Save Time, Money With Single IRB in Multisite Trials, Suggests NIH

Instead of many, one.

That’s the idea behind a draft policy announced by the National Institutes of Health (NIH) to encourage greater use of a single institutional review board (IRB) to oversee clinical trials in the United States conducted at more than one site.

Historically, most research was conducted at a single site, so only one IRB had to approve the protocol. As studies began to include multiple sites, the usual practice was for an IRB at each site to evaluate the study independently. However, that could mean much duplication of effort and wasted time and money that might be better devoted to actual research. Variations in study protocols might also get introduced in the process, muddying the validity of outcomes.

The NIH draft proposes to use a single IRB of record for any domestic multisite trial “to enhance and streamline the process of IRB review and reduce inefficiencies so that research can proceed efficiently without compromising ethical principles and protections.”

A central IRB also might improve research protections for research participants by reducing diffused accountability and institutional conflicts of interest, according to the draft.

The use of a central IRB is not new. In fact, elements of NIH have used the approach for many years. The National Cancer Institute has employed a Central Institutional Review Board since 1999, and the National Institute of Neurological Disorders and Stroke has used single IRBs for certain clinical trials groups.

What has slowed wider implementation of this approach? Several reasons have been proposed for the continued use of IRBs, starting with habit.

“The data [in a study of IRBs] suggest that experience, not simply increased knowledge, is necessary to allow institutional stakeholders to feel more comfortable with central IRB review,” wrote Kathryn Flynn, Ph.D., an adjunct assistant professor of psychiatry and behavioral sciences at Duke University School of Medicine, and colleagues in the January 2013 PLoS One.

Some IRBs may fear that some distant, central IRB could never adequately manage the local research site’s “unique local context,” particularly its “unique patient populations,” Flynn suggested.

That is not an entirely unfounded concern, said William Narrow, M.D., M.P.H., acting director of APA’s Division of Research: “Some states may have their own, more stringent requirements for human research subject protection that have to be accounted for.”

Flynn also found concerns among IRB members about legal liability if something goes wrong in the trial or loss of revenue generated from IRB review of studies funded by industry sponsors.

“Sometimes when problems arise in a research project passed by an IRB, the board focuses more closely on similar issues when assessing the next trial, a process that can lead to variations,” said Narrow.

Splitting responsibility might address both local worries and the need for efficiency.

The central IRB would be the IRB of record for the other participating sites, says the draft. However, the local sites would be responsible for obtaining informed consent, overseeing the implementation of approved protocols, and reporting unforeseen problems and adverse events.

“Using a central IRB would help speed up research and save time and taxpayers’ money,” said Narrow.

NIH is accepting comments on the draft proposal until January 29. Comments should be submitted electronically to the Office of Clinical Research and Bioethics Policy, Office of Science Policy, NIH, via email at [email protected]; by mail to 6705 Rockledge Drive, Suite 750, Bethesda, Md. 20892; or by fax at (301) 496-9839. ■