Exercise Intervention for Late-Life Depression: A Meta-Analysis

Objective:To quantify the association between physical exercise intervention (PEI) and reduction in depressive symptoms in older adults. Data Sources: MEDLINE, PsycINFO, and EMBASE were searched from inception through December 2018 with no language restrictions using keywords related to exercise, depression, elderly adults, and randomized controlled trials. Study Selection: Randomized controlled trials comparing a sedentary control group, with no physically active intervention, to a supervised, moderate-to-vigorous PEI with participants aged >= 60 years and having a primary outcome of depressive symptoms were included. Data Extraction: Data on pre- and post-intervention scores on scales measuring depressive symptoms were extracted using a standard form. Random-effects models were used to pool standardized mean differences (Hedges g) in depressive symptoms across studies. Data Synthesis: Nine studies involving 1,308 participants were included; mean participant age was 82 years. Moderate-to-vigorous PEI was associated with a medium effect size of 0.64 (95% CI, 0.27 to 1.01; z =3.38; P<.001) in reducing depressive symptoms. However, there was considerable heterogeneity (T-2 =0.22, Q = 36.34, P< .0001; I-2 = 78.0%) in the effect of PEI across included studies. Age >80 years, Mini-Mental State Examination (MMSE) score <23, and no depressive symptoms at baseline contributed to heterogeneity. Fitness metrics and adherence to exercise were inconsistently reported, and 5 of 9 studies were deemed at high risk of bias. Conclusions: A moderate reduction in depressive symptoms was seen with PEI among older adults. Nevertheless, more work is needed to support PEI for late-life depression in adults over age 80 years or with MMSE scores <23 suggestive of cognitive decline.