Hilltop to Predict Risk of Avoidable Hospitalizations

The Hilltop Institute has entered into an interagency agreement with the Maryland Primary Care Program (MDPCP) to design and implement a model that will predict the risk of avoidable hospitalizations for individuals in Maryland’s Medicare population.

MDPCP aims to support the delivery of advanced primary care throughout the state and allow community providers to play a vital role in prevention, improving health outcomes, and controlling total health care spending growth. The program is a vital part of Maryland’s Total Cost of Care All-Payer Model, which is transforming the financing and delivery of health care in Maryland to promote greater efficiency, better value, and improved health outcomes.

To develop the model, Hilltop researchers will create an algorithm that identifies a given individual’s relative risk of avoidable hospitalization and design a monthly scoring process that will assist MDPCP providers in triaging care coordination resources. Hilltop will build, train, and host the model, as well as disseminate risk scores to providers in collaboration with CRISP (Chesapeake Regional Information System for our Patients), Maryland’s health information exchange.

The research team is led by Hilltop’s Senior Director of Analytics and Research and Chief Data Scientist, Ian Stockwell, PhD. In his new role, Stockwell is building Hilltop’s expertise in data science and advanced analytics and developing faculty partnerships for research.

Posted: February 4, 2019, 2:35 PM