Asthma exacerbations, meteorological variables, and multipollutant concentrations: Uncovering short-term patterns through Association Rule Mining

A. Betancourt, C. Martínez de Ibarreta Zorita, L.A. Calvo, E. Wirth, S. Budría Rodríguez, D. Antolín-Amérigo, M. Colque-Bayona, I. de la Rocha-Ortiz, M.D. Escalante-Carrero, J. Domínguez-Ortega

Environmental Pollution Vol. 384, pp. 127003

Summary:

Asthma is a major global health issue, affecting millions of people and contributing significantly to disability-adjusted life years. This study investigates the short-term impacts of outdoor multi-pollutant mixture, pollen, and climate conditions on asthma exacerbations by identifying specific interval concentrations of pollutants that may increase asthma risk. Using a time-stratified case-crossover design combined with Association Rule Mining, a data mining technique, we analyzed data from adult patients who visited the emergency departments in 2019 and 2022 of La Paz University Hospital and Ramón y Cajal University Hospital in Madrid, Spain, and the data on PM₁₀, PM_2.5, O₃, NO₂, NO and SO₂ from 52 monitoring stations.

Our findings revealed that specific pollutant concentrations, such as ozone (O₃; 65.3–100 μg/m³) and PM₁₀ (45–50 μg/m³), significantly increased asthma exacerbation risk. These exposures occurred simultaneously or with a time lag of up to three days. Notably, interactions between pollutants, such as O₃ and PM_2.5 or PM_2.5 and pollen, were strongly associated with asthma exacerbations, even when individual pollutant concentrations remained below WHO safety guidelines. Relative risks for these combinations ranged from 1.12 to 1.23, suggesting a cumulative effect.

These results underscore the importance of considering pollutant interactions and their time-lagged effects in understanding asthma risks, rather than evaluating pollutants in isolation. Our findings provide mechanistic evidence that future air quality warning systems should account for multi-pollutant exposure patterns, particularly to protect vulnerable populations during critical exposure windows.

Spanish layman's summary:

Este estudio analiza el efecto conjunto de contaminantes atmosféricos y variables meteorológicas sobre las exacerbaciones asmáticas en la ciudad de Madrid. Se emplea la técnica de Association Rule Mining para identificar patrones y combinaciones de exposición a corto plazo que permiten anticipar picos de riesgo, aportando evidencia útil para el diseño de sistemas de alerta temprana y estrategias de salud pública.

English layman's summary:

This study investigates the combined effects of air pollutant mixtures and meteorological variables on asthma exacerbations in the city of Madrid. Association Rule Mining is used to identify short-term exposure patterns and combinations that help anticipate risk peaks, providing evidence to support early warning systems and public health decision-making.

Keywords: Health effect; Asthma exacerbation; Multi-pollutant mixture; Unsupervised learning; Pollution

JCR-JIF Impact Factor and WoS quartile: 7,200 - Q1 (2025)

DOI reference: https://doi.org/10.1016/j.envpol.2025.127003

Published on paper: November 2025.

Published on-line: August 2025.

Citation:
A. Betancourt, C. Martínez de Ibarreta Zorita, L.A. Calvo, E. Wirth, S. Budría Rodríguez, D. Antolín-Amérigo, M. Colque-Bayona, I. de la Rocha-Ortiz, M.D. Escalante-Carrero, J. Domínguez-Ortega, "Asthma exacerbations, meteorological variables, and multipollutant concentrations: Uncovering short-term patterns through Association Rule Mining", Environmental Pollution, Vol. 384, pp. 127003, November 2025. [Online: August 2025] doi: 10.1016/j.envpol.2025.127003