Exercising alone in men and group exercise in women are cross-sectionally associated with positive mental health among older Japanese

BACKGROUND: Although previous studies have reported that exercising with others has a more beneficial effect on the health of older adults than exercising alone, gender differences in the association between exercise patterns and health are unknown. We investigated the cross-sectional association between exercise patterns and physical and mental health by gender. METHODS: We analyzed 4,211 men and 4,944 women aged ≥65 years without disabilities. Physical and mental health was assessed using the SF-8 Health Survey. Exercise patterns were measured based on three types: participation in exercise groups, non-group-based exercise with others, and exercising alone. Each exercise pattern was classified into five groups according to frequency and continuity: maintained frequent (i.e., daily or weekly), increase in frequency, maintained moderate (i.e., monthly or yearly), decrease in frequency, and continuing non-exercise. Modified Poisson regression with generalized estimating equations was used to estimate adjusted prevalence ratio (APR) for poor physical/mental health. Covariates included age, marital status, education, economic status, body mass index, chronic medical conditions, smoking, dietary variety, cognition, working status, and social participation. RESULTS: For physical health, among both genders, all exercise patterns had a significant dose-response relationship between lower levels of exercise frequency and continuity and a higher prevalence of poor physical health, after adjustment for covariates and mutual adjustment for other exercise patterns (all P for trend <0.036). For mental health, among men, only exercising alone had a significant dose-response relationship (P for trend <0.001) [APR: 1.22 (95% CI: 1.07-1.39) for continuing non-exercise, 1.32 (1.11-1.57) for decrease in frequency, compared to maintained frequent]. Among women, the cross-sectional association with mental health was limited to participation in exercise groups (P for trend: 0.006) [APR: 1.24 (95% CI: 1.09-1.40) for continuing non-exercise, 1.26 (1.09-1.47) for decrease in frequency, and 1.37 (1.13-1.66) for maintained moderate, compared to maintained frequent]. CONCLUSIONS: All exercise patterns were cross-sectionally associated with physical health across genders. However, mental health outcomes varied: Cross-sectional associations with mental health were observed for men when exercising alone, and for women when exercising in groups. These results highlight the importance of gender-sensitive public health strategies, such as tailored community exercise programs. Due to the cross-sectional design, causality cannot be determined. However, the results do suggest that future longitudinal research is needed.