A randomized controlled trial to test the effectiveness of two technology-enhanced diabetes prevention programs in primary care: The DiaBEAT-it study

OBJECTIVE: To evaluate the effectiveness of two technology-enhanced interventions for diabetes prevention among adults at risk for developing diabetes in a primary care setting. METHODS: The DiaBEAT-it study employed a hybrid 2-group preference (Choice) and 3-group randomized controlled (RCT) design. This paper presents weight related primary outcomes of the RCT arm. Patients from Southwest Virginia were identified through the Carilion Clinic electronic health records. Eligible participants (18 and older, BMI ≥ 25, no Type 2 Diabetes) were randomized to either Choice (n = 264) or RCT (n = 334). RCT individuals were further randomized to one of three groups: (1) a 2-h small group class to help patients develop a personal action plan to prevent diabetes (SC, n = 117); (2) a 2-h small group class plus automated telephone calls using an interactive voice response system (IVR) to help participants initiate weight loss through a healthful diet and regular physical activity (Class/IVR, n = 110); or (3) a DVD with same content as the class plus the same IVR calls over a period of 12 months (DVD/IVR, n = 107). RESULTS: Of the 334 participants that were randomized, 232 (69%) had study measured weights at 6 months, 221 (66%) at 12 months, and 208 (62%) at 18 months. Class/IVR participants were less likely to complete weight measures than SC or DVD/IVR. Intention to treat analyses, controlling for gender, race, age and baseline BMI, showed that DVD/IVR and Class/IVR led to reductions in BMI at 6 (DVD/IVR −0.94, p < 0.001; Class/IVR −0.70, p < 0.01), 12 (DVD/IVR −0.88, p < 0.001; Class/IVR-0.82, p < 0.001) and 18 (DVD/IVR −0.78, p < 0.001; Class/IVR −0.58, p < 0.01) months. All three groups showed a significant number of participants losing at least 5% of their body weight at 12 months (DVD/IVR 26.87%; Class/IVR 21.62%; SC 16.85%). When comparing groups, DVD/IVR were significantly more likely to decrease BMI at 6 months (p < 0.05) and maintain the reduction at 18 months (p < 0.05) when compared to SC. There were no differences between the other groups. CONCLUSIONS: The DiaBEAT-it interventions show promise in responding to the need for scalable, effective methods to manage obesity and prevent diabetes in primary care settings that do not over burden primary care clinics and providers. REGISTRATION: https://clinicaltrials.gov/ct2/show/NCT02162901, identifier: NCT02162901.