Machine learning-based prediction of medication refill adherence among first-time insulin users with type 2 diabetes
Authors' conclusion
This is the first study to predict adherence among adult naïve insulin users. The developed model is a potential clinical decision support tool to identify possible non-adherent patients for healthcare providers to design individualized education plans.
Study context
- Population
- Adult T2D naïve insulin users, Taipei Medical University Clinical Research Database (TMUCRD), 2004–2020; adherence = medication possession ratio ≥80%Taiwan cohort
- Participants
- 4,134 users
- Comparison
- Internal test vs External test
- Outcome
- Prediction of insulin adherence (medication possession ratio (MPR) ≥80%)
Methodological detailsmachine learning, Xgboost, internal and external testing
- Machine learning
- Xgboost
- Internal and external testing
Performance
As reported in the abstract
| Test or model | AUC |
|---|---|
| Internal test | 0.782 |
| External test | 0.771 |
AUC
Internal test0.782
External test0.771
Further findings
Predictors (SHAP values)
- Number of prescribed medications, number of outpatient visits and laboratory data were predictive of future insulin adherence
Original publication
Machine learning-based prediction of medication refill adherence among first-time insulin users with type 2 diabetes. Diabetes Res Clin Pract. Published 2 December 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →