As loyal Product Liability Monitor readers know, we keep an eye on developments relating to hydraulic fracturing or “fracking.” Along these lines, a study was published this week in JAMA Internal Medicine exploring associations between unconventional natural gas development (“UNGD”) in Pennsylvania’s Marcellus Shale and asthma exacerbations. The authors note that this was the first study looking at UNGD and objective respiratory outcomes. The study concluded that residential UNGD activity metrics were statistically associated with increased risk of mild, moderate, and severe asthma exacerbations, but acknowledged that further investigation is needed to determine whether the associations are causally related to UNGD.

The study design was a “nested case-control study” that compared patients with asthma with and without exacerbations from 2005 through 2012 who were treated at the Geisinger Clinic, which provides primary care services to more than 400,000 patients in Pennsylvania. Over 35,000 patients with asthma were identified in electronic health records (“EHR”) and patients with exacerbations were frequency matched on age, sex, and year of event to those without. The exacerbations were divided into three categories — mild (new oral corticosteroid medication order), moderate (emergency department encounter), or severe (hospitalization).

On the day before each patient’s index date (cases, date of event or medication order; controls, contact date), the authors estimated activity metrics for 4 UNGD phases — (1) pad preparation, (2) drilling, (3) stimulation (the actual hydraulic fracturing, or “fracking”), and (4) production — using distance from the patient’s home to the well, well characteristics, and the dates and durations of phases.

The study found that those in the highest quartile of residential UNGD activity had significantly higher odds of 3 types of asthma exacerbations (new oral corticosteroid medication orders, emergency department visits, and hospitalizations) than those in the lowest quartile. According to the study, this means that UNGD activity near patient residences was associated with increased odds of mild, moderate, and severe asthma exacerbations.

Strengths of the study, according to the authors, included a large sample size from a population that represents the general population in the region; an exposure assessment metric that accounted for different phases of UNGD; and the use of documented asthma exacerbations (as opposed to self-reporting).

Limitations included the fact that EHR did not contain information on the patients’ occupation and only reflects patients’ most recent address. The authors were also unable to differentiate between asthma exacerbations that were hospitalized from those that occurred while hospitalized. In addition, the UNGD metrics do not provide insight into the mechanism of the associations that were observed.

In the end, the authors conclude that UNGD may increase risk for asthma exacerbations and that there should be further studies with more detailed exposure assessment to better characterize pathways and to identify the phases of UNGD that present the most risk.

While the study has generated attention-grabbing headlines, e.g., “Fracking wells may increase asthma attacks, study says,” and lead-ins noting that “[a]sthma patients are 1.5 to four times more likely to have asthma attacks if they live near bigger or a larger number of unconventional natural gas development wells” – it is important to keep in mind that the findings are not proof of a causal effect. As the medical community acknowledges, asthma is a complex condition and flare-ups can be caused by a number of factors, including air pollution, odors and stress.

And, the Pennsylvania director for Energy In Depth points out, asthma is a condition that can arise for numerous reasons. The study’s authors also acknowledge that they are not able to show causation from oil and gas development, and further acknowledge that cases resulting in hospitalizations included patients who were older, current smokers, and obese.