Prognostic models for stable coronary artery disease based on electronic health record cohort of 102 023 patients

Area of Science:

• Cardiology
• Clinical Epidemiology
• Health Informatics

Background:

• The growing population with stable coronary artery disease (SCAD) lacks validated prognostic models for clinical management.
• Accurate risk stratification is crucial for guiding treatment decisions in SCAD patients.

Purpose of the Study:

• To develop and validate prognostic models for all-cause mortality and non-fatal myocardial infarction (MI) or coronary death in SCAD patients.
• To provide tools for improved risk stratification in clinical practice.

Main Methods:

• Developed prognostic models using linked electronic health records from 102,023 SCAD patients.
• Included predictors such as demographics, comorbidities, and clinical measurements.
• Validated models internally and externally for calibration and discrimination.

Main Results:

• Models demonstrated good calibration and discrimination (C-index 0.811 for mortality, 0.778 for coronary events).
• Identified key predictors including age, sex, CAD diagnosis, deprivation, smoking, and various comorbidities.
• Estimated 5-year risks for mortality (20.6%) and coronary events (9.7%).

Conclusions:

• Validated prognostic models are available for clinical use in SCAD.
• These models can support risk stratification as recommended by clinical guidelines.
• Application of these models may lead to improved life years and event-free years.