South African Research Ethics Committee Review of Standards of Prevention in HIV Vaccine Trial Protocols

Sampling

Respondents ¹ (REC members, Community Advisory Board (CAB) members, site staff, and sponsor representatives) were purposively selected (Mason, 2002) to participate in the larger study (Essack, 2014, 2015). This article reports on the perspectives of REC respondents only (N = 8), who were selected due to their involvement in the review of HVTs. Purposive sampling was “based on careful selection of cases that are typical of the population being studied” and is “often used to create small, relevant samples in qualitative research” (Terre Blanche, Durrheim & Painter, 2006, p. 563). Some respondents were also selected using snowball sampling techniques (Biernacki & Waldorf, 1981), for example, where respondents suggested other suitable interviewees.

At the time this research was undertaken, there were five trial sites conducting prophylactic HVTs in South Africa, and two trials were ongoing. This study explored standards of prevention practices for both trials (Phase I and Phase IIB) at all five sites. The Phase I trial, conducted at two sites, investigated the safety of a vaccine in HIV-uninfected participants at low risk for HIV. The Phase IIB trial, conducted at five sites, investigated the safety and efficacy of a vaccine in HIV-uninfected participants at high risk for HIV. Importantly, standard of prevention concerns may be different for participants at lower risk of HIV acquisition in a Phase I trial compared with high risk participants in later phase studies.

Because this research was qualitative, predefined sample size calculations were not warranted. Rather, sample size was determined by respondents’ willingness to participate. Eight REC members, including four REC chairs participated in in-depth interviews. This study used a combination of face-to-face and telephone interviews to collect data, as the sample was geographically diverse (Opdenakker, 2006). Telephone interviews provided a useful method to minimize the cost of travel and to access participants outside of South Africa (at the research network). An important limitation is that telephone interviews lose access to social cues (e.g., body language). However, telephone interviews “may provide information quite comparable to in-person interviews” (Sturges & Hanrahan, 2004, p. 116).

Documents were analyzed between July 2009 and August 2012, and interviews were conducted between August 2010 and August 2012 (see Essack, 2014, 2015 for a more detailed description of the methodology with the larger sample of study participants). Interviews were semistructured, broadly guided by key questions and offered “an adaptable and reliable means to gather the kind of data needed to conduct empirical bioethics research” (Sankar & Jones, 2008, p. 117). Where relevant to the research questions, additional questions (probes) were used to get respondents to elaborate on particular issues.

The semistructured interview comprised an exploration of interviewees’ perceptions of general issues regarding HIV prevention practices, their perspectives on ethics guidelines, and challenges and successes. Respondents provided their individual informed consent to participate in interviews and for the audio recording of interviews. The interview guides evolved over the duration of the study as findings of new prevention interventions became available and based on important issues identified in earlier interviews. Typically, interviews lasted between 45 and 60 minutes.

Interviews were transcribed verbatim to capture pauses, speech repetitions and overlapping talk. Text was coded using a critical thematic analysis of the data (Braun & Clarke, 2006). Thematic analysis is well suited to revealing the specific way in which individuals or groups conceptualize the phenomenon under study (Joffe, 2011) and is not intrinsically linked to a particular paradigm and can be accommodated by a range of epistemological approaches (Joffe, 2011), including critical realism (Braun & Clarke, 2006) and interpretivism (Terre Blanche, Durrheim, & Kelly, 2006). The analysis broadly followed stages of analysis identified by Braun and Clarke (2006) and Terre Blanche et al. (2006). However, the process was iterative rather than occurring in six discrete phases.

REC members were interviewed on their practices and perspectives pertaining to the review of HVT protocols, particularly concerning standard of prevention considerations. The study was approved by several RECs, namely, the Biomedical Research Ethics Committee, University of KwaZulu-Natal (BE 241/09); Human Research Ethics Committee (Medical), University of the Witwatersrand (M091140); Medunsa Research Ethics Committee, University of Limpopo (Medunsa Campus; MREC/P/13/2010: CR); Human Research Ethics Committee, University of Cape Town (REF 476/2009); and the Committee for Research on Human Subjects (Medical), Walter Sisulu University (089/009).

REC Review of Protocols

All ethics guidelines mandate the ethical review of research protocols (cf. MRC SA, 2003; UNAIDS & AVAC, 2011; UNAIDS & WHO, 2012). RECs review protocols to determine whether they meet ethical and scientific standards for conducting research with human participants. The REC has the power to approve, request modification, or disapprove a research protocol. Independent review is one of eight principles identified for conducting ethical research in developing country contexts (Emanuel et al., 2004). This entails mandatory review by independent bodies, such as RECs and regulatory authorities.

Ethics review of HVT protocols typically considers the prevention package that will be promoted in trials to comply with ethics recommendations and with the scientific imperative to determine the safety and efficacy of the experimental vaccine “against the background of established, ongoing or planned prevention modalities” (Tarantola et al., 2007, p. 4865). Applicable ethics guidelines require that RECs approve the standard of prevention provisions (MRC SA, 2003) and plans for monitoring risk-reduction interventions (MRC SA, 2003; UNAIDS & WHO, 2012).

RECs were perceived by site staff and network representatives as critical stakeholders in the oversight of study protocols, including regarding the standard of prevention (Essack, 2015). REC respondents described that when reviewing protocols they generally considered the “presence or absence of risk-reduction” (Z16, REC). In addition, some REC respondents reported that there was a minimum standard of prevention that must be provided to participants in HVTs, namely condoms, STI treatment, and risk-reduction counseling. This is on par with the minimum standard of prevention offered in most HIV prevention trials (cf. McGrory et al., 2010). REC respondents reported to rarely weigh in on the intricacies of the standard of prevention but described occasionally influencing standard of prevention implementation practices: “. . . we were one of those who actually pushed for the syndromic approach” (C21, REC). Syndromic management of STIs entails identifying symptoms and providing treatment for all infections that could cause those particular symptoms (DoH, 2008). The two protocols reviewed in this study did not specify plans for monitoring risk-reduction interventions and REC interviewees confirmed that they did not review plans for monitoring risk-reduction interventions. More often, RECs sought only generic assurance that researchers would take steps to help keep participants HIV-uninfected and paid less attention to the actual components of the package or related implementation practices. These findings conflicted with assertions that RECs expect researchers to specify “the frequency, mode, and personnel responsible for delivering the service and the infrastructure used” (McGrory et al., 2010, p. 17).

REC respondents generally reported having excellent working relationships with South African HVT investigators, developed on a foundation of trust that investigators “completely want to do the right thing in terms of standards of prevention” with the affirmation that “. . . I haven’t had in . . . my years of being the chair of the committee any acrimonious arguments around standards of prevention issues” (C19, REC). Investigators too, described the relationship with RECs as good and founded on “good communication and mutual respect” (C11, Site Staff, Site 5). Our data showed that in some circumstances investigators have approached RECs prior to protocol development regarding challenges with sponsor policy to “flag what they think we . . . would have a look at” (C17, REC). In cases where sponsors and other stakeholders are unwilling to commit to providing a high standard of prevention, such prior discussions may prompt RECs to probe this reluctance to raise sponsor support of improved standards of prevention.

Comparison of Protocol Review Practices With Ethics Recommendations

Community representatives and local RECs are required to review HVT protocols and informed consent materials regarding the standard of prevention and the monitoring thereof (cf. MRC SA, 2003; UNAIDS & WHO, 2012). In congruence with ethics guideline recommendations, it was reported that for both trials, RECs reviewed the protocols and related consent materials. However, practices deviated from guidelines in terms of REC review of risk-reduction monitoring plans—such plans were not included in protocols or ethics applications. REC review of standards of prevention in protocols and informed consent forms (ICFs) was limited to ascertaining the presence of such a package.

A related study on actual practices at trial sites found that in practice, sites provided substantially more to participants than was specified in consent documents for both trials and in the Phase I trial protocol (Essack, 2014). REC practices for the review of protocols were largely consistent with ethics guideline recommendations (MRC SA, 2003; UNAIDS & AVAC, 2011; UNAIDS & WHO, 2012), except for the requirement that RECs approve plans for monitoring risk-reduction interventions—such plans were not included in study protocols or ethics applications. Therefore, ethics recommendations for protocol drafting and review, and ethical oversight of standards of prevention were only partially achieved. Nevertheless, should RECs review all aspects of the standard of prevention as recommended in guidelines, it is likely that they would face challenges with conflicting guidelines and/or guidelines that impose standards that may be difficult to achieve in practice. Ambiguous and conflicting ethics guidelines are themselves a barrier to ethics review.

Variability in the review of HVT protocols by RECs suggested variable research ethics capacity (among other issues), which limited the ability of RECs to interrogate major substantive issues, such as standards of prevention. In line with NHREC guidelines that REC members receive initial and ongoing research ethics training (Cleaton-Jones & Wassenaar, 2010), RECs could benefit from intensive ethics training including on mechanisms to enhance their interrogation of substantive ethical issues. The lack of accessibility, awareness, and understanding of HVT-specific ethics guidelines (Moorhouse et al., 2014) also emphasizes the need to build ethics competency among all HVT stakeholders as a matter of priority.

REC review of the standard of prevention in protocols and ICFs was limited to ascertaining the presence of such a package, rather than addressing the more substantive issues. Similarly to reports in this study, a previous survey of South African RECs also indicated that consideration of substantive ethical issues—like standards of prevention and care—are impeded by procedural and bureaucratic demands (Moodley & Myer, 2007). Findings of variable research ethics capacity on RECs and the need to build health research ethics capacity on these committees is supported by previous research. Milford, Wassenaar, and Slack (2006) found the majority (73%) of REC members who had previously reviewed HVT protocols “agreed that there was a lack of general and sufficient ongoing training for members in health research ethics” (p. 5). In response to this need, there have been several initiatives to build health research capacity in Africa, detailed in Ndebele et al. (2014).

Our sample size is small and purposive. Therefore, the results presented here may not reflect those of all REC members in South Africa. However, given that the concerns raised by RECs in this study resonate with concerns in the literature as we have shown, it is possible to argue that the types of issues raised by the sample are the types of issues that might be identified in the general context of HVT ethics in South Africa.