Since the declaration of Alma Ata in 1978 the importance of community involvement in health service and policy design has been promoted. Priority-setting for nutrition is no exception and the decisions around which interventions to fund should reflect public values while accounting for scientific evidence, constrained budgets. Individuals, communities and population groups have differing assessments of priorities. Public participation could inform difficult decisions about resource allocation. One approach to involving the public in priority-setting is using public engagement tools such as CHAT, Choosing All Together. CHAT is a simulation decision-making tool that helps laypeople understand that not all services can be provided, and that decisions need to be taken to identify services that are most important for them. We aimed to determine whether this approach was feasible in sub-Saharan Africa.

As part of INPreP, a cross-country study on optimizing nutrition during preconception, pregnancy and infancy in Sub-Saharan Africa, researchers used a participatory approach^{(Reference Tugendhaft, Danis and Christofides1)} to modify CHAT for the first time for nutrition interventions in Ghana, Burkina Faso and South Africa. The aim was to identify a package of nutrition interventions with local communities. Through the participatory modification approach, intervention options for CHAT were identified and costed to allow for meaningful rationing. These options were based on suggestions from communities^{(Reference Compaoré, Ouedraogo and Boua2,Reference Debpuur, Nonterah and Chatio3,Reference Erzse, Goldstein and Norris4,Reference Watson D and Erzse A5)} , local decision makers, and evidence from literature and policy reviews. Community members were purposefully selected and invited to participate in group discussions where they engaged with the game-like exercise. They were able to choose between interventions that they believed should be prioritized to improve maternal and child nutrition in their communities. Participants made their choices using a limited budget that covered 60% of the interventions. Group discussions were conducted in the local languages, audio recorded, transcribed and translated into English where necessary. Transcripts were imported into qualitative data analysis software program (NVivo) for organizing data which are thematically analysed.

A total of 16 group discussions were held in Navrongo and Nanoro, using a semi-structured facilitator script and the CHAT tool. Participants could choose between 8 interventions in Nanoro and 11 in Navrongo. Intervention options included both nutrition-sensitive (i.e.: demonstrations of healthy and affordable recipes using local ingredients) and nutrition-specific interventions (i.e.: food fortification).

This participatory approach led to the successful adaptation of CHAT to the various nutrition contexts in each country and implementation in rural communities with low literacy in Ghana and Burkina Faso. CHAT is currently being implemented in urban South Africa. Findings will provide insight into communities’ priority choices and the justification behind their choices. Furthermore, it will improve our understanding if CHAT is a feasible tool to involve the public in nutrition priority-setting in Sub-Saharan Africa for both researchers and policy leaders.