Adapting and Developing A Diabetes Prevention Intervention Programme for South Africa: Curriculum and Tools

“

TL;DR

This paper describes the systematic, mixed-method development of a culturally tailored diabetes prevention curriculum and tools (facilitator and participant workbooks) for low-income South African communities at high risk of type 2 diabetes. The curriculum, based on the Finnish Diabetes Prevention Study and adapted using focus group input from 68 at-risk adults, achieved high acceptability scores (94.3% for the participant workbook, 89.3% for the curriculum booklet) and was well received in a 6-week pilot study with 10 participants. This work addresses a critical gap in diabetes prevention interventions for African populations and provides a model that could be adapted for other sub-Saharan African countries.

Full summary

4,684 CHARS

**Background:** Type 2 diabetes (T2DM) burden in sub-Saharan Africa is rising rapidly, with an estimated increase from 24 million people in 2021 to 55 million by 2045. South Africa has the second-highest number of people with diabetes in the region (1,826,100 adults by 2017), with prevalence of 11.3% (6.97–16.52%) in the black population and 23.7% (13.93–15.73) in the mixed ancestry population. Despite robust evidence from the Finnish Diabetes Prevention Study and the US Diabetes Prevention Program showing 58% reduction in incident diabetes through lifestyle intervention, no published adaptation of a diabetes prevention lifestyle intervention existed for an African country at the time of this study. The South African Diabetes Prevention Programme (SA-DPP) aims to develop and evaluate a culturally relevant T2DM prevention approach for low socioeconomic communities, based on empowerment ideology and delivered by community health workers through six group sessions complemented by mobile phone messaging.

**Methods:** An 11-step qualitative mixed-method staged approach was followed: (1) reviewing existing evidence; (2) needs assessment via focus group discussions (FGDs); (3) expert input from a multidisciplinary team including three dietitians, a physician, and a psychologist; (4) development of curriculum and tools; (5) expert content evaluation; (6) design and layout; (7) participant readability and acceptability evaluation; (8) refining design and layout; (9) translation into Xhosa; (10) suitability evaluation via a pilot intervention; and (11) finalising the curriculum. The needs assessment comprised 17 FGDs (10 in round one, 7 in round two) with 68 participants (50 female, 58 aged 45–65 years, 42 Xhosa-speaking) recruited from eight resource-poor areas in Cape Town. Readability and acceptability were tested in workshops with 16 participants using a validated 38-question assessment tool with a 3-point Likert scale. The pilot study was conducted over six weekly sessions (4–10 November 2020) with 10 participants (mean age 54.6, SD 10.3) from two sites (Du Noon and Belhar), facilitated by a dietitian due to COVID-19 restrictions.

**Key Results:** The needs assessment revealed that participants understood diabetes as a serious disease but had limited knowledge of its causes, mixing misperceptions with factual information. The primary barrier to healthy eating was financial constraints, while safety concerns limited outdoor physical activity. Participants preferred group sessions held in the community, facilitated by knowledgeable individuals (not necessarily peers), and wanted materials in their home language (particularly Xhosa speakers). The mean acceptability score for the participant workbook was 94.3% (SD 7.6) and for the curriculum booklet was 89.3% (SD 13.3). In the pilot study, Session 2 (Healthy Eating Part 1) had 100% attendance; other sessions averaged ~70% attendance. Most participants reported learning something new in every session and found sessions fun and engaging. Participants successfully applied knowledge to their households: one participant reported her spouse's uncontrolled diabetes became controlled, moving from insulin to oral medication; another reported her son's improved school behaviour after replacing juice with water. Participants struggled with SMART goal setting initially but found it useful, and some did not consistently track goals in their workbooks. The food model exercise for portion sizes was effective but created unrealistic expectations, leading to its removal for ethical reasons. Smoking (n=3) and alcohol reduction remained challenging.

**Clinical Implications:** This study demonstrates a rigorous, participatory approach to developing a culturally tailored diabetes prevention intervention for a low-income African setting, addressing a critical gap in the literature. The high acceptability scores and positive pilot feedback suggest the intervention is well suited to the target population, which may improve adherence and effectiveness. The curriculum aligns with South African Food-Based Dietary Guidelines and addresses context-specific barriers (financial constraints, safety concerns, family dynamics). The next phase will implement the intervention in 16 resource-poor communities with comprehensive process evaluation. While findings may not be generalisable to other contexts, the development process provides a replicable model for other African countries. The study is limited by the small pilot sample (n=10) and COVID-19-related modifications (use of dietitian rather than community health worker as facilitator, centralised venue instead of community locations).