Impact of wildfire-related fine particulate matter on tuberculosis notifications in Brazil: a nationwide panel study, 2003–2023
Preprint
Abstract
Background: Wildfire activity and smoke exposure are increasing worldwide because of climate and land-use change. Although fine particulate matter (PM 2.5 ) may impair pulmonary immune defences against tuberculosis (TB), population-level evidence remains limited. We estimated the effect of wildfire-related PM 2.5 exposure on TB notification rates in Brazil. Methods: We conducted a nationwide panel study linking municipality-level monthly TB notifications from Brazil’s SINAN system with wildfire-related PM 2.5 estimates from GEOS-Chem simulations across 5,545 municipalities (2003-2023). We estimated the impact of high-exposure days (PM 2.5 >25µg/m³) on monthly TB notifications using Poisson regression with fixed effects for municipalities, state-by-year, and state-by-month, controlling for time-invariant differences, secular trends, and seasonality. Distributed lag effects were estimated over 1-24 months before notification. Models accounted for meteorological conditions, GeneXpert diagnostic coverage, and spatial correlation using Conley standard errors. We computed attributable fractions among exposed municipality-months (AFE). Sensitivity analyses evaluated alternative PM 2.5 thresholds (15, 35µg/m³), co-pollutants, and agricultural expansion. Findings: From Jan 1, 2003 to Dec 1, 2023, 1,758,982 TB cases were reported. Of these, 353,319 (20.1%) had at least one high-exposure day (PM 2.5 >25µg/m³) 1-24 months before notification. An additional 14 high-exposure days over the 24-month lag period was associated with an average monthly increase of 2.9% [95% CI: 0.9-4.9%] in TB notification rates. Effects peaked at 13 months (IQR: 11-14) prior to notification. Results showed a dose–response relationship across PM 2.5 thresholds and were robust to controlling for NO2, O3 and agricultural expansion. Overall, wildfire-related PM 2.5 exposure accounted for 2.1% [0.7-3.5%] of TB notifications in exposed municipality-months, corresponding to 7,802 [2,612-12,544] attributable cases. The AFE reached 10.7% [7.1-14.0%] in Pantanal and 7.3% [6.1-8.5%] in Amazônia, areas most impacted by wildfires. Interpretation: Wildfire-related PM 2.5 exposure may represent an increasingly important and modifiable risk factor for TB. As wildfire activity increases across many regions of the world, these findings highlight the need for integrating air quality into climate adaptation and TB control strategies. Funding: National Institutes of Health K01AI173385; University of Utah 3i InitiativeKeywords
Funding Information
- University of Utah 3i Initiative
This publication has 38 references indexed in Scilit:
- Global burden and strength of evidence for 88 risk factors in 204 countries and 811 subnational locations, 1990–2021: a systematic analysis for the Global Burden of Disease Study 2021The Lancet, 2024
- Zonal statistics datasets of climate indicators for Brazilian municipalitiesEnvironmental Data Science, 2024
- Global estimates of daily ambient fine particulate matter concentrations and unequal spatiotemporal distribution of population exposure: a machine learning modelling studyThe Lancet Planetary Health, 2023
- Global and Regional Trends and Drivers of Fire Under Climate ChangeReviews of Geophysics, 2022
- PM2.5 mediated alterations in the in vitro human granuloma and its effect on reactivation of mycobacteriaEnvironmental Science and Pollution Research, 2021
- Risk and burden of hospital admissions associated with wildfire-related PM2·5 in Brazil, 2000–15: a nationwide time-series studyThe Lancet Planetary Health, 2021
- Air pollution and its effects on the immune systemFree Radical Biology and Medicine, 2020
- A systematic literature review and critical appraisal of epidemiological studies on outdoor air pollution and tuberculosis outcomesEnvironmental Research, 2019
- Effect of inhaled air pollution particulate matter in alveolar macrophages on local pro-inflammatory cytokine and peripheral interferon γ production in response to Mycobacterium tuberculosisThe Lancet Global Health, 2018
- GMM estimation with cross sectional dependenceJournal of Econometrics, 1999