Mortality and morbidity associated with winter weather conditions revealed by machine learning models
Environmental Research Communications · 01.11.2026 · 8 (11) · 115010
DOI: 10.1088/2515-7620/aebb07 ↗Kurz gefasst
Fachartikel in der Zeitschrift Environmental Research Communications, online erschienen am 10.11.2026. Verfasst von Jérémie Boudreault und 5 weiteren Autor:innen. Die Arbeit nennt 60 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Sciences and Engineering Research Council of Canada, Canadian Institutes of Health Research, Fonds de recherche du Québec and Gouvernement du Québec.
Abstract (Originalsprache)
Machine learning (ML) models are being increasingly used to predict the health impacts of environmental exposures. Although cold temperatures account for a larger share of the burden of disease than hot temperatures, the use of ML to assess how winter weather conditions affect mortality and morbidity remains limited. In this study, we compared five ML models to predict cases of mortality and morbidity associated with winter conditions during the coldest months of the year in five regions of the province of Quebec, Canada. ML models, including Least absolute shrinkage and selection operator (Lasso), Decision tree (DT), Random forest (RF), Light gradient boosting model (LGBM) and Multi-layer perceptron (MLP), were calibrated to predict all-cause mortality, cardiovascular and respiratory hospitalizations during November to March months, 2000–2019. Lagged weather variables (e.g. temperature, precipitation, humidity, snowfall, freezing rain) and air pollutants (O 3 , PM 2.5 , NO 2 , SO 2 and CO) at up to 21 days were considered, as well as calendar variables (day of week, holiday, month, year). Feature importance and SHapley Additive exPlanations (SHAP) were used to reveal the most important predictors. On the selected test winter years, RF and LGBM performed better than the other models across locations to predict cardiorespiratory hospitalizations. For mortality, while RF was slightly more performant in Montreal, Lasso and calendar-only models were highly competitive in the other locations. Lagged temperatures, precipitation and air pollutants were all important to predict the three health outcomes during winter, with some differences across outcomes. Calendar variables also played a major role, especially for cardiovascular hospitalizations. Public health authorities could benefit from ML models to improve their understanding of winter weather condition associated with different health outcomes. Further studies should compare more thoroughly the performance of ML and non-ML approaches, and extend this associative study to a causal framework.
Förderung
- Natural Sciences and Engineering Research Council of Canada
- Canadian Institutes of Health Research
- Fonds de recherche du Québec and Gouvernement du Québec
- Mitacs
- Ouranos
- Institut national de santé publique du Québec
Zitieren
APA
Boudreault, J., Halladja, A., Lamothe, F., Lavigne, É., Campagna, C., & Chebana, F. (2026). Mortality and morbidity associated with winter weather conditions revealed by machine learning models. Environmental Research Communications, 8(11), 115010. https://doi.org/10.1088/2515-7620/aebb07
Vancouver
Boudreault J, Halladja A, Lamothe F, Lavigne É, Campagna C, Chebana F. Mortality and morbidity associated with winter weather conditions revealed by machine learning models. Environmental Research Communications. 2026;8(11):115010. doi:10.1088/2515-7620/aebb07
BibTeX
@article{Boudreault2026,
author = {Boudreault, Jérémie and Halladja, Adam and Lamothe, Félix and Lavigne, Éric and Campagna, Céline and Chebana, Fateh},
title = {Mortality and morbidity associated with winter weather conditions revealed by machine learning models},
journal = {Environmental Research Communications},
year = {2026},
volume = {8},
number = {11},
pages = {115010},
doi = {10.1088/2515-7620/aebb07}
}
EUVIMED