Exploring Medication Adherence with P2Y12 Inhibitors Using Conditional and Unconditional Quantile Regression Approaches.
Kangho SuhAnirban BasuJosh J CarlsonKelley R BranchPublished in: American journal of cardiovascular drugs : drugs, devices, and other interventions (2021)
Using CQR and UQR allowed for heterogenous assessment of covariates along the adherence distribution, which is not possible with the traditional logistic regression method. The UQR framework revealed patients who initiate prasugrel or ticagrelor generally have lower adherence compared to those treated with branded clopidogrel, especially around the 30th quantile. Using these methods in other types of data sets, such as electronic health records, could help strengthen our results to develop policies to improve antiplatelet adherence in a targeted population.
Keyphrases
- electronic health record
- percutaneous coronary intervention
- acute coronary syndrome
- antiplatelet therapy
- clinical decision support
- public health
- st segment elevation myocardial infarction
- glycemic control
- st elevation myocardial infarction
- coronary artery disease
- adverse drug
- cancer therapy
- big data
- type diabetes
- metabolic syndrome
- machine learning
- deep learning
- newly diagnosed
- insulin resistance
- drug delivery
- artificial intelligence