Obesity, physical activity, and gene–environment interactions: a natural experiment framework

Abstract

Background: Obesity results from the interaction of polygenic susceptibility and environmental factors. Given this complex etiology, physical activity (PA) remains a cornerstone of cost-effective intervention strategies. This longitudinal natural experiment investigated how PA modifies the effects of genetic predisposition on obesity in Chinese youth. Methods: We conducted a 4-year natural experiment leveraging curriculum-driven PA disparities in a specialized arts school (n = 591), creating distinct high-PA (HPA) and low-PA (LPA) exposure groups. Weighted genetic risk scores (WGRSs) were calculated from 13 Asian-derived obesity-related single-nucleotide polymorphisms. Annual anthropometric, metabolic, and lifestyle data were analyzed using generalized linear mixed models to assess gene–PA interactions on obesity. Results: The WGRS predicted baseline obesity measures, with each unit increase associated with a 0.21-kg/m² higher BMI. Over the natural experiment period, BMI increases in the HPA group were smaller than in the LPA group. After adjusting for age, sex, ethnicity, and dietary factors, significant WGRS–PA interactions were observed for BMI trajectories. Participants with higher genetic risk for obesity experienced greater BMI and weight reduction benefits from sustained long-term PA. Conclusions: In summary, the present study identified a significant interaction effect between PA levels and WGRS in modifying BMI trajectories. Genetic susceptibility significantly modifies the protective effects of long-term PA on BMI progression in this cohort of Chinese youth.