Effect of a peer support group intervention on medication adherence, self- care practices, and diabetes knowledge among patients with diabetes in Northern Ethiopia: a randomized controlled trial

BACKGROUND: Diabetes mellitus is a chronic condition that requires consistent adherence to treatment, regular self-care, and adequate knowledge of the disease to prevent complications and improve quality of life. However, many patients struggle to maintain optimal adherence and integrate self-care into their daily routines. Nowadays, many diabetes self-management modalities have emerged as a promising approach to address these gaps. This randomized controlled study evaluated the effect of a structured peer support program on medication adherence, self-care practices, and diabetes-related knowledge among patients with diabetes who attended follow-up at a tertiary hospital. METHODS: A randomized controlled trial was conducted on 97 (intervention group = 48, control group = 49) adults with diabetes who were attending follow-up at Ayder Comprehensive Specialized Hospital (ACSH). Participants were randomly assigned to either a peer-support intervention group or a control group and followed for six months. The intervention comprised three structured peer-led sessions that focused on self-care, medication adherence, and diabetes knowledge. Outcomes were measured at baseline and follow-up using the Morisky Medication Adherence Scale (MMAS-8), adopted Diabetes self-care Activities (SDSCA) scale, and diabetes knowledge test (DKT). Data analysis was performed using chi-square tests, paired-sample t-tests, and independent-sample t-tests to compare baseline characteristics and evaluate intervention effects. Difference-in-differences (DiD) analysis was used to compare the intervention and control groups across outcome variables. RESULT: Compared with the control group, participants in the intervention group demonstrated significant improvements across all outcome measures. Medication adherence increased by a net difference-in-differences estimate of (DiD = + 1.5 points, p < 0.001) on the MMAS-8 scale. Self-care practices improved by (DiD = + 1.3 days per week of recommended behaviors, p < 0.001). Diabetes knowledge scores also showed substantial gains, among both the insulin users (DiD = + 4.9-point, p < 0.001) and non-insulin users (DiD = + 3.4-points, p < 0.001) of the intervention group compare to controls. CONCLUSION: Integrating structured peer-support interventions into routine diabetes care may provide a practical approach to improving patient outcomes while optimizing limited healthcare resources in low-resource settings. Future research should evaluate the sustainability, scalability, and cost-effectiveness of peer-support education model, together with their long-term impact on quality of life.