Factors associated with overweight among adolescents in Bangladesh: a comparative study of Bayesian and frequentist approaches

Bangladeshi adolescents are facing serious health challenges due to the growing burden of overnutrition. Researchers have increasingly focused on malnutrition among children; however, only a few studies have used robust statistical models to assess overweight among adolescents in Bangladesh. The objective of this study was to determine the prevalence of being overweight and identify associated sociodemographic, behavioral, and community-level factors among adolescents aged 15–19 years in Bangladesh. Data were extracted from the nationally representative 2019–20 Bangladesh Adolescent Health and Wellbeing Survey (BAHWS) and were analyzed using a sample of 8,876 adolescents. Using descriptive statistics, Chi-square tests, and logistic regression models under frequentist and Bayesian frameworks, we examined the associations between overweight and sociodemographic, behavioral, and community-level characteristics. Findings depict that the prevalence of being overweight among Bangladeshi adolescents was 10.65%. In both modeling frameworks, ever-married adolescents, those with higher educational attainment, those not currently working, adolescents from wealthier households, those who watched television frequently, and those living in urban areas were significantly associated with increased odds of being overweight. For instance, adolescents in the richest wealth quintile had approximately 2.7 times higher odds of being overweight than those in the poorest wealth quintile. Bayesian estimates were generally consistent with the frequentist estimates and, for several covariates, produced narrower credible intervals, suggesting more stable posterior estimates and lower uncertainty around the estimated associations. The results indicate a high prevalence of being overweight among adolescents in Bangladesh, highlighting it as an important public health concern with significant disparities by marital status, educational attainment, household wealth, employment status, and place of residence. Intervention planning should prioritize ever-married adolescents, those in higher educational grades, nonworking adolescents, those from wealthier households, and urban adolescents who are most vulnerable to being overweight.