IRB disclosure policies, review procedures found lacking
■ Many academic institutional review boards don't require reporting of financial conflicts, a new study finds. A federal probe highlights lapses in the IRB process.
By Kevin B. O’Reilly — Posted April 27, 2009
Two new reports shed harsh light on the institutional review boards that are charged with protecting human research subjects from harm.
The Assn. of American Medical Colleges in 2001 recommended that IRB members disclose any financial conflicts, yet 27.4% of review boards still do not require it, according to a survey of 107 IRB chairs reported in the April Academic Medicine.
"We've been talking about conflicts of interest in clinical research for almost 20 years now, and the fact that nearly 30% of IRBs in medical schools don't have policies that require disclosure is completely astounding to me," said Eric G. Campbell, PhD, senior author of the study and associate professor of medicine at Harvard Medical School in Boston.
More than 20% of IRBs lack a written policy that defines conflicts of interest, and there is confusion about whose responsibility it is to determine when a member's relationship with industry constitutes a conflict.
While IRB chairs said members with a conflict never voted on proposed research protocols, a companion survey of IRB members found that 20% said they always voted, despite the presence of a personal conflict.
"IRBs are, frankly, pretty confused about what they should be doing," said Christine Vogeli, PhD, a study co-author and instructor of medicine at Harvard Medical School. "It really speaks to the need to clearly provide guidance that can be very well-digested."
Last year the AAMC and the Assn. of American Universities issued joint guidelines on reporting financial conflicts, saying that any potential conflict should be reported, no matter how small the amount of money involved. The AAMC did not grant an interview request by this article's deadline.
Sting highlights "IRB shopping"
Meanwhile, a Government Accountability Office investigation released in late March found that "the IRB system is vulnerable to unethical manipulation, which elevates the risk that experimental products are approved for human subject tests without full and appropriate review."
The GAO, in a sting operation, set up a Web site for a bogus IRB and heard from a medical research company seeking its services, "even though GAO's investigators had no medical expertise whatsoever," the agency's report said.
GAO investigators also got approval from an IRB for a fictitious medical device with fake specifications that would fall under the Food and Drug Administration's "significant risk" category of devices.
"We basically said we're shopping for an IRB to give us a quick turnaround on a new device," which was presented as legitimate, said Gregory D. Kutz, the GAO's managing director of forensic audits and special investigations.
Two of three IRBs rejected the protocol, describing it as "awful," a "piece of junk," and "the worst I've seen ... too risky."
But Colorado Springs, Colo.-based Coast Independent Review Board LLC unanimously approved the research protocol. "It was pretty much a rubber stamp," Kutz said.
Coast IRB CEO David Deuber said in a statement, responding to the GAO report, that "across-the-board changes to procedures governing intake and evaluation of clinical trials are needed." The company stopped considering new research protocols for 30 days -- through May 1 -- "as part of an intensive review of Coast IRB's [standard operating procedures] and internal processes," the company's Web site said.