Understanding whether and how a digital health intervention improves transition care for emerging adults living with type 1 diabetes: Protocol for a mixed-methods realist evaluation (Preprint)

BACKGROUND Emerging adults living with type 1 diabetes (T1D) face a series of challenges with self-management and decreased health system engagement, leading to increased risk of acute complications and hospital admissions. Effective and scalable strategies are needed to support this population to transfer seamlessly from paediatric to adult care with sufficient self-management capability. While digital health interventions for T1D self-management are a promising strategy, it remains unclear which elements work, how, and for which group(s) of individuals. OBJECTIVE The present study aims to evaluate the design and implementation of a multi-component digital health intervention to support emerging adults living with T1D in real-world settings. Specifically, the objectives are to identify the intervention components and associated mechanisms that lead to improved user engagement and T1D healthcare transition experiences and determine the individual-level contextual factors that influence the implementation process. METHODS Embedded alongside a randomized controlled trial, this realist evaluation employs a sequential mixed-methods design to analyze data from multiple sources, including intervention usage data, patient-reported outcomes, and semi-structured interviews. In Step 1, we conducted a document analysis to develop a program theory that outlines the hypothesized relationships amongst individual-level contextual factors (C), intervention components and features (I), mechanisms (M), and target outcomes (O) with special attention paid to user engagement. In Step 2, we will conduct semi-structured interviews with the RCT intervention arm participants to validate the hypothesized C-I-M-O configurations. In Step 3, we will triangulate all sources of data using a Coincidence Analysis to identify the necessary combinations of conditions (i.e., factors, pathways, and context) that determine whether and how the desired outcomes are achieved and use these insights to refine the program theory. RESULTS For step 1 analysis, we have developed the initial program theory and the corresponding data collection plan. For step 2 analysis, participant enrollment for the randomized controlled trial started in January 2023. Participant enrollment for the present realist evaluation is anticipated to start in May 2023 and continue until we reach thematic saturation or achieve informational power. CONCLUSIONS Beyond contributing to knowledge on the multiple pathways that lead to successful engagement with a digital health intervention as well as target outcomes in T1D care transitions, embedding the realist evaluation alongside the trial may inform real-time intervention refinement to improve user engagement and transition experiences. The knowledge gained from this study may inform the design, implementation, and evaluation of future digital health interventions that aim to improve transition experiences.