Replication data for: temperature and respiratory mortality in Brazil: daily municipality-level dataset (2010-2020)

Dataset Overview This dataset contains daily municipality-level data on ambient temperature and respiratory mortality for 667 Brazilian municipalities from January 1, 2010 to December 31, 2020, comprising 2,680,006 observations and 1,106,779 respiratory deaths. Data Sources Mortality: Brazilian Mortality Information System (SIM/DATASUS), Ministry of Health. Deaths coded as respiratory diseases (ICD-10: J00-J99). Temperature: National Institute of Meteorology (INMET), interpolated to municipality centroids using inverse distance weighting. Population: Brazilian Institute of Geography and Statistics (IBGE), annual estimates. Inclusion Criteria Population ≥ 50,000 inhabitants Data completeness ≥ 80% for temperature and mortality Average ≥ 1 respiratory death per week during the study period Geographic Coverage 667 municipalities across all five Brazilian macroregions: North (44), Northeast (142), Southeast (298), South (128), and Central-West (55). Main Variables Municipality code (IBGE 6-digit) Date of observation Daily respiratory deaths (ICD-10: J00-J99) Daily mean, minimum, and maximum temperature (°C) Relative humidity (%) Daily precipitation (mm) Annual population estimate Day of week and national holiday indicator Associated Publication This dataset supports the manuscript: "The impact of extreme temperatures on respiratory mortality in Brazil: assessing regional adaptations to different thermal environments" submitted to PLOS Climate (2026). Analytical Methods Data were analyzed using Distributed Lag Non-linear Models (DLNM) with quasi-Poisson regression at the municipality level, followed by random-effects meta-analysis to obtain pooled national and regional estimates of temperature-mortality associations. Ethics Statement This study used aggregated, anonymized secondary data publicly available through DATASUS and IBGE. No individual-level identifiable data were used. According to Brazilian National Health Council Resolution 510/2016, research using publicly available aggregated data is exempt from Research Ethics Committee approval.