Comment on the Proposed Institutional Review Board Retrospective Review of Research

Common Rule Requirements

In the United States, IRBs are governed by the Federal Policy for the Protection of Human Subjects, also known as the Common Rule (U.S. Department of Health and Human Services [DHHS], 2018) and are regulated by the Office for Human Research Protections (OHRP) of (U.S. DHHS, 2011).

The Common Rule requires that no research involving human subjects, unless it is deemed to be exempt from IRB review, can be initiated until it has been reviewed and approved by the IRB (U.S. DHHS, 2018). Thus, IRBs are authorized to conduct prospective review and have the authority to approve, require modifications, or disapprove human research protocols. The Common Rule provides no provision for IRBs to retrospectively review and approve research that has been completed without prior IRB approval in serious noncompliance with federal regulations. Thus, IRBs are not authorized to conduct retrospective review of research. However, the Common Rule does not explicitly prohibit IRBs from carrying out retrospective review and approval of research, so for the purpose of this commentary, I contacted OHRP for policy clarification. The OHRP’s response was:

The IRB does not have the authority to retrospectively review and approve human subjects research that is subject to the HHS regulations for the protection of human subjects at 45 CFR Part 46.

The Common Rule does not provide any direction regarding how data collected from noncompliant research should be handled. Investigators of such research often experience difficulties publishing those data. This is because most, if not all, institutions, as a matter of institutional policy, prohibit data obtained through research conducted without prior IRB approval from being published or used in degree dissertations. Some even require those data to be destroyed (Cooper & McNair, 2019). Likewise, most, if not all, reputable journals require assurance that research was conducted with prior IRB approval as a condition for publication (Rowan-Legg et al., 2009).

Issues With the Proposed Retrospective Review of Research

Cooper and McNair (2019) correctly pointed out that:

There is no code or regulation that explicitly states that if an investigator, institution, or sponsor conducts research without IRB approval, the IRB is not authorized to grant a retrospective approval.

They proposed that IRBs might retrospectively review and approve unintended noncompliant research as a potential pathway to help investigators of such studies to publish their data. However, in making such a proposal, they should have consulted OHRP regarding the regulatory basis for such an approach. Since under the Common Rule, an IRB does not have the authority to retrospectively review and approve human subjects research, any such IRB review and approval would technically be invalid. In addition, host institutions of IRBs that conduct such reviews as proposed by Cooper and McNair would be required to report this serious noncompliance of their IRBs to OHRP along with a remedial action plan to ensure that similar noncompliance would not recur in future (U.S. DHHS, 2011).

Apart from the issue of technical noncompliance with the Common Rule, many troublesome issues that arise from the proposed retrospective review of research were not addressed by Cooper and McNair. For example, what criteria would IRBs use to retrospectively review and approve such research? To approve a human research protocol, IRBs must ensure that the research has met eight Common Rule criteria, namely: requirement for and documentation of informed consent; additional safeguards for vulnerable subjects when appropriate; minimization of risks to human subjects; a reasonable risk-to-benefit ratio; monitoring of subject safety; maintaining subject privacy and the confidentiality of data; and equitable selection of subjects (U.S. DHHS, 2018).

If an IRB uses the above, same criteria for the proposed retrospective review of research, the IRB may find it necessary to recommend protocol modifications to satisfy the Common Rule criteria for approval. However, any protocol modifications will defeat the intended purpose of retrospective review and approval of research that has already been completed without prior IRB approval. For this reason, the IRB may find it easier to just approve the protocol without critical review.

Similarly, retrospective review and approval of a completed study does not correct the initial noncompliance of conducting research without prior IRB approval. Under the current rules, the institution or the IRB is still required to report this serious noncompliance to OHRP along with a remedial action plan (U.S. DHHS, 2011) as mentioned earlier. The only thing that the IRB can accomplish is providing investigators with an invalid document certifying IRB approval of the completed research protocol. Investigators may then use it to mislead journal editors into believing that the research has been carried out with full IRB scrutiny and approval. The potential ramification of this is unfortunate because journal editors may one day no longer be able to rely on the IRB or investigator’s certification that research has been properly conducted with prior IRB approval.

The primary responsibility of IRBs is to protect human subjects participating in research (U.S. DHHS, 2018). The proposed retrospective review and approval of nonexempt research that has been completed without prior IRB approval by Cooper and McNair appears to put investigators’ interests above human subjects protections, raising important ethical concerns. In addition to providing IRBs with the authority to conduct retrospective review and approval of research, Cooper and McNair’s proposal further provides IRBs with the authority to arbitrarily determine which investigator’s noncompliance was due to good faith/unintended error, deserving retrospective IRB review and approval. With so much authority and discretion without built-in oversight, there is a real possibility for abuse by some IRBs.

Conclusion

There are few options to be exercised when nonexempt human subjects research has been conducted without prior IRB approval. Retrospective IRB review and approval of such research as proposed by the Practical Ethicist seems meaningless from the point of view of participant protections, as opposed to investigator condonation. It will not work because it lacks a regulatory basis. Prevention remains the best strategy to ensure that no nonexempt human subjects research is initiated prior to IRB approval. With the recent implementation of the revised Common Rule, which markedly expanded the exempt research categories (Menikoff et al., 2017; U.S. DHHS, 2018), it is hoped that the incidence of nonexempt research conducted without prior IRB approval will be markedly reduced and pressures for retrospective review minimized.