Asthma Exacerbations and Unconventional Natural Gas Development in the Marcellus Shale
Start Date
13-11-2015 8:00 PM
End Date
13-11-2015 9:59 PM
Description
Asthma is a common, chronic disease that can be exacerbated by air pollution and stress. Unconventional natural gas development (UNGD) has documented impacts on air quality and social disruption, and has expanded rapidly. In Pennsylvania, UNGD began in the mid-2000s and by 2012, 6,253 wells were drilled. There are no prior studies of UNGD and objective respiratory outcomes. In this nested case-control study, we evaluated associations between UNGD and asthma exacerbations. We used data from the Geisinger Health System (GHS), which provides primary care services to over 400,000 patients in central and northeastern Pennsylvania. Asthma patients ages 5-90 years (n = 35,508) were identified in GHS electronic health records, and patients with exacerbations were frequency-matched on age, sex, and year of event to patients without exacerbations using incidence density matching. On the day before each patient’s index date, we estimated a UNGD activity metric for four UNGD phases (pad preparation, drilling, stimulation [“fracking”], and production) using distance from the patient’s home to the well, well characteristics, and the dates and durations of the phases. We identified 20,749 mild, 1,870 moderate, and 4,782 severe asthma exacerbations (new oral corticosteroid medication order, emergency department encounter, and hospitalization, respectively). In three-level adjusted models, there was an association between the highest group of the activity metric for each UNGD phase for 11 out of 12 exposure-outcome pairs (odds ratios [95% CI] ranged from 1.5 [1.2-1.7] for the association of the pad metric with severe exacerbations to 4.4 [3.8-5.2] for the association of the production metric with mild exacerbations). Several of the evaluated exposure-effect relations evidence linear trends. Our findings were robust to increasing levels of covariate control and in sensitivity analyses that included evaluation of unmeasured confounding.
Type
Poster
Language
eng
Asthma Exacerbations and Unconventional Natural Gas Development in the Marcellus Shale
Elaine Langone Center, Terrace Room
Asthma is a common, chronic disease that can be exacerbated by air pollution and stress. Unconventional natural gas development (UNGD) has documented impacts on air quality and social disruption, and has expanded rapidly. In Pennsylvania, UNGD began in the mid-2000s and by 2012, 6,253 wells were drilled. There are no prior studies of UNGD and objective respiratory outcomes. In this nested case-control study, we evaluated associations between UNGD and asthma exacerbations. We used data from the Geisinger Health System (GHS), which provides primary care services to over 400,000 patients in central and northeastern Pennsylvania. Asthma patients ages 5-90 years (n = 35,508) were identified in GHS electronic health records, and patients with exacerbations were frequency-matched on age, sex, and year of event to patients without exacerbations using incidence density matching. On the day before each patient’s index date, we estimated a UNGD activity metric for four UNGD phases (pad preparation, drilling, stimulation [“fracking”], and production) using distance from the patient’s home to the well, well characteristics, and the dates and durations of the phases. We identified 20,749 mild, 1,870 moderate, and 4,782 severe asthma exacerbations (new oral corticosteroid medication order, emergency department encounter, and hospitalization, respectively). In three-level adjusted models, there was an association between the highest group of the activity metric for each UNGD phase for 11 out of 12 exposure-outcome pairs (odds ratios [95% CI] ranged from 1.5 [1.2-1.7] for the association of the pad metric with severe exacerbations to 4.4 [3.8-5.2] for the association of the production metric with mild exacerbations). Several of the evaluated exposure-effect relations evidence linear trends. Our findings were robust to increasing levels of covariate control and in sensitivity analyses that included evaluation of unmeasured confounding.