Residential proximity to unconventional oil and gas development and birth defects in Ohio

Abstract

Background: Chemicals used or emitted by unconventional oil and gas development (UOGD) include reproductive/developmental toxicants. Associations between UOGD and certain birth defects were reported in a few studies, with none conducted in Ohio, which experienced a thirty-fold increase in natural gas production between 2010 and 2020.

Methods: We conducted a registry-based cohort study of 965,236 live births in Ohio from 2010 to 2017. Birth defects were identified in 4653 individuals using state birth records and a state surveillance system. We assigned UOGD exposure based on maternal residential proximity at birth to active UOG wells and a metric specific to the drinking-water exposure pathway that identified UOG wells hydrologically connected to a residence ("upgradient UOG wells"). We estimated odds ratios (ORs) and 95% confidence intervals (CIs) for all structural birth defects combined and specific birth defect types using binary exposure metrics (presence/absence of any UOG well and presence/absence of an upgradient UOG well within 10 km), adjusting for confounders. Additionally, we conducted analyses stratified by urbanicity, infant sex, and social vulnerability.

Results: The odds of any structural defect were 1.13 times higher in children born to mothers living within 10 km of UOGD than those born to unexposed mothers (95%CI: 0.98-1.30). Odds were elevated for neural tube defects (OR: 1.57, 95%CI: 1.12-2.19), limb reduction defects (OR: 1.99, 95%CI: 1.18-3.35), and spina bifida (OR 1.93; 95%CI 1.25-2.98). Hypospadias (males only) was inversely related to UOGD exposure (OR: 0.62, 95%CI: 0.43-0.91). Odds of any structural defect were greater in magnitude but less precise in analyses using the hydrological-specific metric (OR: 1.30; 95%CI: 0.85-1.90), in areas with high social vulnerability (OR: 1.27, 95%CI: 0.99-1.60), and among female offspring (OR: 1.28, 95%CI: 1.06-1.53).

Conclusions: Our results suggest a positive association between UOGD and certain birth defects, and findings for neural tube defects corroborate results from prior studies.

Keywords: Birth defects; Congenital anomalies; Congenital malformations; Epidemiology; Fracking; Hydraulic fracturing; Ohio; Oil and gas.

Figure 1.

Stratified analyses of odds of any structural defect in relation to presence of unconventional oil and gas well within 10 km of maternal residence at birth (2010–2017). ^aOdds Ratios and confidence intervals are for logistic regression including the following covariates: year of birth, parity, maternal race, maternal smoking, use of WIC, social vulnerability index, and ambient PM_2.5 concentration. ^bNeighborhoods with “low,” “moderate,” and “high” social vulnerability are defined as those census tracts which had a Social Vulnerability Index in the bottom 33%, the middle 33%, or the top 33% compared to other census tracts across the state, as ranked in the Center for Disease Control 2014 Social Vulnerability Index dataset ^cOdds Ratios and confidence intervals are for logistic regression including the following covariates: year of birth, parity, maternal race, maternal smoking, use of WIC, and ambient PM_2.5 concentration. ^dRural” and “urban” are defined in accordance with 2010 Rural-Urban Commuting Area (RUCA) codes from the United States Department of Agriculture. Individuals living in a census tract with a RUCA code of 1–7 are considered “urban”; individuals living in a census tract with a RUCA code of 8–10 are considered “rural.”