Unpacking the NPRM: Single IRB Review and Continuing Review

Elisa Hurley

On September 8, the Department of Health and Human Services (DHHS) issued a Notice of Proposed Rulemaking (NPRM) for revisions to the Federal Policy for the Protection of Human Subjects, or “Common Rule.” If adopted, the proposals in the NPRM will result in the most substantive revisions to the core regulation governing federally funded human subjects research in the United States since 1981.

In earlier posts, we explored a new category of activities that would be excluded from oversight under the Common Rule and proposed changes to the category of exempt research. We also reviewed the NPRM’s changes to informed consent and its broad consent provisions for biospecimens research.

In today’s post, we examine the NPRM’s proposed changes related to institutional review board (IRB) review.

Overview of NPRM’s Changes to IRB Review
The NPRM proposes two significant regulatory changes related to IRB review.

The first is the elimination of continuing review for certain studies. Unless an IRB decides otherwise, continuing review would no longer be required for: (i) research eligible for expedited review; (ii) research that has progressed to the point that it only involves data analysis (including analysis of “identifiable private information”) and/or obtaining “follow-up clinical data from procedures that subjects would undergo as part of standard care for their medical condition”; (iii) and research reviewed in accordance with a limited IRB review procedure [§__.109(f)].

As a safeguard, the NPRM states an IRB must obtain a confirmation each year that the aforementioned research is ongoing and nothing has changed that “would require the IRB to conduct continuing review” [§__.109(f)(2)]. Citing institutional liability concerns, it also states investigators would still have to submit protocol changes to the IRB (DHHS 2015, 53985). If the aforementioned conditions at §__.109(f) are met and the IRB chooses to conduct continuing review anyway, they must document the rationale for their decision [§__.115(a)(8)].

The second major change is a mandate that any institution located in the U.S. that is “engaged in cooperative research” must rely upon approval by a single IRB for the portion of the research that is conducted in the United States [§__.114(b)(1)]. The single IRB will be chosen in most cases by the funding agency or, if there is no funding agency the principal institution conducting the study (DHHS 2015, 53984).

There are two exceptions to the NPRM’s new mandate. The single IRB requirement does not apply to cooperative research where the law requires more than a single IRB review of the study. The federal department conducting or providing support for the research may also determine that use of a single IRB “is not appropriate” for the given study [§__.114(b)(2)].

Analysis of Proposed Changes
The proposal to eliminate the continuing review requirement for most minimal risk research, and for studies that have reached the stage of analyzing data will be welcome as providing both IRBs and researchers much-needed relief from administrative burdens that do not enhance subject protections. On the other hand, as DHHS points out in the preamble’s discussion of the comments it received on the ANPRM, the single IRB mandate has divided the research community (DHHS 2015, 53983).

The practice of each institution conducting its own IRB review in multi-center trials has been criticized as inefficient and unnecessary. Thus, the stated aim of the single IRB mandate for cooperative research is to avoid redundant review and associated inconsistencies and delays (DHHS 2015, 53982). However, establishing the authority of the central and participating IRBs, the scope of their respective responsibilities (around, for instance, adherence to local law, compliance functions, training, and conflict of interest review), and the means for inter-institutional information-sharing and cooperation, may be costly and time-consuming. Indeed, the NPRM preamble acknowledges that in response to the ANPRM, many commenters expressed concerns about the lack of clear expectations regarding what a local IRB would be responsible for if a single IRB mandate was implemented (DHHS 2015, 53983). Rather than eliminating administrative burdens, single IRB review may create new demands and inefficiencies.

Moreover, research institutions will face heterogeneity among protocols approved by different external IRBs but conducted within their walls. Over the past decade, institutions have improved their programs to promote the ethical conduct of research and increased their ability to conduct high quality review. This often involves the development and dissemination of consistent policies and practices reflecting institutional consensus. In contrast, under the proposed mandate, an institution’s investigators will be accountable to multiple institutions for their systems for protocol submission, requirements for review, standard operating procedures, and ethical guidelines. For example, an investigator overseeing three cooperative research projects under the purview of three different external IRBs may have to apply and enforce differing subject payment rules, ages of assent, and study exit criteria.

Shifting the locus of ethical oversight and responsibility to other institutions or to commercial IRBs also threatens to diffuse institutional responsibility and undermine shared commitment to the very standards that are fundamental to an institution’s culture of accountability, integrity, and pride in its research. Although the NPRM states the proposed changes would not prevent institutions from administering their own “additional internal IRB reviews,” it also notes that “such reviews would no longer have any regulatory status in terms of compliance with the Common Rule” and a local IRB’s “own additional internal review…would not be binding on the local site if not adopted by the single IRB, and the terms of it would not be enforced by OHRP” (DHHS 2015, 53984).

In response to the ANPRM commenters’ liability concerns, the NPRM proposes to allow Common Rule departments the authority to take compliance actions “directly against unaffiliated IRBs” [§__.101(a)]. Currently, OHRP enforces “compliance with the Common Rule through the institutions that are engaged in human subjects research” even if the violation is connected to the external IRB’s responsibilities. The new provision is intended to allow OHRP the flexibility to take compliance actions “directly against the IRB responsible for the flawed review, rather than the institutions that relied on that review” (DHHS 2015, 53983). Yet, at the same time, the NPRM indicates that the single IRB mandate “would not relieve any site of its other obligations under the regulations to protect human subjects” (DHHS 2015, 53984). The lack of clarity regarding roles and responsibilities may lead to decreased subject protections and greater administrative burden.

To be sure, single IRB review may have advantages for particular multi-site studies, based on the number and types of institutions involved, or the nature of the research (for example, when dealing with a rare condition where each site will accrue only a few subjects). Such advantages may often outweigh the complications and costs involved in single IRB review. However, as mentioned earlier, the Common Rule currently permits research sponsors or funders to require a single IRB, and to work with institutions regarding the distribution of the costs of coordination.

Furthermore, the NPRM’s single IRB mandate fails to acknowledge circumstances when review by more than a single IRB might be preferable. For instance, when the number of institutions is small, when the scientific role at each institution is distinct (for example, institution A ascertains subjects and biological samples, while institution B conducts imaging procedures), and when population characteristics demand local consideration, review by each institution’s IRB may be more sensible. Indeed, the preamble acknowledges the ANPRM commenters’ concerns that a single IRB, in another location, might not be in the best position to review social science and behavioral studies where an understanding of local “community norms” is important (DHHS 2015, 53983). Yet it is not clear that such circumstances would fit under the very limited exceptions to the single IRB mandate permitted by the proposed rule. Before mandating a universal shift to external IRB review, careful study of the potential untoward impact on research protections and community confidence would be wise.

Finally, the NPRM is vague about the criteria by which one institution or an independent IRB is to be selected as the single IRB by the funding agency or other sponsor. It only states: “The reviewing IRB is expected to be selected either by the funding agency or, if there is no funding agency, by the lead institution conducting the study.” The agency may, but is not required to, request comment on how to select the IRB of record (DHHS 2015, 53984). Commentators suggest that establishing objective criteria in this area “may help manage concerns about favoritism or selection based on perceived chances of ‘an easy approval.’” For example, objective criteria might focus on the IRB’s substantive experience with relevant research or subject populations and the principal investigator’s home institution. Absent such criteria and a a fully transparent selection methodology, ethical considerations may lose priority to considerations of speed and cost-savings.

Members of the research ethics and oversight community are encouraged to share their thoughts on these topics, as well as other issues raised by the NPRM, by submitting formal comments to DHHS. Comments are being accepted on the proposed rule through January 6, 2016 at 5:00 PM ET.

Department of Health and Human Services (DHHS). 2015. Notice of Proposed Rulemaking: Federal Policy for the Protection of Human Subjects. Federal Register 80 (173): 53982-53985.

Verrill Dana LLP. (2015, Nov. 25). The Common Rule NPRM Blog Series: Part 2 – Single IRB Review. Retrieved from http://clinicalresearchlawblog.com/the-common-rule-nprm-blog-series-part-2-single-irb-review/.