A groundbreaking new study from Boston University School of Public Health (BUSPH) offers fresh insights into the elusive relationship between indoor environmental exposures and childhood asthma. By leveraging advanced computational modeling techniques and integrating electronic health records (EHR) with detailed geospatial housing data, researchers have uncovered robust predictors of allergen exposure and their impacts on lung function. This innovative approach may revolutionize how clinicians diagnose and manage asthma among vulnerable populations, particularly children living in under-resourced urban communities.
Asthma remains the most prevalent chronic pediatric illness in the United States, with a disproportionate burden borne by Black and Latino children. Despite widespread acknowledgment of indoor allergens—such as cockroach and rodent infestations, dust mite populations, and mold—as critical environmental triggers, clinicians often find it challenging to assess these exposures accurately without direct home assessments. The new BUSPH study circumvents this barrier by developing predictive models that estimate the likelihood of allergen presence based solely on residential address-linked data and existing health records, eliminating the need for invasive or costly in-home inspections.
Published in the prestigious journal Annals of Epidemiology, this study represents a technical tour de force, combining machine learning algorithms with vast quantities of clinical and housing information. The research team harnessed EHR data from Boston Medical Center, one of the largest safety-net hospitals serving low-income populations, to analyze lung function metrics against environmental risk factors inferred from neighborhood characteristics and parcel-level housing attributes. These variables included indicators of pest infestation probability tied to historical patterns of structural disinvestment.
.adsslot_j1QwiAREny{width:728px !important;height:90px !important;}
@media(max-width:1199px){ .adsslot_j1QwiAREny{width:468px !important;height:60px !important;}
}
@media(max-width:767px){ .adsslot_j1QwiAREny{width:320px !important;height:50px !important;}
}
ADVERTISEMENT
A key innovation in the study was the use of spatially explicit modeling to predict in-home allergen loads. By incorporating publicly available geospatial datasets detailing housing conditions, neighborhood socioeconomic factors, and historical redlining maps, the researchers were able to link systemic housing inequities with individual health outcomes. The findings revealed that children residing in homes with high predicted probabilities of cockroach and rodent presence exhibited notably poorer lung function compared to peers in less exposed environments. This strongly supports the hypothesis that environmental exposures directly contribute to respiratory compromise in asthma.
Lead author Dr. Patricia Fabian, associate professor of environmental health at BUSPH, emphasized the clinical potential of these predictions. “By inferring allergen exposures through existing medical and geospatial data, physicians can identify at-risk children without the logistics and expense of home visits,” she explained. This method enhances care by enabling targeted interventions—ranging from pest management strategies to refined medical therapy—that address the root cause of asthma exacerbations rather than solely treating symptoms.
The study also casts a stark light on entrenched health disparities fueled by structural racism. The majority of children studied were from Black communities living in historically segregated neighborhoods subject to redlining—a now outlawed discriminatory housing practice that has left enduring legacies of poor housing quality and concentrated poverty. Such environments provide fertile breeding grounds for pests, compounding asthma risk and severity. Black children in the U.S. experience asthma rates twice as high as their White counterparts and suffer asthma-related mortality nearly eight times greater, underscoring the urgent need to confront these social determinants.
Technically, the study builds on previous work by the same team, which initially developed and validated machine learning models capable of estimating the presence of indoor asthma triggers using EHR and geospatial data from over a thousand children. The current research advances this by integrating predicted exposure data with objective lung function tests—specifically spirometry readings captured during routine healthcare visits—to establish direct associations between environmental risk factors and respiratory health outcomes, providing a compelling proof of concept.
Dr. Matthew Bozigar from Oregon State University, a co-corresponding author, highlighted the significance of incorporating measurement error modeling and advanced statistical methods to bolster the robustness of their findings. “Our approach accounts for uncertainty inherent in estimating living conditions indirectly, and still demonstrates strong links between predicted allergen exposure and diminished lung function,” he stated. Such rigor is critical in translating computational predictions into actionable clinical insights.
Beyond the immediate scope of pediatric asthma, the researchers contend that their methodology holds far-reaching implications for public health surveillance and equity-driven interventions. Since EHRs are ubiquitous in modern healthcare systems worldwide, similar predictive models could be tailored to diverse populations and environmental hazards. The expanding availability of high-resolution satellite imagery and environmental datasets further enhances the capacity to couple health outcomes with social and ecological contexts on a global scale.
In practical terms, this technology could prompt healthcare systems to identify clusters of patients living in unsafe housing conditions and collaborate with public health authorities for targeted remediation efforts. It offers a pathway to addressing environmental injustices by uncovering hidden patterns of exposure that exacerbate chronic conditions, particularly in marginalized groups, thereby bridging gaps in preventive care and resource allocation.
The integration of data science, epidemiology, and environmental health embodied in this study exemplifies the power of interdisciplinary research to tackle complex medical and social challenges. As electronic health data grows richer and more accessible, the ability to derive nuanced insights into the interplay between place, environment, and health will likely transform personalized medicine and population health management.
Ultimately, these findings underscore the critical need to include environmental and social determinants in clinical decision-making frameworks. By moving beyond traditional biomedical models to incorporate contextual risk factors, healthcare providers can develop more effective, culturally sensitive, and equitable asthma management strategies. This shift could not only improve quality of life for millions of children but also reduce the disproportionate toll of asthma on communities shaped by historical inequities.
Boston University School of Public Health’s work is funded by the National Institutes of Health and other prominent agencies, reflecting the growing recognition of environmental justice as a cornerstone of public health. As the field advances, such innovative computational approaches could serve as templates for studying a wide array of environmentally linked diseases.
Researchers involved in the study also included experts in biostatistics, environmental health, pediatric pulmonology, and urban housing policy, reflecting the comprehensive multidisciplinary nature necessary for tackling complex exposure-health relationships. The collaborative spirit demonstrated in blending clinical data, environmental science, and social policy analysis paves the way for future explorations into how place influences health at a granular level.
This pioneering research marks a major step forward in elucidating how the invisible burdens of pest allergens within our homes affect children’s respiratory health. By harnessing the digital footprints left in medical and environmental datasets, scientists and clinicians are beginning to unlock the potential for precision public health interventions that address root causes of asthma disparities. The hope is that such efforts will inspire further innovation to protect and promote the respiratory well-being of vulnerable children in cities worldwide.
Subject of Research: People
Article Title: Associations between in-home environmental exposures and lung function in a safety net population of children with asthma using electronic health records and geospatial data
News Publication Date: June 10, 2025
Web References:
Annals of Epidemiology DOI
Boston University School of Public Health
Boston Medical Center
References:
Fabian, P., Bozigar, M., Connolly, C., et al. (2025). Associations between in-home environmental exposures and lung function in a safety net population of children with asthma using electronic health records and geospatial data. Annals of Epidemiology. doi:10.1016/j.annepidem.2025.04.001
Keywords: Asthma, Respiratory disorders, Environmental monitoring, Insects, Pest control, Housing, Children, Public health, Health disparity, Health equity, Urban populations, Rodents
Tags: assessing indoor allergens remotelychildhood asthma environmental factorscomputational modeling in public healthelectronic health records integrationgeospatial housing data analysisindoor allergen exposure predictorsmachine learning in epidemiologypediatric chronic illness preventionresidential location health risksroach and rodent infestation impactsurban health disparities in asthmavulnerable populations asthma management
