Social drivers of type 2 diabetes in Mexico: A spatial multilevel approach to urbanization and unhealthy food expenditure

Type 2 diabetes mellitus (T2DM) is growing worldwide, particularly in Latin American countries. The determinants of this pandemic are diverse and exhibit distinctive spatial variations. Moreover, there is often a mismatch between the scale of disease reporting units and the spatial resolution of disease determinants. This study examines the social drivers of T2DM mortality in Mexican municipalities for 2016 and 2022, focusing on variables related to urbanization and expenditure on unhealthy food. We applied general linear mixed-effects models with a spatial random component to account for state heterogeneity. We found that some urbanization-related variables were significantly associated with T2DM mortality in both years, such as urban land use area per 1000 inhabitants, access to basic housing services, and population employed in informal activities. The effect of unhealthy food access became significant only in 2022, coinciding with an increase in spatial dependence, suggesting a broader geographic diffusion of urban and dietary risk factors. The marginal and conditional R-squared values indicate high explanatory power in both models (marginal: 0.811 in 2016 and 0.808 in 2022; and conditional: 0.834 and 0.832, respectively), with most variance explained by the fixed effects. Notably, the mean mortality rate for T2DM increased by 26 % during the study period. This approach enhances our understanding of how structural and spatial factors interact to shape health outcomes, highlighting the need for targeted public health strategies in rapidly urbanizing regions and emphasizing the multifactorial impact of urban conditions and food environments on health, particularly in guiding dietary interventions to mitigate T2DM.