A cross sectional survey of recruitment practices, supports, and perceived roles for unaffiliated and non-scientist members of IRBs

Abstract

Background

Institutional Review Boards (IRBs) are federally mandated to include both non-scientific and unaffiliated representatives in their membership. Despite this, there is no guidance or policy on the selection of unaffiliated or non-scientist members and reports indicate a lack of clarity regarding members’ roles. In the present study we sought to explore processes of recruitment, training, and the perceived roles for unaffiliated and non-scientist members of IRBs.

Methods

We distributed a self-administered REDCap survey of members of the Association for the Accreditation of Human Research Protection Programs familiar with IRB member recruitment. The survey included closed and open-ended questions regarding: the operation of the HRPP/IRB(s), how unaffiliated and non-scientist members are recruited, whether they had faced challenges recruiting for these roles, and training and mentorship offered. The survey also collected information regarding the perceived value and roles of unaffiliated and non-scientist members.

Results

76 responses were included in the analysis (38% completion rate). The most common approach for recruitment was referral from current IRB members, with almost half of respondents indicating challenges recruiting unaffiliated members. Over 75% indicated no additional training was provided to unaffiliated or non-scientist members compared to affiliated or scientist members. Most common supports provided were travel/parking expenses and honoraria. Commonly perceived roles were to provide an independent voice from the participant perspective, notably regarding consent processes and materials.

Conclusions

Respondents indicated challenges in defining unaffiliated and non-scientist members and limited practices toward recruitment and support. Future work should more closely examine the challenges in defining these roles and applying the definitions in practice, as well as strategies that may improve recruitment and retention of unaffiliated and non-scientist members.

INTRODUCTION

Institutional Review Boards (IRBs) are federally mandated to include both non-scientific (NS) and unaffiliated (UA) representatives in their membership. Specifically, Title 45 Part 46 of The Code of Federal Regulations requires that each IRB must include: 1) at least one member whose primary concerns are in non-scientific areas and 2) at least one member who is not otherwise affiliated with the institution and who is not part of the immediate family of a person who is affiliated with the institution (US Department of Health and Human Services, Office of the Assisstant Secretaryfor Health, & Office for Human Research Protections, 2018). These members are sometimes colloquially known as lay or community members and are thought to play an important role in representing a community voice in research ethics review.

The topic of community representation in research and research ethics broadly, and IRB processes specifically, has been given increasing attention in recent years (e.g., Anderson, 2006; Cargill, 2018; Lidz et al., 2012). Discussions regarding representation in research governance are taking place in parallel to the participatory turn in research, which emphasizes the inclusion of those affected by research in its design, implementation, and dissemination (Canadian Institutes of Health Research, 2011; Cornwall, 1995; Shore et al., 2014). Community/patient engagement and representation on the review board has been proposed as a structural dimension of IRB effectiveness (Berry et al., 2019).

Despite widespread agreement on the importance of including UA and NS members on IRBs across research, federal commissions, and government-sponsored reports, Solomon has pointed to “a lack of clarity at the level of both policy and practice regarding what these members are intended to contribute to IRB deliberations”(Solomon, 2016). Indeed, despite the requirements imposed by the Office for Human Research Protections (OHRP), there is no guidance or policy on the selection of UA/NS members.

The limited amount of empirical research supports this claim regarding ambiguity of UA/NS IRB members. Anderson (2006) found that most UA/NS members were recruited in an ad hoc manner, usually through contact with a serving IRB member or someone at the institution (Anderson, 2006). Training and compensation of UA/NS members have also been found to vary across IRBs (Cargill, 2018).

Some research on the roles and responsibilities of UA/NS members indicates that they see themselves as primarily contributing through reviewing the language of consent forms (Allison, Abbott, & Wichman, 2008; Sengupta & Lo, 2003), drawing attention to issues that might pose challenges for potential study participants (Anderson, 2006), and considering the sensitivity of research with vulnerable populations (Sengupta and Lo, 2003). Others have found that UA/NS members of IRBs are often limited in number in comparison to other IRB members and are rarely designated as primary reviewers (Lidz et al., 2012). Interviews by Klitzman (2012) also found variation in how IRBs incorporate UA/NS members into the review process, as well as confusion across IRBs about the role of such members, and a perception that they are ill-prepared to contribute to IRB deliberations (Klitzman, 2012). However, more recent work examining the roles and voice of community members serving on a behavioral IRB found more active participation and, while a large focus remained on consent procedures, there was also a broader range of input given (Barton, Thominet, Boeder, & Primeau, 2018). This may point to changes in the roles of UA/NS members over the last decade or at least more attention to their potential.

Since the inclusion of UA/NS membership on IRBs is a regulatory requirement, and research indicates a potential changing landscape with respect to the roles and responsibilities of UA and NS members, it is important to understand what the identification, recruitment, training, and roles of UA/NS IRB members look like in practice today. In the present study we sought to shed light on these questions, both updating and expanding on existing research. Specifically, we sought to examine the following questions:

• How often does a single individual serve as both the UA and NS IRB member?

• How do IRBs recruit UA and NS members and are processes formalized?

• How are UA and NS members oriented to their service?

• What roles do UA and NS members play within the IRB?

MATERIALS AND METHODS

We conducted a self-administered online survey of individuals most familiar with IRB member recruitment at institutions across the United States.

Eligibility

Individuals were eligible for inclusion if they were staff members of an HRPP accredited by the Association for the Accreditation of Human Research Protection Programs (AAHRPP) and familiar with IRB member recruitment at their institution. AAHRPP is a global, independent, non-profit accrediting body for Human Research Protection Programs. For the purposes of the present work, we limited our sample to AAHRPP accredited organizations based in the United States and for whom compliance with 45 CFR 46 would be a common standard. In the United States, more than 60% percent of research-intensive universities and 80% percent of medical schools are AAHRPP accredited or have begun the accreditation process and thus AAHRPP members reflect a large proportion of IRBs reviewing research in the United States.

As the division of responsibilities may differ between institutions (i.e., the responsibilities for recruitment may be assigned to different job titles within different institutions), respondents could include members of the Institutional Review Board (IRB) or Human Research Protection Program (HRPP).

Identification and recruitment of participants

Each AAHRP-accredited institution lists one individual as a contact. Individuals listed as institutional contacts consent to receive regular email communications. The President and CEO of AAHRPP, also an author on this study (ES), sent these individuals an invitation email, which included a copy of the participant information sheet and a link to the online survey on behalf of the research team. The initial email was followed by two reminder emails distributed at 2-week intervals. As the research team did not have direct contact information for individuals and because the survey was completed anonymously, reminder emails were sent to all members of the original distribution list.

Invitations asked recipients to complete the survey – if they were the most appropriate individual – or to forward details of the survey to individuals within their organization for whom they felt the survey was most relevant; most likely a member of the HRPP staff or IRB member. Upon clicking the link, participants were presented with a brief synopsis of the study and the option of participating. A copy of the participant information sheet which included the 8 key elements of informed consent was also included as a hyperlink within this introductory text. Continuing onto completion of the survey indicated consent for participation. The sample size for the present study was informed by the available number of individuals as opposed to statistical considerations.

Survey Instrument

The survey was divided into three sections. The first section collected information regarding the operation of the HRPP/IRB(s) for which they were responding, including the type of institution (academic, health system, independent, or other), number of members that sit on full committee meetings, meeting frequency, number of submissions per full committee meeting, and meeting format (in person, virtual, or hybrid).

The second section collected information regarding UA/NS members. Participants were asked to indicate whether their institution routinely recruits one or more than one individual to serve the roles of UA and NS members. When respondents indicated that a single individual served both UA and NS member roles, a single response was obtained for all subsequent questions. When this was not the case, responses were separately collected for those serving in UA and NS member roles. Participants were asked about how UA and NS members are recruited, whether they had faced challenges recruiting for these roles, financial support provided to these members, and training and mentorship offered.

Finally, the survey collected information regarding the roles that UA/NS members play and the perceived value of UA/NS members.

Data collection

Data were collected through a self-administered survey. Study data were collected and managed using REDCap (Research Electronic Data Capture) tools hosted at the Ottawa Hospital Research Institute. REDCap is a secure, web-based software platform designed to support data capture for research studies (Harris et al., 2019; Harris et al., 2009). Survey responses were anonymous with responses identified by participant IDs only.

Data analysis

Survey items were a mix of closed- and open-ended questions. Data from closed-ended questions were exported to SPSS (IBM Corporation, 2021) and summarised using descriptive statistics. Open-ended responses were collected and imported into Excel (Microsoft Corporation, 2018) for analysis. The focus of the present analysis was to simply document the range of items raised with respect to the main open text questions (recruitment challenges, training provided) as opposed to developing an over-arching theory or framework. As such, written comments were analysed using a qualitative description approach (Sandelowski, 2000), a low inference approach in which text is described and coded without seeking to develop an overarching theory. Text was coded in duplicate by two authors (HAT and RJ). Each reviewer took a primary role in reviewing, developing and applying codes to a pre-specified question (recruitment and training, values and contribution) with the second reviewer verifying coding.

RESULTS

The flow of respondents is shown in figure 1. The survey was sent by email to a total of 217 eligible respondents. Of these, 15 of the emails were undeliverable leading to a potential sample of 202 eligible addresses. A total of 83 responses were received from these 202 eligible addresses (41% response rate). Of these, three respondents only completed the consent question and did not provide any further data, and a further 4 responses did not answer questions beyond demographic information about the IRB or institution. This left 76 responses for inclusion in the analysis (38% completion rate).

Figure 1:

Participant flow chart

As indicated in Table 1, the majority of respondents were from academic centres (45, 59%). Just over one quarter of respondents (20, 26%) indicated they were from a health system and 6 (8%) indicated they were from an independent IRB. Those indicating ‘Other’ included research organizations, cancer centers, and government agencies.

Table 1.

IRB/HRPP characteristics (N=76)

Item	N (76)	%
Type of institution
Academic institution	45	59.2
Health system	20	26.3
Independent IRB	6	7.9
Other	4	5.3
Missing	1	1.3
How often does the full committee meet?
Once a week	10	13.2
Every 2 weeks	18	23.7
Once a month	38	50
Other	10	13.2
Number of members on full committee
5	1	1.3
6	4	5.3
7	10	13.2
8	6	7.9
9	8	10.5
10	6	7.9
more than 10	41	53.9
On average, how many submissions does a Full Committee review each time it meets?
1 to 4	24	31.6
5 to 10	28	36.8
More than 10	23	30.
Missing	1	1.3%

The majority of respondents indicated that their full committee met once per month (38, 50%). Of those meeting more often, 18 (24%) met every 2 weeks, while 10 (13%) met weekly. Respondents indicated varying numbers of members: most committees had more than ten members (41; 54%), but a small number of committees had as few as five (1; 1%) or six (4; 5%) members. In contrast, there was a relatively balanced distribution of submissions (including initial reviews, amendments etc.) reviewed per meeting: 1 to 4 (24; 32%), 5 to 10 (28; 37%), more than 10 (23; 30%).

Appointment of the unaffiliated and non-scientist IRB member(s)

Of those who answered the question regarding whether a single or multiple individuals fulfil the roles of UA and NS member, four respondents (5%) indicated that their institution routinely recruits one individual to serve as both UA and NS member on each Full Committee, while 52 (68%) indicated their institution recruits a mix of individuals (i.e., some members are both UA and NS, some serve as one or the other) and 20 (26%) indicated that their institution routinely recruits one or more individual(s) to serve as the UA member and one or more individual(s) to serve as NS member.

In the remainder of the paper, we focus on data pertaining to the institutions in which either a mix of individuals or separate individuals serves as UA or NS members (n=72). These data have been grouped and reported together. Data for the small number of respondents indicating a single individual served in both UA and NS member roles (n=4) are provided in the Supplementary File 1.

Recruitment processes for unaffiliated and/or non-scientist members

Table 2 provides details regarding recruitment (n=72). Most respondents indicated a limited number of approaches to recruitment of UA members (median two approaches, range 0 to 6 approaches) or NS members (median of two approaches, range 0 to 4 approaches).

Table 2:

Approaches to recruitment for unaffiliated and non-scientist members of IRBs (n=72).

Item	Unaffiliated member		Non-scientist member
	N	%	N	%
Seek referrals from current IRB members	62	86.1	57	79.2
Recruit from own institution	24	33.3	48	66.7
Seek referrals from another institution	18	25	14	19.4
Other	17	23.6	8	11.1
Advertise online	10	13.9	10	13.9
Advertise at local events or community centers	4	5.6	2	2.8
Advertise in local print media	2	2.8	1	1.4

Note multiple selections were possible, so percentages sum to greater than 100%.

The most common approach for recruiting a UA member was through referral from current IRB members (62; 86%). This was followed by: recruitment from their own institution (24; 33%); and referral from another institution (18, 25%). Seventeen participants (24%) indicated that they used approaches other than those listed. These approaches also tended to rely on direct personal contacts and included referrals from other faculty and staff at the institution (but who did not serve on the IRB), referrals from other HRPP staff, business or personal interactions, or word of mouth. However, three respondents indicated recruitment was supported by direct community outreach through an Office of Community Engagement, or other community-based committees. One respondent noted that when an IRB member retires, they may continue to serve as an UA.

For NS members the pattern of approaches to recruitment was similar, however, there were substantial changes in the numbers for each response option; 57 (79%) sought referrals from current IRB members, 48 (67%) indicated recruitment of NS members from within the host institution, 14 (19%) sought referrals from another institution, while 10 (14%) advertised online.

For both the recruitment of UA and NS members, formal advertisement of the position appeared to be low: two (3%) and one (1%) respectively for advertisement in print, and four (6%) and 2 (3%) respectively for advertisement at events or community centers.

Challenges with recruitment

Thirty-four respondents (47%) reported challenges in the recruitment of UA members while 20 (28%) noted they had faced challenges when recruiting for NS members. A common challenge related to the time commitment. Several respondents specifically noted the length of the term of service required and the burden of attending monthly (or more frequent) meetings. In addition, the required training was seen to be a barrier in some cases. Others noted that they found it difficult to identify individuals who were not in some way linked to the institution. As, one participant noted:

“We are a large college community. Most people that live here have some affiliation with the university as faculty member, student, children as students, spouse employed by university, etc.” (ID#44)

Similarly, for the recruitment of NS roles, a noted challenge was that those interested in IRB service tended to be scientists and not ‘qualified’ to serve in the non-scientist role. However, an additional challenge appeared to be a lack of clarity around the definition of who a NS was. As one respondent noted:

“Most people interested in this have some kind of education or experience that makes it tough to confidently label them a non-scientist.” (ID#33)

Other challenges pertained to identifying members from diverse backgrounds:

“It is difficult to find ethnically and racially diverse individuals who are able to and interested in joining the IRB.” (ID#85)

This was sometimes explicitly linked to the approach to recruitment:

“We struggle with finding members from diverse backgrounds because we mainly rely on referrals of existing members.” (ID#8)

Other common challenges were broadly construed as the challenges of attracting individuals who are the “right fit” for the role. A full list of the reported challenges is provided in Supplementary Table 1.

Perceived roles and value of unaffiliated and non-scientist members

An independent perspective was a key role for UA members, with respondents noting they were “truly independent” (ID#23), and “not caught up in campus politics.” (ID#44) UA members were seen as having the same responsibilities as NS member, or in a few cases, bringing in specific outside expertise.

For both value and role responses, NS members were seen as having a role of bridging and broadening perspectives, bringing an outsider or contrarian viewpoint. Respondents stated that their NS members: “[are] able to question... policy,” (ID#19) “[break] up the echo chamber,” (ID#74) or that their view “broadens the perspectives of other members.” (ID#5) Comments implied or stated outright an identity and perception of the NS as outsider, with terms such as “bringing in” vs being inside, “outside” vs. Inside, “real world” vs academia, and other contexts that emphasized externalities and differences from mainstream members.

Notably, the NS was viewed as functioning as a proxy for a research subject. This was reflected in the statements about roles, such as: “we want them to have the role of participant” (ID#75) and that the NS “offers insight into the possible perception of the…participant” (ID#10). Indeed, a fundamental responsibility of both the UA and NS roles were ensuring that subjects are protected through informed consent.

Sometimes the status or remit of NS members was limited compared to other IRB members. One respondent, when asked to reflect on the value of NS members, indicated: “the non-scientist... allows the expertise in the room to focus on risk-benefit” (ID#12), explicitly identifying the NS as without expertise in this domain. Another noted that NS “participate nearly the same” (ID#14) (our emphasis). However, other respondents stated that their IRB saw NS members as “a reviewer like all the others” (ID#55) or noted that their NS members had specific regulatory or policy expertise.

Activities of unaffiliated and non-scientist members.

Table 3 provides data on the reported activities of UA and NS members. Forty-two respondents (58%) indicated that an UA member did engage in expedited or delegated review, and 50 (69%) indicated they acted as primary reviewer to some degree. Similarly, 46 respondents (64%) indicated that NS member did engage in either expedited or delegated review and 41 (57%) indicated that they at least sometimes serve as primary reviewer.

Table 3:

Roles of unaffiliated and non-scientist members (N=72)

Item
Are unaffiliated/non-scientist members ever engaged in expedited or delegated reviews?	Unaffiliated member		Non-scientist member
	N	%	N	%
Yes – all the time	11	15.3	7	9.7
Yes- often	6	8.3	16	22.2
Yes – sometimes	10	13.9	11	15.3
Yes – but rarely	15	20.8	12	16.7
No	23	31.9	18	25
Missing	7	9.7	8	11.1
Do unaffiliated/non-scientist members ever act as primary reviewers?
Yes – all the time	12	16.7	7	9.7
Yes- often	14	19.4	9	12.5
Yes – sometimes	15	20.8	18	25
Yes – but rarely	9	12.5	7	9.7
No	11	15.3	22	30.6
Missing	11	15.3	9	12.5

Orientation of unaffiliated and non-scientist members to their service and training provided

Respondents indicated a variety of ways in which new UA or NS members are provided training in order to serve as a member of the IRB. The most common components of training reported were: CITI (36, 50%); general locally developed IRB Member Training (32, 44%); orientation to electronic system used by IRB (20; 28%). Site-specific IRB member orientation (differentiated from training) was noted by 12 respondents (17%). Related, fourteen respondents reported members are mentored in their first reviews by veteran members and eleven respondents noted that new members observe several meetings before becoming a full member of the IRB. Only two respondents reported different training for UA and NS members. One respondent noted that UA members are counselled about their unbiased view as a UA member, and another noted that NS members are oriented to their role of representing a lay perspective. A full list of reported training content is provided in Supplemental Table 2.

Table 4 reports data regarding additional training for UA and NS members. Most respondents (57, 79%) indicated no additional training was provided to UA members above and beyond affiliated members. Similarly, 56 respondents (78%) indicated that no additional training was provided to NS members beyond what was provided to scientist members.

Table 4:

Additional training provided to unaffiliated and non-scientist members (N=72)

Item
Is additional training provided to unaffiliated /non-scientist members that is not provided to affiliated/scientist members of the IRB?	Unaffiliated member		Non-scientist member
	N	%	N	%
Yes	8	11.1	9	12.5
No	57	79.2	56	77.8
Missing	7	9.7	7	9.7

When additional training was provided, this covered role familiarization, training in research and research methods/design, and ethical principles. Notably, comments reflected that this was largely customized according to member needs and delivered one-on-one, sometimes by members of the IRB leadership. Of note, several respondents included references to Institutional policy and/or local laws.

Support for unaffiliated and non-scientist members

Table 5 provides details of support provided for UA and NS members. For UA members the most common financial supports were fixed honoraria (31, 43%), travel or parking expenses (14, 19%), and meals or meal expenses (12, 17%). In 32 cases (44%) respondents indicated that the financial support for UA members differed from the financial support for affiliated members.

Table 5:

Financial supports offered to unaffiliated and non-scientist IRB members (N=72)

Item
Does your institution offer financial support for unaffiliated members?	Unaffiliated member		Non-scientist member
	N	%	N	%
No financial support is provided	19	26.4	35	48.6
Travel/parking expenses	14	19.4	11	15.
Meal or meal expenses	12	16.7	11	15.3
Honoraria (fixed)	31	43.1	17	23.6
Honoraria (variable)	7	9.7	6	8.3
Substitution for lost wages	0	0	0	0
Other	3	4.2	2	2.8
Does your financial support for unaffiliated members differ to that offered to affiliated members?
Yes	32	44.4	15	20.8
No	31	43.1	47	65.3
Missing	9	12.5	10	13.9

For NS members the most common financial supports were the same; fixed honoraria (17, 24%), meal or meal expenses (15%) and travel or parking expenses (15%). Just over one fifth of respondents (15, 21%) indicated that the financial support for NS members differed from that provided to scientist members.

A total of 13 respondents (18%) indicated that changes had been made to facilitate the attendance of UA or NS members. When asked to provide more details, respondents discussed changes to meeting times – by holding meetings during lunch hours or outside of normal business hours – or by holding remote/virtual meetings which reduced travel. In one instance, the respondent noted that while the pandemic had created the need for virtual meetings, the institution planned to retain this format with only occasional in person meetings as it had facilitated attendance. Others noted how virtual meetings had eliminated or reduced parking issues.

DISCUSSION

In the present study we surveyed representatives from HRPPs/IRBs at AAHRPP-accredited institutions regarding the identification, recruitment, and training of UA/NS members of IRBs as well as the perceived roles of these members. Few respondents indicated formal mechanisms of recruitment, with variation in supports. However, there appeared to be core elements of training that were consistently referenced. Training provided was reported to be largely similar to that provided to affiliated and scientist members. Financial support for UA members was reported to differ from that offered to affiliated members to a greater extent than with NS members, yet at the same time three quarters of respondents indicated that they had not made changes to their process to facilitate the attendance of UA/NS members. Taken together, these results suggest that the approaches taken to identify, train and support UA and NS members are largely perfunctory. Perhaps most importantly, the perceived roles for UA/NS members tended to focus on participant-facing materials and the clarity of these, as opposed to larger issues of risks, benefits, or justice considerations, although UA members were perceived to have a more independent perspective. There appears to be a missed opportunity to consider and expand the role of UA/NS members beyond current norms. Further research ought to be conducted to better understand from the perspective of IRB leadership and UA/NS members what ought to be expected from those in the role of UA/NS members and how best to prepare them for that role.

Recruitment approaches tended to be based on existing relationships and networks, including referrals from a current member, recruitment from the host institution, or referrals from another institution as opposed to open advertisements for individuals meeting specific qualifications. The ad hoc nature of recruitment is consistent with the findings of Anderson (2006) and Cargill (2018), as well as work conducted in the UK by Humphreys (2012), indicating that this may be a common practice and one that has changed little over the last decade and a half. Moreover, as noted by Klitzman, while the referral of known individuals, or recruitment of retired former colleagues, may meet the technical requirements of regulations, such practices undermine the spirit in which the regulations were created and reinforce the somewhat tokenistic approach to identifying UA or NS members (Klitzman, 2012).

Particularly concerning was the finding that a third of respondents indicated that UA members were recruited from their own institution in apparent contradiction to OHRP requirements. Comments indicated challenges in determining who would constitute an UA or NS member; an issue not helped by differing guidance from OHRP, the FDA and SACHRP with respect to definitions of these roles (Speers & Rose, 2012). It should, however, be noted that this is not an issue unique to the United States, and that ambiguity regarding who should constitute ‘lay’ members has been raised in other countries, including Canada and New Zealand (Gremillion, Tolich, & Bathurst, 2015). This points to a need for greater clarity regarding these roles, as called for by Solomon (2016) and perhaps, as Lidz et al.,(2012) note, a need for greater attention to defining UA/NS members by who they are and positive qualifications rather than simply who they are not. We thus suggest that greater exploration is needed to understand the potential challenges in defining these roles and applying the definitions in practice, as well as work to improve consistency of the definitions across all regulations and policy recommendations.

Despite reporting many challenges to recruitment and retention of UA and NS members in particular, few respondents indicated broader community outreach, changes to facilitate these members’ participation in meetings, or tailored training – all things that could improve recruitment and retention but would of course need to be further evaluated. We suggest that more deliberate approaches to recruitment may mitigate some challenges but may also serve to create a more considered approach to who is fulfilling the UA and NS roles by requiring explicit delineation of the criteria and qualities being sought in members. Given the comments from some respondents regarding the challenges of recruiting members from diverse backgrounds this may, as per Cargill (Cargill, 2018), have the added benefit of improving the diversity within IRBs that has been noted here as well as in historic (Porter, 1986) and recent (Berry et al., 2019) work.

While many respondents indicated that UA and NS members did serve as primary reviewers at least some of the time, qualitative comments suggested that respondents do see UA and NS members as providing a different kind of a review than scientific members. They were commonly viewed as reflecting a lay or public perspective, as opposed to a specific role in serving as guardians against institutional self-interest (Lidz et al., 2012). Furthering our above suggestion regarding more considered recruitment approaches, we suggest that the envisioned roles and responsibilities of UA and NS members be given due deliberation such that the specific elements that each bring to the IRB are fully articulated. In this respect we point to the developing field of patient engagement in research, and institutions developed to support this, and where there has been great emphasis on developing clear expectations of roles for patient partners as well as methods for engagement (Arkind et al., 2015; Black et al., 2018; Concannon et al., 2019; Paul et al., 2020). Inspiration and models of good practices for recruitment, training, and support could also be taken from tribal and community-led IRBs that prioritize community voice in ethics review processes (Hull & Wilson Dine, 2017; Martin del Campo, Casado, Spencer, & Strelnick, 2013; Shore et al., 2014).

The most common compensation provided to UA/NS members was in the form of honoraria, although the percentage of respondents indicating that honoraria was provided was lower than in previous studies (Sengupta & Lo, 2003). Honoraria was reported to a greater extent for UA members than NS members, possibly reflecting the idea that IRB membership is within the remit of the role of anyone employed by the institution. Yet the relatively low levels of compensation or reimbursement for both roles potentially creates barriers to participation in what effectively amount to volunteer roles; a situation that may exclude certain groups or voices that are unable to provide uncompensated time.

Our study, of course, has several strengths and limitations. A strength is that, despite the relatively low response rate, we had a larger sample than many previous studies, with a well-defined cohort. Despite the large sample we cannot exclude selection bias in terms of the respondents; given the anonymous nature of the survey we cannot report details of who responded because we did not collect demographic data to facilitate response rates. Additional limitations include the fact that we did not have a sample frame available to sample UA/NS members themselves. This may have revealed different perspectives regarding roles and is worth exploring in the future. Finally, given the structured nature of the survey we were reliant on written responses to open questions, and unable to probe these responses for more detail. Further qualitative interviews or focus groups may, therefore, be beneficial in terms of expanding on our findings, particularly regarding the roles of unaffiliated and non-scientist members and the value they bring to the IRB review process.

In conclusion, respondents indicated challenges in defining UA and NS members and limited practices toward recruitment and support of UA and NS IRB members. We suggest that future work should more closely examine the challenges in defining these roles and applying the definitions in practice. Moreover, we advocate for greater consistency of the definitions across all regulations and policy recommendations. Further, we suggest that there is an opportunity to learn from the field of patient engagement in research as well as community-based boards to develop deliberate approaches to recruitment that may mitigate some of the identified challenges through explicit consideration of the criteria and qualities being sought in members. Future research may then include assessment of strategies that may improve recruitment and retention of UA/NS members.