The Ethics of End-of-Trial Obligations in a Pediatric Malaria Vaccine Trial: The Perspectives of Stakeholders From Ghana and Tanzania

Respondent Disposition

Figures 1 and 2 below describe the disposition of respondents. In total, there were 52 key stakeholder interviews. Figure 1 presents respondents’ stakeholder roles in relation to the PMVT. Figure 2 describes the respondents by their country location.

OPEN IN VIEWER

Figure 1.

Doughnut chart presenting number of respondents by stakeholder roles in the pediatric malaria vaccine trial (N = 52).

OPEN IN VIEWER

Figure 2.

Bar graph presenting country locations of respondents.

There were four main result themes: (a) Communicating End-of-Trial, (b) Maintaining Health care Services, (c) Dissemination of Results, and (d) Post-Trial Access.

Qualitative Results

Communicating the End-of-Trial

Clear communication throughout a research study and especially at the end of a trial is necessary for local trial populations to benefit from knowledge generated by the research (negative or positive results), and to support local health care provision, even after the resources provided by a study program have stopped. The end of a trial can present hardship or cause for concern for the communities involved in a research program and the health care setting more generally. Sensitive communication and attention to these issues are respectful and necessary to support the allocation of resources, staff, and health care responsibilities. This is crucial both for the promotion of public health, and the prevention of harm in local populations where health research is conducted. Moreover, a carefully structured exit strategy ought to maintain and instill trust in health care services and encourage future research partnerships. Importantly, if participants or the wider population feel abandoned or exploited by a research program, they are likely to reject and distrust not only research teams but also public health services as a whole. Such a break in trust can be detrimental to an individual’s health, or health of their children, should they fall ill, and no longer have confidence in seeking professional help. Furthermore, a break in professional trust can threaten public health more generally (particularly in a context of infectious disease prevention, reporting, and control).

A number of interview responses, in particular from clinicians and fieldworkers, suggested that participants of the PMVT felt disappointed with the study ending and had concerns about their future health. Vulnerabilities and anxieties of this nature must be appropriately addressed with reassurance and sufficient guidance to transition participants back into available public health services where possible. Moreover, the interview responses highlighted that much of the complex day-to-day ethical decision making is managed by the fieldworker-teams on behalf of the research institute. Institutional support is needed to assist fieldworkers in this role with sufficient training in value-based decision making that best promotes patient safety and supports local health systems. A number of the senior researchers stated that a respectful process of disengagement needed meticulous planning across gradual stages, as was implemented in the PMVT. For example, it was mentioned in the interviews that over the course of the year leading up to the end of the vaccine trial, the participants (mothers of infants) were informed on different occasions that the trial was coming to an end. A method also described by Kamuya et al. (2013).

There can be significant public health benefits from integrating a culture of research into a health care setting and supporting sustained conditions of improved health in a population. There was a divergence of opinion as to whether this benefit could in practice be sustained. On one hand, the health education of the trial remains with community members and thereby improves health-seeking behaviors in the local population. On the other hand, one senior researcher described the participants as having been “over-pampered” and not able to cope with the transition from the controlled research setting to managing their own (and their children’s) health needs. Novel approaches are required to safely support the transition of health care responsibilities of participants (in this case, mothers and infants) and secure beneficial health-seeking behavior accrued during the study. On the whole, it appeared that the end of the trial in the health care settings was managed between researchers and local health care facilities with little, or no, direct input from the sponsor–investigators. As a result, the management of local health needs at the end of the study varied, with no unified agreement on the responsibilities of the research team toward the health of local populations, or how the transition should be handled. This is in contrast to the trial set-up and conduct of the PMVT, which had been highly scrutinized to ensure uniformity across the research centers. For example, one research center took a decision to support local community health by funding health insurance for participants at the end of the trial to reduce disruption to access for health service provision. This was not provided by all research centers or coordinated by the sponsor investigators. Although each group needs to establish contextually relevant approaches, the divergence in management at the end of the trial calls into question whether the health needs of all participants were adequately and consistently addressed by the research program. Moreover, ethically, what are the sponsor–investigator responsibilities toward health care provision when a trial ends? The design of the exit strategy must be appropriately tailored with and for the local setting. Researchers and sponsors must remain committed to supporting the public health goals of the local populations. To a greater extent, the PMVT achieved this through improving local research, clinical and laboratory skills, supporting local health education, and promoting good health-seeking behavior in the community. Establishing the barriers and enablers of sustainability for each of these system strengthening components would help inform end-of-trial responsibilities and further support long-term improvements in local conditions of health.

Maintaining Health Care Services

All respondents agreed that over the course of the PMVT, there had been substantial positive changes in health facilities—improved infrastructure, staff and care standards. This service improvement ensured participants received adequate standards of care as set out by international and national standards (International Council for Harmonization-GCP, 2017; WMA, 2013). Maintaining these beneficial changes beyond the end of the study was a challenge recognized across the different stakeholder groups. These concerns aligned with recent literature calling for more empirical work to fully understand the consequences and benefits of international collaborative health research operating in weak health care settings (Angwenyi et al., 2015). We recommend that future programs of health research and the local health care services design a range of indicators to assess the impact on the health care setting. Involving a capacity developer partner could help with this aspect. For example, the PMVT involved the INDEPTH Network Malaria Clinical Trial Alliance as capacity-developers (Ogutu, Baiden, Diallo, Smith, & Binka, 2010). An alternative option would be to involve an organization such as Council on Health Research for Development (COHRED, 2015), which offers helpful monitoring tools, such as the Research Fairness Initiative (RFI). Health research impact data help inform shared system learning, co-ordinate capacity building efforts and increase the overall social value of a collaborative partnership. Measuring the system impact of a study will identify the positive achievements of integrating a program of research into a health care setting. This impact data can thereby inform and provide an impetus to find new mechanisms for system strengthening, providing quality care services and improving local conditions of health.

Across the interviews, a number of responses from medical professionals stated that the end of the PMVT would result in reduced standards of care, decreased treatment availability and constrained health access. Questions were raised over the issue of whether future maintenance costs for equipment and additional staff could continue. Such fluctuations in services were raised as a concern because of the destabilizing effects this has on the local health system. This system impact at the end of a trial, as previously reported, has the potential to disrupt care and to demoralize hospital staff (Merritt, Katz, Mojtabai, & West, 2016). By contrast, it was also mentioned in a number of interviews that the skills and education training attained by local research teams and health professionals were considered sustainable because these skills had the potential to be passed on between colleagues leading to overall strengthened hospital standards.

The objective of health research and partnership is to improve health and, this needs to be a guiding principle in both the conduct and outcome of the process. An earlier study by Angwenyi et al. (2015) suggested “developing a staggered strategy of exit and hand-over of responsibilities and equipment to MoH (Ministry of Health) or other key local actors. A negotiation process also needs to be in place to handle unmet expectations” (p. 16). Aiming to protect the continuity of care and care standards should be a guiding objective of any handover process. This is an area of further research which would be well served by more research partnerships sharing their experiences, as presented here in respect of this PMVT. Although there is extensive literature and ethics guidance on community engagement at the start of a study, there is very limited guidance on community debriefing at the end of a research study.

Dissemination of Results

The interview responses show that dissemination of results must reach a broad cross-section of society. A clear understanding of the political and cultural setting will inform which institutions, authorities, community representatives, and opinion leaders are to be included and notified of the results. For example, one interview respondent mentioned that it was important to involve the local church and mosque in that particular setting. Identification of the appropriate opinion leaders needs to be supported by community liaison teams, participant advisory groups, health authorities and sociologists (M. Molyneux, 2017; Schulz-Baldes, 2006). Time and careful assessment are needed to understand who represents a community and in what capacity. Moreover, it is important to avoid endorsing inequitable structures in a region, for example, partisan politics, gender inequalities, local rivalries, or discrimination (Alvarez-Castillo & Feinholz, 2006). A further aspect to consider is the order in which you approach different members of a community. Guidance on local hierarchies and social customs are important to maximize the social value of results and the acceptance of findings among local and national health authorities; a vital step for policy change and improved health care practices (Wenner, 2017). The dissemination strategy needs to be directed and appropriately structured to facilitate the translation of results with tailored dissemination tools (accounting for education, training, and background context) and clear research messages.

The content of results is also important. Beyond providing information on the test product, there is often other relevant health-related information that can be shared, such as measurements of quality assurance, standards of care, epidemiology, and public health indicators. Notably, a point that was mentioned in many of the interviews is that a dissemination strategy is an ongoing process extending from community engagement at the start of a trial and continued through system strengthening and sharing in the final results at the end of the trial.

Post-Trial Availability

The central message from the interviews on the topic of post-trial access was that there is a lack of clarity around the methods of implementation. The inability to define an actionable approach for post-trial access to effective proven interventions has become a major barrier to fulfilling the ethical requirement and, benefiting trial populations. In the interviews, one ethics committee member explained that committing to post-trial provisions in an ethics review application form is routine for research groups, but implementation of this benefit in practice is rare. Further meaningful discussion and better guidance is required to support post-trial access and the translation of research into improved local conditions of health (Cook, Snyder, & Calvert, 2015; Haire & Jordens, 2015; Hurst, 2017). Notably, post-trial availability was almost exclusively only mentioned by the ethics committee members; reaffirming the point that it is ethically undisputed but practically challenging. One senior researcher made the point that the posttrial-access concept had no practical application in the context of vaccine research because of the time delay between the trial results and the licensing of a successful product. A single trial conducted in the community will not lead to a product that can immediately be administered to the participants or their wider community. Further steps are needed along the development pipeline before an intervention can and should be made available (London & Kimmelman, 2008). As such, clinical trials and other health interventions will rarely lead directly to post-trial access. It is argued by some commentators that it is important to identify this time-lag in a research proposal and offer other forms of compensation in lieu (Largent, 2017). Others argue that although post-trial availability is not the only condition of ethical acceptability, it is still a “salient” condition of international research in low-resource settings (Haire, 2016). A similar point was made by respondents in the interviews that, on one hand, post-trial availability alone does not ethically justify the conduct of health research. Yet, on the other hand, not considering an actionable post-trial access treatment pathway is unethical. This is especially true in poorer countries where state finances and policy structures are not readily in place to deliver on proven-effective interventions. From a public health perspective, the obligation ensures that the research is driven to improve local conditions of health. Furthermore, extending this principle, Haire and Jordens (2015) argue that “failure to implement research findings, when viewed through a quality improvement lens, is a breach of the social contract made by researchers and policymakers with the communities that participate” (p. 16). In the interviews, it was stated that although post-trial access may be a benefit that is only awarded in the long term, it does need early planning and commitment from a wide range of global and local representatives. As such, this places an onus on countries to strengthen overall administrative and operational systems in preparation to be able to benefit from a preferential rollout strategy (Romore et al., 2016).

A few of the researchers stated that arranging country preparedness falls outside the remit of their responsibilities. The duty of facilitating post-trial availability is therefore argued to be incorrectly levied against the program of research. Commentators have shared this concern before (Nuffield Council on Bioethics, 2002). The main arguments stated in the literature are that the research team does not control the drug approval process in any given country, has limited research funds which are often restricted for defined activities of research and moreover a research team would be conflicted if they attempted to influence the political processes that define government spending decisions (London, 2005). We argue that these are organizational issues, but not sufficient grounds or ethical justification for denying participants and communities benefit for the risks that they are undertaking in research. The latest updates of both the Declaration of Helsinki (2013) and Council for International Organizations of Medical Sciences (CIOMS, 2016). Ethical Guidelines state post-trial access is an ethical requirement that ought to be addressed with a wide range of stakeholders sharing in that responsibility, including for instance “sponsors, researchers and host country governments and other relevant parties” to deliver on this commitment (CIOMS, 2016; WMA, 2013). A major advantage reported by interviewees in the functioning of the PMVT was the early involvement of the national food and drug authority (Food and Drug Administration [FDA]) and governmental officials, so that any positive trial findings could be efficiently scaled to the national level. Further monitoring of this novel partnership model as the vaccine moves through later development and regulatory stages would provide additional helpful insight on the role and responsibilities of collaborative research and post-trial access (World Medical Association, 2013).

A possible alternative is to establish, alongside a study trial, a specific board dedicated to implementing post-trial availability; constituting such a board could become a requirement for ethical research approval. This board may be chaired directly by an ethics committee or an independent mediator and would involve all necessary stakeholders for the rollout of an intervention. At present, researchers seem unprepared and poorly placed to find practical, feasible and legally plausible mechanisms to deliver on post-trial access. This creates a barrier, but the ethical principle is still valid (Cook et al., 2015; Grady, 2013).

Communication

Dissemination of Results

Maintaining Health Care Services

Post Trial