Prediction of coronary artery disease risk based on multiple longitudinal biomarkers

Area of Science:

• Cardiovascular Disease Research
• Biostatistics
• Epidemiology

Background:

• Cardiovascular disease (CVD) risk prediction is a critical area of research.
• High blood pressure (BP) is a major CVD risk factor.
• Current models use single BP measurements, not reflecting longitudinal BP changes.

Purpose of the Study:

• To develop and evaluate joint models for predicting coronary artery disease (CAD) events.
• To assess the predictive performance of models incorporating longitudinal BP data.
• To compare joint models with traditional risk prediction methods.

Main Methods:

• Utilized a primary care cohort with up to 20 years of follow-up.
• Employed joint modeling to analyze time-to-event data (CAD) and longitudinal BP measurements (systolic and diastolic).
• Applied novel prediction metrics and simulations to evaluate model performance.

Main Results:

• Longitudinal BP trajectories demonstrated significant predictive power for CVD events.
• Joint models incorporating BP trajectories showed improved predictive performance compared to traditional models.
• Simulations confirmed the enhanced predictive capabilities of the proposed joint models.

Conclusions:

• Longitudinal analysis of blood pressure provides superior cardiovascular risk prediction.
• Joint modeling offers a more accurate approach to assessing CVD risk by integrating time-dependent BP data.
• These findings can inform clinical guidelines for more effective CVD prevention strategies.