Stakeholder Experiences with the Single IRB Review Process and Recommendations for Food and Drug Administration Guidance

ABSTRACT

The revised Common Rule requires using a single institutional review board (sIRB) for U.S.‐based, multisite, nonexempt, federally conducted or supported research with human participants. The 21st Century Cures Act directs the Department of Health and Human Services (HHS) to harmonize differences between HHS and the U.S. Food and Drug Administration (FDA) regulations governing research with humans. Anticipating that the FDA may update its 2006 centralized IRB guidance, we conducted interviews with 34 stakeholders engaged in FDA‐regulated clinical research to identify benefits and challenges of using sIRBs and to gather recommendations for revising the FDA's guidance. The main benefits were consistency and standardization, speed and efficiency, and streamlining and simplification. The main challenges were uncertainty at local institutions, including addressing local context; decreased timeliness of the research review process; variable processes; and insufficient communication. Several recommendations for FDA guidance focused on the local context and communication plans. Findings suggest that the sIRB review process may be gaining efficiency although challenges remain.

Ethics review of clinical research with human participants in the United States is changing. As of January 25, 2018, investigators submitting grant applications and contract solicitations to the National Institutes of Health (NIH) became subject to a new policy: U.S.‐based, multisite, nonexempt research must use a single institutional review board (sIRB) for ethics review, with limited exceptions. ¹ An sIRB review process allows a single IRB to be formally responsible for the ethical review of a research study on behalf of all participating institutions and sites. As of January 20, 2020 (the compliance date for the sIRB provision in the revised Common Rule), the sIRB review process is now required for all U.S.‐based, multisite, nonexempt, federally conducted or supported human subjects research subject to Common Rule subpart A (45 C.F.R. 46). ² The NIH sIRB policy aligns with the revised Common Rule, ³ and both are based on the premise that having a single IRB review multisite research increases the efficiency of ethics review while maintaining participant protections. ⁴

The U.S. Food and Drug Administration (FDA) has separate regulations for the protection of human participants in FDA‐regulated clinical investigations, including for informed consent (21 C.F.R. part 50) and IRBs (21 C.F.R. part 56). These regulations generally conform to the Common Rule. ⁵ The FDA has supported the use of what is called “centralized IRB review” (another term for an sIRB review process) for multicenter clinical research since 1981 and has encouraged the use of sIRB review since 2006, when it issued guidance on using a centralized IRB review process for multicenter clinical trials. ⁶ As described in the introductory section, the guidance is directed toward sponsors, institutions, IRBs, and clinical investigators and provides recommendations on complying with 21 C.F.R. part 56 when using a centralized IRB review process for multicenter clinical research. The guidance includes five sections that (1) describe the roles and responsibilities of individuals who facilitate the central IRB process (i.e., the institution, the institution's IRB, the sponsor, the investigator, and the central IRB); (2) provide guidance for considering local aspects (e.g., locally relevant information) within a central IRB process; (3) provide recommendations for documenting agreements among local institutions/IRBs and the central IRB as well as written procedures for implementing the central IRB process (e.g., about communication among IRBs and the roles of the local IRB versus the centralized IRB); (4) provide recommendations for documenting the use of central IRBs for sites that are not affiliated with a local IRB; and (5) provide examples of cooperative IRB review models (i.e., models in addition to the primary model of multiple sites relying on a centralized IRB for a single trial).

In 2016, the 21st Century Cures Act was passed by the U.S. Congress. Section 3023 directs the secretary of Health and Human Services (HHS) to harmonize differences between HHS and FDA human subjects regulations to the extent practicable and consistent with other statutory provisions, including considerations of the sIRB process. ⁷

In anticipation that the FDA may, among other regulatory actions, update its 2006 centralized IRB guidance document in response to the Cures Act, the Clinical Trials Transformation Initiative (CTTI) solicited information about experiences with the sIRB review process from stakeholders engaged in FDA‐regulated clinical research so that CTTI could offer the FDA another source of evidence to consider if they choose to revise their guidance. CTTI is a public‐private partnership between Duke University and the FDA, and it seeks to identify and drive adoption of practices that increase the quality and efficiency of clinical trials.

When the FDA issued their guidance in 2006, independent IRBs (commercial IRBs that are independent of academic and medical institutions) primarily facilitated the centralized IRB process. Now, academic and community‐based health system IRBs, in addition to independent IRBs, frequently serve as central IRBs; these academic and community‐based IRBs have therefore gained substantial experience in the sIRB review process and can provide additional and possibly different perspectives to consider when revising the guidance. The research we present here focused on identifying (1) the benefits and challenges experienced by a variety of stakeholders when using an sIRB review process for multisite, FDA‐regulated clinical research so that the day‐to‐day experiences from those with firsthand knowledge of implementing this process are illuminated and better understood and (2) stakeholders' suggestions for revising the FDA's 2006 guidance document. We describe these data, building the evidence base on the use of the sIRB review process in practice, a need previously identified in the literature. ⁸ CTTI has shared the findings with the FDA.

The benefits of the sIRB review process that many stakeholders described suggest that it is gaining in efficiency for some institutions. However, challenges remain, particularly with regard to uncertainty at local institutions, decreased timeliness of the research review process, variable processes, and insufficient communication.

Study Methods

We conducted a qualitative descriptive study ⁹ using in‐depth interviews. Interviews were conducted from March 29 to August 31, 2018. We purposefully selected ¹⁰ four groups of stakeholders to participate in the interviews: (1) representatives of U.S.‐based IRBs, including independent, academic health systems, and community‐based health system IRBs; (2) representatives of U.S.‐based pharmaceutical, device, and other, similar companies; (3) investigators who conduct multisite, U.S.‐based research regulated by the FDA; and (4) regulatory and study coordinators of FDA‐regulated multisite clinical research.

We used multiple methods to identify stakeholders to invite for participation. One method was searching conference presentations and attendance rosters (e.g., from the Public Responsibility in Medicine and Research [PRIM&R] Advancing Ethical Research Conference) to identify IRB representatives who have experience with the sIRB review process. We also searched ClinicalTrials.gov for investigators who have led multisite studies, and we reviewed participant lists for industry and IRB representatives who participated in previous CTTI‐sponsored meetings on the use of sIRBs. We also worked with CTTI staff members, who referred investigators, study coordinators, and regulatory administrators from other CTTI projects or their professional networks, and the CTTI Steering Committee members, who used their professional networks to identify potential stakeholders, particularly investigators and study coordinators. We invited these individuals to be screened and to recommend other individuals to invite for participation. To capture multiple perspectives and experiences, we selected individuals to participate in the interviews if they had experience with FDA‐regulated clinical trials and had some engagement with the sIRB review process.

We explored the following topics in the in‐depth interviews as they relate to FDA‐regulated clinical research, tailoring the questions for each stakeholder group based on their role in the sIRB review process: (1) benefits and challenges of the sIRB review process, directing industry representatives and investigators to reflect on their experiences with using an sIRB for multisite studies and directing IRB representatives and regulatory and study coordinators to reflect on their experiences with relying on another IRB for ethics review and/or serving as the reviewing IRB for multisite studies; and (2) suggested modifications to the FDA's 2006 guidance document. ¹¹

We explored the benefits of the sIRB review process in detail because of the underlying premise of the revised federal regulations that the sIRB review process is beneficial. To facilitate the discussion, we first asked stakeholders to describe the benefits they had experienced with the sIRB review process. Then, as follow‐up probes, we asked stakeholders to reflect on specific benefits described in the peer‐reviewed literature, if their initial responses were not similar to these specific benefits, to learn whether stakeholders had in fact experienced these benefits. The benefits included improvements in (1) administrative burden among local IRBs, ¹² (2) duplication of effort across site IRBs, ¹³ (3) disparities or subjective variations in IRB review between site IRBs, ¹⁴ (4) cost effectiveness of IRB review, ¹⁵ (5) human subjects protections, ¹⁶ (6) ease of monitoring, ¹⁷ (7) ease of analyzing adverse events at multiple sites, ¹⁸ (8) ability to select an IRB with the necessary expertise to review the trial, ¹⁹ (9) time of initial review, ²⁰ (10) efficiency in submitting protocol amendments and continuing reviews, ²¹ and (11) ability to maintain version control due to the centralization of regulatory documents. ²² We tailored the follow‐up probes for each stakeholder group so that we asked about only relevant benefits. We asked industry representatives and regulatory and study coordinators about all 11 benefits, IRB representatives about 10 of the benefits, and investigators about 6 of the benefits.

To facilitate stakeholders' identification of recommendations for modifications to the FDA's 2006 centralized IRB guidance document, we emailed a copy of the guidance to all stakeholders before the interview. The following sections were discussed during the interview: (1) “Roles in Ensuring IRB Review”; (2) “Addressing Local Aspects of IRB Review”; (3) “IRB Records—Documenting Agreement and Procedures,” which we combined with the section “Using a Central IRB at Unaffiliated Sites”; and (4) “Examples of Cooperative IRB Review Models.” During the interview, we asked stakeholders to review each section of the guidance and answer the following questions for each section: (1) What, if anything, is unclear in the existing guidance, and why? (2) What information should be added in a revision, and why? (3) What information should be removed from the existing guidance, and why? We tailored the review of the sections for each stakeholder group to align with their specific roles: all stakeholders reviewed the sections on roles in ensuring IRB review and addressing local aspects of IRB review; IRB and industry representatives reviewed examples of cooperative IRB review models; and IRB representatives reviewed the sections on IRB records and using a central IRB at unaffiliated sites.

All interviews were conducted by telephone, were digitally audio‐recorded with stakeholders' permission (except for one stakeholder who wished not to be audio‐recorded and for whom detailed notes were taken instead), and were transcribed verbatim following a transcription protocol. ²³ Demographic information was collected at the beginning of each interview.

We used applied thematic analysis to analyze the data. ²⁴ We used a two‐stage deductive and inductive approach. First, two analysts developed and then applied a priori structural codes using NVivo 12 (QSR International Pty Ltd.) qualitative data analysis software to segment the stakeholders' narratives into broad categories (e.g., all text describing the benefits of using an sIRB). Second, after the structural coding was complete, analysts identified and applied emergent content‐driven codes to the text in each of the structural coding reports (e.g., potential themes related to the benefits of using an sIRB). We assessed intercoder agreement for 20% of the transcripts. We resolved discrepancies in code application through group discussion, and we subsequently made edits to the codebook. We applied the updated codebook to previously coded transcripts and used it to apply codes to subsequent transcripts. Once coding was completed, analysts reviewed the content‐driven codes by examining code frequencies across transcripts to identify salient factors (i.e., themes and subthemes), including only narratives that described perspectives based on experiences with implementing the sIRB review process, as opposed to stakeholders' aspirations or hypothetical concerns about the process. While our sample size does not allow any meaningful comparison among stakeholder groups, analysts initially explored any potential differences in responses based on the stakeholders' role in the sIRB review process (e.g., using, relying, or reviewing) and stakeholder group. The overall broad themes were similar, and, hence, we grouped them together. We then described all themes, together with illustrative quotations, in analytical memos that we used to summarize the results below.

The Duke University Health System IRB determined that the research was exempt from further IRB review. Prior to study participation, all stakeholders were provided with an informational sheet that explained the study in detail, including its purpose, risks, and benefits.

Study Results

We interviewed 34 stakeholders, including 10 IRB representatives, 9 industry representatives, 9 investigators, and 6 regulatory and study coordinators (see table 1, which is available online, along with the other tables and the appendix; see the “Supporting Information” section at the end of this article). The IRB representatives included executive officers, directors, chairs, and managers, and all were from different institutions. Eight of the IRB representatives were affiliated with academic or community‐based health system IRBs, and 2 were affiliated with independent IRBs. Six were affiliated with IRBs that had five or more years of experience using sIRBs. All the IRBs had been involved in the sIRB review process for FDA‐regulated clinical trials, though their roles varied (i.e., serving as the reviewing IRB, the relying institution, or both). Industry representatives included directors and managers, primarily of clinical operations, and all were from different companies. Five were from large companies (i.e., with a market cap over $10 billion), and the companies represented a variety of medical products (e.g., drugs or devices). All companies had experience using the sIRB review process for FDA‐regulated clinical research, and six of the representatives were from companies with five or more years of experience using sIRBs. All investigators were from different institutions, primarily academic and community‐based health systems. They represented a range of therapeutic areas and years of experience, and all had conducted phase III, multisite, FDA‐regulated clinical trials. All the investigators also had experience with the sIRB review process, including eight who had sIRB experience with FDA‐regulated studies. All but two of the regulatory and study coordinators were from different academic or community‐based health systems. All reported that their institutions had been involved in the sIRB review process for FDA‐regulated clinical trials, and all had been involved in FDA‐regulated clinical trials that relied on another IRB. Years of coordinating experience with FDA‐regulated clinical trials ranged from 8 to 25, and experience with the sIRB review process ranged from less than 1 to more than 5 years.

Benefits of the sIRB Review Process

The three most common themes stakeholders expressed about the benefits of the sIRB review process were (1) consistency and standardization, (2) speed and efficiency, and (3) streamlining and simplification. Table 2 provides quotations from the stakeholders on these topics.

Consistency and standardization

Nearly all the stakeholders noted that consistency, standardization, and coordination across sites are benefits of the sIRB review process, primarily because the process ensures that all participating institutions are following the same study protocol. As an example, several stakeholders mentioned the benefits of the sIRB review process for the informed consent process, stating that using an sIRB enhances research participant protections by ensuring that participants are provided with consistent information across sites and by minimizing unnecessary consent language through the use of a template. Some stakeholders also elaborated on the variations in review they experienced when working with multiple local IRBs in the past, which the sIRB process eliminates (see table 2, section 1).

Speed and efficiency

Nearly all the stakeholders described speed and efficiency as benefits of the sIRB review process. Many of these stakeholders focused their narratives on the faster review and approval process during the initial review, as well as for amendments. They described that the initial sIRB review process reduces time to site start‐up and leads to more rapid initiation of trials than does working with multiple local IRBs. For protocol amendments, stakeholders described that new sites can be rapidly added and approved because the sIRB is already knowledgeable about the protocol and a template informed consent form is available. Stakeholders also described that amendments are approved for all sites at the same time, allowing for all sites to implement the protocol changes simultaneously, enhancing the efficiency of study conduct. Stakeholders attributed the faster review process to a variety of factors, including review of protocol documents by a single institution on behalf of all sites rather than by multiple sites operating on their own timelines, negotiation of amendments and other contingencies with the sIRB at one time instead of having discussions with multiple local IRBs, the established relationship between the sponsor and the sIRB, the experience level of the reviewing IRB in working with multiple sites, and the more frequent meetings of sIRBs, primarily among independent IRBs (see table 2, section 2).

Streamlining and simplification

About two‐thirds of stakeholders, across all stakeholder groups, said the sIRB review process streamlined and simplified ethics review. Most of these stakeholders focused their explanations on the decreased administrative burden and workload, including fewer administrative resources and personnel needed. Stakeholders who spoke about relying on an sIRB said the process leads to less unnecessary work, including the amount of paperwork. They also described decreases in the burden of regulatory documentation and less effort for local IRB board members during initial review, amendments, and continuing reviews. Several stakeholders also noted the helpfulness of having a single study‐wide process for IRB submissions for all sites (see table 2, section 3).

Additional benefits

Often in response to the follow‐up probes, and with less description than for the benefits above, numerous stakeholders also identified the quality of review and cost effectiveness as benefits of the sIRB review process. Many stakeholders described a higher quality of ethics review with sIRBs because of the sIRB members' expertise. Stakeholders explained that sIRBs—often citing independent IRBs but not always—are familiar with research regulations and requirements, have diverse membership and can access necessary specialized experts, and include full‐time professionals rather than volunteers. Stakeholders also said that sIRBs' greater familiarity with a specific type of research or content area increases the quality of their ethics review and that using an sIRB can address previous concerns with inadequate ethics reviews by some local IRBs across sites (see table 2, section 4). With regard to cost effectiveness, stakeholders indicated that the sIRB review process reduces expenses at relying institutions, reduces overall study costs by saving time, and reduces administration costs. The sIRB review process is also easier to predict, stakeholders explained, making budgeting for expected costs easier.

Challenges

The four most common challenges in implementing the sIRB review process were uncertainty at local institutions, decreased timeliness of the research review process, variable processes, and insufficient communications. Table 3 provides quotations from stakeholders about these challenges.

Uncertainty at local institutions

About two‐thirds of stakeholders, across all stakeholder groups, described challenges related to local institutions responding cautiously to their new, unfamiliar role and proceeding with nascent procedures. Many of these stakeholders, across all groups but mostly IRB representatives, spoke about challenges related to losing control of the review process, particularly when relying institutions were not willing to cede control, either fully or with certain aspects of the review process. Stakeholders described excessive customization of study consent forms and re‐review of protocols and amendments as examples of local institutions' efforts to maintain some control or involvement in the review process.

Several stakeholders focused on duplication of effort by local institutions. For example, stakeholders described what they referred to as “shadow reviews” and “administrative reviews,” wherein local institutions or IRB chairs review entire applications that have already been reviewed by the sIRB. Some stakeholders described that local IRBs limit their redundant reviews to certain sections of the application (e.g., reviews by specialty committees, such as radiation safety reviews). Stakeholders also described duplication of effort by study teams, such as a requirement to submit all or part of an application to their local institution to obtain permission to rely on an outside IRB, a step that requires more work by the investigator.

Several stakeholders also commented on challenges associated with considering “local context” during the sIRB review process. They described difficulties in conforming to local laws, standards, and regulations and spoke of a lack of standards for what constitutes local context. Institution‐specific consent processes, requirements of the Health Insurance Portability and Accountability Act (HIPAA), and state laws were given as examples of challenges that took considerable effort to surmount (see table 3, section 1).

Decreased timeliness of the research review process

About half of the stakeholders, across all stakeholder groups, spoke about sIRB challenges that negatively affect the research timeline. They explained that delays occur when working with inefficient sIRBs and when extra time is needed to discuss the research due to a lack of understanding of the relying IRB on institutional and site practices; to fix errors and oversights by the sIRB; and to complete sequential, ancillary, and duplicative reviews by the local IRB, particularly when such reviews are conducted sequentially as opposed to in parallel with other reviews (see table 3, section 2).

Variable processes

About half of the stakeholders, across all stakeholder groups, remarked on the variability of sIRB review processes among different IRBs, primarily the difficulty of managing inconsistencies and the lack of policies and procedures across institutions. Many stakeholders highlighted the limited standardization within sIRB workflows and/or policies, such as division of responsibilities. Several also pointed out that each institution follows its own system, which is different from other institutions; therefore, continuity and common processes are lacking.

Several stakeholders also discussed the lack of IRB readiness to engage in the sIRB review process. They noted an observable inconsistency in the capacity of reviewing IRBs, and several stakeholders recounted instances of delays or issues with reviews because the sIRB did not have the bandwidth, infrastructure, or expertise needed to meet the demands of the relying institutions (see table 3, section 3).

Insufficient communications

About one‐third of the stakeholders, across all stakeholder groups except investigators, emphasized challenges with communication. Many described insufficient communication from reviewing IRBs to relying sites, including irregular feedback and updates, the lack of disseminated IRB meeting minutes, no notice of process changes that occurred during a review, lack of an established point of contact, the lack of familiarity with IRB portals among study team members, and the lack of transparency regarding the way determinations were made by the reviewing IRB. Stakeholders also noted poor communication between investigators and their local IRBs as a challenge (see table 3, section 4).

Infrequent challenges

A small number of stakeholders described time delays due to negotiating reliance agreements and concerns about the quality of ethics review, and one stakeholder mentioned liability concerns. Only a single stakeholder mentioned cost concerns, and a few noted that they have incurred some costs in preparing for implementation of the sIRB review process, such as to account for transitioning roles.

Recommendations for Revising the FDA's sIRB Guidance Document

The appendix lists all stakeholder recommendations for revising the FDA's sIRB guidance. ²⁵ The most common recommendations from each section were (1) to make the language in the section on IRB roles more directive and less suggestive; (2) to include a matrix illustrating the institutional roles and responsibilities of each entity involved in the sIRB review process; (3) to clarify the most relevant aspects of local context that should be considered by the reviewing IRB; (4) to describe the process for reviewing local context information, including how local information should be shared with the reviewing IRB, who should oversee the process to ensure local context is being considered, and how to document consideration of local context; (5) to clarify communication plans in the section on IRB records, including how relying institutions will assess the qualifications of the reviewing IRB, procedures for reporting noncompliance or unanticipated problems, and who is responsible for establishing the reliance agreement; (6) to add templates, including reliance agreements and memoranda of understanding; and (7) to describe how communication should occur between the reviewing and relying institutions in the section on IRB review models.

Discussion

We conducted this research to illuminate the benefits and challenges of implementing the sIRB review process for FDA‐regulated, multisite clinical trials from the perspective of stakeholders who are engaged in FDA research and to gather recommendations for revising the FDA's guidance on the use of sIRBs. The benefits of the sIRB review process that many stakeholders described—consistency and standardization, speed and efficiency, and streamlining and simplification—suggest that the process is gaining in efficiency for some institutions. Future FDA guidance can encourage processes that support such benefits. However, challenges remain, particularly with regard to uncertainty at local institutions, decreased timeliness of the research review process, variable processes, and insufficient communication. The FDA could provide additional guidance in these areas.

Not surprisingly, similar concepts were mentioned as both benefits and challenges. For example, an experienced sIRB can contribute to a speedy and efficient process, whereas an inexperienced sIRB can lengthen the process. Stakeholders also spoke about unanticipated consequences of the sIRB review process for the study timeline, such as additional and sometimes duplicative reviews by local IRBs, which lengthen the overall process. For the “consistency and standardization” benefit and the “variable processes” challenge, the stakeholders' experiences suggest that some institutions have achieved standardization and others have not. An institution's success in standardizing their sIRB review process, potentially within their own institution and in relation to other institutions, may determine whether the sIRB review process is beneficial or challenging.

Stakeholders identified several concerns with the sIRB review process that have been previously identified in the literature, including consideration of local context, ²⁶ reluctance to cede control, ²⁷ and duplicate reviews. ²⁸ Stakeholders gave examples of difficulties in considering local context, described relying IRBs' unwillingness to relinquish control of ethics review to another institution, and emphasized the current lack of standards for what constitutes local context. Cost, ²⁹ liability concerns by the sIRB, ³⁰ and delays in executing reliance agreements ³¹ did not emerge as concerns or were mentioned by only a few stakeholders.

The challenges stakeholders identified and the recommendations they offered for FDA guidance involved two important areas: the local context and communication plans. Incorporating local context into the sIRB ethics review process has been a significant concern since the idea of requiring an sIRB for multisite studies as part of the Common Rule was introduced in the Advance Notice of Proposed Rulemaking in July 2011. ³² At that time, stakeholders believed that the sIRB review process would lead to a lack of representation of the communities where the research would be conducted. Since then, research has shown that concerns about local knowledge may have been overstated. ³³ Yet our findings show that stakeholders continue to be concerned and experience difficulties in considering local context. Guidance exists on specific information to gather and review when considering local context. For example, the National Cancer Institute's (NCI's) long‐established sIRB review process requires institutions to complete four worksheets that describe “local context details, including state and local laws, conflict of interest policies, management plans, standard institution language to be added to consent forms, and a description of potential study participant populations and plans to safeguard vulnerable populations.” ³⁴ SMART‐IRB provides guidance on entering local context information and concerns in their online reliance agreement systems (SMART‐IRB also has other tools to facilitate the sIRB review process). ³⁵ Klitzman et al. ³⁶ suggest consideration of the following areas: linguistic and cultural issues, geographic and socioeconomic issues, researcher‐specific issues, and institution‐specific issues.

To address communication challenges, the NCI established a comprehensive online communication system where investigators have access to a toll‐free number and email and responses are provided within several days. ³⁷ Although over 80% of users have reported being very satisfied with different aspects of the system, reproducing this process at academic or health system IRBs might not be feasible. Nonetheless, sIRBs can establish processes that allow for clear channels and timely responses, such as identifying administrative support or a communication liaison ³⁸ whose primary responsibility is to communicate with investigators and relying IRBs, a direct cost that is allowed by the NIH. ³⁹

Stakeholders in this study also suggested clarification from the FDA on the roles and responsibilities of institutions involved in the sIRB review process, another area of concern mentioned in the literature. ⁴⁰ Overall, the ethics review is only one component of ethics oversight. Even when an sIRB review process is used, multiple required tasks remain with the local IRB. In an earlier sIRB project, CTTI developed a matrix of responsibilities to distinguish local institutional responsibilities from the ethics review responsibilities of the sIRB. ⁴¹ The matrix outlines 17 roles, such as “assess investigator qualifications” and “conduct security and privacy review for HIPAA,” and suggests the institution that should be responsible for each: the sIRB, the relying institution, both, or either. Similar guidance by the FDA may be helpful.

The strength of our research was the inclusion of stakeholders who had experience in FDA‐regulated, multisite clinical trials and the sIRB review process, thereby allowing the documentation of benefits and challenges to be based on experience rather than hypothetical responses. Yet, as with all qualitative research, our results must be interpreted with consideration of the study population from which the data were collected. We interviewed a purposeful sample; another group of stakeholders may have had different experiences, and the benefits and challenges they identified could have differed. Stakeholders may have focused several of their narratives on working with independent IRBs as part of a central IRB process, which has been used for many years, rather than working with academic IRBs who may be newer to serving as reviewing IRBs. In addition, although we focused on gathering stakeholders' perceptions of benefits and challenges that were grounded in their day‐to‐day experiences with implementing the sIRB review process, other objective metrics of efficiency and effectiveness are needed to complement stakeholders' descriptions of their experiences. Creative, alternative evaluation frameworks of the sIRB review process are needed, given that sIRB use is now required for U.S.‐based, multisite research with human participants that is subject to the revised Common Rule. Previous studies that compared the sIRB review process to a local IRB process will likely no longer be possible. ⁴²

We shared the findings of this study with the FDA in November 2018. We are unaware of their plans for revising their guidance and whether they will require the use of the sIRB review process for all FDA‐regulated, multisite clinical trials. We hope our findings are useful to them as they consider possible changes to the regulations and guidance applicable to FDA‐regulated, multisite clinical trials.

Additional studies should be conducted to build the evidence base on stakeholders' experiences with the sIRB review process, including its effectiveness and quality, so that the benefits of the sIRB review process can be highlighted, solutions to challenges can be identified, and best practices can be established.

Acknowledgments and Disclaimers

Funding for this work was made possible, in part, by the U.S. Food and Drug Administration through cooperative agreement U18FD005292 and grant R18FD005292. Views expressed in written materials or publications and by speakers and moderators do not necessarily reflect the official policies of the Department of Health and Human Services, nor does any mention of trade names, commercial practices, or organization imply endorsement by the United States Government. Partial funding was also provided by pooled membership fees from the Clinical Trials Transformation Initiative's member organizations.

CTTI thanks all stakeholders for sharing their experiences. CTTI also thanks Katelyn Blanchard of Duke University for assistance in implementing the research, Brian Perry of Duke University for assistance with data analysis, and Damon M. Seils, also of Duke University, for editorial assistance. All staff who contributed to implementation of the research or manuscript preparation were compensated as part of their salaries.

Supporting information

Supporting Information—Tables and Appendix

The tables and appendix are available in the “Supporting Information” section for the online version of this article and via Ethics & Human Research's “Supporting Information” page: http://www.thehastingscenter.org/supporting-information-ehr/.

Supporting Information