Association between physical activity and suicidal ideation: Findings from the 1982 Pelotas (Brazil) birth cohort and an updated systematic review and meta-analysis.

内容摘要

Evidence on the impact of physical activity on suicidal ideation remains inconsistent. We examined this association in the 1982 Pelotas Birth Cohort (Brazil) and updated a systematic review and meta-analysis. In 1982, all births in Pelotas (southern Brazil) were identified and prospectively followed up. Leisure-time physical activity at age 23 was measured using the International Physical Activity Questionnaire, and suicidal ideation at age 30 was assessed using the Mini International Neuropsychiatric Interview. Odds ratios (OR) were estimated using logistic regression. For the systematic review, PubMed, CINAHL, and PsycArticles were searched for studies that assessed the relationship between physical activity as exposure and suicidal ideation as outcome. We estimated pooled OR using a random-effects model of studies reporting OR and prevalence ratios (PR). In the cohort (n = 3268), being sufficiently active was associated with lower odds of suicidal ideation in crude analysis (OR = 0.56, 95% CI 0.41-0.78) but the association vanished after adjustment for confounders (OR = 1.01, 95% CI 0.68-1.50). Seventy-two studies met inclusion criteria for the systematic review, of which 50 studies were meta-analyzed, yielding a pooled OR of 0.81 (95% CI 0.76-0.87; I2 = 94.9%). Associations were observed in cross-sectional studies (ORpooled = 0.82, 95% CI 0.77-0.87) but not in cohort studies (ORpooled = 1.00, 95% CI 0.80-1.25). Analysis conducted in the 1982 Pelotas Birth Cohort showed that controlling for confounding is crucial. Moreover, pooled estimates from prospective studies do not support a protective association, suggesting that physical activity may not be independently associated with suicidal ideation.