**Background:** Priority setting for nutrition interventions is essential given limited resources, and community engagement can improve the relevance and effectiveness of policy. The 'Choosing All Together' (CHAT) tool, a board-game-like deliberative engagement method, has been used for health insurance benefit package decisions in several countries but had not been applied to nutrition interventions. This study aimed to use CHAT to identify community priorities for nutrition interventions and explore reasons for those choices in rural northern Ghana, a region with the highest malnutrition levels in the country (stunting 22.4%, wasting 9.0%).
**Methods:** The study used an exploratory cross-sectional design with mixed methods, conducted between December 2020 and February 2021 in the Kassena-Nankana East and West Districts of the Upper East Region, Ghana. Eleven nutrition interventions were identified through a desktop review of policies (1998–2018), 11 in-depth interviews with stakeholders, and 10 focus group discussions with community members. Interventions were costed using an ingredients approach and grouped into 6 themes: community nutrition education, youth education, male involvement, livelihood empowerment, health system strengthening, and micronutrient supplementation. Two interventions were nutrition-specific (micronutrient supplementation, food fortification) and nine were nutrition-sensitive. The modified CHAT board represented interventions as color-coded pie slices with costs shown as sticker holes (104 total holes). Participants (n=53, in 6 groups of 6–12) received 60 stickers (covering ~58% of holes) and completed two rounds: first in pairs, then as a whole group to reach consensus via majority vote. Facilitators guided deliberations using hypothetical scenarios. Discussions were audio-recorded, transcribed, and thematically analyzed using NVivo 12. Quantitative sticker choices were recorded on paper and analyzed in Microsoft Excel.
**Key Results:** The average age of participants was 39 years (range 24–50); 52.8% were women; 92.5% were married; 26.4% had no formal education and 37.7% had primary education. All 6 groups selected all three livelihood empowerment interventions (water wells/tanks, agricultural inputs, livelihood skills training), requiring 32% of total stickers. All 6 groups also selected both micronutrient supplementation interventions (iron-folate supplementation and food fortification), requiring 45% of total stickers. Male involvement in nutrition education was selected by 5 of 6 groups (only one women's group aged 35–50 did not select it). Food demonstration in durbars and youth nutrition education were each selected by 4 groups. Radio broadcasting was selected by 2 groups, health system strengthening by 1 group, and SMS/posters by 0 groups. Participants justified livelihood choices by citing poverty, subsistence farming, erratic rainfall, and the need for year-round agricultural output and income. Micronutrient supplementation was prioritized because community members recognized that their diets lack essential nutrients, leading to anemia in children and pregnant women. Male involvement was valued because men are household heads and main decision-makers. Less-prioritized interventions (nutrition education via SMS, radio, posters) were seen as having no direct benefit and faced barriers including inability to read and lack of money for phones/radios. Participants reported positive experiences with the CHAT exercise, noting smooth deliberations, consensus-building, and satisfaction with their choices.
**Clinical Implications:** The study demonstrates that CHAT is feasible for nutrition priority-setting in low-literacy rural settings. Community priorities—livelihood empowerment, micronutrient supplementation, and male involvement—reflect the local context of poverty and food insecurity. These findings suggest that policy makers should invest in sustainable agriculture, water supply, income-generating activities, and micronutrient programs to address malnutrition in similar settings. Engaging men in nutrition education may improve household dietary outcomes. The study supports the use of deliberative community engagement to enhance the relevance and implementation of nutrition interventions, consistent with evidence that community engagement improves health outcomes. Limitations include the small number of groups (6), potential translation issues (mitigated by local facilitators), and the 3-hour time commitment for participants.