‘Bogus analysis’: Trump-appointed EPA advisers criticize study linking air pollution and coronavirus deaths

Source: By Abby Smith, Washington Examiner • Posted: Tuesday, April 28, 2020

Trump administration appointees to key panels that advise the Environmental Protection Agency on science are criticizing a study that concludes areas with higher levels of air pollution will experience higher death rates from the coronavirus.

The study, from the Harvard T.H. Chan School of Public Health, has led to increased public pressure on the EPA in recent weeks, threatening the agency’s deregulatory agenda on air pollution rules.

The advisers to the Trump administration, including the chair of the committee advising the agency on air pollution, say they are skeptical of the Harvard study’s methodology and conclusions. Those same appointees have previously raised questions about the strength of connections between fine particle pollution and health effects.

The Harvard study hasn’t been peer reviewed yet, but it has been submitted to the New England Journal of Medicine for review.

96% of nearly 3,300 inmates in four state prisons lack symptoms but tested positive for coronavirus

“To me, this whole thing is a bogus piece of analysis,” said Tony Cox, who chairs the EPA’s Clean Air Scientific Advisory Committee. “It was rushed out without being properly vetted. It’s technically unsound. It has sensational policy implications, none of which are trustworthy.”

“I would have zero confidence in the published results of this study because its interpretation, design, and analysis are fundamentally flawed,” Cox, a risk analyst and president of Cox Associates, said in an interview with the Washington Examiner.

Francesca Dominici, a senior author on the study, said the criticisms are inaccurate and invalid, and pointed to multiple other preliminary studies from researchers around the globe connecting long-term exposure to air pollution and increased risks from the virus.

The criticism comes as the EPA has convened a “rapid review” science advisory panel to inform the agency as it improves its ability to address the environmental and human health effects of COVID-19. The panel first meets Thursday.

One of the research questions they will consider deals directly with whether exposure to air pollution increases susceptibility to respiratory viruses or exacerbates existing COVID-19 infections.

The Harvard study concludes that just a small increase in long-term exposure to fine particulate matter, emitted primarily from fossil fuel combustion, leads to an 8% increase in the COVID-19 death rate. Thus, areas with higher levels of historical pollution are likely to experience higher death rates from the pandemic, according to the research.

Fine particulate matter has been linked to health effects such as nonfatal heart attacks, aggravated asthma, increased respiratory problems, and premature death in people with heart or lung disease, according to the EPA.

There is “definitely enough there to pose a red flag,” said Dominici, who co-directs Harvard’s Data Science Initiative. Several other preliminary studies in recent weeks have also linked higher levels of particle pollution with higher death rates in China and in Italy.

“We have a virus that is affecting our lungs, and I think it’s pretty undeniable that that’s happening,” Dominici told the Washington Examinerin an interview. Not prioritizing addressing air pollution “seems to me a very unwise decision.”

The Harvard study has already made waves in policy discussions. It’s been cited by Democratic lawmakers in letterscalling on the EPA to reverse policy decisions, by environmentalists in a lawsuit challenging the agency’s memo on enforcement flexibility during the virus, and by presumptive Democratic nominee Joe Biden in an updated fact sheet on his climate plan.

The research has also received heightened attention as the Trump EPA has continued its deregulatory agenda amid the pandemic. Just in the last few weeks, the EPA relaxed Obama-era fuel economy standards, reversed the legal underpinning of mercury limits for power plants, and proposed to maintain current levels for national limits on fine particle pollution, despite recommendations from the agency’s career staff to strengthen them.

Trump administration allies say the study shouldn’t be used in policy making decisions given its preliminary nature.

“The study does not appear to be ready for prime time and, if misused by policymakers, could needlessly divert critical health care and regulatory resources from areas that need them most,” a former EPA official told the Washington Examiner.

Stanley Young, a member of the EPA’s Science Advisory Board appointed by Trump officials, raised concerns the report’s language suggests a causal link between fine particulate matter pollution and COVID-19 deaths. The study, though, only concludes an association, he said.

“This is the proverbial butterfly in China causing a tornado in Alabama, and generally the world doesn’t work that way,” Young, a statistician and CEO of CGStat, said in an interview. “The paper is written in a way that strongly implies cause and effect,” he said, adding the authors “could have been a lot more circumspect than that.”

Cox suggested the Harvard study doesn’t control for the “urban-rural continuum,” which takes into account local-level population density as opposed to county averages.

“Everything in the Harvard paper to me appears to be completely consistent with the hypothesis that crowding matters, but there’s nothing to suggest, other than wishful thinking perhaps, that [fine particulate matter] has any effect at all,” Cox said.

Dominici, however, said Cox’s claim isn’t true. The researchers adjusted for population density and conducted separate analyses for urban counties and rural counties, she said. Those separate analyses had “very similar” results in terms of how fine particulate matter pollution affected the COVID-19 death rate, she added.

The researchers also accounted for more than 20 other confounding variables, such as age, stage of the epidemic, date of first confirmed case, and behavioral conditions like obesity and smoking, Dominici said.

Young noted the coronavirus outbreak is an “incredibly fluid” situation. Scientists are learning more about how many people the virus has infected as testing becomes available. Tests might show the virus isn’t as deadly as previously thought, which would affect the data in the Harvard study, he said.

Dominici, though, said a change in the death rate from COVID-19 wouldn’t affect the conclusions of their study. That’s because the researchers use whatever the death rate is as a baseline to examine whether long-term fine particulate matter exposure had an effect on it, she said.

Overall, Cox and Young both suggested the Harvard study doesn’t make a direct enough link between higher levels of fine particulate matter and deaths from the coronavirus to influence policy decisions.

“Now of all times is the time to really pay attention and be very responsible in distinguishing between associations that are quite trivial to produce and don’t necessarily have implications for sound policy, and real valid scientific information empirically backed by data showing how changing exposure changes health consequences,” Cox said.

Dominici, though, said it’s always possible to find a hole in research because no data will ever be perfect. “It’s much harder to provide rigorous science at the time of a pandemic than criticize the science,” she said.