Work

Unmasking spatial disparities and survival bias in the United States diabetes prevalence: A multilevel analysis of individual and ecological determinants using the 2024 behavioral risk factor surveillance system

Year: 2026

Type: article

Abstract: Objective: Diabetes prevalence varies across the United States, but individual versus state-level contributions remain unclear. We examined these predictors and survival bias. Methods: Using 2024 data from 406,782 adults in the United States Behavioral Risk Factor Surveillance System, we fitted multilevel logistic models for individual and state-level covariates. Age-stratified analyses tested survival bias. Results: Individual obesity was the strongest predictor (OR = 2.33; 95% CI: 2.28, 2.38). Asian adults exhibited the highest risk (OR = 2.05; 95% CI: 1.91, 2.20), followed by Black (OR = 1.76) and Hispanic adults (OR = 1.54). State-level obesity independently predicted diabetes (OR = 1.38 per 10-point increase; 95% CI: 1.17, 1.63), explaining 57% of between-state variance. Individual obesity effects attenuated in higher-obesity states. Inverse associations for smoking (OR = 0.89; 95% CI: 0.86, 0.92) and binge drinking (OR = 0.62; 95% CI: 0.60, 0.65) were null in young adults but strengthened with age. Conclusions: State-level context independently elevates diabetes risk. Asian Americans face disproportionately high risk, supporting lower body mass index screening thresholds. Apparent protective effects of smoking and binge drinking reflect survival bias.

Source: Preventive Medicine Reports

DOI: 10.1016/j.pmedr.2026.103688