Predictive modelling of medication adherence in post-myocardial infarction patients: a Bayesian approach using beta-regression
Authors' conclusion
In addition, simplified models, with 90-days adherence as single predictor, had similar performance compared with full complex models.
Study context
- Population
- Adult post-MI patients followed for 1 year; adherence to statins, P2Y12 inhibitors and aspirin
- Participants
- 3,692 patients
- Comparison
- Simplified model vs full multi-variable model
- Outcome
- Proportion of days covered (PDC) from pharmacy records, modelled as a proportion
- Follow-up
- 1 year
Methodological detailsBayesian beta-regression, multi-level models, cross validation
- Bayesian beta-regression
- Multi-level models
- Cross validation
Performance
As reported in the abstract
Model performance
Bayesian R2, statins model
61.4%
Bayesian R2, P2Y12 inhibitors model
71.2%
Bayesian R2, aspirin model
55.2%
Further findings
Median
| Outcome | Statins | P2Y12 inhibitors | Aspirin |
|---|---|---|---|
| 1-year PDC by medication | 0.8 (IQR 0.33–1.00) | 0.79 (IQR 0.23–0.99) | 0.79 (IQR 0.23–0.99) |
Original publication
Predictive modelling of medication adherence in post-myocardial infarction patients: a Bayesian approach using beta-regression. Eur J Prev Cardiol. Published 4 October 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →