Longitudinal Association Between Physical Exercise and Depressive Symptoms in Older Adults: The Prospective Explanatory Role of Loneliness and the Moderating Role of Cognitive Emotion Regulation

Objective: This study examined the prospective association between self-reported physical exercise and depressive symptoms among older adults, and further tested whether loneliness statistically accounted for this association and whether baseline cognitive emotion regulation strategies moderated the exercise-loneliness pathway. Methods: A two-wave prospective survey with a six-month interval was conducted among 980 community-dwelling older adults in Zhengzhou, China. Baseline data were collected in September 2024, and follow-up data were collected in March 2025. Physical exercise was assessed using the Physical Activity Rating Scale-3, depressive symptoms using the 10-item Center for Epidemiologic Studies Depression Scale, loneliness using the 8-item UCLA Loneliness Scale, and cognitive emotion regulation strategies using the short version of the Cognitive Emotion Regulation Questionnaire. Cross-lagged models were used to examine the residual prospective association between physical exercise and depressive symptoms. A two-wave prospective explanatory model was then tested to examine the role of follow-up loneliness, and moderated explanatory (half-longitudinal) analyses were conducted using baseline adaptive and maladaptive cognitive emotion regulation strategies as moderators. Results: Baseline physical exercise was significantly associated with lower depressive symptoms at follow-up after controlling for baseline depressive symptoms and covariates (β = -0.182, p < 0.001). In contrast, baseline depressive symptoms were not significantly associated with follow-up physical exercise (β = -0.016, p = 0.633). Baseline physical exercise was negatively associated with follow-up loneliness (β = -0.267, p < 0.001), and follow-up loneliness was positively associated with follow-up depressive symptoms (β = 0.324, p < 0.001). The indirect association through follow-up loneliness was significant (indirect effect = -0.087, 95% CI [-0.112, -0.065]). Baseline adaptive cognitive emotion regulation strategies strengthened the association between physical exercise and lower loneliness (interaction β = -0.076, p < 0.001), whereas baseline maladaptive strategies weakened this association (interaction β = 0.059, p = 0.004). Conclusions: Self-reported physical exercise was prospectively associated with fewer depressive symptoms among older adults six months later. This association was statistically accounted for, in part, by lower follow-up loneliness, and baseline cognitive emotion regulation strategies were associated with the strength of the exercise-loneliness association. Because this study used a two-wave observational design with the explanatory variable and outcome measured at the same follow-up wave, the findings should be interpreted as prospective associations rather than evidence of causal or temporal mediation.