Who owns your consent? How REBs give away participants’ agency

Abstract

We draw on three illustrative vignettes to examine how REBs manage participants’ agency in the context of qualitative research. We ask: Who owns a participant’s consent? Central to informed consent is the principle of Respect for Persons, which privileges the autonomy of individuals to make decisions about what happens (or not) to them. Yet, REBs sometimes require researchers to get permission from organizations to conduct research on their current and former members, even when the research is not about those organizations. Our aim is to raise awareness about the inherent contradictions of this practice and to consider guidelines for determining the appropriateness of involving organizations that may be tangentially connected to the research objectives or potential participants.

Introduction

[Respect for Persons] individuals should be treated at autonomous agents. ;. ;.capable of deliberation about personal goals. ;. ;. ;. To show lack of respect for an autonomous agent is to repudiate that person’s considered judgments, to deny an individual the freedom to act on those considered judgments, or to withhold information necessary to make a considered judgment, when there are no compelling reasons to do so (National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, 1978: 4–6).

Respect for Persons is one of three principles for ethical research. This principle underpins the informed consent process and privileges the rights of individuals to make judgments about what happens to them, including whether or not to participate in research. Respect for Persons triggers a host of procedural requirements, including communicating to participants their rights at multiple stages of the process (e.g. letter of information, consent form) and allowing participants to withdraw their consent (e.g. Guillemin and Gillam, 2004; Marzano, 2012; Thorpe, 2014). Some projects also build in practices such as member checking and debriefing, giving participants additional opportunities to review their data and withdraw their consent. Broadly, these practices attempt to maximize the “degree of control [participants have] over their participation” (Conolly, 2008: 206; for further discussions, see Locke and Ramakrishna Velamuri, 2009; Roulet et al., 2017).

Despite these requirements, Research Ethics Boards (REBs), which is the term for research ethics committees (RECs) or institutional review boards (IRBs) in Canada, sometimes award gatekeeper status to formal organizations that potential participants are or were afilliated with. While the role of gatekeepers has long been a prominent feature in discussions of research methods, they are typically focused on researchers negotiating entry directly with the individuals or groups who mediate access to a study’s setting and participants (e.g. Lincoln and Guba, 1985). Whyte’s (1943) access to the “corner boys” through Doc and Liebow’s (1967) “fortuitous” entry into the world of “corner life” are well-known examples.

We draw on three illustrative vignettes from our own research to raise questions about the degree to which REBs may compromise the autonomy and self-determination of individuals. Each vignette captures a different type of relationship between the participant and organization that REBs sometimes “give away” participants’ consent to. Through these examples, we problematize how awarding gatekeeper status to outside organizations at the expense of individuals’ agency may violate a fundamental principle that REBs are tasked to protect. Our aim is to raise awareness about the inherent contradictions of this practice.^1,2

The Respect for Persons principle

Respect for human dignity underpins modern day research ethics along three interrelated principles: beneficence, justice, and respect for persons (National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, 1978). In Canada, beneficence has been recast as Concern for Welfare (Canadian Institutes of Health Research, Natural Sciences and Engineering Research Council of Canada and Social Sciences and Humanities Research Council, 2022). Concern for persons attends broadly to the duty of REBs and researchers to protect the “physical, mental, and spiritual health, as well as their physical, economic, and social circumstances” of participants (Canadian Institutes of Health Research, Natural Sciences and Engineering Research Council of Canada and Social Sciences and Humanities Research Council, 2022: 8; emphasis added). This principle includes ensuring participants are not exposed to unnecessary harm and providing them with enough information to make an informed assessment about potential risks and benefits. Researchers must consider the broader impacts that participation in a study might have in a participant’s life, including harms that might arise along biological, psychological, and social dimensions. The principle of Concern for Welfare requires researchers to take a broad view of “risk of harm” and assess privacy and self-determination rights beyond the actual research intervention.

Justice, on the other hand, refers to the fair and equitable treatment of people to ensure that “no segment of the population is unduly burdened” by the risks associated by the research enterprise or denied access to participation or the benefits of research results (Canadian Institutes of Health Research, Natural Sciences and Engineering Research Council of Canada and Social Sciences and Humanities Research Council, 2022: 8). Researchers are required to ensure the fair and equitable recruitment of research subjects in light of the research question and any vulnerabilities that may compromise participants decision-making capacity. Researchers must also ensure that no group is arbitrarily excluded from participating or benefiting from the research.

Our paper addresses a third principle, Respect for Persons. Respect for Persons “implies that individuals who participate in research should do so voluntarily, understanding the purpose of the research, and its risks and potential benefits, as reasonably as possible” (Canadian Institutes of Health Research et al., 2018: 27). Respecting the autonomy of participants is a core tenet of this principle and “means giving due deference to a person’s judgement and ensuring that the person is free to choose without interference” or coercion (Canadian Institutes of Health Research et al., 2018: 6). The application of Respect for Person is often limited to the consent process at the point at which the researcher has engaged participants in some manner at the beginning, during or after data collection. Researchers are required to provide participants with a reasonable description of what their participation will entail, the scope of the research inquiry, and how their data will be used. In practice, consent is also specified as ongoing throughout the research process, starting with the initial contact with potential participates to providing them with an opportunity to withdraw their consent in future analyses or knowledge mobilization efforts (e.g. publications) (Canadian Institutes of Health Research et al., 2018).

The principle of Respect for Persons, however, is more broadly defined and does not assume that it begins with a researcher’s initial contact. Instead, it privileges the “moral obligation to respect autonomy” (Canadian Institutes of Health Research et al., 2018: 6), including, but not limited to, providing individuals with the “opportunity to choose what shall or not shall happen to them” (National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, 1978: 3). It asserts that individuals—even those who lack the capacity to provide consent—should “have the opportunity to participate in research that may benefit themselves or others” provided that the guidelines of research ethics are met (Canadian Institutes of Health Research et al., 2018: 27). According to these principles, inclusion of third parties is limited to people who are entrusted to make decisions on behalf of those individuals who are not able to exercise informed consent (e.g. minors).

Yet, REBs sometimes grant gatekeeping status to outside organizations even when potential participants are capable of self-determination. Awarding this gate-keeper status may be justified if the research is about the organization in question, and if the type of data generated by the research protocol may potentially compromise organizational activities (e.g. trade secrets). However, it is unclear why they are granted ownership of participants’ consent by REBs when the research does not directly consider the organization or when there are already safeguards in place to protect the organization from harm.

If individual autonomy is situated as a fundamental right and condition for honoring human dignity, it should supersede potential participant relationships with an organization. In the absence of foreseeable harm to the organization, what is the ethical basis of having participants’ agency managed by an external gatekeeper? For example: Should an employer have the right to gatekeep access to employees who are not “on the clock” and when the research is not directly about the workplace organization? Should an organization have claim over people who are no longer members? These examples are not hypothetical but derived from situations the authors encountered during the process of conducting research in the fields of criminology, education, and sociology. In this paper we provide three examples from our own research to problematize how awarding gatekeeper status to organizations at the expense of individuals’ agency may violate the fundamental principles that the REB are tasked to protect.

The use of vignettes in the social sciences

In the context of research, vignettes are used in a variety of ways. As an elicitation tool, they consist of short stories or hypothetical scenarios that research participants are asked to discuss. Vignettes create a context through which participants can explore the situational and contextual meanings derived from specific social interactions (Finch, 1987). Used in this manner, vignettes “allow actions in contexts to be explored; to clarify people’s judgements; and [. ;. ;.] provide a less personal and therefore less threatening way of exploring sensitive topics” (Barter and Renold, 1999: 1; see also Aujla, 2020; Jenkins et al., 2010). They also allow for comparison across similar cases and for the “features of contexts to be specified” such that attitudes, meanings, and normative judgments about specific situations can be thematically elucidated (Finch, 1987: 105; Iafolla, 2018).

Vignettes, however, can also be used outside of the research setting to illustrate real-life phenomena in service of a broader theoretical or methodological agenda. Accounts from battle-worn researchers about “the field” can illustrate the “underside” of our work (Hughes, 1971; see also Fine, 1993; Lareau and Shultz, 1996). We can learn about the production of data through our interactions with research participants, the methodological tools employed, the lines of questioning pursued, ethical or methodological issues that arise, and the factors that influenced the collection of data. We can also gain insight into the realities of data collection, including mistakes that are made along the way and strategies for resolving them (e.g. Langer, 2016; Sampson and Johannessen, 2020).

Beyond the possibility of improving the quality of our data and data collection techniques, vignettes can also serve to illuminate another dimension of the business of research; not only what happens “out there” in the field, but the process by which the enterprise of research unfolds. In the case of university research, REBs play a big role in this process by structuring the manner in which research is performed. This realization has inspired a cottage industry of articles about their reach (or ‘creep) (Haggerty, 2004), how they define harm and risk (Bracken-Roche et al., 2017), and how ethical guidelines are implemented (Karram Stephenson et al., 2020). The authors of this paper have had encounters with REBs that reflect this creep of REB into the research process, and we see the use of vignettes as a useful tool for comparing the (over)reach of the REB into the business of research in different substantive contexts.

In this paper we use vignettes to explore three types of relationship between researchers, organizations, and prospective research participants. These stories are drawn from our experiences negotiating access to our participants with our respective REBs and our reflections about how they award gatekeeper status to external bodies like work and schooling organizations. Through these vignettes, we reflect on the issue of participant agency, or the principle of Respect for Persons, and consider our own interactions with research ethics boards and with organizations related to our different programs of research.

We offer this paper as an entry point into problematizing the management of the process of consent when human subjects research involves, however directly or tangentially, the engagement of an organization. We explore different kinds of relationships between research participants and the organizations that structure their involvement in research. Exploring the relationships that individuals have as research participants (and researchers!) to the organizations that make up their fields of research yields important insights about risks and benefits to participants that arise in the context of research.

Active, arms-length and terminated relationships: Three vignettes

The vignettes detailed below capture three types of relationships that a potential participant may have with an organization: active, arms-length, and terminated. Active relationships are direct and ongoing connections between the potential participants and the organization that researchers may be required to gain permission from (e.g. workplace). Arms-length relationships may or may not be active. These relationships sometimes generate the involvement of third parties (e.g. workplaces, volunteer sites) when researchers want to speak to potential participants in their capacity as a professional or knowledgeable actor (e.g. interviewing a teacher about the state of education). Terminated relationships are relationships between individuals and organizations that are no longer active (e.g. former clients, former volunteers).

Our examples illustrate how the research process may be complicated by REBs when they require researchers to obtain permission to recruit participants from third parties. Former or current employers, instructors, medical professionals, or Boards of Directors are examples of third parties from whom researchers might be required to seek approval. These third parties (in our examples, organizations) are not the subject of scholarship, yet, in order to conduct research these third parties must approve of the research proposal as a condition of REB approval. In other words, the relationship between the participant and the researcher is mediated by the third party, which acts as the gatekeeper of the participants’ consent. As we discuss, when third parties are required to approve proposed research, they are given ownership of participants’ consent. Even if individuals are interested in participating (or even if they have), their consent can be withheld or withdrawn if third party organizations or institutions do not wish to approve the research. This gatekeeper status bestowed upon third party organizations has important implications for consent that, as we argue, undercut the agency and autonomy of participants.

Active relationships: The case of bank tellers

Active relationships are those relationships in which there is a direct, ongoing, and immediate relationship between the (potential) research participant and the organization from which consent must be obtained. Employment relationships are an obvious example of an active relationship. Individuals who have immediate and sustained relationships with organizations are conceived of as having active relationships precisely because of the continued commitment between the parties to the relationship in question. Students enrolled in universities, police officers working for police services, and judges officiating in the court system are examples of active relationships.

In some cases, it may appropriate for the organization to assent or consent to the research protocol, particularly in cases when the process of recruitment is facilitated by the organization or when the organization itself is either directly or indirectly “participating” by serving as the subject of the research. If the organization’s policies, procedures, or practices are the subject of the research, or if proprietary information is involved in data collection, it is reasonable that an organization would consider whether it wants its members to participate. ³ However, when the research is not about the specific organization or does not impose any reasonable risk to the organization, it is less clear why they are awarded gate-keeping status by REBs.

Vignette 1

A researcher who had previously worked in banking wants to conduct a study into the use of Canadian anti-money laundering legislation by front-line bank employees.

The research protocol specifically states that no one should disclose the name of their employer or any details which might identify a particular financial institution. The protocol promises anonymity to participants, including using pseudonyms and the de-identification of details that might reveal where a participant worked. No branch identification numbers, addresses, or participant names will be recorded. The researcher proposes to record consent orally, transcribe the interview, and destroy it after one year (or earlier if consent is withdrawn). The interview schedule does not include questions that are related to obtaining information about specific clients or uncovering specific trade secrets or to identifiable processes linked to any institution. Instead, the questions are couched more generally in relation to the laws governing the process of detecting suspicious financial transactions in Canada which are implemented in a highly standardized way across reporting institutions. As such, they are about a legally mandated process that is essentially public in nature.

The organization has protections in place to ensure sensitive or confidential information is not asked for, revealed, or reported in the context of the data collection or the dissemination of findings (e.g., peer reviewed papers). As a condition of employment, the potential participants are already contractually obligated to ensure client privacy and are prevented from divulging information that may compromise their employer or its clients. Having worked in the banking industry, the researcher is well aware of these parameters, as well as the processes through which she must request permission had she been studying the organization’s policies and processes. In this case, although the bank does not have an REB, it requires that any research be approved by the Vice-President of Community Relations and Public Engagement; ⁴ in more involved cases, approval may be required from more senior offices, including the Board of Directors. She is aware that other banks likely have such processes as well.

Despite the safeguards afforded to participants through anonymity and confidentiality requirements in research, the employees’ own contractual obligations related to confidentiality and privacy, and the fact that the research focuses on the use of the law and not on a particular organization, the researcher is informed by REB that prior to recruitment, she needs to obtain permission directly from each financial institution that employs potential participants. The researcher complies and obtains the cooperation of the participating institutions, however, rather than sending out an invitation to participate, she is informed that she will be assigned specific branches that have been randomly selected from which to recruit participants. To do so, she must contact the branch manager who will approve the participation of branch-level employees.

As a consequence, the institutions that employ participants become aware of recruitment efforts, as the research must be facilitated through the financial institution, which must identify branches where the research might be undertaken. The employer now knows specifically which branches have been invited to participate, and the manager of selected branches knows employees have agreed to participate. These developments mean that the REB becomes even more concerned that the anonymity of participants may be compromised, which then requires the researcher further assuage the REB’s concerns that the employer will not know which employees have participated – a concern the researcher had wanted to obviate in the first place by recruiting directly from personal networks. The requirements made of the researcher by the REB to protect the participants in the end was counterproductive as its intended result—improved confidentiality and anonymity—was in fact further complicated by the very process that was instated to protect participants. In the end, the research was approved but required much more engagement with the REB to assure the review board that participant privacy and anonymity were protected.

If the organization is protected and is not the subject of scholarly inquiry, it is unclear why it is positioned as a gatekeeper between the researcher and the research participant. Subject to constraints placed upon them by employment contracts or professional standards, people are free to talk about their experiences, including employment, academic or medical experiences as they see fit and with whom they see fit, in social settings ranging from conferences to cocktail parties. In this context, the participant is aware of his or her own obligations and what should be discussed and has the agency to decide what to say and to whom. Gatekeeping participation in active relationships presumes that researchers are the only people to whom individuals might speak with about the work they do for the organization in question. This is simply not realistic.

Certainly, not all third-party organizations have REBs—those organizations that do not conduct research as a part of their regular practices or mandates are unlikely to have a process for ethical review. Still, many organizations that conduct research at the level of the organization or otherwise engage with outside bodies do delegate such decision-making to internal bodies of oversight, such as Vice-Presidents, Boards of Directors, or marketing and communications departments, as the case may be. Businesses are keenly attuned to issues of risk—in particular, legal risk and reputation risk—and understand the risks and benefits that participating in research can bring at an organizational level.

Herein lies the key difference: the unit of analysis. The focus of the study is the use of the law within a business sector, not an evaluation of institutional practices or policies. The REB in this case has made a categorical error and attributed the authority for consent to participation to the institution on the basis of the risk to the organization, not the risk to the individual of participating. Indeed, requiring the researcher to obtain consent from the employer created an additional layer of risk to employees as employers were made aware the research was in progress. ⁵ In such cases, the gatekeeping of consent can undermine the goals of scholarly inquiry in addition to undercutting the agency of research participants, who may wish to share their knowledge with the researcher.

Arms-length relationships: The case of teachers

Arms-length relationships are those relationships in which participants are asked to speak to their professional experience or expertise. Individuals may have active and continuous ties with an organization; however, it is removed from the research objectives. The day-to-day functioning of a particular organization, for example, is not central to the research at hand. Talking to lawyers who work for firms about the practice of law, musicians who work for an opera house about the music industry, and doctors who work at a hospital about how medical education has changed are all examples of participants who can offer researchers with insight and “know-how” into their professional worlds without detailed discussion of a particular workplace. Yet, in some cases the employment or organizational relationship is seen to complicate the process of securing consent and trigger their involvement even when that relationship does not bear directly on the research.

Vignette 2

A student wants to write a paper about after-school extracurricular activities (e.g., robotics clubs, drama clubs, and sports teams). At the height of COVID-19, these activities are cancelled. The student wants to talk to teachers who typically coach these activities. The goal of the research is not to examine board practices but to consider teacher perceptions of the value of after-school activities and the implications of canceling them for students. From the instructor’s perspective, it would give the student an opportunity to gain hands-on experience gathering and analyzing data.

The student proposes to recruit a small sample of ten to twelve teachers through her personal networks and emails (e.g., Gmail). Teachers would be conducting interviews from their home since schools are closed at the time. The student and her supervisor undertake the ethics approval process through their university’s REB. They learn that their university REB approval may not be sufficient and are directed to every school board REB affiliated with the student’s personal networks. ⁶ The university’s REB cannot initially determine whether the project can proceed in the absence of multiple approvals from different school boards even though the project engages general questions about after-school activities and clubs. Since the student’s personal networks are spread widely, the project will require REB approval from at least four school boards. The research is dropped because further approvals from individual REBs would delay the research and compromise the student’s academic progress.

As in the case of active relationships, it is not the case that all third-party organizations have their own REBs through which to seek approval. Some organizations have alternate processes through which to seek approval for engagement with institutions, as discussed above, and these can range from formal presentations to boards of directors or Vice Presidents, as one of the authors has experienced while obtaining permission from organizations to engage in research that involves their employees. Other partnerships may take an ad hoc approach to ethics and discuss community- or group-specific issues or norms that may impact the research process. Like active relationships, it may be appropriate for the organization to assent or consent to the research protocol if they are involved (e.g. asked to facilitate recruitment) or implicated (e.g. the inclusion of interview questions that are about the organization) in some manner. It is similarly reasonable to include an organization if the research is potentially seeking proprietary or other sensitive information, or access to vulnerable populations (e.g. interviews with participants who are incarcerated, or surveys of children in school). But in the absence of direct or indirect inclusion in the research protocol or process, what role should third-party organizations play in overseeing the participation of people when the organizations are not the subject or focus of research? On what grounds should they be awarded gate-keeping status when the research is arms-length to the organization? ⁷ Yet, by virtue of their status as members of a particular organization, REBs are sometimes reluctant to extend to participants’ the ability to decide to participate in a scholarly project even when the research does not involve their contractual duties.

These relationships also raise questions related to role of the REB in terms of looking out for participants’ welfare. In Canada, one of the specific concerns that REBs must address is that the research is consistent with the principles of respect for persons and concern for welfare. In principle, the participant makes the decision to participate. The inclusion of a third party should be awarded in limited circumstances.

Authorized third parties acting on behalf of these individuals may decide whether participation would be appropriate. For the purposes of this Policy, the term “authorized third party” (also known as “authorized third party decision makers”) refers to any person with the necessary legal authority to make decisions on behalf of an individual who lacks the capacity to decide whether to participate or to continue to participate in a particular research project. These decisions involve considerations of Concern for Welfare and Justice (Canadian Institutes of Health Research, Natural Sciences and Engineering Research Council of Canada and Social Sciences and Humanities Research Council, 2022: 31; emphasis added).

Thus, the principle of Concern for Welfare requires that researchers and REBs consider the risks and benefits to participants across a variety of social, psychological, and physical dimensions, leaving the ultimate decision to participate to the individual taking part in research. ⁸ In conjunction with Concern for Welfare, Respect for Persons requires that the autonomy of the individual should be respected. In the examples of active relationships or arm’s length relationships outlined above, it is beyond the scope of REBs to delegate this authority to employers as decision-makers.

Terminated relationships: The case of former organizational members

Terminated relationships are relationships that no longer exist with the organization including former clients, employees, students, or club members.

Vignette 3

A researcher is interested in speaking to a specific type of professional who has left one career for another. To recruit participants, the researcher’s initial REB application proposes to contact the former colleagues (who are still working at the organization) of participants by email, asking them to pass along a ‘letter of information’ to their contacts. The protocol uses active consent; after the letter of information is passed along, only interested participants will contact the researcher. The researcher does not have access to the intended participants’ contact information. This protocol receives REB approval, and the researcher begins to collect data.

During a follow-up conversation with REB, the recruitment process is casually mentioned. The REB realizes it missed this detail about the recruitment process and goes into full-blown damage control mode.

The REB convenes a meeting that the researcher must attend in order to redress the issue to the REB’s satisfaction. This meeting is attended by employees in administrative positions of some note within the university. The researcher is informed that he has violated an REB requirement, placing his funding and reputation in jeopardy. The violation in question is not the nature of the research or a breach of the REB protocol. Instead, the issue in question was the method of recruitment that the researcher had received REB approval for. Despite approval to obtain active consent from potential participants, the REB is concerned that passing the letter of information on to former colleagues may result in participation due to social pressure. In other words, the REB is concerned that one former colleague forwarding a letter of invitation to another will result in the person who has left the organization feeling pressure to participate due to social desirability, even though the person forwarding the email has no real influence on the recipient’s career.

Further, the REB holds that contacting current members by email and asking them to pass along a letter of information to former members is considered ‘a use of an organizational resource,’ and that contacting former colleagues via their public emails is not the problem; however, asking them to contact their former colleagues is not appropriate, and is informed that approval notwithstanding, he and his collaborators should have known that the approved method of recruitment was not appropriate and was in fact a violation of ethical principles.

As a consequence of employing this method of recruitment—a method of recruitment which received approval from the REB—the researcher is informed that a letter regarding this violation will be sent to the agency funding this research. The researcher must revise his internal REB application, and he will also need to obtain REB approval from each organization through which he intends to contact former members, even though those organizations and their current members are not the subject of the research. The researcher is further required to destroy all data collected prior to getting REB approval from these organizations despite receiving active and informed consent from participants. As a consequence, the researcher is forced to destroy over 100 surveys and work on this project is halted for several months pending further REB approval.

The authors found the response of this REB to be quite dramatic: convening meetings, requiring the destruction of data collected in good faith by people who consented according to the terms of the REB’s original approval, and taking action that jeopardizes research funding is quite extraordinary. Yet, despite multiple meetings and attempts to reach a mutually satisfactory solution, the researcher was required to comply with the REB’s requirements or risk their funding, professional status, and the completion of this project.

This gatekeeping complicates the process of conducting research in numerous ways. We raise five key issues. First, the REB demanded additional REB approval from organizations that potential participants are no longer members of. While technically sending an employee an email to their work email does use an organizational resource, it demands significantly less resources than completing a full-blown REB review to access people who are no longer members. ⁹ Reading, forwarding, responding to, and deleting emails are routine activities by the members who work for the organizational setting in question, and it is difficult to see how a forwarded invitation to participate in a study imposes a risk that requires additional REB approvals.

The second issue relates to the appropriateness of REBs to withdraw participants’ consent retroactively. We ask: how is the destruction of data generated through informed consent consistent with the principle of Respect for Persons? In this example, the researcher was required to destroy all data generated through an active and informed consent process. The REB did not simply manage participants’ agency: it overrode it. In the process, not only did the REB violate participants’ agency to determine what does or does not happen to them, it demonstrated a lack of respect for the time and effort participants spent in good faith contributing to scholarship.

Third, there are broader methodological consequences to consider. Former organizational members are a type of “hidden population” that can be difficult to find, particularly if they have left no organizational or digital footprint (see Aurini et al., 2022). While the literature has tended to focus on hidden populations who engage in stigmatized or illegal behavior (e.g. Heckathorn, 1997; Watters and Biernacki, 1989), former employees, clients, students, or club members are also examples of populations that have may have no sampling frame or registry. In these cases, researchers may need to rely on knowledgeable informants to find participants. Informants, who are not the focus of the study, may be scattered across multiple organizations and sites that, if we rely on these guidelines, could trigger a cascade of REB reviews. In some cases, informants may not be affiliated with organizations that have REB offices in the first place.

Fourth, requiring researchers to gain REB approval from organizations that are only tangentially affiliated with the research project may dissuade researchers from attempting larger-scale qualitative or quantitative research. In this case, the researcher eventually had to complete an additional twelve REB applications. This is an onerous requirement and excessive oversight for what amounts to gaining permission to ask a current member to forward a former member an email for their consideration. This requirement also ignores the reality that REB approval is required by organizations that may yield no participants.

Finally, the ultimate course of action required by the REB suggests that REBs such as this one may be particularly attuned to the issue of concern for welfare—however, these concerns may not necessarily relate to concerns for the welfare of the participants but to the institution and the REB. These actions are protective of the institution and represent efforts toward damage control in favor of the university and REB. None of the steps taken addressed the purpose of the data collection, sought to mitigate the harm done to the researcher or the research process, or limit the overriding of the participants’ autonomy. For example, the REB could have undertaken a secondary review at the institutions and requested the research be retroactively covered under REB (which is permissible when scholars change institutions, provided they have REB approval at their original school). Additionally or alternatively, the REB could have required the researcher to provide the participants a letter offering them destruction of the data if they had concerns about the process through which it was collected. These actions would have respected the autonomy of participants instead of abrogating it entirely.

The authors do not wish to trivialize the work of the REB; the input of REBs can have beneficial effects on the research process and ensure that participants are treated fairly and are able to exercise their autonomy. The steps this REB took, as outlined here, suggest that the REB is working to serve a different function other than minimizing risks to participants and ensuring that research is conducted in an ethical manner. As such, this case points to an important latent purpose of the REB, which we suggest is legal and reputation risk management for the institution.

Discussion: The role of gatekeepers in research

The role of the REB is to ensure research meets ethical requirements. Requiring additional REB approvals across multiple organizations—when those organizations are tangential to the research—ignores the fundamental purpose of REBs: to ensure that the research conducted is ethical, and that the rights of participants are protected. These requirements can sometimes increase risks to participants by requiring researchers to inform employers of employee participation. They can also prevent research from taking place. In extreme cases, it can result in the destruction of data collected in good faith by researchers, itself a serious adverse impact on scholarship. In any case, requiring consent by institutions that are not involved in or the subject of scholarship before participants can consent to their involvement creates a gatekeeping role that undercuts the fundamental principle of Respect for Persons.

There is a rich literature on the role of gatekeepers—individuals who mediate access to sites and participants—in the context of research (e.g. Broadhead and Rist, 1976; Gouldner, 1954; Hammersley and Atkinson, 1995; Malinowski, 1967; Morrill et al., 1999; Patton, 2002; Shaffir and Stebbins, 1991; Shenton and Hayter, 2004; Van Maanen, 1998; Whyte, 1943). This literature tends to focus on the act of convincing the people “on the other side of the door” to let researchers into their social worlds (Feldman et al., 2003: 4). People on the “other side” are usually intimately tied to the site, people, or unobtrusive sources (e.g. documents) the researcher is interested in gaining access to (e.g. Grant, 2017). This literature also includes tales from the field as well as practical strategies for “getting in” (e.g. Clark, 2010; Marland and Esselment, 2019; Monahan and Fisher, 2015; Smith, 2012; Vuban and Eta, 2019). Researchers have documented the importance of fostering relationships with key informants, perverseness, and the role of luck (e.g. Jackall, 2009; Whyte, 1943).

By requiring third-party mediation of recruitment in the ways that we have outlined, the REB may run afoul of its own requirement to demonstrate Respect for Persons. In essence, by assigning gatekeeper status to organizations, the REB may allow gatekeepers to exercise control and agency over the autonomy of prospective participants. We see this maneuver as part of the increasing bureaucratization of universities (Heimer and Petty, 2010) and ethics “creep” that is based on hypothetic harms and risks (Haggerty, 2004).

We argue that the onus should be the REB to justify its inclusion in the first place and the grounds by which a participants’ consent should be managed by a third-party. Is the research directly about the organization including its culture, policies, or practices? If so, REBs should consider the following questions:

• ○ How do we maximize participants’ agency while also protecting the organizations that they engage with?

• ○ Are these risks already mitigated through other contractual or legal channels? Would additional organizational or REB oversight duplicate safeguards that are already in place?

• ○ Can potential risks to the organization be pre-emptively mitigated without organizational gatekeeping (e.g. de-identifying the organization and using pseudonyms)?

• ○ What are the foreseeable and reasonable risks to the organization? Are we unfairly assuming potential risks that are based on extraordinary examples that are completely misaligned with the study at hand (e.g. Stanford experiment) or that impose requirements based on “hypothetical” risks that have never happened? ¹⁰

• ○ Would the inclusion of the third party potentially place participants at greater risk? How can we mitigate these risks?

The authors value the work of the REBs in protecting participants, researchers, and organizations from adverse outcomes related to the research. Our paper does not question the necessity of REBs or critique standard requirements such as reviewing the aims, methods, and outcomes of research with participants, building in safeguards to protect all parties, and developing data management and storage procedures. The authors also acknowledge that the REB application process requires researchers to articulate a clear plan of action that is applied consistently and anticipate potential problems. The process of developing a standardized roadmap in advance can greatly enhance the quality and trustworthiness of research. There is value in an ethical review of research proposals, as it promotes the protection of participants and can identify methodological or procedural issues that researchers should address.

Yet, the intervention of REBs into research can undercut the fundamental ethical values that researchers are required to uphold. We question whether ceding control to outside organizations may “show a lack of respect for autonomous agent” and “deny an individual freedom to act on those considered judgements” (National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, 1978: 4). Informed consent is intimately tied to self-determination, autonomy, an understanding of one’s own best interests and welfare, and is a fundamental principle individuals enjoy not only in research, but more broadly in their lives. We agree that research which focuses on members of vulnerable populations may bring additional considerations—fundamentally, individuals at an enhanced risk of harm can benefit from additional protections and considerations that a broader assessment of risks might engage. However, the examples we have discussed demonstrate that instead of enhancing considerations and protections for participants, REBs may undercut the autonomy and dignity of participants when they act as gatekeepers and require researchers to obtain approval from other organizations.

Conclusion

A more realistic assessment of harms and risks is crucial if the autonomy of the participants is to be respected. When REBs situate third party organizations as the gatekeepers of consent, they run the risk of undercutting the fundamental principles of the informed consent process. The principle of Respect for Persons, coupled with Concern for Welfare and Justice, requires that potential participants exercise their autonomy to the extent they are capable. Requiring the approval of third parties who are not the subject of research erodes the informed consent process and runs the risk of preventing participants from making their own decisions about participation. Delegating approval can also affect research in other ways. We have discussed real adverse impacts on research, including recruitment, the destruction of data collected in good faith, and the obstruction of research altogether. Requiring researchers to obtain approval from third parties may also indicate that REBs are more focused on managing the risks (e.g. legal, reputation) than they are on supporting research that aligns with ethical norms.

The purpose of developing standards for informed consent was to honor the agency and autonomy of people who are invited to participate in research. Autonomy is central to the informed consent process. Demonstrating Respect for Persons in this way can encourage a happier marriage between REBs and researchers, preserve the integrity and purpose of the research design and process, and more importantly, better align the work of REBs with the spirit of the three pillars that underpin ethical guidelines.