Table 2.
Main results of multiple linear and logistic regressions analyzing the association of physical activity or exercise with symptoms of depression or anxiety in the included studies
| Author | Predictor | Depressive symptoms | Anxiety symptoms | ||||
|---|---|---|---|---|---|---|---|
| Beta (95% CI) | OR (95% CI) | Beta (95% CI) | OR (95% CI) | ||||
| Volume | |||||||
| Bauer et al. [11] | EX (minutes/week) | 0.00# (NR; NR) | 0.01# (NR; NR) | ||||
| Callow et al. [26] | PA (PASE score) | − 0.22*** (NR; NR) | − 0.02 (NR; NR) | ||||
| Chen et al. [28] | EX (hours/day) | 0.68* (0.47; 0.97) | 0.66* (0.45; 0.96) | ||||
| Deng et al. [30] | EX (> 60 min/day; Ref: < 60 min/day)) | − 0.08*** (NR;NR) | − 0.05 (NR; NR) | ||||
| Cheval et al. [29] | PA (minutes/day) | NR# (NR; NR) | NR# (NR; NR) | ||||
| Jacob et al. [35] | EX (minutes/day) | 0.88° (0.8; 0.97) | 0.85° (0.79; 0.97) | ||||
| Khan et al. [36] | EX (any amount; | − 2.1* (− 4.02; − 0.17)a | − 0.55 (− 1.92; 0.82)a | ||||
| Ref.: No EX) | |||||||
| Moreira et al. [38] | EX (hours) | − 1.17° (NR; NR)a | − 0.81° (NR; NR) | ||||
| Plomecka et al. [40] | EX (> 15 min/day; | − 0.13*** (NR; NR) | NR# (NR; NR) | ||||
| Ref.: ≤ 15 min/day, < 60 min/day) | |||||||
| EX (≥ 60 min/day; | − 0.15*** (− 0.18; − 0.12) | NR# (NR; NR) | |||||
| Ref.: ≤ 15 min/day) | |||||||
| Schuch et al. [8] | PA (minutes/day; per 10 min increase) | − 0.03 (− 0.1; 0.03) | − 0.05 (− 0.13; 0.02) | ||||
| Frequency | |||||||
| Deng et al. [30] | EX (1 to 2 times/week; Ref: < 1x/week) | − 0.11*** (NR; NR) | − 0.09** (NR; NR) | ||||
| EX (> 2 times/week; Ref: < 1x/week) | − 0.15*** (NR; NR) | -0.12** (NR; NR) | |||||
| EX (every day; Ref: < 1x/week) | − 0.11*** (NR; NR) | -0.09* (NR; NR) | |||||
| Filgueiras and Stultz-Kolehmainen [32] | EX (frequency/week) | − 2.68** (NR; NR)a | − 1.64*** (NR; NR)a | ||||
| Fullana et al. [34] | EX (Unclear) | 0.93 (NR; NR) | 0.95 (NR; NR) | ||||
| Lebel et al. [37] | EX (Godin Shephard Leisure-Time Exercise Score) | − 0.01* (NR; NR) | 0.99 (0.99; 0.99) | − 0.01** (NR; NR) | 0.99 (0.99; 1.0) | ||
| Planchuelo-Gómez et al. [39] | EX (1–2 times/week; Ref.: No EX) | − 0.17 (NR; NR)a | |||||
| EX (3–5 times/week; Ref.: No EX) | − 0.85* (NR; NR)a | ||||||
| EX (6–7 times/week; Ref.: No EX) | − 1.29*** (NR; NR)a | ||||||
|
Filgueiras and Stultz-Kolehmainen [31] |
EX (frequency/week) | NR# (NR; NR) | NR# (NR; NR) | ||||
| Intensity | |||||||
| Callow et al. [26] | Light PA (PASE score) | 0.12** (NR; NR) | |||||
| Moderate PA (PASE score) | − 0.01 (NR; NR) | ||||||
| Vigorous PA (PASE score) | 0.09* (NR; NR) | ||||||
| Schuch et al., 2020 [8] | Vigorous PA (minutes/day) | − 0.19* (− 0.34; − 0.04) | 0.6** (0.44; 0.83) | − 0.22* (− 0.4; − 0.03) | 0.71** (0.52; 0.96) | ||
| Moderate PA (minutes/day) | 0.00 (− 0.09; 0.09) | 0.77 (0.57; 1.02) | − 0.03 (− 0.14; 0.08) | 0.75 (0.58; 1) | |||
| Regular/guideline conforming | |||||||
| Chen et al. [27] | EX (regular; | 0.37 (NR; NR)b | 0.43 (NR; NR)b | ||||
| Ref.: not regular) | |||||||
| Deng et al. [30] | EX (regular; | − 0.2*** (NR; NR) | − 0.14*** (NR; NR) | ||||
| Ref.: not regular) | |||||||
| Fu et al. [33] | EX (not regular; | 1.71*** (1.28; 2.29) | 1.45* (1.08; 1.93) | ||||
| Schuch et al. [8] | PA (≥ 30 min/day; Ref.: < 30 min/day) | 0.72* (0.54; 0.96) | 0.72* (0.54; 0.96) | ||||
| Change | |||||||
| Bauer et al. [11] | EX (less; equal; more)c | − 0.08*** (NR; NR) | − 0.05*** (NR; NR) | ||||
| Deng et al. [30] | EX (no change; Ref.: large change) | − 0.27*** (NR; NR) | − 0.21*** (NR; NR) | ||||
| EX (little change; Ref.: large change) | − 0.22*** (NR; NR) | − 0.17*** (NR; NR) | |||||
| Filgueiras et al. [31] | EX (none, increase, decrease) | NR# (NR; NR) | NR# (NR; NR) | ||||
| Meyer et al. [7] | PA (increased; Ref.: maintained high) | − 0.01 (− 0.05; 0.02) | 0.00 (− 0.03; 0.04) | ||||
| PA (decreased; Ref.: maintained high) | 0.09*** (0.05; 0.13) | 0.03 (− 0.01; 0.07) | |||||
| PA (maintained low; Ref.: maintained high) | 0.04 (0.00; 0.07) | 0.02 (− 0.02; 0.05) | |||||
| Stanton et al. [41] | PA (negative change; Ref.: no change/positive change) | 1.08*** (1.06; 1.11) | 1.09*** (1.05; 1.13) | ||||
| Zhang et al. [42] | PA (per 100 MET increase) | − 0.04* (− 0.08; 0) | − 0.03 (− 0.07; 0) | ||||
| Zheng et al. [43] | PA (decrease vs. no change/increase) | 2.07** (NR; NR) | 1.24 (NR; NR) | ||||
AOR adjusted odd's ratio, EX exercise, MET metabolic equivalent of tasks, NR not reported, OR odd's ratio, PA physical activity, PASE Physical activity Scale for the Elderly, Ref. reference category
*p < 0.05; **p < 0.01; ***p < 0.001; °significant association, p value not reported; #no significant association, p value not reported
aUnstandardized regression coefficient
bOdd's ratio calculated from case counts
cPost hoc analysis revealed that a decrease in exercise was significantly associated with less depression compared to stable exercise and increase. No other comparison reached significance