By Michael Bastasch • The Daily Caller
One of the Environmental Protection Agency’s (EPA) longest and most successful air pollution standards is based on a taxpayer-funded study plagued by “data fabrication and falsification,” according to a veteran toxicologist.
Toxicologist Albert Donnay says he’s found evidence a 1989 study commissioned by EPA on the health effects of carbon monoxide, which, if true, could call into question 25 years of regulations and billions of dollars on catalytic converters for automobiles.
“They claimed to find an effect when there wasn’t one,” Donnay told The Daily Caller News Foundation. “They even fabricated the methods they used to get their results.”
“They were spinning this to give EPA what they wanted and commissioned,” Donnay said. “They reported results that could not have come from human beings.”
EPA gave “primary consideration” to 1989 study put together by the Health Effects Institute (HEI) to replace a previous 1981 study that relied on fabricated data.
In 2011, EPA did what it’s done since 1971 and left the carbon monoxide standard unchanged at 9 parts per million measured over an 8-hour time period and 35 parts per million over 1 hour.
And why not? EPA’s been wildly successful in reducing carbon monoxide over the years. Levels were 85 percent below EPA standards in 2016, making it the second-most reduced air pollutant, according to agency data.
“Carbon monoxide is a colorless, odorless gas emitted from combustion processes” that can cause serious health effects and even death at extremely high levels, according to EPA.
Accidental carbon monoxide poisoning kills between 400 and 500 people a year, according to government data.
Dr. Aronow
But Donnay said his research shows the HEI study is full of “scientific fraud” that should be investigated by EPA and HEI.
“If you commission someone to replicate a fraud, you have to commit fraud to do it,” Donnay told TheDCNF when discussing the results of his reanalysis.
In 1983, EPA realized its carbon monoxide standard was based on “questionable, perhaps falsified, research” conducted by Dr. Wilbert Aronow in the 1970s, The New York Times reported at the time.
The Federal Drug Administration found Aronow had been falsifying drug trial evidence for patients at a Veterans Affairs hospital in the Los Angeles area. EPA also used Aronow’s research.
The agency acted quickly to replace Aronow’s fraudulent data, funding a massive human exposure study with HEI — a research group funded funded by EPA and the auto industry.
HEI put together a study to test human carbon monoxide on men with coronary artery disease, observing how long they could exercise before they succumbed to angina after being exposed to the gas. HEI hired three teams from different universities to help with the research.
The HEI study was published in 1989 and cost $2.5 million to finish. The New England Journal of Medicine also published the study in 1989, and medical journal Environmental Health Perspectives published the study two years later.
Environmental Health Perspectives declined to further comment on Donnay’s request. Neither HEI nor the New England Journal of Medicine responded to TheDCNF’s request for comment.
HEI researchers, unsurprisingly, found that “low levels of carboxyhemoglobin [carbon monoxide] exacerbate [angina] during graded exercise in subjects with coronary artery disease.”
EPA heralded the HEI study, and relied on it to promulgate its 1994 carbon monoxide standard.
EPA has leaned on it to keep its outdoor carbon monoxide in place ever since. The agency gave the HEI study “primary consideration” in its 2011 review of carbon monoxide regulations.
Donnay joined a lawsuit brought by environmentalists to challenge EPA’s carbon monoxide decision. The courts ruled against activists, so, in 2014, Donnay decided to reanalyze the landmark study EPA had relied on all these years.
“I was eager to challenge that ruling,” he said.
‘Extensive Evidence Of Data Fabrication’
First Donnay would need the raw data HEI researchers used in their study, but that proved more difficult than expected.
In 2012, HEI president Daniel Greenbaum told Donnay raw data for the HEI study was “discarded” four years earlier since neither EPA nor any other researchers questioned its results.
So, he pulled together raw data from the three journals where the HEI study was published, and relied on some raw data featured in a HEI annual report from 1985.
In 2014, Donnay was finally ready to present his re-analysis. He “found extensive evidence of data fabrication and falsification,” he wrote in a draft abstract presented to his University of Maryland advisors.
“The most obvious evidence of deliberate scientific fraud in the HEI study is that Allred et. al. printed two different sets of summary results in their HEI report and a third in their New England Journal of Medicine article that came out the same week,” Donnay told TheDCNF.
“In neither of these versions do they explain how they ended up with different sets of results, or even acknowledge that they did,” he said.
“Looking at what they chose to include and exclude tells you they are hiding something,” he said.
Donnay’s University of Maryland academic advisors, however, told him he could not use his affiliation with the university if he published his findings. The school’s research integrity officer even said his work could “put UMB at risk,” according to emails obtained by TheDCNF.
“There was a lot of resistance to the fact this could be fraud,” Donnay said. “I realized it might be hard to get these findings published.”
But Donnay persisted and in May 2014 sent a letter to all three academic journals where the study was published, asking them to issue a retraction. Donnay also notified EPA’s science integrity office and the inspector general.
All three journals rejected his request. EPA’s inspector general found Donnay’s accusations didn’t rise to levels of criminal charges, and EPA Scientific Integrity Official Francesca Grifo declined to review Donnay’s work since the HEI study was written before 2012 — before the agency’s scientific integrity policy went into place. EPA’s science integrity office did not respond to TheDCNF.
“Given the extraordinary care with which this study was conducted and independently reviewed, they have recommended, and we agree, that there is no reason to retract this study,” HEI’s chairman of the board Richard Celeste wrote in a response letter to Donnay.
Donnay held onto his result, and sought input from Roger McClellan, the former chairman of the EPA science committee that relied on HEI’s study to set the 1992 carbon monoxide standard. McClellan also served on the HEI oversight committee for the carbon monoxide study.
McClellan told Donnay that while the HEI study had some “warts and blemishes,” the “core findings of the study remain sound,” he wrote in a 2015 letter.
McClellan wrote the “only serious finding you reported that is disappointing to me is my learning that the original study records were destroyed by the HEI in 2008.”
Donnay presented his research in the form of a poster at the 2015 Society of Toxicology meeting and received some “perfunctory replies,” he said, but “nobody was willing to address my specifics.”
McClellan did not respond to TheDCNF’s request for comment.
Below are links to Donnay’s evidence:
Roger McClellan’s Letter — Feb 23, 2015
Summary Of Responses To Donnay’s Work
Donnay’s disclosure statement:
Mr. Donnay earns income as a consulting toxicologist specializing in carbon monoxide poisoning and also has patents-pending related to CO testing and treatment from which he does not yet earn income. The review of the HEI MultiCenter Study on CO that he conducted as part of his dissertation research was self-funded.