On April 30, 2018, the Environmental Protection Agency (EPA) published in the Federal Register a proposed rule titled, Strengthening Transparency in Regulatory Science. The rule would prohibit the EPA from basing regulatory action on scientific studies for which the underlying raw data and models are not publicly available “in a manner sufficient for validation and analysis.” Comments on the proposed rule are due May 30.
According to the notice, the proposed rule is “designed to increase transparency in the preparation, identification, and use of science in policymaking.” On its face, this sounds good. Many of us support the idea of increased transparency in science—and, in particular, data sharing—as a means to shoring up accountability for the use of public funds and resources, promoting reproducibility, augmenting the impact of resources spent on science and research, and enhancing public trust in and support for science as an endeavor that helps to improve human lives. Indeed, many of these concepts are specifically cited in the proposed rule notice.
However, many in the scientific community are expressing deep concern that this proposed rule is actually a way for EPA to limit the types of science it can and will use in regulatory decision-making, to the detriment of environmental policy and the public’s health.
More specifically, critics argue, this move would disqualify from use in policymaking a lot of high-quality, peer-reviewed science for which the underlying raw data has not been made publicly available for legitimate reasons. Doing so would restrict the EPA, many worry, from having access to and considering a complete body of research, and from basing environmental policy on the best available scientific evidence. It’s worth noting that much existing science for which original data sets were not maintained or were stored in outdated formats would be eliminated under the proposed rule. Critics also argue that the rule is a solution to a problem that doesn’t exist, given how transparent EPA already is about the scientific bases for its policy decisions.
Reproducibility is cited as a driving concern behind this and other efforts over the years, by the agency and Congress, to restrict the types of science EPA will consider. But as an April letter to EPA Administrator Scott Pruitt from the Union of Concerned Scientists noted, “many public health studies cannot be replicated, as doing so would require intentionally and unethically exposing people and the environment to harmful contaminants or recreating one-time events (such as the Deepwater Horizon oil spill).” Furthermore, we disrespect the contribution of those who lived through such circumstances and agreed to take part in research so that we might gain important knowledge from their experiences, if that research is then unavailable as a basis for future policies.
These aren’t the only human subjects protections concerns raised by the proposed rule. Much of the research that has informed or could inform EPA policy has collected sensitive health and other information about individuals, sometimes over a long period of time. People often agree to be in studies on the understanding that their information will be kept confidential and their privacy protected—indeed, that is one reason that data underlying much of the most important existing science in this area is not currently publicly available. The proposed rule will thus put scientists doing work that could inform important environmental and health policy in a bind: either make their science available to policymakers, or keep the promises made to their study subjects when they collected their personal information.
The EPA has said it will redact such private information, but it will have to spend “hundreds of millions of dollars” to do so. Not only would such redacting slow down the rulemaking process and use significant resources, some argue that, given the nature of much epidemiological and other public health research, such redacted data will not be useful for future analysis. There is of course also the concern that if people know their information will be shared, even if they are told their privacy will be protected, they may worry about privacy breaches and be reluctant to participate in important research.
Finally, there are questions about who will have access to the data that is made available if the rule goes through, and what they will do with it. The new rule asks for input on mechanisms and strategies for making data publicly available in a way that properly protects confidential information. Some suggest there should be restrictions on who can access the data to validate the findings, and mechanisms for vetting data access requests. From an ethical perspective, whoever has access to the data must be held to the same standards for use of personal data that were agreed to by the participants when they were recruited into the study. How that ethical requirement can be operationalized within the scope of this EPA rule remains to be seen.
EPA is accepting public comments on the proposed rule until May 30, 2018, and so far, over 42,000 comments have been submitted. If you want to share your thoughts with the EPA, you can submit comments by going to regulations.gov and searching for Docket ID No. EPA-HQ-OA-2018-0259 and clicking on “comment now.”