Recordings of IRB meetings show gaps in discussions
Don't always discuss all Common Rule criteria
There have been many surveys of IRBs and their operations — gauging how long protocols take in their journey through review, how much it costs to operate a human subjects protection program, and IRB members' attitudes about various aspects of their work.
But it's hard for those types of studies to reveal how protocols are actually reviewed on a day-to-day basis in IRBs across the country.
A recent study published in the journal Academic Medicine gives us an unusual look behind the closed doors of IRB meetings, to see what gets discussed in review — and what doesn't.1
Charles Lidz, PhD, a professor of psychiatry at the University of Massachusetts Medical School in Worcester, and his colleagues gained permission from 10 academic medical centers to audio-record a total of 20 IRB meetings, along with reviewing the 104 study applications discussed during those meetings.
Lidz's team analyzed the recordings to see whether IRBs were discussing all of the criteria required by the Common Rule: risk minimization; risk/benefit comparison; equitable subject selection; informed consent; data monitoring to ensure safety; privacy protection and confidentiality; and protection of vulnerable subjects. Lidz says the team determined whether the elements were either adequately addressed in the application or discussed during the meeting.
Equitable selection
The results showed that while some elements were routinely addressed, others surfaced less often in discussions. For example, the team found that equitable subject selection was adequately addressed in only half of the submitted applications. IRBs raised the issue in only 40% of the remaining studies.
In contrast, 98% of the reviews included discussion of informed consent.
"I think there are a number of these things that are simply not thought about very much, particularly the equity of including or excluding groups," Lidz says. "Sometimes, for example, the issue of protecting vulnerable populations gets in the way of equitable subject selection."
He notes that studies sometimes exclude a group without explaining why it is necessary to do so and that decision often is not questioned by the IRB.
"We found a number of places where they said, 'We're excluding all vulnerable subjects,'" he says. "That takes thinking. These are complicated questions — do you want to include them in a trial or are we excluding them for the right reason? Those questions about vulnerable subjects and equitable subject selection tend to take a second place, or sometimes simply get ignored. And I think that's too bad."
The team found that only 13 of the 104 total applications adequately addressed a risk/benefit comparison, and IRBs only discussed that criterion in 39 of the remaining studies. Lidz says this generally involved a failure to consider potential benefits of a study alongside its risk.
"A lot of the time, the benefit is taken for granted — that the research is a good thing," he says. "You wish there was a closer look at that: Is this research going to learn anything? But that's a really hard question for an IRB to tackle successfully."
Checklist may help
IRB discussion in the meetings did not always lead to a decision on a particular element.
In the area of risk minimization, for example, the team determined that 82 of the 104 applications lacked enough information about minimizing risks. IRBs discussed it in 65 of those 82 cases, but in two of them, the issue was raised in the meeting without being resolved.
Lidz says his group interviewed the IRB staff about their work on the applications. "But the way we tended to see this was that the committee really needs to assess these things."
He says the solution may be as simple as a checklist kept by the IRB chairman to ensure that each element is addressed and decided before moving on.
"I don't think this is rocket science — it's just making sure we've gone over these things because that's what's expected of us," Lidz says.
And he does believe some of the work of addressing these criteria can be handled by staff, as long as there is oversight of some sort by the IRB.
"If they decide, for example, 'We don't want to talk about informed consent, we'll let the staff do that,' that's okay, if they have a policy and a review procedure where the chair reviews it," he says. "You can imagine privacy and confidentiality protections being a carefully designed checklist of what the application has to say, and have that be a staff-reviewed process."
That would leave IRBs with more time to devote to important questions about safety and risk, he says.
Lidz says the team did not set out to compare the IRBs in the study to one another, or to tell the IRBs involved that they're doing a bad job.
"We were consistently impressed with the quality of the discussions — they're thoughtful, they're serious," he says. "They just sometimes don't cover everything."
Reference
- Lidz CW, Appelbaum PS, Arnold R, et al. How Closely Do Institutional Review Boards Follow the Common Rule? Acad Med 2012 May 22. (epub).
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