Table 3.

Outcomes of the included meta-analyses.

References	Results/Main Outcomes
Adamson et al. (50)	Results from the meta-analysis yielded a small but statistically significant effect of 0.28 for depression reduction overall (SE = 0.07; 95% CI: 0.15 to 0.41; p = 0.00). The effect of exercise on depression is slightly larger (0.36) in the MS studies compared with the other studies (0.20), but this difference was not statistically significant.
Bergenthal et al. (46)	The pooled result of three trials (N = 249) for depression shows a statistically significant benefit for the exercise arm (SMD = 0.25; 95% CI:−0.00 to 0.50; p = 0.05).
Brown et al. (24)	Exercise provided a small overall reduction in depressive symptoms compared to standard care among all types of cancer ds = −0.13, 95% CI: −0.26 to −0.01). Collectively, the 40 effect sizes lacked homogeneity I2 = 55%, 95% CI: 35 to 68, p < 0.001. Subgroup analysis by cancer type revealed significant reductions in depressive symptoms among breast cancer survivors d = −0.17, 95% CI: −0.32 to −0.02, but no significant difference in depressive symptoms among prostate, leukemia, lymphoma, and colorectal cancer survivors.
Buffart et al. (25)	After excluding outliers, yoga resulted in significant large reductions in depression (d = −0.69; 95% CI: 1.02 to −0.37).
Carayol et al. (17)	Exercise resulted in an improvement of depressive symptoms (d = −0.275, 95% CI: −0.457 to −0.094, p = 0.003). I2 = 39% (p = 0.09).
Chung et al. (38)	Exercise improved depressive symptoms (d = −1.233, SE = 0.482, 95% CI: −2.177 to −0.289, p = 0.010). Heterogeneity: χ2 = 23.80,p < 0.001; I2 = 87.40%
Craft et al. (26)	Mean ES of (−0.22, 95% CI: −0.43 to −0.009; p = 0.04) under a random effects model, when comparing exercise interventions to control groups.
Cramer et al. (19)	Evidence for large short-term effects were found for depression (SMD = −1.59, 95% CI: −2.68 to −0.51; p < 0.01). Heterogeneity: Chi2 = 84.03, df = 5 (p < 0.00001); I2 = 94% Long-term effects (2 studies, N = 43) of up to 24 weeks did not show significant results.
Cramer et al. (18)	Comparison of yoga vs. no therapy: Yoga did not appear to reduce depression (pooled SMD = −0.13, 95% CI −0.31 to 0.05; seven studies, 496 participants; low-quality evidence). Comparison of yoga vs. psychosocial/educational interventions provided moderate-quality evidence indicating that yoga can reduce depression (pooled SMD = −2.29, 95% CI: −3.97 to −0.61; 4 studies, 226 participants).
Dalgas et al. (48)	In summary the meta-analysis indicated a small beneficial effect of exercise on depressive symptoms in people with MS. The SMD across studies was(g = −0.37, 95% CI: −0.56 to −0.17).
Eng and Reime (52)	Overall, physical exercise resulted in less depressive symptoms over 13 studies involving 1022 patients (SMD = −0.13, 95% CI: −0.26 to −0.01, I2 = 6%, p = 0.03) with low heterogeneity. Ten studies evaluated the effect on depressive symptoms after a period of time had elapsed following the exercise sessions (range from 10 weeks to 9 months). Physical exercise did not change depressive symptoms over these 10 follow-up studies involving 889 patients (SMD = −0.04, 95% CI: −0.17 to 0.09, I2 = 1%, p = 0.53).
Ensari et al. (49)	The weighted mean ES was small, but statistically significant (g = 0.36, SE = 0.09, 95% CI: 0.18 to 0.54, Z = 3.92, p < 0.001), indicating the exercise training resulted in an improvement in depressive symptoms compared to control.
Fong et al. (27)	Measured by the Beck depression inventory, physical activity was associated with reduced depression (−4.1, 95% CI: −6.5 to −1.8; p < 0.01) in survivors of mixed types of cancer. Four other studies in the sample used the HADS (two) or POMS (two). The results for HADS were (−0.5, 95%CI: −2.8 to 1.7, p = 0.64 and for POMS −7.5, 95 CI:−16.0 to 1.0, p = 0.09).
Furmaniak et al. (20)	Exercise may lead to little or no improvement in depression (SMD = −0.15, 95% CI: −0.30 to 0.01, test for overall effect: Z = 1.86; p = 0.062). Heterogeneity: Tau2 = 0.0; Chi2 = 3.73, df = 5; p = 0.59; I2 = 0.0%.
Gomes Neto et al. (39)	Exercise lead to a reduction in depression symptoms (−7.32; 95% CI: −9.31 to −5.33). Test for overall effect size Z = 7.21, p < 0.00001.
Graven et al. (53)	When the data from the two studies was pooled; (SMD = −2.03, 95% CI: −3.22 to −0.85) immediately after the intervention phase (note, different time points were used in the two studies for the follow-up assessment).
Herring et al. (16)	Exercise training significantly reduced depressive symptoms by a heterogeneous mean effect of 0.55 (95% CI: 0.31 to 0.78, p < 0.001). A significant improvement in fatigue (β = 0.37, Z = 2.21, p ≤ 0.03) accounted for significant variation in the overall effect of exercise on depressive symptoms.
Langhorst et al. (41)	Meditative movement therapies improved depressive symptoms (SMD = −0.49, 95% CI: −0.76 to −0.22, p = 0.0004 Heterogeneity I2 = 27%; Tau2 = 0.03). 2 studies evaluated the follow-up (N = 132), with no significant improvement of depressive symptoms. In subgroup analyses, only Yoga yielded significant effects on depression at final treatment.
Liang et al. (43)	A statistically significant difference was observed (MD = −2.31, 95% CI: −3.33 to −1.30, p = 0.001), which indicated that home-based exercise interventionsreduced the depression scores, compared to the control groups. No significant heterogeneity between home-based exercise groups and control groups (p = 0.24, I2 = 30%).
Lin et al. (29)	Improvement of depressive symptoms (−0.95, 95% CI: −1.55 to −0.36, test for overall effect: Z = 3.15, p = 0.002). Heterogeneity: τ2 = 0.63, χ2 = 66.81, df = 7, p < 0.00001); I2 = 90%.
Lin et al. (28)	No significant effects were found for depression and health-related quality of life. (No SMD or CI provided) Liu et al. (35)	The Thai Chi group had a significantly lower level of depression (SMD 9.42, 95% CI: 13.59 to 5.26, p < 0.001, I2 = 81%, N = 168) compared with the non-active control groups.
Newby et al. (31)	Exercise interventions significantly reduced depressive symptoms (Point estimate −0.961, SE = 0.319, CI 95% −1.585 to −0.337, Z = −3.017, p = 0.003).
O'Brien et al. (47)	One meta-analysis was performed and demonstrated a significant improvement in the depression-dejection subscale of the Profile of Mood States Scale (POMS) by a reduction of 7.68 points for participants in the aerobic exercise intervention group compared with the non-exercising control group (95% CI: −13.47 to −1.90, p = 0.009, I2 = 94%, p < 0.0001).
O'Dwyer et al. (44)	A meta-analysis including three studies found significantly lower depression scores in the exercise groups compared to controls (SMD = −0.40 SD; 95% CI: −0.71 to −0.09, test for overall effect size Z = 2.54, p = 0.01).
Patsou et al. (21)	Reduction in depressive symptoms showed a small to moderate effect of depressive symptoms in favor of the exercise g = −0.38 (95% CI −0.89 to 0.13, p = 0.14). With regard to the type of the exercise intervention, aerobic interventions yielded a large and significant effect on depression at the last follow-up (3–6 months) measurement compared with the control groups (g = −1.23, 95% CI: −1.97 to −0.49, p = 0.001).
Perry et al. (54)	This represents a statistically significant, positive small to medium overall effect size of physical exercise to reduce depressive symptoms in people following TBI (SMC = 0.48, 95% CI = 0.16 to 0.81). Tests of heterogeneity were significant (p < 0.01), confirming that heterogeneity was present amongst the studies included in the analysis.
Samartzis et al. (36)	Interventions using exercise training appeared more effective compared to usual care (SMD = 0.391, 95% CI: 0.213 to 0.569). There was a trend for SSRI superiority compared to exercise training for improving depression (Q = 3.257, df = 1, p = 0.071).
Singh et al. (22)	Large effect in favor of exercise SMD = 0.66, 95%, CI: 0.52 to 0.80, p < 0.01, I2 = 90%; high heterogeneity. Intervention duration had an effect on depression (χ2 = 7.93, df = 1, p < 0.01), with interventions lasting longer than 12 weeks producing a large effect (SMD = 0 .84, 95% CI: 0.65 to 1.03, p < 0.01) and interventions lasting 12 weeks or less having a moderate effect (SMD = 0.44, 95% CI: 0.23 to 0.65, p < 0.01).
Song et al. (51)	A fixed-effect model indicated that TCQ significantly reduced depression scores compared to control groups, with an overall medium effect size (g = −0.457, 95% CI: −0.795 to −0.118, p = 0.008). Q- value (p = 0.739) and I2 (0%) indicated limited heterogeneity.
Song et al. (40)	Exercise training was able to reduce depression in HD patients (SMD = -0.95, 95% CI: −1.18 to −0.73; Z = 8.33, p < 0.00001).
Sosa-Reina et al. (42)	There is strong evidence from intention-to-treat and per protocol analysis that exercise reduces symptoms of depression (−0.40, 95% CI: −0.55 to −0.24; p < 0.001). Values of Cohen's g suggested that exercise had a small effect on symptoms of depression.
Tu et al. (37)	Strong evidence of a decrease in the symptoms of depression with exercise (SMD −0.38, 95% CI: −0.55 to −0.21, p < 0.00001) in 3–6 months follow-up.
Vashistha et al. (32)	The pooled data did not reveal a significant improvement in depression (−3.02, 95%CI: −7.83 to 1.79, test for overall effect: Z = 1.23, p = 0.22). Heterogeneity: Tau2 = 9.80; Chi2 = 4.64, df = 1, p = 0.03; I2 = 78%.
Wang et al. (34)	The HAMD scores of patients performing TCEs improved (MD −3.97, 95% CI: −5.05 to −2.89, p < 0.001; I2 = 0, p = 0.91) compared with those of patients in the control group, based on a random-effects model. The POMS depression scale scores of the patients performing TCEs significantly improved (MD −3.02, 95% CI: −3.50 to −2.53, p < 0.001; I2 = 0%, p = 0.76) compared with those of patients in the control group, based on a random-effects model.
Wayne et al. (30)	The overall effect size based on a random-effects model favors TCQ on depression in cancer patients (g = −0.27, 95% CI: −0.44 to −0.11, p = 0.001). A subgroup meta-analysis limited to five RCTs using an active control group showed a statistically non-significant trend toward TCQ improving depression (g = −0.22, 95% CI:−0.47 to 0.02, p = 0.080). A subgroup meta-analysis limited to the three RCTs with a no-treatment control group showed a statistically positive effect of TCQ.
Ying et al. (33)	No obvious difference in mitigating depression (SMD = −0.18, 95% CI: −0.54 to 0.17, p = 0.31).
Zhu et al. (23)	Exercise intervention reduced depression, (SMD = −2.08, 95% CI: −3.36 to −0.80, p = 0.001, I2 = 2%, p = 0.41).
Zhou et al. (45)	Based on the data for depression, there were no significant differences in these parameters between the exercise and control groups (SMD = −0.15, 95% CI: −0.51 to 0.22, p = 0.28, p for heterogeneity = 0.57, I2 = 0%).