Remodelling IRBs

Hastings Center Report, The, July-August, 1998 by Joanne Silberner

The first major overhaul in the design of institutional review boards since they came into being in the 1970s could be on the horizon. There's been a steadily increasing hum of concern about IRBs in Washington. In 1995, a presidential advisory committee questioned the adequacy of IRBs. So did a 1996 General Accounting Office analysis. The GAO study detailed a number of abuses, including one case in which three times the number of people were enrolled in a trial as were approved by the IRB. The National Bioethics Advisory Commission has been working on its own report, and is expected to recommend, among other things, more education for IRB members and a more thorough informed consent process.

The hum grew louder in June when the Inspector General (IG) of the Department of Health and Human Services released its own critique, which included a summary ominously titled "A System in Jeopardy." The IG was careful to note that investigators found no widespread abuses, but made it clear that there is reason for concern. The IG found that IRBs are overworked and in danger of becoming an anachronism. The result is that potential subjects may not understand the distinction between research and therapy.

The IG's office reviewed federal records, interviewed government officials and IRB members, visited six academic health centers, attended IRB meetings, and went with FDA inspectors on site visits. Among other things, the report concludes that IRBs "review too much, too quickly, with too little expertise." They don't follow up on research once it's started. Board members face a conflict of interest as members of the institutions that will ultimately benefit from the research if it's approved. Most board members are not well trained, and boards do little quality control.

The recent emergence of independent boards, most of them for profit, deals with some of the concerns, according to the IG's report. With a paid rather than volunteer staff their turnaround is much quicker--11 days as opposed to 37 days at academic centers. Since the staff is not drawn from the institutions being judged, there is less chance for a conflict of interest. And having one IRB simplifies review for multi-site trials.

But independent IRBs raise concerns of their own. One of the strengths of conventional IRBs, the IG notes, is that they're familiar with the institution, the investigators, and the community--not necessarily true for an independent IRB, which may try too hard to please its clients by approving questionable trials. And an institution that's turned down by one commercial IRB may simply try another with a reputation for being more lenient.

All this concern is aimed at a system that was designed for the way clinical research was done in the 1970s. As the IG report observes, it used to be a matter of a government grant going to a chief investigator for a relatively small study at a single institution. In recent years, the number of trials going on at any one institution has increased, and multi-institutional studies are common. If the NIH budget doubles within the next five years, as politicians in Washington are calling for, there would be a five-fold increase in the number of people in cancer trials, further burdening IRBs.

There's already been some action at the state level. A New York court recently ruled that incapacitated people could not be enrolled in any trial that has a nontherapeutic element (see HCR, November-December 1997 for a full discussion). While the case was brought on behalf of six people with mental illnesses, the ruling could be applied more broadly.

The IG report offered up a series of suggestions, mostly calling for more: More nonscientific and noninstitutional members on the boards. More independence. More resources. It also called for dropping some (unnamed) procedural requirements and for instituting quality controls. Bioethicists got a plug, with a caveat. They were seen to "help broaden an IRB's perspective" and to be "sensitive to subtle yet significant ethical issues." But at the same time, the report noted that they're "more likely to focus on obscure ethical concerns."

So with a presidential commission, the General Accounting Office, the Health and Human Services IG, and the National Bioethics Advisory Commission all calling for change, how soon is something going to happen? Don't expect much from Congress in the short run. Not that there's opposition, but they're busy with re-election campaigns. The only health care issue on Capitol Hill as of this writing is what sort of rights consumers should have in managed care. Without a scandal, protecting research subjects isn't a vote-getter. The Department of Health and Human Services is already in the process of beefing up its human subject protection regulations regarding fetuses and newborns. The big push for modernizing IRBs is likely to come from there.

Joanne Silberner is a Kaiser Family Foundation Fellow on leave from National Public Radio.