In August, the RAND Corporation, a policy research organization, released the first comprehensive profile of IRB demographics and function in 20 years, investigating questions about IRB practices, efficiency, and effectiveness on the cusp of implementation of the revised federal regulations, including single IRB review. (In December of 2017, the investigators solicited PRIM&R’s help to get word out to potential survey respondents.)

The “Profile of Institutional Review Board Characteristics Prior to the 2019 Implementation of the Revised Common Rule,” which we’ll refer to as the RAND Report, surveyed two groups of respondents: IRB representatives (further split into chairs, administrators, and members) and PIs. The survey included questions about the types of research protocols processed by each board, how the various parties viewed the effectiveness of IRBs in protecting human subjects, the impact that single IRB regulation would have, and much more.

Part of the investigators’ aim was to follow up on a 1998 evaluation of the implementation of human subjects protections legislation; that study would come to be known as the “Bell Report,” the first systematic, large-scale evaluation of IRB operations. Many of the findings of the RAND Report are juxtaposed against the Bell findings, despite differences in the methodologies of the two reports (a difference that stems mostly from a lack of agreed-upon measures of IRB practices, efficiency, or effectiveness.) This juxtaposition illustrates the evolution of the responsibilities and functions of IRBs in the era of the Common Rule, and lays the groundwork for further evaluation once the revised Common Rule has fully taken hold.

Study Conclusions

Since the Bell report was published in 1998, the approximate number of IRBs, IRB personnel, and protocols reviewed has increased substantially. While this fact is not surprising, the actual numbers are impressive: at the time of the RAND survey, the number of IRBs had increased from 491 sampled in 1998 to 3,411 IRBs registered with OHRP. IRBs in 1998 conducted 284,000 reviews; in 2017 OHRP reported 818,125 active protocols. And the approximate number of IRB members conducting those reviews increased from 6,923 to 30,229. The RAND authors also noted the strong variation among IRBs and the institutions they serve. For their analysis, they divided IRBs into five general categories (university medical, university social/behavioral, hospital, government and private research institutions, and independent/central), and found substantial differences in the number and types of protocols these IRBs reviewed. This variability among IRBs, according to the researchers, “suggests a need for strong support in terms of providing information” about regulatory, compliance, and ethical issues, especially for low-volume IRBs.

The study also found that IRBs are taking on more and more responsibilities beyond what is expressly required of them by the regulations, including functions like HIPAA review, multisite IRB coordination, and ethics review for research not involving human subjects. Attitudes varied among respondents about the degree to which this expansion of responsibilities existed, and moreover about how appropriate it is, but most agreed that it had happened to some extent.

A majority of IRB representatives (defined to include members, chairs, and administrators) and PIs alike believe their IRB “runs with reasonable efficiency,” and this figure is trending upward across all respondents. Additionally, fewer than half of PIs “reported perceived problems obtaining IRB reviews of their protocols” and “nearly two-thirds agreed or strongly agreed that the IRB views its role as being a PI’s ally rather than as a hurdle to clear.”

Regarding specific changes brought by the revised Common Rule (including new exemption categories, single IRB review of multisite studies, no continuing review of expedited studies, and others), expectations are decidedly mixed. Generally speaking, the further respondents were from the day-to-day operations of the IRB—ranging from PIs (most distant) to IRB members, to chairs, to administrators (the closest)—the more likely they were to believe that the regulatory changes would happen efficiently and benefit human subjects protections. Though not entirely without optimism, IRB administrators demonstrated skepticism that the transition to the new regulatory environment will bring positive changes.

Other findings of note

• Only 35% of PIs agree or strongly agree that “IRBs should play a key role in ensuring diversity among study participants.” Nearly twice the proportion of IRB representatives (61%) agree or strongly agree that this is a key IRB function.
• IRB administrators “claim greatest familiarity” with the revised Common Rule: 26.1% of administrators say they’re extremely familiar with the new regulations, while only 7.1% of chairs and 2.7% of members said the same about themselves.
• Two-thirds of IRB administrators believe that single IRB review for multisite research will make IRBs less efficient, or will have no impact on efficiency. Only around one-third of IRB chairs and members are similarly skeptical. Nearly 40% of administrators think the change will weaken human subjects protections.

Do these findings match your experience? Which survey responses surprised you?

The full report is free to read and can be found here.