Influence of climate and heatwaves on dengue transmission in Sao Paulo and Natal, Brazil

PLOS One | https://doi.org/10.1371/journal.pone.0334838 October 24, 2025 5 / 13 values, selecting the model with the lowest AIC as the best fit. Results from the binomial regression are expressed as incidence rate ratios (IRRs), calculated as the exponentiated beta coefficients (β). The IRR provides a relative measure for comparing incidences between two events [32]. Results During the 417 epidemiological weeks of the study, a total of 149,468 dengue cases were reported in São Paulo city, while Natal recorded 80,999 cases. The years 2020 and 2021 were excluded from the analysis to mitigate potential under- reporting caused by the COVID-19 pandemic. Moreover, a substantial amount of data was missing from conventional meteorological stations during this time frame. São Paulo averaged 335 dengue infections per week, with a high standard deviation of 896, indicating substantial weekly variability. The minimum number of infections was 0, and the maximum was 9,319 cases in a week, highlighting significant peaks of dengue during the period (Fig 2). Maximum and minimum tem- peratures remained fairly consistent throughout the weeks, exhibiting only moderate fluctuations. However, the thermal amplitude indicated a notable variation in daily temperatures during many weeks. Total precipitation showed great vari- ability, with many dry weeks and some weeks with high rainfall volumes, which may be related to seasonal variability and its impact on the spread of dengue, considering the role of rainfall in the proliferation of the Aedes aegypti mosquito, the vector of the disease.

In Natal, the weekly average of dengue infections was 189, with a standard deviation of 250, showcasing substan- tial variability in case numbers over time, though not as extreme as in Sao Paulo. Weekly maximum and minimum Fig 2. Time trend for weekly dengue infections, maximum and minimum temperature, thermal amplitude and rainfall in São Paulo/SP, Brazil, 2014–2023. *The years of 2020 and 2021 were excluded. https://doi.org/10.1371/journal.pone.0334838.g002