Grant: Project Grant

In recent years, there is a growing call for personalized approaches to cardiovascular disease management which aims to tailor treatments to an individual's needs. Clinical prediction models, which are mathematical algorithms that use patients' information to estimate their risk of an important outcome (e.g., survival), are essential to clinical decision making in the management of individuals with cardiovascular diseases. These tools are increasingly being used to guide decisions in clinical care to aid disease diagnosis, evaluate treatment options, and inform other important clinical decisions. But, the majority of these models are developed based on patients' risk factors collected at baseline and rarely incorporate changes in information about patients' risk factors to provide updated (real-time) risk projections over time. In addition, these models are mostly developed based on demographic and clinical risk factors but ignore patients' appraisals of their own health status and well-being (i.e., patient-reported outcomes [PROs]), which are known to modify their future risks of mortality and future cardiovascular events. Consequently, risk assessments based on existing prognostic tools can be less accurate and might misinform clinical decision making. This study addresses these limitations by developing prediction models that incorporate PROs into risk estimation to aid personalized real-time risk assessment and inform clinical decision making in the management of cardiovascular diseases. The prognostic tools developed in our study will provide more accurate personalized risk assessment/projections and improve clinical decision making related to the management of cardiovascular diseases. These methods will also enhance the implementation of PROs in clinical settings and facilitate the use of these outcomes in disease monitoring and improvement of patient outcomes