Abstract
BACKGROUND: Depression and anxiety are psychiatric disorders that significantly affect patients' well-being, morbidity, and mortality. Aerobic exercise (AE) and resistance exercise (RE) have been proposed as interventions to alleviate depressive and anxiety symptoms; however, their comparative effectiveness remains unclear. This study compared the associations of AE, RE, and their combination (total exercise; TE) with depression and anxiety in a large representative sample of the Korean population. METHODS: Data from the Korean National Health and Nutrition Examination Survey (KNHANES) were used to analyze 21,298 participants for depression (measured using the Patient Health Questionnaire-9) and 8,707 participants for anxiety (measured using the Generalized Anxiety Disorder-7). The participants were divided into four exercise groups: TE, AE, RE, and non-AE-RE. Propensity score based exact matching was applied using a 1:1:1:1 ratio across the four exercise groups to control for confounding factors such as age and sex. Multivariate linear and logistic regression analyses were conducted to evaluate the effects of each exercise type on depression and anxiety. RESULTS: After propensity score matching, the RE group had the lowest depression (β = -0.534; 95% confidence interval [CI], -0.765 to -0.303; P < 0.001) and anxiety (β = -0.459; 95% CI, -0.808 to -0.110; P = 0.010) scores compared with the non-AE-RE group. The TE group also showed favorable associations with depression (β = -0.473; 95% CI, -0.706 to -0.241; P < 0.001), although to a lesser extent than the RE group. AE was not significantly associated with either of the outcomes. Subgroup analyses indicated stronger associations of RE in women and older adults. CONCLUSION: RE was more strongly associated with lower depression and anxiety levels than AE, particularly in women and older adults. This study underscores the importance of incorporating resistance training into exercise prescriptions and public health interventions to enhance mental health outcomes.