This week, researchers from the University of Pennsylvania and Columbia University released a study, which asserts a correlation between natural gas development and an increase in hospitalizations in three Pennsylvania counties: Bradford, Susquehanna, and Wayne. However, in the study’s press release, the authors are clear that, “the study does not prove that hydraulic fracturing actually causes these health problems.” In the actual study the researchers also admit,
“The precise cause for the increase in inpatient prevalence rates within specific medical categories remains unknown.” (Page 14)
Of course, one wouldn’t know that from the media’s response to the findings, which suggests a causal link between fracking and hospital visits. Here are the top five things to keep in mind when reading this report.
Fact #1: The county with the highest number of wells (Bradford) actually had the lowest overall inpatient occurrences
Using the data in the chart below, the ranking of the counties with the highest number of incidents by population is as follows: Susquehanna (1.92), Wayne (1.69) and Bradford (1.57).
If we are to follow the researchers’ logic and assume a link between well density and hospitalizations, shouldn’t Bradford County, which had the highest well density, have the highest number of incidents? One would think so, but actually the authors note that the zip code with the highest incidences of hospitalizations was found in Wayne County, where no natural gas development occurred due to a moratorium put in place by the Delaware River Basin Commission (DRBC). From the report:
“In our analysis, one particular zip code had extremely high inpatient prevalence rates compared to other zip codes. Thus, a sensitivity analysis was performed (data not shown). This zip code is located within Wayne County and had no active wells from 2007 to 2011.” (Page 12)
Interestingly, that zip code was removed from the analyses with no explanation as to why. What’s more baffling is that the authors also removed a zip code from Bradford County that had “extremely high wells/km2 in 2010 and 2011.” From the report:
“In our analysis, one particular zip code had extremely high inpatient prevalence rates compared to other zip codes. Thus, a sensitivity analysis was performed (data not shown). This zip code is located within Wayne County and had no active wells from 2007 to 2011. Removal of this zip code from the analysis had little effect on either the number of wells or the quantile analyses, and there was no change in inference and the estimated risk ratios. Next, a zip code in Bradford had extremely high wells/km2 in 2010 and 2011, 16.9 wells/km2 and 23.4 wells/km2, respectively. Consequently, we explored both sets of analyses without this zip code to determine whether removal of this zip code changed inference. Like the first sensitivity analysis, removal of the Bradford zip code had little effect on inference.” (Pages 12-13)
What the authors don’t make clear is what the incidence rates for hospitalization looked like in the zip code with “extremely high wells” or why it was removed. Could it be because the hospitalization rates were very low or stable?
Fact #2: Hospitalization rates by zip code remained stable or declined
The researchers state,
“Fig. 3 also shows that, within each zip code, the contribution by year was comparable, suggesting that within each zip code, the inpatient rates are relatively stable from 2007–2011. Indeed, the average overall inpatient prevalence rates for 2007–2011 are, respectively, 15.18, 15.30, 14.86, 14.00, 14.25. This indicates that on average, zip code overall inpatient prevalence rates were relatively stable or possibly declining from 2007 to 2011, which mirrors national trends.” (Page 9)
What’s more is these occurrences remained stable or even at times decreased, despite significant growth in the numbers of wells in each county, as Figure 2 from the report demonstrates.
If the wells being placed near residences occurred at this high a rate and a link did exist between this and hospitalizations, shouldn’t the incidences of those hospitalizations also have risen? One would think so, and yet they didn’t.
Fact #3: Researchers don’t account for the fact that cardiovascular and neurologic disorders from outside exposures take decades to develop – or any other causes of health problems for that matter
Although the authors’ data show that overall incidents weren’t impacted by well density, they proceed to apply obscure modeling exercises, breaking health incidents down into 25 categories – and from that, they claim cardiologic and neurologic incidences increased more in zip codes with higher well densities.
But, of course, making such a correlation isn’t that simple. The researchers never address the specific diseases within these categories or if the patients suffered from a history of heart disease, heart attacks, stroke, diabetes, artery obstruction, or neurologic disorders. Were these new incidences? Did these individuals smoke? Were they overweight?
Further, age and sex (i.e. being male) increases the risk of some heart conditions such as heart attacks according to the American Heart Association. Did the male population increase? Were these conditions occurring in younger individuals or the elderly who are more prone to heart problems?
In a recent radio interview, Dr. Theodore Them, the Chief of Occupational and Environmental Medicine for Guthrie Health Systems in Bradford Count,y noted that studies on shale development often leave out the very crucial element of “confounders.” As Dr. Them put it,
“And there can be confounders such as smoking habits, drinking habits, drug use that never get accounted for in these studies and cause people to come to the wrong conclusions.” (28:36-30:09)
And this is confounded by the fact that the study focuses on two health categories (heart conditions and neurological problems) that are well-known to take decades to develop, rather than occur instantaneously from short-term exposures. The authors acknowledge this limitation in the study, noting,
“We recognize that a five-year observation period may limit our ability to discern a direct impact on health in the surrounding community but may offer an opportunity to assess hospital utilization rates over time.” (Page 13)
What can occur from short term effects, however, are pulmonary conditions (lung problems), but the researchers did not see any increase in these. As the Philadelphia Inquirer reported, quoting Reynold A. Panettieri Jr., a “senior author” in the study,
“Panettieri said that he had expected the analysis to find an increase in pulmonary admissions – it did not.”
Fact #4: Previous health studies by medical professionals in these counties contradict the researchers’ findings
This study is also not in line with previous studies conducted in the state, or with the experiences of medical professionals in these counties. The Center for Rural Pennsylvania recently conducted a study that looked at inpatient hospitalizations in Lycoming and Bradford Counties in northwestern Pa., and Green and Washington Counties in the southwestern part of the state. These are some of the most heavily drilled counties in the Marcellus. It found:
“Inpatient hospitalizations in the four counties and the two regions increased slightly in the northern tier and decreased slightly in the southwest, but it is not possible to directly connect this to Marcellus Shale drilling.”
“There are no overall trends for injuries in the four study counties or across the two regions; however, there are noticeable increases in injuries associated with falls, motor vehicle accidents, and accidents involving motorcycles. These types of injuries could be related to any type of large-scale construction activity and not necessarily to Marcellus Shale drilling.“
In the radio interview mentioned before, Dr. Them provides a different view altogether on what the health system has seen in the years since natural gas development took off in the county. From that interview (emphasis added):
“As far as the remainder of the population, in the six or eight years that we’ve been having drilling ongoing around us, I’ve had two people come to me or were referred to me. Two total. One was a gentleman who lived between a limestone quarry and a well and had his water quality tested and found there were relatively high levels of barium… This gentleman’s problem was not from the gas wells. It was from the limestone underlying his home through which his water well went and through which the groundwater flowed. And the solution was to put a reverse osmosis unit on his drinking water supply. And he went away. I never heard from him again.”
“The other one was a case of a middle aged woman who came to me and her demand was, “I want compensation.” And when asked why she wanted compensation, it was “for the chemical exposure” which she was alleging she experienced from hydrofracturing water tankers that went passed her house on a daily basis. Now this is a woman with 30 years of allergies to environmental materials such as dust, trees, mold spores, pollen, grasses and so on. And her complaints were nosebleeds and congestion. And her complaint was that the trucks were causing her nosebleeds and congestion because they were going by her house and carrying chemicals, when in fact they were carrying water… She wanted money from the gas companies from her alleged suffering from nosebleeds and congestion from which she had suffered the prior 30 years without any change.
That’s the extent of the people that I have either had referred to me or I have seen in my practice in occupational medicine since 2006-2007-2008. I’m not seeing a flurry of people coming to see us along these lines, despite the fact that the 275 other physicians in my group know that, I and my core colleagues, are the experts in these matters and that such patients should be sent to us. We’re not seeing them.” (18:30-22:18)
Dr. Them went on to say:
“I have yet to see any conclusive study, any absolutely conclusive study, on confirmed adverse human health impacts from drilling, particularly in the Marcellus Shale. The Barnett Shale in Texas has been produced for 20 years—Dr. Zucker referred to that—no confirmable documented adverse human health effects there, despite the fact that the Barnett Shale formation is very similar structurally, geologically to the Marcellus Shale.” (29:32-30:06)
Fact #5: At least one of the researchers “aims to stop fracking”
“To protect our water, air, health, biodiversity, climate and communities, we aim to stop fracking.”
And while we’d like to give her the benefit of the doubt to be able to be unbiased in her research, it’s a little difficult to do when she has made her agenda to stop fracking very clear on multiple occasions, such as following video:
“And I will say that until we have credible science that will help us answer these individuals and their communities, there must be a moratorium on all steps that are involved in the high volume and extreme methods of natural gas extraction, processing and transport.” (1:53)
Her mission extends to a wider anti-fossil fuel agenda. As she puts it in one of her blog posts,
“There is a Dakota saying that goes like this: ‘When you find yourself riding a dead horse, dismount.’
My name is Dr. Poune Saberi and I am a physician in Occupational and Environmental Medicine and a faculty member at the Hospital of the University of Pennsylvania. I am here to say: fossil fuels are dead and we must dismount.”
The bottom line is that the researchers’ data show hospitalization rates remaining stable, with high oil and gas developing counties having the lowest hospitalization rates, while the county with no wells had the zip code with one of the highest hospitalization rates.
Health studies are important and should add to the growing dialogue and to improve best management practices for multiple industries. But this report appears to be yet another spearheaded by anti-fracking activists that resorts to claims of “needing more study” because it fails to find the scientific evidence to malign the development that’s bringing economic, and indeed, environmental and health benefits to Pennsylvanians and families across the country.