Associations between COVID-19 incidence and environmental stressors in Brazil: a nationwide study from 2020 to 2022

Abstract

Background: Environmental stressors such as temperature (Temp), relative humidity (RHumid), and fine particulate matter (PM2.5) may influence the incidence of COVID-19. While many studies have examined these associations in Europe and Asia, research in Brazil—a country with diverse climatic zones and a high burden of COVID-19—remains limited. Objective: This study aimed to assess the associations between environmental stressors and COVID-19 incidence at the municipality level across Brazil over a three-year period, differentiating between climate zones and pre-Omicron/Omicron periods. Methods: We utilized a generalized additive model (GAM) framework to analyze monthly COVID-19 incidence while adjusting for population size, spatial structure, and temporal trends. Distributed lag nonlinear models (DLNM) were used to evaluate lagged exposure-response associations. Separate models were fitted for five climate zones to assess regional variations. Results: In the overall analysis, Temp was positively associated with COVID-19 incidence [relative risk RR 2.47, 95% confidence interval (2.04, 2.91)], while PM2.5 [RR 1.03 (0.95, 1.11)] and RHumid [RR 1.02 (0.91, 1.13)] did not demonstrate clear effects. Climate zone-specific analyses revealed diverse effects: Temp had a positive association with COVID-19 in Temperado (TE) [RR 17.9 (15.26, 22.19)] and Tropical Brazil Central [RR 1.87 (1.57, 2.10)], but a negative association in Tropical Nordeste Oriental [RR 0.008 (0.004, 0.012)] and Tropical Zona Equatorial (TZE) [RR 0.12 (0.08, 0.15)] climate zones. RHumid showed varying positive and negative associations depending on the climate zone, while high levels of PM2.5 are positive associated with COVID-19 incidence in zones TE [RR 2.10 (1.93, 2.28)] and TZE [RR 1.87 (1.54, 2.31)]. DLNM results revealed parabolic lag response curves, with extreme values of Temp and RHumid raising risks in certain zones. Significance: Our study provides a comprehensive, long-term analysis of environmental stressors and COVID-19 incidence across diverse climate zones in Brazil. The results reveal considerably spatial and temporal variations in how Temp, RHumid, and PM2.5 influence COVID-19 incidence. These findings emphasize the importance of considering regional climatic conditions when assessing environmental risk factors for COVID-19. Understanding these associations can inform targeted public health interventions and preparedness strategies for future respiratory disease outbreaks.