Machine learning uncovers drivers of asthma prevalence across US communities

Published online April 29, 2026

Asthma affects millions of people in the United States, but some communities experience a much greater burden than others. Where people live, work, and grow up may influence asthma risk through factors such as air pollution, housing conditions, and social and economic challenges. Understanding how these factors work together may help improve asthma prevention and reduce health disparities.

In The Journal of Allergy and Clinical Immunology (JACI), Nguyen et al explored how social and environmental conditions are linked to asthma rates across communities in the United States. The investigators used information from the Centers for Disease Control and Prevention Environmental Justice Index, a national database containing measures of environmental exposure and community characteristics.

Using machine learning, a type of computer-based analysis that can identify patterns in large amounts of data, the researchers first grouped communities with similar asthma burdens, social and environmental conditions, and related health outcomes. They then developed a random forest machine learning model, a computer tool that learns from many patterns at once, to predict asthma prevalence and identify which social and environmental factors were most strongly linked to asthma burden across communities.

Nguyen et al identified nine distinct community patterns across the United States, including three clusters with the highest asthma burden. Communities with greater asthma burden often faced overlapping social and environmental challenges. These communities experienced greater social and economic challenges, including higher poverty, housing burden, and language barriers, together with higher exposure to air pollution and nearby environmental hazards. The random forest machine learning model showed strong performance in predicting asthma prevalence and identified social and economic conditions as among the strongest factors linked to asthma burden across communities.

This study suggests that asthma burden is influenced not only by individual or biological factors, but also by the social and environmental conditions in which people live. By combining environmental and community data with machine learning, Nguyen et al demonstrated a new approach to identifying communities at higher risk for asthma. These findings may help guide more targeted public health strategies and policies aimed at reducing asthma disparities and improving health equity.

The Journal of Allergy and Clinical Immunology (JACI) is an official scientific journal of the AAAAI, and is the most-cited journal in the field of allergy and clinical immunology.

Full Article and Graphical Abstract