Against Armchair Ethics: Some Reflections from Petrie-Flom

As followers of my Twitter feed will know, I spent Friday and the first half of the Saturday at The Future of Human Subjects Research Regulation, a conference sponsored by the Petrie-Flom Center for Health Law Policy, Biotechnology, and Bioethics at Harvard Law School.

The conference organizers, led by Professor Glenn Cohen, did a magnifent job bringing together participants with varied views and backgrounds, and as a group we discussed ideas ranging from treating research subjects as workers to including adolescents on IRBs. Those who want the full story should be able to watch videos of the presentations before too long.

I was struck, however, by one recurring theme: the distinction between evidence-based approaches and armchair ethics.

Some examples:

• Osagie Obasogie argued that the Institute of Medicine's report on prison research is inferior to the National Commission's 1976 report on the grounds that the National Commission members visited more prisons, while the IOM authors relied more on the writings of ethicists.
• Ellen Wright Clayton argued that the ANPRM's proposals for biospecimen privacy--particularly the "standard, brief general consent form allowing for broad, future research"--would not adequately balance subjects' wishes and researchers' needs. Her strongest argument against such an approach was empirical: at Vanderbilt, 12 percent of clinical patients opt out of the BioVU DNA databank, evidence that a more layered yet still workable system of choice is up and running. She thinks the risks of inappropriate reidentification of biobank samples much smaller than other privacy risks.
• Michael McDonald, Susan Cox, and Anne Townsend reported on the findings of their multi-year study in which they asked research subjects about their experiences and found that participants did not use ethics-committee terms like risks and benefits, but instead discussed the "impacts" of research, both positive and negative. (McDonald also referred conference participants to their website, QI4 Research Ethics, which has survey tools for soliciting participant views.)
• Carol Weil noted that while the ANPRM frets about privacy, that's not the main concern of people with rare diseases. They want research on those diseases, and they want to share information when they think it will lead to better medicine.

The problem is that the current regulatory system is not set up to get empirical information into the hands of ethics boards, researchers, or subjects. McDonald declared himself "astounded" at the resistance to evidence-based ethics review, and he was clearly dismayed when Laura Stark told him that the only time the IRBs she observed soliciting participants' views was as a "punitive" action against a researcher. If you don't look at outcomes, McDonald noted, "there's no such thing as getting it right."

Cynical as I am, I am not astounded. If IRBs existed to give researchers the information they need to conduct ethical research, we wouldn't have the mortifyingly stupid CITI Program. And if one understands the IRB's primary role as protecting an institution's "eligibility to receive federal funds for research," then there's not much reason for IRBs to take the trouble to learn about how they offend research participants.

Clayton and McDonald both want to preserve prospective oversight for social science research; I do not. They think it appropriate to deploy the coercive power of the state in an effort to improve research ethics; I consider coercion counterproductive except for avoiding true atrocities.

But I think we agree on more than might be apparent. We are not fans of the current system, we consider TCPS2 an advance over both TCPS1 and the Common Rule, and we would like to see additional reforms based on what we have learned over the decades.

If policy reform were an ongoing process, we could try easing up on review or abolishing it and see what happens. But since the proposed reforms, if they go through, may be the last in our lifetimes, the high stakes exaggerate differences of opinion.