Abstract:
|
In the clinical studies of cardiovascular diseases, multiple longitudinal variables (e.g., blood pressure, cholesterol) are measured and recurrent events (e.g., stoke and coronary heart disease) are recorded. Personalized prediction of the next occurrence of recurrent events is of great clinical interest because it enables physicians to make more informed decisions and recommendations for patients, leading to improved outcomes and increased benefit. We propose a joint model of longitudinal and recurrent event data. We develop a Bayesian approach for parameter estimation and a dynamic prediction framework for predicting target patients' future outcome trajectories and risk of next recurrent events, based on their data up to the prediction time point. Our method development is motivated by and applied to the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT, n=42,418), the largest clinical study to compare the effectiveness of medications to treat hypertension.
|