Fracking and Public Health: Finding the Best Balance

By Mark Wolverton

Just a few years ago, the term fracking might have drawn a perplexed stare from most of the general population. But today, particularly in large parts of the United States such as rural Pennsylvania, the word isn’t just a figure of speech but a real and immediate curse word. Although fracking in common parlance has come to represent the entire range of unconven­tional gas and oil drilling activities, technically it refers only to the technique of hydraulic fracturing or hydrofracturing: ex­tracting natural gas from deep within the Earth by cracking open rock layers with high pressure liquids.

It’s become one of the most contentious political and environ­mental issues of the past decade. It is a focus of bitter contro­versy, from the national news all the way down to the grassroots community level, as the lives and perhaps the health of residents in previously sleepy rural communities are disrupted and threatened. Meanwhile, gas well drilling in states such as Penn­sylvania and North Dakota continues to expand and accelerate at so fast a pace that the concerns of residents, environmen­talists, and other involved parties seem almost an afterthought. 

That’s certainly true from a scientific and public health standpoint, particularly in Pennsylvania, where the prolifera­tion of well drilling seems to be inversely proportional to the available data on its environmental and health effects. At Penn Medicine, studying those effects falls under the aegis of the Center of Excellence in Environmental Toxicology (CEET). Part of the Perelman School of Medicine, CEET is also one of only twenty Environmental Health Sciences Core Centers (EHSCCs) in the United States, as designated by the National Institute of Environmental Health Sciences (NIEHS). In collab­oration with the Mailman School of Public Health at Columbia University, Penn’s center recently published a major study identifying an association between fracking and increased hospitalization rates. Their work is already having a major impact on the fracking controversy, introducing some sorely needed facts into an issue too often muddied by passionate but ill-informed rhetoric on all sides.

As the only EHSCC in the Environmental Protection Agency’s Region III (Pennsylvania, Delaware, Maryland, Washington DC, Virginia, and West Virginia), Penn’s center could be ex­pected to take a particular interest in the fracking controversy. Most of those states lie firmly atop the Marcellus Shale forma­tion, one of the major sources of natural gas to be developed and explored in recent decades. Half of Pennsylvania’s land mass, in fact, rests under Marcellus Shale. At a 2012 meeting of all the EHSCCs, Trevor Penning, Ph.D., a professor in the Department of Systems Pharmacology and Translational Ther­apeutics and director of CEET, gave a presentation on potential public health impacts of hydraulic fracturing in Pennsylvania. “I made the point that this was such a com­plex issue that no single environ­mental sciences center could tackle this by itself. What we needed was to pool our resources so we could work collaboratively together. That led to the establishment of an in­ter-center working group on hy­draulic fracturing.” 

And led as well to the recent study, published in the journal PLOS One in the summer of 2015. The lead author was Reynold A. Panettieri Jr., M.D. ’83, G.M.E. ’90, deputy director of CEET and professor of medicine. The study, “Unconventional Gas and Oil Drilling is Associated with Increased Hospital Utiliza­tion Rates,” he notes, was rated among the journal’s top six viewed publications online, garnering about 14,000 views in only a month and a half. “So it’s obviously a passionate area.”

Although the commotion and controversy over hydraulic frac­turing might be fairly recent, the practice itself is not. It was first used in Kansas in 1947. As men­tioned, it’s only one part of an elaborate process that includes the selection of the well site, pad construction, drilling, and encas­ing the borehole in concrete. A mile or more deep, the drilling changes to a horizontal direction, into and along the shale lay­ers. Then the actual hydraulic fracturing process begins, in which the horizontal borehole is perforated by holes at vari­ous spots and millions of gallons of fracturing flu­ids are pumped through them at high pressures. That forces the rock layers to crack and fractures to open, releasing their coveted stores of natural gas, to be recovered and stored back on the surface.

Why Fracking, Why Now?

If fracking is nothing new, why is everyone sud­denly so concerned about it? The main reason is that in the past, hydraulic fracturing tended to be done in places where oil and gas drilling were al­ready well established. But in recent years, the economics of energy production and demand, dwindling supplies of more “conventionally” ob­tained gas, the desire to reduce our reliance on dirtier carbon-based fuels such as coal, and the development of improved drilling techniques that make accessing formations such as the Marcellus Shale cheaper and more practical have all come together to spur a burgeoning new enterprise. That means that drilling is now being done in thousands of locations that have never before seen anything more industrial than a passing diesel truck. Now these com­munities are being invaded by hundreds of trucks, heavy equipment, and noisy and smelly activities that used to hap­pen hundreds or thousands of miles away. All these develop­ments have literally brought home the oil and gas industry to many people who previously would encounter it only at their local gas stations.

“While hydraulic fracturing is not new,” Panettieri explains, “the magnitude of the drilling is certainly different. So we have to take that into account and the proximity to popula­tions which didn’t necessarily exist before. The other part of the equation obviously is, well, if we don’t use this fuel, we would simply continue to do coal burning. Isn’t this better than coal burning? That becomes a societal issue.”

Engineers and geologists prefer to use the term fracking (with a “k”) to mean the overall drilling process and the word fracing (or frac’ing) to refer solely to the hydraulic fracturing technique. They make that distinction because some phenom­ena that have been blamed on fracking, such as recent earth­quakes in Oklahoma, are not actually the result of hydraulic fracturing but other processes, such as the injection of waste water into deep underground wells. Whatever one’s preferred terminology, it’s the possible release of toxic materials into air or water that’s of prime concern, especially for the people liv­ing near well sites. A 2013 Wall Street Journal report esti­mated that more than 15 million Americans live within one mile of a well drilled since the year 2000. In the rush to find and develop resources to satisfy America’s ever-increasing thirst for energy, the people most directly affected by it have more and more questions – but definitive answers remain frustratingly elusive.

“How do we have no data on an enterprise of this magnitude?” asks Aubrey K. Miller, M.D., M.P.H., a senior medical advisor at National Institute of Environmental Health Sciences. “How do we have 15 million people living within one mile of a well, and we can’t answer their questions?” The reassurances from the oil and gas industry that well operators are scrupulously observing best practices often ring hollow to mothers whose children have developed mysterious rashes and other ailments or homeowners who have found their drinking water wells contaminated by unknown substances. Anecdotes and horror stories abound, as do confident proclamations of safety, but with a dearth of hard data, it’s virtually impossible to sift out truth from the rhetoric.

Assessing the Risks

“There are different ways to approach the problem,” Pen­ning says. “One is a classical risk-assessment approach,” with four components. “Risk identification, meaning what are the hazards that we have to be concerned about as it relates to human health? I think we have a good handle on what the possible air pollutants might be. We don’t have a strong han­dle on the pollutants that might contaminate the water sup­ply.” Although the hydraulic fracturing fluid used by well op­erators is about 99% water, the remainder is a mixture of other chemicals that varies among different companies, and the specific formulation of a company’s “fracking fluid” is con­sidered proprietary information – a trade secret to be pro­tected and guarded. “Therefore,” notes Penning, “we don’t know what we’re looking at.”

A second component of risk assessment is to establish a solid baseline of data to identify what’s normal and what’s not. “Because of the lack of baseline data both in air and water quality, it’s tough to attribute contamination or air pollution to the process itself, because there could be other sources,” Penning explains. That, of course, provides a convenient out for industry representatives to discount the complaints and concerns of local residents. 

Once a particular hazard has been identified, it needs to be characterized. What are its effects? How much exposure is needed to induce what kind of responses? Again, in the frack­ing debate, answers to these questions are few and far be­tween. As Penning observes, “Most things follow a dose re­sponse curve, and you need to know what the margin of safety is between the exposure and the levels that would cause a health effect. And that’s also obviously very difficult when you can’t do the hazard identification in the first place.”

The final piece of risk assessment is figuring out how to mitigate risks and how to communicate information effectively both to policy makers and to residents. “So it’s a four-part ap­proach, but it’s also sequential,” Penning says. “And we’re stuck on one and two right now. That’s what led us to actually think about designing the PLOS One study. We felt we needed some kind of broad-stroke approach to determine whether there was a signal worth looking at.”

What’s Real and What Isn’t Real?

The CEET/Columbia study looked at two counties in northeastern Pennsylvania where unconventional oil and gas drilling has blossomed and is continuing (Bradford and Susquehanna) and one in which no active wells are located (Wayne). Researchers looked for any correlation between the density of wells in each area and the inpatient admission rates from 2007 to 2011. “Our whole approach is an unbiased ex­amination of the data,” says Panettieri. “What’s most import­ant is that we look at this with glasses off, what’s real and what isn’t real.”

Penning, Panettieri, and their collaborators were startled by what they found. Using data from 25 specific medical catego­ries (e.g., cardiology, dermatology, gynecology, gastroenterol­ogy, neurology), they examined hospitalization rates within each category and county, noting any association with well activity. They expected to find fairly stable rates, considering that they were working with a time span that was relatively short, in epidemiological terms. 

Instead, the study found significant associations between inpatient admissions and fracking activity in a number of medical categories, particularly for cardiology and neurology. Seeing such a strong statistical signal over such a short time is considered fairly remarkable. Says Panettieri, “It was shocking to us that we saw a signal after five years. Many epidemiological studies of toxic exposure take decades. This was fast and really speaks to maybe some major issues that need to be addressed.”

But just as with any other controversial public issue, people tend to hear what they want to hear, and nowhere is that more prevalent than with fracking. “Obviously when the data came out, we were embraced quickly by all of the pundits who were against fracking, while we made no friends in the oil industry,” says Panettieri with a laugh. 

Careful scientists that they are, however, Panettieri and his colleagues emphasize the study’s limitations and caution against drawing premature and unfounded conclusions. The study states that “the precise cause for the increase in inpa­tient prevalence rates . . . remains unknown” and that “the clinical significance of the association remains to be shown.” 

As Penning puts it, “This is just a drop in the ocean in terms of where we could go with this approach.” He points out that there are subcategories that need to be mined to see whether certain subsets of disease show up. In addition, he says, “so far we’ve only looked at hospitalization rates. We need to look at outpatient data as well.”

Weighing the Impact on Wells and Springs

Aside from the CEET study, another draft report released in 2015 has caused a mighty stir: a study by the U.S. Environ­mental Protection Agency that examines the potential impact of hydraulic fracturing on drinking water sources. It focused specifically on Bradford and Susquehanna Counties in Penn­sylvania, the same counties studied by CEET. Analyzing more than 225 water samples taken over a year and a half from about 40 different locations, the EPA found no evidence of any “widespread” impact on homeowner wells or springs from hydraulic fracturing, although it did find some evidence of fracking-associated methane and ethane in several wells.

It was predictable that the EPA study (which remains to be finalized after review and public comment) was interpreted differently by pro- and anti-fracking camps. While those in industry saw themselves vindicated, fracking opponents criti­cized the EPA for pandering to oil and gas companies. The EPA has acknowledged some definite limitations in the study, and scientists such as Penning have criticized it on that score. 

As he explains, “we sent in a 22-page commentary that was actually not too pleasant.” The CEET experts identified what they considered serious gaps in knowledge that should have been taken into account in the agency’s executive summary. 

Even more emphatic is Marilyn V. Howarth, M.D., an ad­junct associate professor of emergency medicine and pharma­cology at the Perelman School who also serves as director of CEET’s Community Outreach and Engagement Core. “The EPA’s conclusions are premature,” she says. “Their retrospec­tive studies found impacts on drinking water such as in­creased sediment and problems with clarity. No one should be expected to drink water that contains debris whether or not the debris is toxic. In addition, there has been no prospec­tive study on drinking water throughout the life of a nearby well. Water quality may significantly yet transiently change.” Howarth asserts that there is no data to determine if these changes are occur­ring and, if they are, what are the impacts on peo­ple’s health.

Still, Penning emphasizes that the best way to resolve such contentious issues is to strive for a scrupulous scientific objectivity, which means tak­ing biases into account, both conscious and un­conscious. Many studies are not adequately de­signed, or their results are exaggerated or misrep­resented by the press or overzealous advocates on both sides. “We get these one-liners, but you have to look at the details,” he observes. “One thing that made the PLOS One study so powerful is it’s unbiased. We didn’t have any preconception of what we should be measuring. Some of the other health studies have been focused on one or more particular health point that’s been pre-selected, so you don’t know if, based on that pre-selection, other things could have been missed.” 

What Remains to Be Done?

Because so much remains unknown and uncer­tain, there’s much more work to be done. While many researchers are focusing on toxicological ef­fects of air and water pollution from hydraulic fracturing, little attention has been given as yet to the psychosocial aspects, which may be even more important in the long run. 

Says Penning, “If you look at the PLOS One study, you’ll see there was a trend but not a specific association between psychological disorders as well, but specifically issues of depression, anxiety, and so on. I think psychosocial stress is a very im­portant component of this, along with sleep deprivation.” 

As Panettieri points out, “These areas are pretty bucolic, they’re rural. And in comes a lot of industry very quickly set­ting up camp, disrupting that lifestyle, and it engenders a lot of stress. We know that chronic stress can cause cardiovascu­lar disease and impair the immune system. We believe that with this short observation period and the rapidity and ro­bustness of the data, there’s a stress response from noise, dis­ruption of lifestyle, uncertainty, and air pollution from diesel truck exhaust. Our hypothesis is that a combination of chronic stress on top of air pollution rather than water pollu­tion might be responsible for our findings.”What should be done now, while we continue to debate and study and assess and define the risks of hydraulic fracturing? It’s a question that depends heavily upon the political climate, which in Pennsylvania is undergoing a shift from the unabash­edly pro-industry stance of the administration of the former governor, Tom Corbett, to the apparently more cautious admin­istration of Governor Tom Wolf. One measure that CEET and other scientific and public health experts recommend is estab­lishing a health registry to track and monitor the health of residents living near fracking sites. But while the current Wolf budget has set aside funding for a registry, it’s not nearly enough.

“It was only funded to a level of $100,000 a year, which is basically one person at a computer,” Penning notes, adding, “a registry is not really a valuable tool unless there is appropriate exposure data to go with it.” That data, he says, should be based on a particular ill effect or symptom as diagnosed by a physician trained in taking exposure history. “Unfortunately, what we have in Pennsylvania – which is not unusual for the U.S. as a whole – is a lack of physicians that are actually trained in occupational environmental medicine.”