Engaging the public in priority setting for health in rural South Africa

Abstract

Introduction: The importance of public engagement in health priority setting is widely recognised as a means to promote more inclusive, fair, and legitimate decision-making processes. This is particularly critical in the context of Universal Health Coverage, where there is often an imbalance between the demands for and the available health resources. In South Africa, public engagement is protected in the Constitution and entrenched in policy documents; yet context specific tools and applications to enable this are lacking. Where public engagement initiatives do occur, marginalised voices are frequently excluded, and the process and outcomes of these initiatives are not fully evaluated. This hampers our understanding of public engagement approaches and how to meaningfully include important voices in the priority setting agenda. The aim of this doctoral (PhD) research was to investigate the feasibility and practicality of including the public in resource allocation and priority setting for health in a rural setting in South Africa using an adapted deliberative engagement tool called CHAT (Choosing All Together). Methods: The PhD involved the modification and implementation of the CHAT tool with seven groups in a rural community in South Africa to determine priorities for a health services package. For the modification of CHAT, desktop review of published literature and policy documents was conducted, as well as three focus group discussions, with policy makers and implementers at national and local levels of the health system and the community, and modified Delphi method to identify health topics/issues and related interventions appropriate for a rural setting in South Africa. Cost information was drawn from various national sources and an existing actuarial model used in previous CHAT exercises was employed to create the board. The iterative participatory modification process was documented in detail. The implementation process was analysed in terms of the negotiations that took place within the groups and what types of deliberations and engagement with trade-offs the participants faced when resources were constrained. In terms of the outcomes, the study focused on what priorities were most important to the rural community within a constrained budget and the values driving these priorities, but also how priorities might differ amongst individuals within the same community and the characteristics associated with these choices. Qualitative data were analysed from the seven group deliberations using the engagement tool. Content analysis was conducted, and inductive and deductive coding was used. Descriptive statistics was used to describe the study participants using the data from a demographic questionnaire and to show the group choices from the stickers allocated on the boards from the groups rounds. The investment level (sticker allocation) of all study participants was recorded at each stage of the study. From these the number of stickers allocated to each topic by the participants was calculated by adding up the number of stickers across interventions selected by the participant by topic. The median and interquartile range across study participants was calculated for the topic totals. To examine differences in sticker allocations, Wilcoxon rank sum tests were performed for differences across participant categories and sticker allocations in the final round of CHAT. Findings: Based on the outcomes, seven areas of health need and related interventions specific for a rural community context were identified and costed for inclusion in the CHAT board. These include maternal, new-born and reproductive health; child health; woman and child abuse; HIV/AIDS and TB; lifestyle diseases; quality/access; and malaria. The CHAT SA board reflects both priority options of policymakers/ experts and of community members and demonstrates some of the context specific coverage decisions that will need to be made under NHI. The CHAT implementation shows that the rural communities mostly prioritised curative services over primary prevention due to perceived inefficacy of existing health education and prevention programmes. The exercise fostered strong debates and deliberations. Specifically, the groups engaged deeply with trade-offs between costly treatment for HIV/AIDS and those for non-communicable disease. Barriers to healthcare access were of particular concern and some priorities included investing in more mobile clinic. The individual level priorities were mostly aligned with societal ones, and there were no statistically significant differences between the individual and group choices. However, there were some statistically significant differences between individual priorities based on demographic characteristics such as age. The study demonstrates that giving individuals greater control and agency in designing health services packages can increase their participation in the priority setting process, align individual and community priorities, and enhance the legitimacy and acceptability of the decision-making process. In terms of reconciling plurality in priority setting for health, group deliberative approaches help to identify social values and reconcile some of the differences, but additional individual voices may also need to be considered alongside group processes, especially among the most vulnerable. Conclusion: This research marks the first instance of modifying and implementing a deliberative tool for priority setting in a South African rural context. The findings shed light on the process and some of the outcomes of this approach within a vulnerable community, offering insights into public engagement in priority setting more broadly. The study demonstrates that participatory methods are feasible in modifying public engagement tools such as CHAT and can be adapted to different country contexts, potentially enhancing the priority setting process. Regarding the implementation of CHAT, the study provides an example of how a rural community grappled with resource allocation decisions, considered different perspectives and societal implications, and set priorities together. The research also highlights the priorities of this rural community, the social values driving their choices, and individual characteristics that are important to consider when setting priorities. The work demonstrates that meaningful public engagement includes various factors that interrelate and impact one another and that could inform a dynamic and cyclical approach going forward, as well as the importance of transparency during all stages of the process.