Background: Diabetes self-management education, including nutrition education (NE) is an essential component of diabetes management. Effective NE can assist individuals with type 2 diabetes mellitus (DM) in resource limited settings to improve their dietary self-care; an area cited among the most difficult with consequent improvement in health outcomes. Aim: To develop a NE programme that is tailored to the needs of adults with type 2 DM in a resource limited setting and to evaluate the programme's effectiveness on health outcomes. Setting: Makapanstad and Mathibestad community health centres in the Moretele sub-district, North West Province (South Africa). Methods: The study was done in three phases employing mixed methods research. Qualitative methods, using focus group discussions with 31 diabetic patients (a convenience purposive sample), and an open ended self-administered questionnaire with ten health professionals serving them, assessed the NE needs and preferences (phase 1). The data were analysed according to the framework approach. The results from the needs assessment were used to plan a tailored NE programme (phase 2). A randomised controlled trial (quantitative) with a sample of 82 patients (with HbA1c ≥ 8), allocated to either intervention or control groups, evaluated the effect of the NE programme (phase 3). Outcomes [HbA1c, dietary behaviours, blood lipids, blood pressure, body mass index (BMI), diabetes knowledge and attitudes towards diabetes and its treatment] were assessed at baseline, six months and 12 months respectively. An analysis of covariance (ANCOVA) compared the groups on measured outcomes using baseline values, age, gender, and clinic as covariates. Rank ANCOVA was used for dietary intake. The level of significance for all tests was set at α < 0.05 for a two-tailed test. Results: Needs assessment Diabetes related knowledge deficits and inappropriate dietary practices, including food portion control problems, inadequate intake of vegetables and fruits and unbalanced diets, were observed. Eight barriers and two facilitators to dietary adherence were identified. Financial constraint was the major barrier while social support was the major facilitator. NE recommendations included content related to the disease and diet, group education at the clinic, a competent educator, provision of education materials and inclusion of family members. The planned NE programme consisted of eight weekly training sessions and six follow-up sessions (monthly and bi-monthly), vegetable gardening demonstrations and education materials. Nutrition education programme effects: Seventy six participants (38 per group) completed the study. The differences in HbA1c (primary outcome) between the intervention and control groups were -0.62% (p=0.15) at six months and -0.67% (p=0.16) at 12 months. Few participants, four from the intervention group and one from the control group, achieved HbA1c target (<7%) at both six and 12 months, [(p=0.20), (p=0.36)] respectively. There were no significant between group differences in BMI, lipid profile and blood pressure at six months and 12 months. Starchy foods intake (median servings) were significantly lower in the intervention group compared to the control group, 9.3 vs. 10.8 (p=0.005) at six months and 9.9 vs. 11.9 (p=0.017) at 12 months. The proportion of participants growing own vegetables significantly increased in the intervention group compared to the control group 17/41 vs. 5/40 (p=0.003) at six months and 16/38 vs. 5/38 at 12 months. No significant group differences in the intake of energy, macronutrients, vegetable and fruits, sodium, cholesterol and fibre were observed at six and 12 months. Diabetes knowledge improved in the intervention group +0.95 (p=0.033) and +2.2 (p=0.000) when compared with the control group at six and 12 months respectively. There were no significant group differences in the attitudes towards diabetes and its treatment. Conclusions: The qualitative needs assessment provided insight for planning a tailored NE programme. The NE improved some dietary behaviours (starchy foods portion control and growing own vegetables) and diabetes knowledge. A non-significant lowering of HbA1c was observed.