User Engagement and Feature Preferences in an AI-powered mHealth Intervention for Diabetes Prevention: Secondary Analysis of a Randomized Controlled Trial (Preprint)

Abstract

<sec> <title>BACKGROUND</title> Prediabetes is highly prevalent and increasing globally, yet lifestyle interventions remain underutilized. Artificial intelligence (AI)-driven mobile health tools can help scale diabetes prevention efforts, but the key factors driving their success are not well understood. </sec> <sec> <title>OBJECTIVE</title> This prospective study aims to characterize the most valued features and the role of user engagement on outcomes in a fully automated mHealth intervention for diabetes prevention. </sec> <sec> <title>METHODS</title> Data from 151 participants with prediabetes and overweight or obesity assigned an AI-based Diabetes Prevention Program (Sweetch, Sweetch Ltd.) in a parent RCT (NCT05056376) were analyzed. Engagement (defined as total days where app was used) was categorized into tertiles (low, medium, high). Baseline characteristics were compared across engagement groups using ANOVA, Kruskal-Wallis, and chi-square tests, and regression models assessed the association between engagement and achievement of diabetes risk reduction outcomes (≥5% weight loss, ≥4% weight loss with ≥150 min/week of activity, or ≥0.2-point A1C reduction at 12 months). Perceived usefulness of intervention features was surveyed at 12 months. </sec> <sec> <title>RESULTS</title> At 12 months, median engagement was 98 days (IQR: 34–232), with most participants (75.5%) demonstrating a decreasing engagement trajectory over time. Older age (p &lt; 0.001) and lower baseline BMI (p &lt; 0.05) were significantly associated with higher engagement. High engagement was significantly associated with achieving the composite diabetes risk reduction outcome (OR: 2.59; 95% CI: 1.11–6.01), ≥5% weight loss (OR: 3.31; 95% CI: 1.16–9.42), and ≥0.2% A1C reduction (OR: 3.57; 95% CI: 1.19–10.75) compared to low engagement. The app features perceived most useful in achieving participant health goals were weight tracking, activity tracking, and the digital scale. </sec> <sec> <title>CONCLUSIONS</title> Higher engagement with an AI-driven intervention requiring no human intervention was associated with improved diabetes risk reduction. Contrary to concerns about lower digital literacy, older adults engaged with the intervention the most. Features related to weight and physical activity tracking were most valued by patients in the program. </sec> <sec> <title>CLINICALTRIAL</title> ClinicalTrials.gov Identifier: NCT05056376 </sec>

Ähnliche Arbeiten