In a groundbreaking study, researchers have developed advanced electronic health record-based predictive models for worsening heart failure (WHF) events and all-cause mortality. The study, spanning from 2011 to 2019, surveyed 338,426 adults within the Kaiser Permanente integrated health care system.
Alan S. Go, MD, Kaiser Permanente Bernard J. Tyson of Medicine (KPSOM) Professor of Health Systems Science, and Andrew P. Ambrosy, MD, Associate Professor of Health Systems Science, coauthored the article, “Developing Clinical Risk Prediction Models for Worsening Heart Failure Events and Death by Left Ventricular Ejection Fraction,” recently published in the Journal of the American Heart Association.
The study focused on diverse left ventricular ejection fraction (LVEF) categories: reduced EF, mildly reduced EF, and preserved EF. Results unveiled the one-year risks for WHF events and death. For reduced EF, the risks were 22.3% and 13.0%, for mildly reduced EF, 17.0% and 10.1%, and for preserved EF, 16.3% and 10.3%. Notably, the models displayed impressive predictive accuracy, with the WHF model achieving an area under the curve of 0.76, and the death model reaching 0.83. The consistency in performance and predictors across different WHF encounter types and LVEF categories offer a significant leap in risk assessment, providing valuable insights for clinicians and shaping the future of heart failure management.