CHARGE-AF Risk Calculator – Atrial Fibrillation Prediction

About the CHARGE-AF Model

The Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE-AF) model estimates the 5-year risk of new-onset atrial fibrillation in individuals without previously diagnosed AF. The original simple model uses routinely available demographic and clinical variables.

Model calculation

Model predictors

• Age, race, height, and weight.
• Systolic and diastolic blood pressure.
• Current smoking and antihypertensive medication use.
• Diabetes mellitus, heart failure, and prior myocardial infarction.

Population and performance

The derivation pooled 18,556 participants aged 46–94 years from ARIC, CHS, and the Framingham Heart Study. External validation included 7,672 participants from AGES-Reykjavik and the Rotterdam Study. The simple model's derivation C-statistic was 0.765; discrimination in the external cohorts was lower (0.664 and 0.705, respectively).

Important Limitations

• Use only for individuals without previously diagnosed AF.
• The original derivation population was aged 46–94 years. Estimates outside this age range are extrapolations.
• The race term was derived in cohorts that were predominantly White, with African-American participants also represented. Calibration in Asian and other populations is uncertain.
• External validation has shown that discrimination and calibration can vary by population and setting.
• The model estimates risk; it does not diagnose AF, determine an indication for anticoagulation, or establish a screening or treatment threshold.

References