Promoting Problem-Solving Among Low-Income Adults with Type 2 Diabetes: A Cluster-Randomized Trial of Mobile Diabetes Detective (Preprint)

BACKGROUND Problem-solving skills are essential to the successful management of type 2 diabetes but challenging to develop for low-resource medically underserved individuals. mHealth interventions have shown positive impact on diabetes management but few focus specifically on problem-solving. OBJECTIVE To evaluate the efficacy of an mHealth intervention, Mobile Diabetes Detective (MoDD), for facilitating problem-solving and self-care among underserved adults with type 2 diabetes. METHODS This cluster-randomized non-blinded clinical trial included 219 participants with type 2 diabetes (HbA1c > 7.5%) from 8 Federally Qualified Health Centers that serve low-resource communities. The intervention arm (n=111) used MoDD for 4 weeks and up to 1 year. The control arm (n=108) received standard diabetes education and usual care. The primary hypotheses were improvement in HbA1c, diabetes problem-solving abilities, and diabetes self-care behaviors from baseline to 4 weeks, 3 months, and 12 months. Secondary hypotheses were improvement in diabetes self-efficacy and diabetes distress. Data were analyzed for both within group change and between group differences at each study milestone. RESULTS Participants (n=219) were predominantly female (67%), ethnically and racially diverse (51% Hispanic, 42% African American), with mean glycated hemoglobin HbA1c=9.9% at baseline. There was no change in HbA1c at 4 weeks in either arm; both arms reduced HbA1c from baseline to 3 months (intervention: -0.58% (p=0.005); control: -1.03% (p<0.001)); only the intervention arm reduced HbA1c from baseline to 12 months (intervention: -0.45% (p=0.01); control: -0.37% (p=0.047). There was no improvement in diabetes problem-solving, and mixed results in diabetes self-care behaviors. Both arms improved diabetes self-efficacy at 3 and 12 months; the intervention arm reported increase in diabetes distress at 3 and 12 months. The only difference in outcomes between study arms was in following a diabetes-specific diet at 12 months with the control arm reporting higher adherence (intervention: 4.48, control: 5.17, p=0.004). CONCLUSIONS All participants improved blood glucose management at 3 months but only the intervention arm sustained improvement over time; there was no change in diabetes problem solving abilities. These findings suggest potential benefits of using mHealth interventions for underserved individuals with type 2 diabetes and the need for further research on improving problem-solving with mHealth. CLINICALTRIAL NCT02021591