Association of Early Sports Participation With Sedentary Behavior in Community-Dwelling Adults-The Role of Sociodemographic Factors in a Retrospective Epidemiological Study

Abstract

Background: The role of sociodemographic factors in the association between early sports participation (ESP) and sedentary behavior is unclear. We analyzed the association of ESP with sedentary behavior and identified the influence of sociodemographic factors in adulthood. Methods: A sample of 264 community-dwelling adults was randomly assessed (42.6 [17.0] y old, 57.6% women). Sociodemographic factors (age, gender, and socioeconomic status) and ESP (retrospectively) were assessed by questionnaire. Sedentary behavior was measured by accelerometer and self-reported for domains identification. The association between ESP and sedentary behavior domains and sociodemographic factors was analyzed by Poisson regression and presented in prevalence ratio (PR). Sociodemographic factors were separately included as covariates to identify their role in the main association analysis. Results: The ESP prevalence was 56.4% (n = 149). ESP participants included a higher proportion of men (59.7% vs 20.0%), had a lower age (36.8 [15.3] vs 50.8 [15.9]), and included a smaller proportion of individuals with low socioeconomic status (24.8% vs 43.5%) compared with those without ESP. The ESP was inversely associated with older age (PR = 0.58, P < .001 for middle aged; PR = 0.34, P < .001 for older participants), female gender (PR = 0.79, P < .001), low socioeconomic status (PR = 0.63, P = .036), and TV watching (PR = 0.67, P = .011). ESP was associated with driving (PR = 1.50, P = .028), office/paper work (PR = 1.63, P = .012), and using a cellphone (PR = 1.60, P = .009). The age was the main confounding factor of association between ESP and sedentary behavior, followed by socioeconomic status. No mediation role was identified. Conclusion: The ESP was associated with mentally active behavior domains and inversely associated with mentally passive sedentary behavior domains, but this association was majorly affected by sociodemographic factors, mainly by age.