7633  0.8262    
## bsln_adjusted    0.0175  0.0216   0.8085  0.4188  -0.0249  0.0599    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
15.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

15.2.8 Type of HIIE

15.2.8.1 Forest plot

15.2.8.2 R output
##                            SMD             95%-CI %W(fixed) %W(random) HIIE
## Abdelbasset 2020       -0.0000 [-0.7044;  0.7044]      12.2       12.2 HIIT
## Eguchi 2012            -0.0554 [-0.9321;  0.8213]       7.8        7.8 HIIT
## Honkala 2017 (Healthy) -0.7456 [-1.5117;  0.0205]      10.3       10.3  SIT
## Honkala 2017 (T2D)      0.2285 [-0.7624;  1.2194]       6.1        6.1  SIT
## Madssen 2014           -0.1727 [-0.8365;  0.4911]      13.7       13.7 HIIT
## Maillard 2016          -0.2152 [-1.1980;  0.7676]       6.2        6.2 HIIT
## Matsuo 2015            -0.1827 [-0.9845;  0.6192]       9.4        9.4 HIIT
## Motiani 2017           -0.9164 [-1.7245; -0.1083]       9.2        9.2  SIT
## Ramos 2016a            -0.1349 [-0.7335;  0.4638]      16.8       16.8 HIIT
## Sjöros 2018            -0.0817 [-0.9384;  0.7750]       8.2        8.2  SIT
## 
## Number of studies combined: k = 10
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.2352 [-0.4807; 0.0104] -1.88  0.0605
## Random effects model -0.2352 [-0.4807; 0.0104] -1.88  0.0605
## 
## Quantifying heterogeneity:
##  tau^2 = 0 [0.0000; 0.2142]; tau = 0 [0.0000; 0.4628];
##  I^2 = 0.0% [0.0%; 44.9%]; H = 1.00 [1.00; 1.35]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 33.9%]; H = 1.00 [1.00; 1.23]
## 
## Test of heterogeneity:
##     Q d.f. p-value
##  6.15    9  0.7249
## 
## Results for subgroups (fixed effect model):
##        k     SMD             95%-CI    Q   I^2
## HIIT   6 -0.1191 [-0.4211;  0.1829] 0.20  0.0%
## SIT    4 -0.4398 [-0.8628; -0.0169] 4.06 26.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 1.46    1  0.2265
## Within groups  4.26    8  0.8330
## 
## Results for subgroups (random effects model):
##        k     SMD            95%-CI  tau^2    tau
## HIIT   6 -0.1191 [-0.4211; 0.1829]      0      0
## SIT    4 -0.4228 [-0.9169; 0.0713] 0.0663 0.2575
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   1.06    1  0.3041
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
15.2.8.3 Meta-regression
## 
## Mixed-Effects Model (k = 10; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0794)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 8) = 4.5804, p-val = 0.8013
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.5690, p-val = 0.2104
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.1228  0.1541  -0.7974  0.4252  -0.4248  0.1791    
## HIIESIT   -0.3316  0.2647  -1.2526  0.2104  -0.8505  0.1873    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
15.2.8.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

16. HOMA-IR

16.1 Overall

16.1.1 Forest plot

16.1.2 R output

##                      SMD             95%-CI %W(fixed) %W(random)
## Abdelbasset 2020  0.0000 [-0.7044;  0.7044]       7.5        7.4
## Ciolac 2010      -0.0896 [-0.7829;  0.6037]       7.7        7.7
## Fisher 2015      -0.2395 [-1.0668;  0.5878]       5.4        5.4
## Gillen 2016       0.1059 [-0.7953;  1.0071]       4.6        4.6
## Grieco 2013       0.3271 [-0.5176;  1.1719]       5.2        5.2
## Hovanloo 2013     1.1129 [ 0.0598;  2.1660]       3.3        3.4
## Lunt 2014        -0.3860 [-1.2308;  0.4588]       5.2        5.2
## Lunt 2014        -0.2755 [-1.1167;  0.5656]       5.2        5.3
## Matsuo 2015      -0.0673 [-0.8677;  0.7331]       5.8        5.8
## Mitranun 2014     0.0348 [-0.7060;  0.7757]       6.7        6.7
## Ramos 2016a      -0.3407 [-0.9430;  0.2615]      10.2       10.0
## Robinson 2015     0.1583 [-0.4706;  0.7872]       9.4        9.2
## Sandvei 2012     -0.3121 [-1.1352;  0.5110]       5.5        5.5
## Sawyer 2016       0.1095 [-0.8151;  1.0342]       4.3        4.4
## Skleryk 2013     -0.3525 [-1.3400;  0.6351]       3.8        3.8
## Trapp 2008        0.6178 [-0.1148;  1.3503]       6.9        6.9
## Winn 2018        -1.1850 [-2.2475; -0.1225]       3.3        3.3
## 
## Number of studies combined: k = 17
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0388 [-0.2313; 0.1537] -0.40  0.6927
## Random effects model -0.0388 [-0.2345; 0.1568] -0.39  0.6972
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0049 [0.0000; 0.2894]; tau = 0.0701 [0.0000; 0.5379];
##  I^2 = 2.9% [0.0%; 52.5%]; H = 1.01 [1.00; 1.45]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  16.48   16  0.4202
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)

16.1.3 Sensitivity analysis

16.1.3.1 Forest plot

16.1.3.2 R output
## 
## Influential analysis (Random effects model)
## 
##                                 SMD            95%-CI p-value   tau^2     tau   I^2
## Omitting Abdelbasset 2020   -0.0403 [-0.2407; 0.1600]  0.6932  0.0000  0.0000  0.0%
## Omitting Ciolac 2010        -0.0331 [-0.2338; 0.1675]  0.7462  0.0000  0.0000  0.0%
## Omitting Fisher 2015        -0.0262 [-0.2244; 0.1719]  0.7953  0.0000  0.0000  0.0%
## Omitting Gillen 2016        -0.0440 [-0.2412; 0.1533]  0.6624  0.0000  0.0000  0.0%
## Omitting Grieco 2013        -0.0566 [-0.2545; 0.1413]  0.5751  0.0000  0.0000  0.0%
## Omitting Hovanloo 2013      -0.0741 [-0.2701; 0.1218]  0.4585  0.0000  0.0000  0.0%
## Omitting Lunt 2014          -0.0190 [-0.2169; 0.1789]  0.8509  0.0000  0.0000  0.0%
## Omitting Lunt 2014          -0.0247 [-0.2227; 0.1733]  0.8069  0.0000  0.0000  0.0%
## Omitting Matsuo 2015        -0.0356 [-0.2342; 0.1629]  0.7252  0.0000  0.0000  0.0%
## Omitting Mitranun 2014      -0.0425 [-0.2421; 0.1571]  0.6766  0.0000  0.0000  0.0%
## Omitting Ramos 2016a        -0.0035 [-0.2069; 0.2000]  0.9735  0.0000  0.0000  0.0%
## Omitting Robinson 2015      -0.0573 [-0.2597; 0.1452]  0.5794  0.0000  0.0000  0.0%
## Omitting Sandvei 2012       -0.0221 [-0.2203; 0.1762]  0.8274  0.0000  0.0000  0.0%
## Omitting Sawyer 2016        -0.0437 [-0.2408; 0.1533]  0.6635  0.0000  0.0000  0.0%
## Omitting Skleryk 2013       -0.0256 [-0.2221; 0.1709]  0.7982  0.0000  0.0000  0.0%
## Omitting Trapp 2008         -0.0846 [-0.2843; 0.1151]  0.4065  0.0000  0.0000  0.0%
## Omitting Winn 2018          -0.0015 [-0.1974; 0.1944]  0.9880  0.0000  0.0000  0.0%
##                                                                                    
## Pooled estimate             -0.0388 [-0.2345; 0.1568]  0.6972  0.0049  0.0701  2.9%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

16.1.4 Small-study effects

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

Left panel shows the contour-enhanced funnel plot for the meta-analysis. The shaded areas represent the p-value: light gray p < 0.01, gray 0.05 > p > 0.01, dark gray 0.1 > p > 0.05. The standard error of each study is plotted as a function of the effect size (Cohen’s d). Negative and positive x-axis values represent a favorable effect for MICT and HIIE, respectively. Right panel shows the radial plot, with the standardized treatment effect (z-score) plotted as a function of the inverse of the standard error. The dashed line represents the regression line, and the continuous line represents the regression line from the Egger Test.

16.2 Subgroups

16.2.1 Overall

16.2.1.1 Forest plot

16.2.1.2 R output
##                        SMD            95%-CI     meta-analysis
##                    -0.0388 [-0.2345; 0.1568]           Overall
## Healthy             0.2277 [-0.1342; 0.5896]        Population
## Overweight/obese   -0.3300 [-0.6995; 0.0396]        Population
## Metabolic Syndrome -0.0922 [-0.4744; 0.2901]        Population
## T2D                 0.0161 [-0.4944; 0.5266]        Population
## < 30 y              0.1619 [-0.1690; 0.4928]               Age
## 30 - 50 y          -0.2937 [-0.6607; 0.0733]               Age
## > 50 y             -0.0513 [-0.3824; 0.2799]               Age
## < 5 weeks           0.0367 [-0.5651; 0.6386] Training Duration
## 5 - 10 weeks       -0.0735 [-0.3979; 0.2509] Training Duration
## > 10 weeks         -0.0706 [-0.3759; 0.2346] Training Duration
## < 0.5               0.0282 [-0.2511; 0.3075]         Men Ratio
## > 0.5              -0.1569 [-0.4724; 0.1586]         Men Ratio
## Cycling             0.1548 [-0.1024; 0.4120]  Type of Exercise
## Running            -0.2835 [-0.5745; 0.0076]  Type of Exercise
## < 3                 0.0138 [-0.2081; 0.2358]   Baseline Values
## > 3                -0.1939 [-0.5823; 0.1944]   Baseline Values
## HIIT               -0.0952 [-0.3336; 0.1432]      Type of HIIE
## SIT                 0.0715 [-0.3040; 0.4470]      Type of HIIE
## 
## Number of studies combined: k = 17
## 
##                          SMD            95%-CI     z p-value
## Random effects model -0.0388 [-0.2345; 0.1568] -0.39  0.6972
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0049; tau = 0.0701; I^2 = 2.9% [0.0%; 52.5%]; H = 1.01 [1.00; 1.45]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  16.48   16  0.4202
## 
## Results for meta-analyses (random effects model):
##                     k     SMD            95%-CI  tau^2    tau     Q  I^2
## Overall            17 -0.0388 [-0.2345; 0.1568] 0.0049 0.0701 16.48 2.9%
## Population         17 -0.0388 [-0.2345; 0.1568] 0.0049 0.0701 16.48 2.9%
## Age                17 -0.0388 [-0.2345; 0.1568] 0.0049 0.0701 16.48 2.9%
## Training Duration  17 -0.0388 [-0.2345; 0.1568] 0.0049 0.0701 16.48 2.9%
## Men Ratio          17 -0.0388 [-0.2345; 0.1568] 0.0049 0.0701 16.48 2.9%
## Type of Exercise   17 -0.0388 [-0.2345; 0.1568] 0.0049 0.0701 16.48 2.9%
## Baseline Values    17 -0.0388 [-0.2345; 0.1568] 0.0049 0.0701 16.48 2.9%
## Type of HIIE       17 -0.0388 [-0.2345; 0.1568] 0.0049 0.0701 16.48 2.9%
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2

16.2.2 Population

16.2.2.1 Forest plot

16.2.2.2 R output
##                      SMD             95%-CI %W(fixed) %W(random)         population
## Abdelbasset 2020  0.0000 [-0.7044;  0.7044]       7.5        7.4                T2D
## Ciolac 2010      -0.0896 [-0.7829;  0.6037]       7.7        7.7            Healthy
## Fisher 2015      -0.2395 [-1.0668;  0.5878]       5.4        5.4   Overweight/obese
## Gillen 2016       0.1059 [-0.7953;  1.0071]       4.6        4.6            Healthy
## Grieco 2013       0.3271 [-0.5176;  1.1719]       5.2        5.2            Healthy
## Hovanloo 2013     1.1129 [ 0.0598;  2.1660]       3.3        3.4            Healthy
## Lunt 2014        -0.3860 [-1.2308;  0.4588]       5.2        5.2   Overweight/obese
## Lunt 2014        -0.2755 [-1.1167;  0.5656]       5.2        5.3   Overweight/obese
## Matsuo 2015      -0.0673 [-0.8677;  0.7331]       5.8        5.8 Metabolic Syndrome
## Mitranun 2014     0.0348 [-0.7060;  0.7757]       6.7        6.7                T2D
## Ramos 2016a      -0.3407 [-0.9430;  0.2615]      10.2       10.0 Metabolic Syndrome
## Robinson 2015     0.1583 [-0.4706;  0.7872]       9.4        9.2 Metabolic Syndrome
## Sandvei 2012     -0.3121 [-1.1352;  0.5110]       5.5        5.5            Healthy
## Sawyer 2016       0.1095 [-0.8151;  1.0342]       4.3        4.4   Overweight/obese
## Skleryk 2013     -0.3525 [-1.3400;  0.6351]       3.8        3.8   Overweight/obese
## Trapp 2008        0.6178 [-0.1148;  1.3503]       6.9        6.9            Healthy
## Winn 2018        -1.1850 [-2.2475; -0.1225]       3.3        3.3   Overweight/obese
## 
## Number of studies combined: k = 17
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0388 [-0.2313; 0.1537] -0.40  0.6927
## Random effects model -0.0388 [-0.2345; 0.1568] -0.39  0.6972
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0049 [0.0000; 0.2894]; tau = 0.0701 [0.0000; 0.5379];
##  I^2 = 2.9% [0.0%; 52.5%]; H = 1.01 [1.00; 1.45]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 41.6%]; H = 1.00 [1.00; 1.31]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  16.48   16  0.4202
## 
## Results for subgroups (fixed effect model):
##                      k     SMD            95%-CI    Q   I^2
## Healthy              6  0.2221 [-0.1128; 0.5569] 5.78 13.5%
## Overweight/obese     6 -0.3300 [-0.6995; 0.0396] 3.00  0.0%
## Metabolic Syndrome   3 -0.0922 [-0.4744; 0.2901] 1.22  0.0%
## T2D                  2  0.0161 [-0.4944; 0.5266] 0.00  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  4.84    3  0.1842
## Within groups  10.01   13  0.6934
## 
## Results for subgroups (random effects model):
##                      k     SMD            95%-CI  tau^2    tau
## Healthy              6  0.2277 [-0.1342; 0.5896] 0.0277 0.1666
## Overweight/obese     6 -0.3300 [-0.6995; 0.0396]      0      0
## Metabolic Syndrome   3 -0.0922 [-0.4744; 0.2901]      0      0
## T2D                  2  0.0161 [-0.4944; 0.5266]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   4.58    3  0.2054
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
16.2.2.3 Meta-regression
## 
## Mixed-Effects Model (k = 17; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0670)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            100.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 13) = 11.0804, p-val = 0.6041
## 
## Test of Moderators (coefficients 2:4):
## QM(df = 3) = 5.3966, p-val = 0.1450
## 
## Model Results:
## 
##                           estimate      se     zval    pval    ci.lb    ci.ub 
## intrcpt                     0.2341  0.1705   1.3731  0.1697  -0.1001   0.5683    
## .byvarOverweight/obese     -0.5825  0.2539  -2.2944  0.0218  -1.0801  -0.0849  * 
## .byvarMetabolic Syndrome   -0.3282  0.2590  -1.2671  0.2051  -0.8359   0.1795    
## .byvarT2D                  -0.2176  0.3113  -0.6989  0.4846  -0.8277   0.3926    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
16.2.2.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

16.2.3 Age

16.2.3.1 Forest plot

16.2.3.2 R output
##                      SMD             95%-CI %W(fixed) %W(random) category_age
## Abdelbasset 2020  0.0000 [-0.7044;  0.7044]       7.5        7.4       > 50 y
## Ciolac 2010      -0.0896 [-0.7829;  0.6037]       7.7        7.7       < 30 y
## Fisher 2015      -0.2395 [-1.0668;  0.5878]       5.4        5.4       < 30 y
## Gillen 2016       0.1059 [-0.7953;  1.0071]       4.6        4.6       < 30 y
## Grieco 2013       0.3271 [-0.5176;  1.1719]       5.2        5.2       < 30 y
## Hovanloo 2013     1.1129 [ 0.0598;  2.1660]       3.3        3.4       < 30 y
## Lunt 2014        -0.3860 [-1.2308;  0.4588]       5.2        5.2    30 - 50 y
## Lunt 2014        -0.2755 [-1.1167;  0.5656]       5.2        5.3    30 - 50 y
## Matsuo 2015      -0.0673 [-0.8677;  0.7331]       5.8        5.8    30 - 50 y
## Mitranun 2014     0.0348 [-0.7060;  0.7757]       6.7        6.7       > 50 y
## Ramos 2016a      -0.3407 [-0.9430;  0.2615]      10.2       10.0       > 50 y
## Robinson 2015     0.1583 [-0.4706;  0.7872]       9.4        9.2       > 50 y
## Sandvei 2012     -0.3121 [-1.1352;  0.5110]       5.5        5.5       < 30 y
## Sawyer 2016       0.1095 [-0.8151;  1.0342]       4.3        4.4    30 - 50 y
## Skleryk 2013     -0.3525 [-1.3400;  0.6351]       3.8        3.8    30 - 50 y
## Trapp 2008        0.6178 [-0.1148;  1.3503]       6.9        6.9       < 30 y
## Winn 2018        -1.1850 [-2.2475; -0.1225]       3.3        3.3    30 - 50 y
## 
## Number of studies combined: k = 17
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0388 [-0.2313; 0.1537] -0.40  0.6927
## Random effects model -0.0388 [-0.2345; 0.1568] -0.39  0.6972
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0049 [0.0000; 0.2894]; tau = 0.0701 [0.0000; 0.5379];
##  I^2 = 2.9% [0.0%; 52.5%]; H = 1.01 [1.00; 1.45]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 43.2%]; H = 1.00 [1.00; 1.33]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  16.48   16  0.4202
## 
## Results for subgroups (fixed effect model):
##             k     SMD            95%-CI    Q   I^2
## < 30 y      7  0.1584 [-0.1520; 0.4688] 6.77 11.4%
## 30 - 50 y   6 -0.2937 [-0.6607; 0.0733] 3.33  0.0%
## > 50 y      4 -0.0513 [-0.3824; 0.2799] 1.33  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  3.41    2  0.1818
## Within groups  11.43   14  0.6517
## 
## Results for subgroups (random effects model):
##             k     SMD            95%-CI  tau^2    tau
## < 30 y      7  0.1619 [-0.1690; 0.4928] 0.0227 0.1508
## 30 - 50 y   6 -0.2937 [-0.6607; 0.0733]      0      0
## > 50 y      4 -0.0513 [-0.3824; 0.2799]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   3.27    2  0.1951
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
16.2.3.3 Meta-regression
## 
## Mixed-Effects Model (k = 17; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0609)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            100.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 15) = 14.5449, p-val = 0.4847
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.9320, p-val = 0.1645
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt    0.3425  0.2914   1.1754  0.2398  -0.2286  0.9135    
## age       -0.0095  0.0068  -1.3900  0.1645  -0.0228  0.0039    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
16.2.3.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

16.2.4 Training Duration

16.2.4.1 Forest plot

16.2.4.2 R output
##                      SMD             95%-CI %W(fixed) %W(random) category_duration
## Abdelbasset 2020  0.0000 [-0.7044;  0.7044]       7.5        7.4      5 - 10 weeks
## Ciolac 2010      -0.0896 [-0.7829;  0.6037]       7.7        7.7        > 10 weeks
## Fisher 2015      -0.2395 [-1.0668;  0.5878]       5.4        5.4      5 - 10 weeks
## Gillen 2016       0.1059 [-0.7953;  1.0071]       4.6        4.6        > 10 weeks
## Grieco 2013       0.3271 [-0.5176;  1.1719]       5.2        5.2         < 5 weeks
## Hovanloo 2013     1.1129 [ 0.0598;  2.1660]       3.3        3.4         < 5 weeks
## Lunt 2014        -0.3860 [-1.2308;  0.4588]       5.2        5.2        > 10 weeks
## Lunt 2014        -0.2755 [-1.1167;  0.5656]       5.2        5.3        > 10 weeks
## Matsuo 2015      -0.0673 [-0.8677;  0.7331]       5.8        5.8      5 - 10 weeks
## Mitranun 2014     0.0348 [-0.7060;  0.7757]       6.7        6.7      5 - 10 weeks
## Ramos 2016a      -0.3407 [-0.9430;  0.2615]      10.2       10.0        > 10 weeks
## Robinson 2015     0.1583 [-0.4706;  0.7872]       9.4        9.2         < 5 weeks
## Sandvei 2012     -0.3121 [-1.1352;  0.5110]       5.5        5.5      5 - 10 weeks
## Sawyer 2016       0.1095 [-0.8151;  1.0342]       4.3        4.4      5 - 10 weeks
## Skleryk 2013     -0.3525 [-1.3400;  0.6351]       3.8        3.8         < 5 weeks
## Trapp 2008        0.6178 [-0.1148;  1.3503]       6.9        6.9        > 10 weeks
## Winn 2018        -1.1850 [-2.2475; -0.1225]       3.3        3.3         < 5 weeks
## 
## Number of studies combined: k = 17
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0388 [-0.2313; 0.1537] -0.40  0.6927
## Random effects model -0.0388 [-0.2345; 0.1568] -0.39  0.6972
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0049 [0.0000; 0.2894]; tau = 0.0701 [0.0000; 0.5379];
##  I^2 = 2.9% [0.0%; 52.5%]; H = 1.01 [1.00; 1.45]
## 
## Quantifying residual heterogeneity:
##  I^2 = 3.2% [0.0%; 55.1%]; H = 1.02 [1.00; 1.49]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  16.48   16  0.4202
## 
## Results for subgroups (fixed effect model):
##                k     SMD            95%-CI    Q   I^2
## < 5 weeks      5  0.0675 [-0.3191; 0.4541] 8.98 55.5%
## 5 - 10 weeks   6 -0.0735 [-0.3979; 0.2509] 0.70  0.0%
## > 10 weeks     6 -0.0706 [-0.3759; 0.2346] 4.79  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.38    2  0.8287
## Within groups  14.47   14  0.4155
## 
## Results for subgroups (random effects model):
##                k     SMD            95%-CI  tau^2    tau
## < 5 weeks      5  0.0367 [-0.5651; 0.6386] 0.2563 0.5062
## 5 - 10 weeks   6 -0.0735 [-0.3979; 0.2509]      0      0
## > 10 weeks     6 -0.0706 [-0.3759; 0.2346]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.11    2  0.9463
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
16.2.4.3 Meta-regression
## 
## Mixed-Effects Model (k = 17; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0141 (SE = 0.0664)
## tau (square root of estimated tau^2 value):             0.1189
## I^2 (residual heterogeneity / unaccounted variability): 7.77%
## H^2 (unaccounted variability / sampling variability):   1.08
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 15) = 16.2639, p-val = 0.3647
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.1764, p-val = 0.6745
## 
## Model Results:
## 
##           estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt     0.0430  0.2204   0.1950  0.8454  -0.3890  0.4750    
## duration   -0.0089  0.0213  -0.4200  0.6745  -0.0506  0.0328    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
16.2.4.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

16.2.5 Men Ratio

16.2.5.1 Forest plot

16.2.5.2 R output
##                      SMD             95%-CI %W(fixed) %W(random) category_men_ratio
## Abdelbasset 2020  0.0000 [-0.7044;  0.7044]       7.5        7.4              > 0.5
## Ciolac 2010      -0.0896 [-0.7829;  0.6037]       7.7        7.7              < 0.5
## Fisher 2015      -0.2395 [-1.0668;  0.5878]       5.4        5.4              > 0.5
## Gillen 2016       0.1059 [-0.7953;  1.0071]       4.6        4.6              > 0.5
## Grieco 2013       0.3271 [-0.5176;  1.1719]       5.2        5.2              < 0.5
## Hovanloo 2013     1.1129 [ 0.0598;  2.1660]       3.3        3.4              < 0.5
## Lunt 2014        -0.3860 [-1.2308;  0.4588]       5.2        5.2              < 0.5
## Lunt 2014        -0.2755 [-1.1167;  0.5656]       5.2        5.3              < 0.5
## Matsuo 2015      -0.0673 [-0.8677;  0.7331]       5.8        5.8              > 0.5
## Mitranun 2014     0.0348 [-0.7060;  0.7757]       6.7        6.7              < 0.5
## Ramos 2016a      -0.3407 [-0.9430;  0.2615]      10.2       10.0              > 0.5
## Robinson 2015     0.1583 [-0.4706;  0.7872]       9.4        9.2              < 0.5
## Sandvei 2012     -0.3121 [-1.1352;  0.5110]       5.5        5.5              < 0.5
## Sawyer 2016       0.1095 [-0.8151;  1.0342]       4.3        4.4              < 0.5
## Skleryk 2013     -0.3525 [-1.3400;  0.6351]       3.8        3.8              > 0.5
## Trapp 2008        0.6178 [-0.1148;  1.3503]       6.9        6.9              < 0.5
## Winn 2018        -1.1850 [-2.2475; -0.1225]       3.3        3.3              < 0.5
## 
## Number of studies combined: k = 17
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0388 [-0.2313; 0.1537] -0.40  0.6927
## Random effects model -0.0388 [-0.2345; 0.1568] -0.39  0.6972
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0049 [0.0000; 0.2894]; tau = 0.0701 [0.0000; 0.5379];
##  I^2 = 2.9% [0.0%; 52.5%]; H = 1.01 [1.00; 1.45]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 48.8%]; H = 1.00 [1.00; 1.40]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  16.48   16  0.4202
## 
## Results for subgroups (fixed effect model):
##         k     SMD            95%-CI     Q   I^2
## < 0.5  11  0.0338 [-0.2096; 0.2772] 12.92 22.6%
## > 0.5   6 -0.1569 [-0.4724; 0.1586]  1.04  0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.88    1  0.3482
## Within groups  13.96   15  0.5283
## 
## Results for subgroups (random effects model):
##         k     SMD            95%-CI  tau^2    tau
## < 0.5  11  0.0282 [-0.2511; 0.3075] 0.0500 0.2237
## > 0.5   6 -0.1569 [-0.4724; 0.1586]      0      0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.74    1  0.3893
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
16.2.5.3 Meta-regression
## 
## Mixed-Effects Model (k = 17; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0015 (SE = 0.0612)
## tau (square root of estimated tau^2 value):             0.0386
## I^2 (residual heterogeneity / unaccounted variability): 0.89%
## H^2 (unaccounted variability / sampling variability):   1.01
## R^2 (amount of heterogeneity accounted for):            69.77%
## 
## Test for Residual Heterogeneity:
## QE(df = 15) = 15.1344, p-val = 0.4418
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.3259, p-val = 0.2495
## 
## Model Results:
## 
##            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt      0.1373  0.1820   0.7543  0.4507  -0.2195  0.4941    
## men_ratio   -0.3591  0.3119  -1.1515  0.2495  -0.9703  0.2521    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
16.2.5.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

16.2.6 Type of Exercise

16.2.6.1 Forest plot

16.2.6.2 R output
##                      SMD             95%-CI %W(fixed) %W(random) type_exercise
## Abdelbasset 2020  0.0000 [-0.7044;  0.7044]       7.5        7.4       Cycling
## Ciolac 2010      -0.0896 [-0.7829;  0.6037]       7.7        7.7       Running
## Fisher 2015      -0.2395 [-1.0668;  0.5878]       5.4        5.4       Cycling
## Gillen 2016       0.1059 [-0.7953;  1.0071]       4.6        4.6       Cycling
## Grieco 2013       0.3271 [-0.5176;  1.1719]       5.2        5.2       Cycling
## Hovanloo 2013     1.1129 [ 0.0598;  2.1660]       3.3        3.4       Cycling
## Lunt 2014        -0.3860 [-1.2308;  0.4588]       5.2        5.2       Running
## Lunt 2014        -0.2755 [-1.1167;  0.5656]       5.2        5.3       Running
## Matsuo 2015      -0.0673 [-0.8677;  0.7331]       5.8        5.8       Cycling
## Mitranun 2014     0.0348 [-0.7060;  0.7757]       6.7        6.7       Running
## Ramos 2016a      -0.3407 [-0.9430;  0.2615]      10.2       10.0       Running
## Robinson 2015     0.1583 [-0.4706;  0.7872]       9.4        9.2       Cycling
## Sandvei 2012     -0.3121 [-1.1352;  0.5110]       5.5        5.5       Running
## Sawyer 2016       0.1095 [-0.8151;  1.0342]       4.3        4.4       Cycling
## Skleryk 2013     -0.3525 [-1.3400;  0.6351]       3.8        3.8       Cycling
## Trapp 2008        0.6178 [-0.1148;  1.3503]       6.9        6.9       Cycling
## Winn 2018        -1.1850 [-2.2475; -0.1225]       3.3        3.3       Running
## 
## Number of studies combined: k = 17
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0388 [-0.2313; 0.1537] -0.40  0.6927
## Random effects model -0.0388 [-0.2345; 0.1568] -0.39  0.6972
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0049 [0.0000; 0.2894]; tau = 0.0701 [0.0000; 0.5379];
##  I^2 = 2.9% [0.0%; 52.5%]; H = 1.01 [1.00; 1.45]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 28.1%]; H = 1.00 [1.00; 1.18]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  16.48   16  0.4202
## 
## Results for subgroups (fixed effect model):
##           k     SMD            95%-CI    Q  I^2
## Cycling  10  0.1548 [-0.1024; 0.4120] 6.55 0.0%
## Running   7 -0.2835 [-0.5745; 0.0076] 3.40 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                   Q d.f. p-value
## Between groups 4.89    1  0.0270
## Within groups  9.95   15  0.8227
## 
## Results for subgroups (random effects model):
##           k     SMD            95%-CI tau^2 tau
## Cycling  10  0.1548 [-0.1024; 0.4120]     0   0
## Running   7 -0.2835 [-0.5745; 0.0076]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   4.89    1  0.0270
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
16.2.6.3 Meta-regression
## 
## Mixed-Effects Model (k = 17; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0 (SE = 0.0608)
## tau (square root of estimated tau^2 value):             0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability):   1.00
## R^2 (amount of heterogeneity accounted for):            100.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 15) = 11.1226, p-val = 0.7439
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 5.3543, p-val = 0.0207
## 
## Model Results:
## 
##                       estimate      se     zval    pval    ci.lb    ci.ub 
## intrcpt                 0.1620  0.1310   1.2361  0.2164  -0.0949   0.4188    
## type_exerciseRunning   -0.4579  0.1979  -2.3139  0.0207  -0.8458  -0.0701  * 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
16.2.6.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

16.2.7 Baseline Values

16.2.7.1 Forest plot

16.2.7.2 R output
##                      SMD             95%-CI %W(fixed) %W(random) category_bsln
## Abdelbasset 2020  0.0000 [-0.7044;  0.7044]       7.5        7.4           > 3
## Ciolac 2010      -0.0896 [-0.7829;  0.6037]       7.7        7.7           < 3
## Fisher 2015      -0.2395 [-1.0668;  0.5878]       5.4        5.4           < 3
## Gillen 2016       0.1059 [-0.7953;  1.0071]       4.6        4.6           < 3
## Grieco 2013       0.3271 [-0.5176;  1.1719]       5.2        5.2           < 3
## Hovanloo 2013     1.1129 [ 0.0598;  2.1660]       3.3        3.4           < 3
## Lunt 2014        -0.3860 [-1.2308;  0.4588]       5.2        5.2           < 3
## Lunt 2014        -0.2755 [-1.1167;  0.5656]       5.2        5.3           < 3
## Matsuo 2015      -0.0673 [-0.8677;  0.7331]       5.8        5.8           > 3
## Mitranun 2014     0.0348 [-0.7060;  0.7757]       6.7        6.7           < 3
## Ramos 2016a      -0.3407 [-0.9430;  0.2615]      10.2       10.0           < 3
## Robinson 2015     0.1583 [-0.4706;  0.7872]       9.4        9.2           < 3
## Sandvei 2012     -0.3121 [-1.1352;  0.5110]       5.5        5.5           < 3
## Sawyer 2016       0.1095 [-0.8151;  1.0342]       4.3        4.4           > 3
## Skleryk 2013     -0.3525 [-1.3400;  0.6351]       3.8        3.8           > 3
## Trapp 2008        0.6178 [-0.1148;  1.3503]       6.9        6.9           < 3
## Winn 2018        -1.1850 [-2.2475; -0.1225]       3.3        3.3           > 3
## 
## Number of studies combined: k = 17
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0388 [-0.2313; 0.1537] -0.40  0.6927
## Random effects model -0.0388 [-0.2345; 0.1568] -0.39  0.6972
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0049 [0.0000; 0.2894]; tau = 0.0701 [0.0000; 0.5379];
##  I^2 = 2.9% [0.0%; 52.5%]; H = 1.01 [1.00; 1.45]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 48.9%]; H = 1.00 [1.00; 1.40]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  16.48   16  0.4202
## 
## Results for subgroups (fixed effect model):
##       k     SMD            95%-CI     Q  I^2
## < 3  12  0.0138 [-0.2081; 0.2358] 10.31 0.0%
## > 3   5 -0.1939 [-0.5823; 0.1944]  3.71 0.0%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.83    1  0.3626
## Within groups  14.01   15  0.5245
## 
## Results for subgroups (random effects model):
##       k     SMD            95%-CI tau^2 tau
## < 3  12  0.0138 [-0.2081; 0.2358]     0   0
## > 3   5 -0.1939 [-0.5823; 0.1944]     0   0
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.83    1  0.3626
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
16.2.7.3 Meta-regression
## 
## Mixed-Effects Model (k = 17; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0130 (SE = 0.0650)
## tau (square root of estimated tau^2 value):             0.1140
## I^2 (residual heterogeneity / unaccounted variability): 7.31%
## H^2 (unaccounted variability / sampling variability):   1.08
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 15) = 16.1829, p-val = 0.3700
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.3098, p-val = 0.5778
## 
## Model Results:
## 
##                estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt          0.0776  0.2329   0.3329  0.7392  -0.3790  0.5341    
## bsln_adjusted   -0.0464  0.0834  -0.5566  0.5778  -0.2099  0.1170    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
16.2.7.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

16.2.8 Type of HIIE

16.2.8.1 Forest plot

16.2.8.2 R output
##                      SMD             95%-CI %W(fixed) %W(random) HIIE
## Abdelbasset 2020  0.0000 [-0.7044;  0.7044]       7.5        7.4 HIIT
## Ciolac 2010      -0.0896 [-0.7829;  0.6037]       7.7        7.7 HIIT
## Fisher 2015      -0.2395 [-1.0668;  0.5878]       5.4        5.4  SIT
## Gillen 2016       0.1059 [-0.7953;  1.0071]       4.6        4.6  SIT
## Grieco 2013       0.3271 [-0.5176;  1.1719]       5.2        5.2 HIIT
## Hovanloo 2013     1.1129 [ 0.0598;  2.1660]       3.3        3.4  SIT
## Lunt 2014        -0.3860 [-1.2308;  0.4588]       5.2        5.2 HIIT
## Lunt 2014        -0.2755 [-1.1167;  0.5656]       5.2        5.3  SIT
## Matsuo 2015      -0.0673 [-0.8677;  0.7331]       5.8        5.8 HIIT
## Mitranun 2014     0.0348 [-0.7060;  0.7757]       6.7        6.7 HIIT
## Ramos 2016a      -0.3407 [-0.9430;  0.2615]      10.2       10.0 HIIT
## Robinson 2015     0.1583 [-0.4706;  0.7872]       9.4        9.2 HIIT
## Sandvei 2012     -0.3121 [-1.1352;  0.5110]       5.5        5.5  SIT
## Sawyer 2016       0.1095 [-0.8151;  1.0342]       4.3        4.4 HIIT
## Skleryk 2013     -0.3525 [-1.3400;  0.6351]       3.8        3.8  SIT
## Trapp 2008        0.6178 [-0.1148;  1.3503]       6.9        6.9  SIT
## Winn 2018        -1.1850 [-2.2475; -0.1225]       3.3        3.3 HIIT
## 
## Number of studies combined: k = 17
## 
##                          SMD            95%-CI     z p-value
## Fixed effect model   -0.0388 [-0.2313; 0.1537] -0.40  0.6927
## Random effects model -0.0388 [-0.2345; 0.1568] -0.39  0.6972
## 
## Quantifying heterogeneity:
##  tau^2 = 0.0049 [0.0000; 0.2894]; tau = 0.0701 [0.0000; 0.5379];
##  I^2 = 2.9% [0.0%; 52.5%]; H = 1.01 [1.00; 1.45]
## 
## Quantifying residual heterogeneity:
##  I^2 = 0.0% [0.0%; 49.6%]; H = 1.00 [1.00; 1.41]
## 
## Test of heterogeneity:
##      Q d.f. p-value
##  16.48   16  0.4202
## 
## Results for subgroups (fixed effect model):
##        k     SMD            95%-CI    Q   I^2
## HIIT  10 -0.0952 [-0.3336; 0.1432] 6.38  0.0%
## SIT    7  0.0717 [-0.2556; 0.3990] 7.81 23.2%
## 
## Test for subgroup differences (fixed effect model):
##                    Q d.f. p-value
## Between groups  0.65    1  0.4191
## Within groups  14.19   15  0.5112
## 
## Results for subgroups (random effects model):
##        k     SMD            95%-CI  tau^2    tau
## HIIT  10 -0.0952 [-0.3336; 0.1432]      0      0
## SIT    7  0.0715 [-0.3040; 0.4470] 0.0594 0.2437
## 
## Test for subgroup differences (random effects model):
##                     Q d.f. p-value
## Between groups   0.54    1  0.4625
## 
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Cohen's d (standardised mean difference)
16.2.8.3 Meta-regression
## 
## Mixed-Effects Model (k = 17; tau^2 estimator: DL)
## 
## tau^2 (estimated amount of residual heterogeneity):     0.0081 (SE = 0.0633)
## tau (square root of estimated tau^2 value):             0.0899
## I^2 (residual heterogeneity / unaccounted variability): 4.67%
## H^2 (unaccounted variability / sampling variability):   1.05
## R^2 (amount of heterogeneity accounted for):            0.00%
## 
## Test for Residual Heterogeneity:
## QE(df = 15) = 15.7346, p-val = 0.3999
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.7131, p-val = 0.3984
## 
## Model Results:
## 
##          estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt   -0.1014  0.1251  -0.8107  0.4176  -0.3467  0.1438    
## HIIESIT    0.1784  0.2113   0.8445  0.3984  -0.2357  0.5926    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
16.2.8.4 Bubble plot
Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.

Bubble plot for the meta-regression for the subgroup analysis. The slope of the meta-regression (β) as well as the associated p-value are printed at the top of the graph.