Taxpayers Beware: Broomfield Health Survey Ignores Key Context
Researchers with a long history of publishing deeply flawed studies targeting Colorado’s oil and natural gas industry – who have been previously criticized by state regulators – are out with a $55,217 survey that polled residents in Broomfield, Colo. on the health impacts of development that relies on faulty methodology.
A public meeting to present the findings to funders (i.e. taxpayers) is scheduled for Thursday, although the researchers acknowledge (slide 30) that the findings have not been fully released to the public or peer-reviewed, explaining:
“It is not standard to share a manuscript before submission to a journal as this could jeopardize the integrity of the study and potential for publication.”
The survey was funded by the City and County of Broomfield and polled 427 people who lived within 1 mile, 1-2 miles, or more than 2 miles from the six well pads operated by a Denver-based E&P “to determine if residents living near [unconventional oil and gas development] sites within [City and County of Broomfield] experience a greater frequency and number of health symptoms than residents living at further distances.”
However, a review of the results of the survey, which have only been partially published, shows that this research ignores the impacts of COVID-19 and other local economic activities, hinges on incomplete information, and was likely undermined by perception bias.
Here’s what you need to know about the survey:
#1: Ignores the Late 2021 COVID-19 Surge
The survey was conducted between October and December 2021 – the same time the COVID-19 omicron surge was on the rise, including increasing cases in Broomfield County, according to New York Times data.
Yet, the survey never mentions the COVID-19 surge nor does it apparently screen respondents for exposure or symptoms. This is noteworthy as the survey polled respondents on “significantly greater frequencies of upper respiratory and acute symptoms” – the same ailments associated with the disease.
Additionally, the survey included a disproportionally large number – 58 percent – of respondents age 65 and older who lived within one mile of a well site – an age group at particular risk for COVID-19.
#2: Lacks Data Around Children
The reported findings from the survey provide incomplete information regarding where the parents who responded about their children’s symptoms live, and whether or not there is correlation between the parents’ observed concerns about “odors, noise, and air pollution” and the reported symptoms.
According to the survey, 59 “respondents living within 2 miles of a UOGD site also reported that their children experienced significantly greater frequencies of lower respiratory, gastrointestinal, and acute response symptoms than those living greater than 2 miles from UOGD sites.”
Additionally, the study does not share specifically how parents’ concerns with odors, noise, and air pollution impacted the frequency of symptoms reported on behalf of their children.
#3: Builds In “Participation Bias”
Among the limitations of the survey, the researchers acknowledge a “potential for participation bias,” as the survey identified a correlation between respondents’ “level of concern with odors, noise and air pollution” and the frequency of reported symptoms, regardless of distance from a UOGD site.
Yet, without the raw data behind the survey, it’s impossible to find if a correlation between perceived environmental concerns and reported symptoms is greater than the correlation between proximity to a UOGD site and reported symptoms.
Further, respondents knew they were reporting symptoms for a study “to better understand the health implications of living near oil and gas development,” which may have impacted the number and type of respondents, hence the researchers acknowledgement of a “potential for participation bias.”
This is compounded by the fact that the researchers did not get the initial response needed to complete the study and had to expand the random selection:
“Due to a lower than expected response rate, the University of Colorado’s Multiple Institutional Review Board (COMIRB) has approved an amendment to the IRB package for this research to increase the population that receives postcard surveys. Approximately 2,300 randomly selected households will also be receiving health surveys in the mail in the near future. We encourage all residents who receive a postcard to participate.”
#4: Cherry Picks Existing County Air Monitoring
While the researchers include data from Broomfield’s ongoing air monitoring that’s part of its Oil and Gas Chapter of Broomfield’s Comprehensive Plan, researchers fail to mention that this air monitoring found:
“No measured pollutant concentrations captured to date have surpassed CDPHE’s health guideline levels.”
This means that there isn’t a pathway to exposure defined that would be causing symptoms as a result of nearby oil and gas operations, an issue that is prevalent in McKenzie’s previous research as well (more on that shortly).
#5: Numerous Factors Unaccounted For In Findings
Beyond COVID-19, the researchers also did not account for additional environmental factors which may contribute to health effects in the Broomfield community. This includes, but is not limited to:
- Proximity to Interstate highways
- Industrial warehouses in the general area that operate 24 hours a day, 7 days a week
- Large regional landfill approximately 2.25 miles north of the development area
- Local, regional, and international airports
- Wind direction and speed is not held constant in the survey and is highly variable over short periods of time
- Lack of activity at well sites near residents who filed complaints
- High altitude that commonly causes nose bleeds during seasonal changes. It impacts 1 in every 5,000 people
- Residential housing development
- Underground gasoline storage tanks for nearby gas stations
#6: Researchers’ History of Bias
One of the researchers who conducted the survey, Dr. Lisa McKenzie with the University of Colorado School of Public Health, has a long history of publishing research targeting the oil and natural gas industry that was later rejected by public health officials. This includes:
- A 2014 study where McKenzie claimed to find the “an association between those who lived within a 10-mile radius of (oil and gas development) and congenital heart defects and possibly neural tube defects.”
- The Colorado Department of Public Health and the Environment responded to the study with the statement: “[W]e disagree with many of the specific associations … [and] a reader of the study could easily be misled to become overly concerned.”
- A 2017 study where McKenzie claimed to find an association between oil and natural gas development and childhood cancer.
-
- CDPHE responded to the study with the statement: “We support studies that evaluate the potential impact of environmental contaminants on public health, and certainly, Benzene exposure has been proven to increase risk of certain types of cancers, including leukemia. However, this study’s conclusions are misleading in that the study questions a possible association between oil and gas operations and childhood leukemia; it does not prove or establish such a connection.”
- A 2018 study where McKenzie claimed to find the “lifetime cancer risk of those living within 500 feet of a well was eight times higher than the EPA’s upper level risk threshold.”
- CDPHE’s former Chief Medical Officer and Executive Director, Dr. Larry Wolk responded to the study with the statement: “This study confirms our 2017 findings of low risk for cancer and non-cancer health effects at distances 500 feet and greater,” referring to the 2017 CDPHE Report: Assessment of Potential Public Health Effects from Oil and Gas Operations in Colorado, which found, based on 10,000 air samples, a “low risk for cancer and non-cancer health effects at distances 500 feet and greater.”
Further, in addition to McKenzie citing her own disavowed research as resources for the preliminary findings, she cites several others that have also been identified as having limitations or researcher bias, particularly on the Marcellus comparison slide.
For instance, Rabinowitz et al, 2014 claimed that people in Washington County, Pa., living within a kilometer of oil and gas wells suffer from more respiratory and skin conditions than those who live farther away.
The study was released on the heels of a Pennsylvania Department of Environmental Protection list of instances where drinking water supplies may have been affected throughout the state, but none were identified for Washington County during the survey period (2011-2012). In fact, there were only two cases the DEP identified – one in 2008, outside the study area and one in 2014, after the survey was completed.
Further PADEP conducted extensive air monitoring in southwestern Pennsylvania, where Washington County is located, and concluded:
“Sampling for carbon monoxide, nitrogen dioxide and ozone, did not detect levels above National Ambient Air Quality Standards at any of the sampling sites.” Read more on this study here.
Similarly, Steinzor et al, 2013 was a study conducted in collaboration with Wilma Subra that claimed people living next to oil and gas infrastructure in Pennsylvania experienced higher instances of throat, nasal and sinus symptoms. Both Steinzor and Subra are associated with the group Earthworks that has likened fracking to “rape”. Earthworks also has at least one researcher on its Board of Directors – Anthony Ingrafea – that calls himself “a self-admitted advocate” against fracking and told Coloradans:
“I’d be lying if I said that I don’t… that I haven’t entered any of the studies that I have participated in in the last five years that led to my being able to publically speak or publish in peer-review papers… I’d be lying if I told you I went into every one of those with an entirely objective, blank opinion. I didn’t know. I didn’t care. Whatever happened when the numbers were done, I’d be [inaudible]. I’d be lying if I said that. I’m sorry. [cross talk] It would be untrue.”
And Tustin et al, 2017 is one part of a series of studies from a group of researchers at Johns Hopkins University using data from Pennsylvania’s Geisinger Health System. The team has claimed its research establishes links between fracking and increased depression, asthma, premature births, and radon in Pennsylvania homes, despite failing to take measurements that would prove concentration and exposure levels that would establish a causal link between shale development and negative health outcomes. In Tustin 2017, the claim was sinus conditions, migraines and fatigue. Their work has been criticized for making “unwarranted causal interpretation of associational results.” The most vocal of the group is Brian Schwartz – a fellow at the Post Carbon Institute, an anti-fossil fuel organization that has called fracking a “virus,” who called for a fracking ban in Pennsylvania in 2019.
Conclusion
It appears that this research is yet another in a long line from McKenzie that looks to blame oil and natural gas development for any issues occurring in a community. And yet again, the research leaves more questions than conclusions on what the findings actually show, particularly since only partial information was released – this time leaving taxpayers on the hook for inadequate, flawed research.