Asthma is a chronic heterogeneous airway disorder characterized by inflammation, mucus hypersecretion, airway hyperreactivity, and impaired airflow. Severe exacerbations of asthma occur frequently in children and require immediate use of systemic steroid therapy to prevent serious outcomes such as hospitalization or death. In addition to direct health risks, pediatric asthma exerts a substantial cost burden, as asthma exacerbations are a leading cause of emergency department visits, hospitalization, and missed school days. Multiple environmental factors are purported to play a role in asthma symptoms, including aeroallergens, pollutants, weather changes, and community viral outbreaks such as influenza. Additionally, asthma prevalence is greater in children of low socioeconomic status (SES) and in African-American and Hispanic/Latino children, suggesting both environmental and genetic effects on asthma incidence and severity. The existence of geographical asthma ?hotspots? indicates that asthma prevalence and severity are influenced by place-based risks, including local air quality, built environment factors, access to health care providers, socioeconomic factors, culture, and behavior. To effectively prevent and treat pediatric asthma attacks, it is necessary to understand how patient-specific characteristics interact with environmental factors to render an individual susceptible to severe asthma exacerbations. Lacking sufficient power, previous studies have largely examined suspected asthma triggers in isolation; thus, there is a significant knowledge gap regarding how environmental factors interact with each other and with patient-level factors to promote severe asthma exacerbations in pediatric populations. We hypothesize that a longitudinal analysis of environmental exposures and patient-level factors will elucidate new multifactorial causes of severe asthma exacerbations. To elucidate the contributions and interactions of environmental and patient-level factors, we will apply machine learning approaches to a longitudinal (2007-2017) geocoded database of patient electronic health records detailing asthma-related health encounters and publicly available, overlapping spatiotemporal environmental data. Further, we will evaluate the interactions between person-level clinical factors, including obesity, history of premature birth/bronchopulmonary dysplasia, and atopy, to determine their effects on susceptibility to selected environmental triggers. These analyses will 1) provide an analysis of the relative contribution and interactions of environmental factors to pediatric asthma exacerbations, 2) identify geographic hotspots of asthma prevalence and severity, and 3) determine how person-level clinical factors influence susceptibility to different asthma triggers. Our findings will provide new insights into risk factors for severe asthma exacerbations, spur new studies into the biological mechanisms that underlie the interactions between human biology and the environment, inform preventive strategies and patient education efforts, and serve as a model that can be expanded to larger cohorts.
Asthma affects 1 in 10 children in the United States, and severe asthma exacerbations require immediate medical attention to prevent hospitalization or death. A wide variety of environmental triggers such as pollutants and allergens are associated with asthma symptoms, but the interaction between different environmental triggers and person-level characteristics such as obesity, allergy, and history of premature birth are unclear. These studies will use longitudinal, overlapping asthma patient and environmental data to understand the role of different environmental triggers in asthma exacerbations in order to develop new preventive and therapeutic strategies for asthma.
