Study of Premature Births Fails to Show Fracking Connection

Published November 16, 2015

Serious methodological errors render unreliable the findings of a recent study titled “Unconventional Natural Gas Development and Birth Outcome in Pennsylvania, USA,” which suggested pregnant mothers living near hydraulic fracturing sites could be at a higher risk of giving birth to premature babies.

The study, which was published in the journal Epidemiology, was conducted by researchers at the Johns Hopkins-Bloomberg School of Public Health. It relies exclusively on retrospective data analysis of pregnancy and birth statistics of just under 11,000 pregnant women in hydraulic fracturing-rich areas of Pennsylvania between 2009 and 2013. The data were used to find statistical associations between the distance a pregnant mother lived from a fracking site and whether or not she gave birth prematurely.

Study Cannot Account for Key Issues

Dr. Gilbert Ross, senior director of medicine and public health at the American Council on Science and Health, says although these methods sound sophisticated, they tell us very little about whether fracking had any impact on the pregnancies.

“There is no possible way this retrospective study could have accounted for key issues, such as genetic factors, history of prior pregnancy issues, [or] drug or alcohol use in the parents, all of which have a large influence on birth weights and the duration of pregnancy,” Ross said.

“Realistically, there is no way hydraulic fracturing could have had an impact on pregnancy outcomes,” said Ross. “Fracking occurs two miles below the surface, and the Environmental Protection Agency found no evidence that hydraulic fracturing is causing widespread water contamination. There is no plausible mechanism to explain how pregnant women might have been exposed to anything related to fracking to cause attenuated gestation duration.

“Indeed, the authors didn’t measure any levels of possible toxicant exposure,” said Ross. “How could they, of course, in a retrospective analysis such as this?”

Results Undermine Authors’ Claims

Some of the study’s results fail to support claims fracking could be leading to an increase in premature births, and in some instances, the results indicate just the opposite.

According to the study, premature birth rates near fracking activity were actually lower than the national average. According to the Centers for Disease Control 11.5 percent of all births in the United States are premature. By comparison, the study found only 11 percent of the births near fracking sites were premature.

To study the effect of natural gas development on pregnancy, the researchers divided mothers into four groups, or quartiles, based on their distance from a fracking well. Mothers in quartile one lived farthest from the sites, and mothers in quartile four lived closest. Although mothers living in quartiles one and two had the lowest rates of premature births, mothers in quartile three had higher rates than the mothers who lived closest to the fracking activity.

“If living close to fracking activity were truly influencing premature births, we would expect to see the highest incidences of early deliveries in the areas nearest fracking wells,” Ross said. “One of the key criteria in assessing whether a possible factor is causally related to an outcome is called ‘dose-response,’ meaning if A causes B, then more of A should cause more of B.”

Isaac Orr ([email protected]) is a research fellow for energy and environmental policy at The Heartland Institute.