Intelligent Care Platform - Quasi-Experimental Trial

Overview

This study will measure the implementation and effectiveness of an intervention, the use of a digital health tool, Haibu Diabetes, for the management of pediatric type 1 diabetes (T1D), on clinical outcomes. The investigators will use both quantitative (i.e., surveys) and qualitative (i.e., interviews) methods to assess implementation. To measure effectiveness, the investigators will compare users of the technology to a matched comparison group of non-users on relevant clinical outcomes using clinical data. Overall, the investigators hope to understand whether use of this digital tool improves clinical outcomes while also considering factors that affect implementation and clinical operations.

Purpose: The purpose of this trial is to implement the use of a digital tool designed to be used for the management of pediatric type 1 diabetes by patients, their parents, and their HCP. The investigators will assess the implementation of this digital tool to understand how the platform is utilized, satisfaction with the platform for this purpose, acceptability, feasibility, and efficiency for managing T1D as reported through surveys and interviews with participants. Importantly, the investigators will assess the effectiveness of the digital tool by examining its impact on clinical outcomes including diabetes management habits (insulin dose adjustments, reviewing glucose data), A1C, TIR, diabetes distress, and quality of life. Hypotheses: 1. Using Haibu Diabetes in the management of type 1 diabetes will lead to improved clinical outcomes, as evidenced by: * Greater proportion of Haibu Diabetes users who adjusted insulin dose since last visit compared to non-Haibu Diabetes users (+10% difference) * Greater proportion of Haibu Diabetes users who reviewed glucose data since last visit compared to non-Haibu Diabetes users (+10% difference) * Lower average A1C among Haibu Diabetes users compared to non-Haibu Diabetes users (-0.5% difference) Increased TIR in Haibu Diabetes users compared to non-Haibu Diabetes users (+10% difference) * Lower diabetes distress scores in Haibu Diabetes users compared to non-Haibu Diabetes users (-10% difference) * Higher quality of life scores in Haibu Diabetes users compared to non-Haibu Diabetes users (+10% difference) 2. Haibu Diabetes will be feasible to implement, acceptable to both patients/families and HCP, and will demonstrate sustained use. 3. Haibu Diabetes patient/family and HCP users will be satisfied with using Haibu Diabetes and will report a perceived improvement in their care experience. Haibu Diabetes will show promise in enhancing operational clinical workflow efficiency. Justification: Digital health tools that are used collaboratively by patients, families, and providers are not part of the standard of care. However, these tools may benefit patients and improve the management of T1D. Objectives: (The outlined objectives correspond to the 3 hypotheses outlined in order above) 1. To evaluate the effectiveness of Haibu Diabetes in improving clinical outcomes in the management of type diabetes by comparing diabetes self-care habits (adjusted insulin since last visit, reviewed glucose data since last visit), A1C, time in range (TIR), quality of life and diabetes distress in Haibu Diabetes users compared to non-Haibu Diabetes users. 2. To assess the reach, feasibility, adoption and sustained implementation of Haibu Diabetes among patients, caregivers, and HCP. 3. To describe the impact of Haibu Diabetes on healthcare improvement including patient experience, HCP experience, and operational workflow. Research Design: This is a quasi-experimental type 1 hybrid implementation-effectiveness trial assessing implementation metrics using surveys and interview methods, and comparing users of a digital health platform to non-users on various clinical outcomes. Statistical Analysis: Data analysis will involve both descriptive and comparative statistical methods to evaluate clinical outcomes between Haibu Diabetes users and non-users. Descriptive statistics, including means, standard deviations, medians, proportions, will summarize baseline characteristics, key metrics such as platform reach, fidelity of implementation, and patterns of usage, and summarize patient/caregiver-reported outcomes (e.g., satisfaction, perceived ease of use, patient-centeredness), and healthcare team-reported outcomes (e.g., satisfaction with the platform, workload, stress levels, job satisfaction). To determine the effectiveness of Haibu Diabetes using comparative statistical methods, the investigators will conduct propensity score matching to select patients from the BC-PDR (non-Haibu Diabetes users) that are as close as possible to those in the study (Haibu Diabetes users). Several matching algorithms will be compared, and the one yielding the best balance as measured by lowest standardized mean differences will be used. Regression analyses of the matched sample will include covariate adjustment for the propensity score to provide 'doubly robust' estimates. Confidence intervals will be computed via the nonparametric bootstrap. Secondary analyses may include the use of entire BC-PDR non-Haibu Diabetes users as a control group with regression adjustment for confounders. All analyses will be conducted using R statistical software or STATA 15.1 StataCorp. Outcomes for which comparative analyses will be conducted include: diabetes self-care habits, glycemic control (HbA1c levels, time-in-range), and TIDAL and diabetes distress scores. The planned study sample size is 80 Haibu Diabetes patient/caregiver users and approximately 30 HCP users.