New Health Services Research Article Documents Accuracy of Hilltop Pre-AH™ Model
A new study just published in Health Services Research by Hilltop researchers Morgan Henderson, PhD, Fei Han, PhD, and Ian Stockwell, PhD, and their partners at the Maryland Primary Care Program (MDPCP) Chad Perman, MPP, and Howard Haft, MD, documents the underpinnings, development, and performance of the Hilltop Pre-AH Model™. The Hilltop Pre-AH Model™ is a risk prediction model that uses a variety of risk factors derived from Medicare claims data to estimate the probability that a given individual incurs an avoidable hospital event in the near future. These risk scores are intended to assist MDPCP practices with the identification of patients that have a high risk of incurring an avoidable hospitalization or emergency department visit and are currently estimated for almost 400,000 individuals spanning over 500 primary care practices in Maryland. The authors found that 48.7% of avoidable hospital events are incurred by the top 10% riskiest individuals as of April 2020, indicating strong model accuracy, and that the scores significantly outperform the Centers for Medicare & Medicaid Services hierarchical condition category risk scores in terms of predictive power. Read the abstract.
November 11, 2021