The rigorous vetting of unpublished research by independent, qualified experts—what’s often called “peer review”—is an undisputed cornerstone of modern science. Central to the competitive clash of ideas that moves knowledge forward, peer review enjoys so much renown in the scientific community that studies lacking its imprimatur meet with automatic skepticism. Academic reputations hinge on an ability to get work through peer review and into leading journals; university presses employ peer review to decide which books they’re willing to publish; and federal agencies like the National Institutes of Health use peer review to weigh the merits of applications for federal research grants. When members of Congress make an end run around this vetting process and pump R & D cash directly into their home districts, they’re widely disparaged for supporting a particularly odious and anti-scientific version of pork-barrel politics.
Given all this, you might expect that a recent White House Office of Management and Budget proposal to expand the use of peer review in the evaluation of scientific research conducted by federal agencies would find a warm welcome from scientists. You’d be dead wrong. Scientific heavyweights like the American Public Health Association, the Association of American Medical Colleges, and the Federation of American Societies for Experimental Biology have issued scathing critiques of the proposal (the latter two jointly), as have a range of other organizations and experts. The hallowed American Association for the Advancement of Science—which publishes the preeminent peer reviewed journal Science—also has worries about the idea. A group of Democratic members of Congress even dubbed it a “wolf in sheep’s clothing.” What’s going on here?
The trouble is that peer review comes in many forms, and the devil is in the details. For instance, while scientific journals can usually take their time in peer reviewing article submissions, serious consequences may result from hampering the ability of federal regulatory agencies, like the Food and Drug Administration or the Environmental Protection Agency, to make science-based determinations. Similarly, while academic peer review merely affects individual reputations, the peer review of government regulatory information implicates the economic fortunes of numerous major corporations—especially big polluters who want environmental and public health rules softened. These companies would love to influence scientific determinations made by agencies that regulate them, and have plenty of money at their disposal for lobbying and litigation. Any attempt to change the current system of regulatory peer review thus has major economic and political implications.
The specific peer review proposal in question—published last September in the Federal Register by the Office of Management and Budget’s Office of Information and Regulatory Affairs (OIRA)—begins from the premise that OIRA has a newfound role in improving the quality of scientific information released by various wings of the federal government. This authority, allegedly based upon the so-called Information Quality Act (IQA) of 2001 (previously known as the ”Data Quality Act”), has been widely contested; many government science watchers consider the IQA an anti-democratic ruse designed to help corporations undermine government science that hurts them in the pocketbook. The suspicion arises because the two paragraph “Act,” allegedly drafted by a corporate lobbyist, was quietly inserted into an appropriations bill by a single Republican legislator with most of Congress unawares.
Nevertheless, the tiny but influential OIRA—whose job is second-guessing rules promulgated by federal agencies, and which is sometimes called the “black hole” of the regulatory process—appears hell bent upon asserting its new science cop powers. Noting that peer review can provide a “vital second opinion on the science that underlies federal regulation,” but charging that federal agencies often apply it insufficiently, OIRA wants to institute a new, government-wide apparatus to scrutinize key scientific information that forms the basis of regulatory policy. First of all, agencies would be required to rigorously peer review all “significant regulatory information” that’s intended for public dissemination. And for “especially significant regulatory information”—defined as data “with a possible impact of more than $ 100 million in any year”—additional hurdles would exist. In a laborious-sounding “formal” process, carefully chosen reviewers would have to sift through public comments before submitting final peer review reports—to which the regulating agency would, in turn, have to respond. The head of OIRA, currently the former Harvard professor John Graham, would decide when information counts as “especially significant”; in such cases, agencies would have to consult with OIRA to ensure they’re conducting adequate peer review.
Perhaps because it promises to slow down the regulatory system, corporate-sponsored groups have celebrated this policy proposal. Scientists, however, denounce the idea as an overbroad and expensive plan likely to stymie agency efforts to protect public health and the environment. Most pointedly, critics are outraged that OIRA’s proposal blocks academic scientists from serving as reviewers if they have received research funding from the government agency in question (a condition likely to exclude leading academic experts in the field), yet fails to bar the participation of industry scientists (who are, if anything, far more conflicted). Scientists also fear the proposal would place control over the new peer review structure in the hands of a political agency, the White House Office of Management and Budget, at a time of increasing politicization of science across the federal government. It hardly helps that Graham’s nomination to his post was itself highly controversial, as environmental groups attacked the corporate ties of his Harvard Center for Risk Analysis, as well as Graham’s own commitment to a form of cost-benefit analysis that critics consider arbitrary and industry-friendly.
Alongside these worries, the inevitable negative peer reviews produced in this process could also bolster the legal cases of corporations engaged in never-ending litigation to block environmental or other rules. Additional peer review procedures wouldn’t just consume time and money; they would also create novel opportunities for interested parties to raise challenges at various stages of the process, such as over the selection of reviewers or the composition of their reports.
In this sense, the whole OMB “peer review” episode, which has both garnered significant media coverage and mobilized the scientific community, says a great deal about present day, state-of-the-art strategies for politicizing science. After all, just as no one opposes “data quality” or “information quality” in the abstract, everyone agrees that peer review represents a noble goal. But such lofty objectives can be easily subverted, if you’re clever or cynical enough. That’s exactly what’s happening with the OMB proposal: An established and well respected scientific norm is being used to actually undermine science.
In fact, “peer review” turns out to be highly susceptible to such subversion. For one thing, it’s only as reliable as the individuals chosen to serve as “peers” and the formal process within which peer review takes place. Peer review doesn’t automatically serve as a “guarantor of truth,” note Rutgers University science policy scholars Stuart Shapiro and David Guston, and it also won’t necessarily quench controversy in highly politicized scientific areas. Much regulatory information is already controversial, due to the amount of money and, in some cases, the number of lives at stake. Increased scrutiny of such information could simply fan the flames of disputation and prevent government action to protect the public health.
Furthermore, industry groups have already demonstrated a canny ability to exploit the existing academic peer review process to their own advantage. A telling example comes from the testimony of George Washington University epidemiologist David Michaels, a former Clinton administration Energy Department official and chief critic of OIRA’s “peer review” proposal:
Earlier this year, scientists hired by the beryllium industry published a re-analysis of a study done by CDC [Centers for Disease Control] scientists, in which, by changing some parameters, the statistically significant elevation of lung cancer in beryllium-exposed workers was no longer statistically significant. It was published in a peer reviewed journal—never mind this is not a journal that published much epidemiology. But it was peer reviewed, and has the opposite conclusion of other peer reviewed studies. The beryllium industry is now promoting this study as evidence that NIEHS [National Institute of Environmental Health Sciences] and IARC [International Agency for Research on Cancer] are wrong.
OMB’s new peer review proposal would only facilitate this kind of clever assault on well established scientific information. Indeed, corporate groups that generally support the OMB proposal, like the Center for Regulatory Effectiveness (CRE), actually object that the proposal gives too much respect and credence to standard academic peer review. The CRE’s demand for “peer review” in the government regulatory sphere thus appears less based on a respect for science itself than on a desire to open up the process to a wide variety of potential objections and challenges.
So while “peer review” may be important as a direction or ideal, it’s not a process that can be automatically and uncritically trusted to strengthen science. In closing, another telling example of the ambiguities inherent in peer review comes to mind. Last year the Stanford Law Review, a non-peer reviewed journal, published a devastating critique of economist John Lott’s famed “More Guns, Less Crime” hypothesis, which had generally been supported by a series of peer reviewed studies that, in retrospect, seem severely flawed. Not surprisingly, Lott now criticizes the Stanford Law Review study for not being peer reviewed. But the substance of the argument is what actually counts, and economists seem increasingly convinced that Lott has lost on the merits—peer review notwithstanding.
Perhaps there’s a lesson there when it comes to the OMB/OIRA proposal. Before automatically embracing “peer review,” we should be exceedingly careful to determine what the phrase actually means, who’s using it—and why they care.