In his recent comments to the EPA Science Advisory Board (SAB), EPA Administrator Andrew Wheeler demonstrated that he is doubly blind. He is blind to the definition of a double-blind study, which he completely misinterpreted, and he is blind to the value of accurate scientific advice.
First, some background: The Food and Drug Administration (FDA) depends on double-blind studies for its decisions. But EPA is not FDA, and limitations on the ability of environmental health scientists to do double-blind studies are a major reason.
EPA has proposed a rule, inaccurately calling it a “transparency rule.” It virtually requires EPA to discard any study for which the raw data is not made publicly available from consideration in agency rulemaking. Transparency, of course, sounds like a good idea as does the ability to replicate studies. But those who support the so-called transparency rule implausibly argue that EPA should align with the FDA’s rules for studies submitted in support of new drugs or medical devices. This is totally misleading.
What do health scientists mean by a double-blind study? Imagine a study in which patients with an existing disease are asked to volunteer for a study — a clinical trial — of a potentially valuable new therapeutic drug and the individual is randomly assigned to the group receiving the placebo or the drug. A single-blind study is usually one in which the patient does not know whether she or he is receiving the new drug or a placebo. But the gold standard is a randomized double-blind study in which neither the patient nor the observing health professional is aware of which group the patient is in until the study is completed and analyzed.
Why won’t this work for almost any environmental health studies? Simply put, would you willingly volunteer to be exposed to a pollutant, not to see what good it does, but what harm it might cause?
Environmental health studies look at what evidence is available, generally studies of populations that have been exposed to pollutants as a result of bad practices or ignorance, to determine the health effects of pollutants on humans. We must depend on the published scientific web of multiple studies done by different authors on different populations using different research approaches.
To study pollutant effects, we take advantage of differences in exposure due to geographical factors, such as a source of a pollutant located in one city but not another; temporal factors such as evaluating communities before and after hydrofracturing for shale gas; or idiosyncratic differences in exposure perhaps related to a pollutant-laden tanker overturning in a community.
FDA drug studies are also different in other ways. Consider the many possible difficulties in measuring actual human exposure under environmental situations as compared to FDA studies in which the dose of a new drug and the health impact of concern is specified in the study design. In addition, FDA studies specify the required number of study subjects through a power calculation based upon relatively rigid determinations of the differences among the subject groups that would be meaningful — something much more subject to argument after environmental health studies are completed.
The likely goal of the transparency rule promoted by Wheeler and his industry supporters is to obtain raw data that can be used by paid industry consultants who then nitpick to find minor blemishes in each study. These blemishes are then falsely magnified into issues that seemingly affect the validity of the study, as occurred with the study of the hazards of exposure to formaldehyde.
So how did Wheeler react to criticism that he is misapplying FDA’s approach to transparency? First, it seems clear that his objective is to marginalize science review. At the long-delayed June 5 meeting of the SAB, Wheeler finally allowed the members to consider the transparency rule long after it was released for public comment. Normally, the SAB provides input before the agency issues a proposed rule for public comment. Wheeler is reshaping the process to turn SAB review into just another public comment, contrary to longstanding EPA practice under multiple Republican and Democratic administrations.
Then Wheeler showed his ignorance by saying “FDA routinely uses the double-blind scientific studies for their work where they have two teams of researchers replicating the same data, both teams don’t know each other exist and where they are.” He added, “that federal research would be more accepted by the public if you used the double-blind standard for everything,” providing a rationale that misstates how research on human exposure to pollution is conducted.
The problem with this mistaken interpretation goes beyond its implications for the proposed EPA rulemaking on transparency. The definition of a “double-blind study” is not insider baseball known only to a few technical experts. It is widely known and used within the scientific community. One can be certain that virtually all of EPA’s scientific staff knowledgeable about health science would be able to correctly define the term. It’s likely Wheeler is too aware of his personal scientific limitations to have made this mistake on his own. He arguably needed someone whom he thought was scientifically knowledgeable to feed him this ludicrous misinformation about double blind studies. This is further evidence that EPA leadership is taking advice from political outsiders, not experienced EPA career scientists. One possible interpretation is that Wheeler is a gullible tool of the industry for whom he was a lobbyist before coming to EPA.
This obvious mistake also highlights the failure of Wheeler and former EPA Administrator Scott Pruitt to successfully recruit a nominee for the position, in which I formerly served, of assistant administrator for Research and Development to be submitted by President Trump for confirmation by the Senate. In the past, such assistant administrators had both the scientific knowledge and the gravitas to keep reputable science from being ignored. Unfortunately, the current assault on science at EPA makes it unlikely that any credible scientist would accept this nomination.
Dr. Bernard Goldstein is an emeritus professor at the University of Pittsburgh Graduate School of Public Health and the former EPA assistant administrator for Research and Development appointed by President Reagan.