Epidemiology
Poster Session 4
Emily Root, MD (she/her/hers)
Resident
University of Louisville
Louisville, KY, United States
Brian Guinn, MPH, PhD
Assistant Professor in Epidemiology and Population Health
University of Louisville
Louisville, KY, United States
David Harris, MS
Professor and Director, University of Arizona Biobank
University of Kentucky, Kentucky Geological Survey
Lexington, KY, United States
William Haneberg, PhD
University of Kentucky, Kentucky Geological Survey
Lexington, KY, United States
Edward Miller, MD (he/him/his)
Division Director Maternal Fetal Medicine
University of Louisville Hospital
Louisville, KY, United States
Caitlin Thomas, MD
University of Louisville Hospital
Louisville, KY, United States
Exposure to hydraulic fracturing (HF), an oil and gas extraction method, may be associated with adverse birth outcomes. Nitrogen (N2) HF uses nitrogen gas to extract natural gas, whereas slick water HF uses a combination of chemicals, water and sand. By differentiating outcomes by type of HF performed, recommendations can be made on the methods of gas extraction which confer the lowest risk to pregnant patients. Our aim is to demonstrate the impact of residential proximity to slick water versus N2 based HF wells and preterm birth and low birth weight.
Study Design:
This is an ecological study. Ten eastern Kentucky (KY) counties that contain both slick water and N2 fracking were identified. Data from the KY Public Health Department included: demographics, gestational age (GA), and birth weight for singleton pregnancies over 20 years. Individual births were categorized by term (GA > 37weeks) or preterm (GA < 37 weeks) as well as LBW (< 2500g) or VLBW (< 1500g). The KY Geologic survey provided location data for HF wells in 10 eastern KY counties. KY county shapefiles were downloaded from the US Census Bureau’s TIGER/Line Shapefile repository. Data for percent poverty was downloaded from the American Community Survey 2019 5-year estimates. Choropleth maps are expressed by quantiles. All maps were produced in ESRI ArcGIS Pro version 2.8.4.
Results: There were 29,360 unique births during the study period. A total of 398 HF wells were identified in all counties, 45.7% were slick water HF and the vast majority were in Lawrence county (n=160). Lawrence county had the highest prevalence of low birth weight (15.37%, n=166). The 9 remaining counties did not show a spatial association between residential proximity and low birth weight. There was no association between preterm birth and slick water or N2 well location.
Conclusion:
There appears to be a spatial association between slick water HF, and low and very low birth weight. This spatial correlation is not present for preterm birth.