Associations of particulate matter and meteorological factors with respiratory hospitalizations in Rio de Janeiro: A distributed lag nonlinear model study

Abstract

Understanding the combined effects of air pollution and meteorological conditions on respiratory health is essential for urban policy, particularly in tropical megacities. This study investigated the nonlinear and delayed effects of air pollution and weather variables on respiratory hospitalizations in Rio de Janeiro, Brazil, from 2015 to 2024. Daily hospital admissions for respiratory diseases (ICD-10: J00–J99) were analyzed in relation to mean temperature, relative humidity, precipitation, and PM2.5 and PM10 concentrations. Distributed lag nonlinear models (DLNM) with a negative binomial distribution were applied over a 14-day lag period. Particulate matter showed significant acute cumulative effects, with relative risks (RR) of 1.264 (95% CI: 1.12–1.43) for PM2.5 and 1.304 (95% CI: 1.16–1.47) for PM10 at the 90th percentile compared with the median. Temperature exhibited a nonlinear U-shaped association, with increased cumulative risk at high levels (RR = 1.196; 95% CI: 1.08–1.33). In contrast, higher relative humidity (RR = 0.837; 95% CI: 0.79–0.89) and precipitation (RR = 0.716; 95% CI: 0.63–0.82) were associated with reduced hospitalization risk. These findings indicate that particulate matter is a major driver of respiratory morbidity in this tropical urban setting, supporting stricter air quality standards and integrated environmental health surveillance.