Study analyzes impact of investigator attendance on IRB review efficiency
Findings are mixed
IRB members have different opinions and policies on whether to permit investigators to attend IRB meetings. But are these policies based on any evidence that one way works better than another?
Maryland researchers decided to examine this question with a study that evaluated the effect of investigator attendance on IRB review efficiency.1
"Some IRBs in this country have the main researcher come and talk to the committee as they do their deliberations, and other committees do not and learn about the study through a paper trail," says Nancy E. Kass, ScD, a Phoebe R. Berman professor of bioethics and public health at Johns Hopkins University's Berman Institute of Bioethics and Bloomberg School of Public Health in Baltimore, MD.
"We were interested as authors who have sat on IRBs or staff IRBs in what kinds of differences it makes," Kass explains.
So Kass and co-investigators conducted a retrospective review of 125 protocols submitted to four IRBs at Johns Hopkins Medical Institutions between March 12, 2002, and June 30, 2005. Two of the IRBs invited principal investigators (PIs) to meetings, and two did not.1
They tested hypotheses about the IRBs' efficiency in handling the protocols, singling out the factor of principal investigator attendance at the IRB meetings.
"One hypothesis would say that bringing a principal investigator to the IRB meeting would result in a more efficient review process," says Holly A. Taylor, PhD, MPH, an assistant professor in the department of health policy and management at Johns Hopkins University's Bloomberg School of Public Health.
For example, it could be argued that having a PI at the meeting would enable IRB members to ask questions directly and obtain answers at the meeting rather than having to delay approval of a protocol until IRB staff obtain answers to those questions through phone calls, letters, or e-mails, Taylor says.
"But you could flip that around and say bringing a PI to the IRB meeting would make the process less efficient," Taylor says. "If you had a very busy academic clinical investigator come to the IRB meeting, then this could add time at the front end of the process, and that might result in some inefficiencies."
What investigators found was that there is no simple answer to this question. There appeared to be no difference in review efficiency between the IRBs that invited PIs to attend and those that did not. But there was a sizable increase in efficiency at one IRB when a PI was present, according to historical data.1
A change of course—no longer inviting PIs
"It happened that our natural experiment developed even more opportunity for study because — completely unrelated to our interest — one of the committees that had always had a PI come to meetings decided to change the approach and not have PIs come," Kass says. "So we had a comparison of two IRBs that invited PIs and two IRBs that did not, and we had one IRB that was compared to itself before and after the change."
When efficiency factors were compared for the two versus two IRBs, there wasn't a significant difference in how long it took for the committee to make its final decision, Kass says.
"But when we looked at the IRB that changed its approach, it was quite different," she adds. "When they had a PI there it was quite a bit more efficient."
Ideally, such a study would randomize protocols sent to IRBs with half going to IRBs that have PIs present and half going to IRBs where the PI is not present, Taylor notes.
"That would help control for some factors, such as the IRB chairs' preference for how they like to do their work," Taylor adds.
For the retrospective review, investigators looked at these indicators of efficiency: review time to approval, number of pieces of correspondence, and number of IRB reviews before approval.1
For all of the 125 protocols included in the study, the mean time from date of submission to date of approval was 75 days; the median was 64 days. Also, there was an average of 5.6 letters and e-mail messages between the IRB and PI, and the average protocol was reviewed at 1.8 IRB meetings before it was approved.
There were no significant differences in these averages between the two groups of IRBs.
However, for the one IRB that changed its policy from permitting PI attendance to not inviting PIs, the approval time was considerably longer when the PI did not attend. The mean number of days to approval was 70 when the PI attended and 114 days on average when the PI did not attend.
"If I was the administrator of a large academic medical center, similar to Hopkins, and I thought about whether to bring in PIs, I have no reason to think it makes the process less efficient," Taylor says. "We know it likely does not lead to a more inefficient system, and there may be other benefits to having the PI present."
For instance, IRBs that invite PIs to attend might make their review process more transparent to the PI, and this could improve IRB-PI communication and relations, Taylor says.
"The PI gets to see how the IRB has good intentions," Taylor adds.
Different strokes for different folks
It's possible that a larger study would show significant differences in efficiency, Taylor notes.
But it's also possible that IRBs that don't invite PIs to attend use different methods of obtaining information in an efficient manner.
Prior to IRB review meetings, IRB staff and chairs could send e-mails to investigators who have gaps in their protocol submissions. When investigators answer these questions, their answers are included with the protocol at the IRB review, Taylor explains.
"That might be as efficient as having the PI come, and that may be one reason why we didn't find a difference," Taylor says.
"It turns out that in one of the IRBs [studied] where the PI was not routinely invited to attend, the IRB chair took it upon himself before each meeting to personally call many of the investigators to get a lot of questions answered," Kass explains. "That struck him as one way to get an efficient and thorough review."
The IRB chair believed that by getting questions answered early it might be more efficient than having a back-and-forth exchange through e-mail, and he believed that calling the PIs himself was more efficient than having them answer questions during committee time, Kass adds.
"That is not only another efficiency approach, but it might have contributed to why those committees [that didn't invite PIs to meetings] weren't slow," she says.
IRB administrators who do not believe their IRBs have good communication and public relations between the board and investigators might consider inviting PIs to IRB review meetings to make the process more transparent and possibly improve their relationship, Taylor suggests.
"I'm a researcher and ethicist so I want to be careful to not overreach in terms of this very small study," Kass says. "But it's the kind of study that suggests it would be important to repeat this research with much larger numbers of IRBs and see if there's a more consistent answer."
Future research could focus on other outcomes related to having PIs attend IRB meetings, Kass suggests.
"For example, would this impact whether members of the IRB feel like they understand the protocol better and therefore feel like they are conducting an even higher quality review?" Kass asks. "And I'd love to know whether having the PI present changes the relationships and the public relations between the IRB and the researchers, because there is a human interaction."
To test these hypotheses there would need to be a larger study that includes interviews with IRB members and PIs to assess their feelings toward each other and to test whether they understood the protocol and their sense of satisfaction, Kass says.
"Our study was intended to be a first look, a pilot study that leads me to think it'd be important to go on and do a larger study," she adds. Reference
- Taylor HA, Currie P, Kass NE. A study to evaluate the effect of investigator attendance on the efficiency of IRB review. IRB2008;30:1-5.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content