Predicting Lipid-Lowering Medication Persistence after the First Cardiovascular Disease Hospitalization
Related Experiment Video
Updated: Dec 13, 2025
Differential Effects of Lipid-lowering Drugs in Modulating Morphology of Cholesterol Particles
Published on: November 10, 2017
Feiyu Hu1, Jim Warren1, Daniel J Exeter2
1School of Computer Science, University of Auckland, Auckland, New Zealand.
Insights
Predicting medication persistence after cardiovascular disease (CVD) hospitalization is crucial for secondary prevention. Machine learning models, particularly Boosted Regression Trees, show promise in identifying patients likely to adhere to lipid-lowering therapy.
Area of Science:
- Cardiology
- Health Informatics
- Pharmacoeconomics
Background:
- Medication persistence is vital for secondary prevention of cardiovascular disease (CVD).
- Predicting adherence to lipid-lowering therapy post-hospitalization is challenging but essential.
Purpose of the Study:
- To analyze patient factors influencing medication persistence after a first CVD hospitalization.
- To predict 1- to 2-year persistence to lipid-lowering therapy using machine learning.
Main Methods:
- Analysis of a large cohort (313,207 patients) post-first CVD hospitalization.
- Application of proportional hazard models, logistic regression, support vector machines, artificial neural networks, and Boosted Regression Trees (BRT).
Main Results:
- Factors like age, diabetes, comorbidities, hospitalization duration, CVD type, procedures, and new therapy status significantly impacted persistence.
- BRT models achieved the highest predictive performance for 1-year (c-statistic 0.811) and 2-year (c-statistic 0.793) persistence.
Conclusions:
- Machine learning models can effectively predict medication persistence in CVD patients.
- A clinical decision support tool based on these findings could enhance secondary CVD prevention efforts.
Objectives:
This study analyzed patient factors in medication persistence after discharge from the first hospitalization for cardiovascular disease (CVD) with the aim of predicting persistence to lipid-lowering therapy for 1 to 2 years.
Methods:
A subcohort having a first CVD hospitalization was selected from 313,207 patients for proportional hazard model analysis. Logistic regression, support vector machine, artificial neural networks, and boosted regression tree (BRT) models were used to predict 1- and 2-year medication persistence.
Results:
Proportional hazard modeling found significant association of persistence with age, diabetes history, complication and comorbidity level, days stayed in hospital, CVD diagnosis type, in-patient procedures, and being new to therapy. BRT had the best predictive performance with c-statistic of 0.811 (0.799-0.824) for 1-year and 0.793 (0.772-0.814) for 2-year prediction using variables potentially available shortly after discharge.
Conclusion:
The results suggest that development of a machine learning-based clinical decision support tool to focus improvements in secondary prevention of CVD is feasible.
Related Concept Videos
01:20Lipid-Lowering Drugs: Statins and Miscellaneous Agents
01:26Atherosclerosis III: Management
01:26Coronary Artery Disease IV: Preventive Measures
01:21Drug Accumulation During Multiple Dosing: Repetitive IV Injections
01:27Coronary Artery Disease V: Interprofessional Care
01:01Dosage Regimen: Fixed Dose
Fixed-dose regimens can be used for various routes of administration, including intravenous (IV) injections and oral medications. For IV administration, a predetermined amount of the drug is...