Community health worker intervention to prevent incident type 2 diabetes: A cluster randomized trial

OBJECTIVE: Type 2 diabetes mellitus (T2DM) affects millions of Americans, and prediabetes significantly increases progression to T2DM. Community Health Workers (CHWs) may help prevent diabetes through behavioral interventions, but their effectiveness in real-world settings remains uncertain. We evaluated whether a CHW intervention prevents incident T2DM among adults with prediabetes. RESEARCH DESIGN AND METHODS: Pragmatic, cluster-randomized controlled trial in primary care clinics at the Veterans Affairs NY Harbor Healthcare System (Manhattan) and Bellevue Hospital, New York City. The study included adults aged 18-75 years with HbA1c 5.7-6.4% and no prior diabetes or use of diabetes medications other than metformin. Primary care teams were cluster-randomized to a 12-month CHW program-health action plans, motivational interviewing, health education, and community referrals-versus usual care. The primary outcome was incident diabetes during up to 6 months' follow-up after the 12-month intervention. Secondary outcomes were changes in HbA1c, weight, blood pressure, lipids, and health care utilization. RESULTS: Incident diabetes occurred in 5.9% of intervention patients versus 4.3% of controls (risk difference 1.6%; 95% CI -3.68% to 4.10%). No significant differences were observed in HbA1c, weight, blood pressure, or lipids. The intervention group had more primary care encounters (77.6% vs 62.5%) and emergency department encounters (35.0% vs 29.1%). CONCLUSIONS: In this pragmatic trial, a 12-month CHW program did not reduce incident diabetes among adults with prediabetes but was associated with greater health care engagement (more primary care visits). More intensive or standardized CHW models should be tested to determine whether increased engagement translates into long-term metabolic benefit. TRIAL REGISTRATION: The study protocol was registered at ClinicalTrials.gov (NCT03006666, 12/27/2016). https://clinicaltrials.gov/study/NCT03006666.