Work

Digital Engagement Phenotypes and 6-Month Weight Loss in a Tirzepatide-Supported Digital Weight-Loss Program: A Retrospective Cohort Study

Year: 2026

Type: article

Abstract: Background and Objectives: Real-world adherence to medication-supported weight management programs is often low. While digital weight loss services (DWLS) provide multi-modal digital supports to improve engagement and counter attrition, existing literature frequently relies on unidimensional or binary classifications of user engagement. This study used unsupervised machine learning to identify distinct digital engagement phenotypes and evaluated their independent associations with 6-month weight loss outcomes in patients prescribed tirzepatide. Materials and Methods: This retrospective cohort study analyzed deidentified data from 9,470 medication-adherent, complete-case adult patients within an Australian DWLS who initiated tirzepatide between May 20 and December 2, 2025. K-means clustering was performed on four continuous, longitudinal usage metrics: weekly app logins, health coach messaging, automated assistant (JuneBot) messaging, and weight tracking. To evaluate the primary clinical endpoint - 6-month percentage weight loss - unadjusted pairwise comparisons (Tukey HSD) and a fully adjusted ordinary least squares multivariate linear regression model were executed to control for baseline demographic, clinical, and interim behavioral covariates. Results: Four stable engagement phenotypes emerged: non-engaged (n = 1,772), high health-coach engagement (n = 1,141), high JuneBot engagement (n = 1,472), and passive self-trackers (n = 5,085). In the multivariate regression model (Adjusted R2 = 0.3606, p < 0.001), all active phenotypes were strong independent predictors of weight loss success relative to the non-engaged baseline. The adjusted incremental weight loss premiums were similar between the automated conversational cohort (β = 2.28%, SE = 0.24, p < 0.001) and the human health-coaching cohort (β = 2.17%, SE = 0.26, p < 0.001), while passive self-trackers demonstrated a smaller independent premium (β = 1.67%, SE = 0.20, p < 0.001). Conclusions: In a real-world per-protocol cohort, active engagement in a DWLS is associated with significantly improved 6-month weight loss outcomes. Under strict clinical safety guardrails, structured automated conversational support may achieve outcome parity with human coaching, offering a highly scalable mechanism to complement pharmacologic obesity care.

Source: Medicina

DOI: 10.3390/medicina62091647