Importance: Community health workers (CHWs) improve chronic disease management in underserved populations, but scalable integration strategies are limited. Objective: To evaluate whether a multidimensional intervention incorporating telementored CHWs and a structured participant-CHW-clinician feedback loop can improve diabetes outcomes. Design, Setting, and Participants: This 12-month randomized clinical trial was conducted at 3 institutionally and geographically diverse community clinics in Texas between September 1, 2023, and April 30, 2025, and included low-income, uninsured White Hispanic adults with type 2 diabetes identified through clinic databases. Data were analyzed between May 1 and July 31, 2025. Intervention: Individuals were randomized 1:1 to the intervention or control. For the intervention, CHWs delivered (1) group diabetes education, (2) individualized telehealth-based coaching, and (3) a novel participant-CHW-clinician feedback loop to facilitate communication, address participant concerns, and improve care coordination. The control was usual care (quarterly clinician visits and access to multidisciplinary and social services). Main Outcomes and Measures: The primary outcome was baseline to 12-month change in hemoglobin A1c (HbA1c) level. Secondary outcomes included changes in cholesterol levels, American Diabetes Association (ADA) guideline adherence, participant recruitment, intervention fidelity, and feedback loop issue resolution. Results: Of 257 participants included in the intention-to-treat analysis (mean SD age 54 11 years; 166 64.6% female), 129 were in the intervention group and 128 were in the control group. The intervention reduced HbA1c (net difference, -1.0 95% CI, -1.5 to -0.4 percentage points pp; P = .001), total cholesterol (net difference, -35.4 mg/dL; 95% CI, -54.6 to -17.2 mg/dL; P = .02), and low-density lipoprotein cholesterol (net difference, -29.7 mg/dL; 95% CI, -44.5 to -14.9 mg/dL; P < .001) levels compared with control. ADA guideline adherence improved for foot examinations (absolute risk AR, 19.2 95% CI, 7.4-30.9 pp; relative risk RR, 1.65 95% CI, 1.19-2.27; P = .03) and urine microalbumin screening (AR, 15.8 95% CI, 5.3-26.3 pp; RR, 1.24 95% CI, 1.07-1.43; P = .048). CHWs addressed 490 participant concerns (87.2%) via the feedback loop, including medication refills, glucose management, and access to care. Conclusions and Relevance: In this randomized clinical trial, the CHW intervention significantly improved diabetes outcomes in low-income settings, potentially by reducing fragmentation through structured feedback. Findings also highlight limitations in usual care, underscoring the need for scalable strategies to strengthen chronic disease management in low-income populations. Trial Registration: ClinicalTrials.gov Identifier: NCT04835493.
Vaughan et al. (Mon,) studied this question.