Analyzing EMRs help validate atrial fibrillation risk prediction models

The incidence of atrial fibrillation has grown in the past few years—and it is now the most common sustained cardiac arrhythmia in the world. Healthcare experts don’t see things getting much better, either. By 2050, the number of patients with atrial fibrillation is expected to double to an estimated 12 million to 16 million, according to recent data.

And so, researchers are focusing on developing prediction tools for atrial fibrillation, which could help identify high-risk patients and manage patients with the disease. One such risk model, though, might not be as effective as once thought.

Lead researcher Matthew J. Kolek, MD, of Vanderbilt University in Nashville, Tennessee, and colleagues recently evaluated more than 30,000 decertified electronic medical records (EMRs) of adults who were at least 40 years old and had no history of atrial fibrillation. They also assessed a risk model that researchers from the CHARGE-AF (Cohorts for Heart and Aging Research in Genomic Epidemiology–Atrial Fibrillation) trial developed in 2012 using pooled data from prospective cohort studies. The results were published in JAMA Cardiology Oct. 12.

After a mean follow-up period of 4.8 years, 7.3 percent of patients developed atrial fibrillation. The researchers found that the full and simple CHARGE-AF risk models had poor calibration in this patient population. The models underpredicted atrial fibrillation in low-risk patients and overpredicted atrial fibrillation in high-risk patients. The simple model did not include electrocardiogram predictors.

Based on their findings, the researchers mentioned that EMRs could be used to evaluate existing prediction models and develop new models and be incorporated into clinical practice to prospectively identify people at high risk of atrial fibrillation and other diseases.

“The models performed poorly in our EMR cohort, illustrating the difficulty of applying risk models developed within prospective cohort studies to a real-world EMR context,” the researchers wrote. “Risk models for the development of [atrial fibrillation] or other complex disorders are unlikely to be widely used in clinical care unless they can be incorporated into EMR systems. Risk models, therefore, should be derived from and validated in different EMR cohorts, with the goal of prospectively and automatically identifying individuals at high risk for [atrial fibrillation] and implementing personalized strategies for primary prevention.”