Scientists from IBM Research and Sacramento, Calif.-based Sutter Health created an artificial intelligence model to predict heart failure using primary care patient information from EHRs.
The team identified 1,684 heart failure cases and 13,525 patient controls to develop the model, according to a study published in Circulation: Cardiovascular Quality and Outcomes. Using the model, the scientists found only six of 28 established risk factors consistently predicted heart failure.
The model's accuracy increased based on time before heart failure diagnosis, degree of historical data included in EHRs and how frequently patients visited their physician. The model also improved when EHRs included diagnosis, medication order and hospitalization information.
"These empirical findings suggest possible guidelines for the minimum amount and type of data needed to train effective disease onset predictive models using longitudinal electronic health records data," study authors concluded.