Community-Based Peer Support for Diabetes Management: 24-Month Changes Relative to Comparison Communities

OBJECTIVE To evaluate the effectiveness of community-based peer support for diabetes self-management with HbA1c and other clinical and psychosocial outcomes over 24 months. RESEARCH DESIGN AND METHODS This study used an intervention comparison design with 12 intervention communities and 4 comparison communities matched according to location in urban or suburban areas. A community organization approach was used to integrate standardization of key messages and patient education protocols, along with adaptation and innovation among multiple community partners. The primary outcome was HbA1c; secondary outcomes included BMI, fasting plasma glucose (FPG), systolic and diastolic blood pressure, LDL cholesterol (LDL-C), depressive symptoms (PHQ-8), diabetes distress, and general quality of life (EQ-5D). RESULTS The analyses included 967 participants completing both the baseline and follow-up assessment. Intervention communities versus comparison communities were older (mean age 66.43 vs. 63.45 years), included more women (57.1% vs. 45.5%), and had longer diabetes duration (mean 7.95 vs. 6.40 years). Significant improvements were found for HbA1c (7.42% [58 mmol/mol] vs. 7.95% [63 mmol/mol]), BMI (25.31 vs. 25.94 kg/m2), FPG (7.91 vs. 8.59 mmol/L), and depressive symptoms (PHQ-8 score 1.39 vs. 1.41), favoring intervention communities, after adjusting for baseline values of outcome measures and confounders (P ≤ 0.028). No interactions were found with age (<65 vs. ≥65 years). Men showed modestly greater diastolic blood pressure reduction, and women showed a minor increase of LDL-C in intervention communities. These analyses by age or sex and sensitivity analyses with missing data imputation supported the robustness of findings. CONCLUSIONS Culturally adapted and appropriate community-based peer support for diabetes management may improve clinical and psychosocial outcomes at 24 months among people with diabetes.