Climate influences on hospitalization patterns in Mexico: Evidence from 30 million records

Abstract Climate change is expected to have wide-ranging effects on human health, yet the extent to which environmental factors drive health outcomes is poorly understood, particularly in tropical locations. Here, we leverage a large dataset of approximately 30 million individual-level hospitalizations from Mexico, linked with locally resolved climate data, to understand the seasonality of morbidity and the role of climate in driving these patterns. We first apply a Fourier transform to identify disease categories that exhibit significant seasonal signals. Next, we apply fixed effect regression models to identify climate drivers of these seasonal patterns for both broad disease categories specified by the International Classification of Diseases (ICD) and a comprehensive range of specific disease subcategories defined by the World Health Organization (WHO). We found that half of the ICD disease category hospitalizations had a significant seasonal signal. Among these, 89% exhibited a significant positive association with temperature, 33% exhibited a significant positive association with precipitation, and 11% exhibited a significant negative association with precipitation. Overall, we found that temperature is a significant driver of 26% of disease subcategories defined by the WHO. The disease areas most influenced by climate are infectious, cardiovascular, respiratory, injury, and maternal conditions. These findings highlight how precipitation and temperature drive seasonal hospitalization patterns for communicable diseases, non-communicable diseases, and injuries in tropical and temperate climates.