Artificial Intelligence (AI) Patient Support and 6‐Month Medication Adherence in a Digital Obesity Program: A Retrospective Analysis
Main finding · 6-month medication adherenceResult — go to the resultOR of attrition 1.178 95% CI 1.052–1.318Authors' conclusion
Integrating an AI digital assistant did not independently improve medication adherence; programmatic costs, program pauses and initial tracking anxiety were stronger predictors of long-term retention. After matching post-Junebot cohorts, moderate, active digital engagement appeared to correlate with 6-month retention.
Study context
- Population
- Adults prescribed semaglutide within an Australian digital weight loss service, May 2024–October 2025
- Participants
- 16,556 adults
- Exposure
- Post-Junebot vs pre-Junebot
- Outcome
- Medication adherence (≥ 6 orders fulfilled within 183 days)
- Follow-up
- 6 months
Methodological detailsmultivariate binary logistic regression, 1:1 propensity score matching sensitivity analyses
- Multivariate binary logistic regression
- 1:1 propensity score matching sensitivity analyses
Main finding
6-month medication adherence
OR of attrition
1.178
95% CI 1.052–1.318
P = 0.004
By group
Post-Junebotvs Pre-Junebot
OR of attrition 1.178 (95% CI 1.052–1.318); P = 0.004
As reported in the abstract
Observational data: associations, not effects.
Further findings
| Outcome | OR of attrition | P value |
|---|---|---|
| Attrition, era-matched PSM modelPropensity score matched | 1.309 | P = 0.038 |
Predictors of attrition
- High-intensity tracking in month 1 (> 25 tracks), OR
- 15.753; p < 0.001
- Program pauses, OR
- 2.508; p < 0.001
- Higher program cost, OR
- 2.458; p < 0.001
Adherence odds with structured digital engagement in 50%-74.99% of weeks (within post-Junebot cohort)
- Medication-only cohort, OR
- 32.016
- Medication plus health coaching cohort, OR
- 13.311
Original publication
AI Patient Support and 6‐Month Medication Adherence in a Digital Obesity Program: A Retrospective Analysis. Diabetes Obes Metab. Published 10 August 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →