Physical activity and incident type 2 diabetes: the potential mediating role of sleep medication use in the population-based Azar Cohort

Abstract

Physical activity (PA) is associated with lower risk of type 2 diabetes (T2D), but the pathways are unclear. We examined the association between PA and incident T2D and evaluated the indirect association through sleep medication use in the Azar Cohort Study. This cohort study included 13,259 adults aged 35–70 years who were free of T2D at baseline. PA and sleep medication use were assessed at baseline. PA was expressed as metabolic equivalent task hours (MET-h/week). Modified Poisson regression with robust variance was used to estimate risk ratios (RRs). Mediation analysis used logistic and Poisson regression within the potential-outcomes framework. Compared with high or very high PA, moderate or lower PA was associated with greater risk of incident T2D in the unadjusted model (RR = 1.49, 95% CI:1.03–2.14), the DAG-adjusted model (RR = 1.31, 95% CI:1.01–1.70), and the expanded model (RR = 1.23, 95% CI:1.01–1.50). Moderate or lower PA was associated with sleep medication use (adjusted OR = 1.93, 95% CI:1.23–3.02). The indirect effect through sleep medication use was significant in the DAG-adjusted model (RR = 1.09, 95% CI:1.01–1.16) and expanded model (RR = 1.04, 95% CI:1.01–1.06), with mediated proportions of 37.4% and 20.8%, respectively. Continuous PA showed an inverse association with incident T2D, with no significant evidence of nonlinearity. Higher PA was associated with lower occurrence of incident T2D. Sleep medication use showed a modest indirect association. Because PA and sleep medication use were assessed at baseline, the mediation findings should be interpreted as statistical rather than temporally established causal mediation.