ABSTRACT Youth-onset type 2 diabetes (T2D) represents a growing threat to health equity. Continuously increasing in incidence, T2D is up to 9-fold more common among racial-ethnic minoritized youth, and approximately half of affected youth live in poverty. T2D diagnosis in childhood often results in severe complications by early adulthood, as well as reduced life expectancy. Modification of diabetes-related health behaviors, including nutrition and physical activity, is central to management, but several barriers exist for adolescents and young adults (AYA) with T2D. These include hyperglycemia unawareness, delayed or burdensome glycemic feedback, and the difficulty of attending intensive in-person interventions. Fortunately, wearable continuous glucose monitors (CGMs) enable the development of remote just-in-time adaptive interventions that use real-time data at the right time, in the right context to promote behavior change. With the transition to adult care, AYA with T2D are particularly well suited to benefit from a CGM-supported intervention that supports development of diabetes self-management skills and engagement in health behaviors. To lay the foundation for such an intervention, we have developed Healthmine, a mobile app that integrates CGM, activity trackers, and diet logging for use by individuals with diabetes. The objective for this R01 proposal, REFLECT2D, is to evaluate the time-dependent glycemic impact of Healthmine app-based prompts. Aligned with the COM-B Model for Behavior Change, Healthmine prompts are designed to foster motivation and actual engagement in health behaviors by providing of an opportunity to review real-time data and by supporting the development of an improved capability to interpret and act on such data. We will enroll 100 AYA ages 16-24 with T2D in a clinical trial, beginning with a baseline study visit and educational session led by a certified diabetes education and care specialist and registered dietitian. After a 10-day run-in with CGM to evaluate baseline glycemic control, we will conduct a 90- day micro-randomized trial (MRT), in which each AYA will be randomized with equal (1/3) probability once daily to receive a physical activity-focused prompt, a nutrition-focused prompt, or no prompt. In MRT, each participant acts as their own control. After MRT, AYA will attend a second study visit to assess intermediate outcomes and will continue to wear CGM for a 90-day observational period without app prompts, followed by final study visit, to assess duration of effect on motivation and glycemic outcomes. Our approach will allow us to determine the effects of prompts on short-term CGM-measured glycemia during MRT (Aim 1), characterize ongoing CGM and app use and longer-term impact on glycemia during an observational period (Aim 2), and evaluate changes in outcomes from baseline to follow-up, including glycemia and behavioral measures, as well as the role of motivation as a mediator (Aim 3). Findings will inform the development of a CGM-based just-in-time adaptive intervention, which will directly address barriers to in-person engagement in health behavior change, allowing for scalability and ultimately reducing the widening health disparities caused by youth-onset T2D.

NARRATIVE Youth-onset type 2 diabetes is rapidly increasing in incidence, carries a high risk of severe complications in early adulthood, and disproportionately impacts marginalized communities. Adolescents and young adults with type 2 diabetes often face barriers to healthy lifestyle change and diabetes self-management, including limited or delayed feedback on how such health behavior changes affect blood sugar. Our study, REFLECT2D, will evaluate use a novel smartphone app pairing real-time health behavior and glycemic data, laying the foundation for future interventions that use the right type of data at the right time to support health behavior change, ultimately improving outcomes for adolescents and young adults with type 2 diabetes.