Predictive Modeling
Hilltop is at the forefront in the use of artificial intelligence (AI) and machine learning in applied health services research and analytics. Hilltop is collaborating with the Maryland Department of Health to build on the success of the Maryland Primary Care Program (MDPCP) and support the delivery of advanced primary care throughout the state through Maryland’s AHEAD Primary Care Programs. These programs provide funding and support for the delivery of advanced primary care, allowing providers to play an increased role in the prevention and management of chronic disease, as well as in the prevention of unnecessary hospital utilization.
To support providing better care at lower cost, Hilltop developed a suite of predictive risk stratification models, called the Hilltop Pre- Models, that are designed to identify patients at high risk for potentially preventable health care utilization in order to help providers target care resources to the patients who need them most. Hilltop currently generates these predictive model risk scores for over 2 million Marylanders with Medicare and Medicaid each month.
Hilltop launched its predictive modeling work in 2019 with the Pre-AH Model, which identifies Maryland Medicare beneficiaries at risk for avoidable hospitalizations using administrative claims and publicly available social and environmental data. Since then, Hilltop has expanded its portfolio to include models that predict severe diabetes complications requiring inpatient or emergency department care (Pre-DC), and all-cause mortality (Pre-HE). Hilltop has also developed versions of these models tailored for Maryland’s Medicaid HealthChoice and fee-for-service populations. Learn more about these models and how they are used in our Risk Score Specifications and Codebook.
03/9/2026
Risk Score Specifications and Codebook for The Hilltop Institute’s Pre- Models, Version 4
03/01/2026
Hilltop Prediction Tools Frequently Asked Questions
03/01/2026
The Hilltop Pre-AH Model: Avoidable Hospitalizations Fact Sheet
03/01/2026
The Hilltop Pre-DC Model: Severe Diabetes Complications Fact Sheet
03/01/2026
The Hilltop Pre-HE Model: Hospice Eligibility & Advanced Care Planning Fact Sheet
11/22/2024
Evaluating a Predictive Model of Avoidable Hospital Events for Race- and Sex-Based Bias
11/01/2024
The Hilltop Pre- Models: In Brief
08/04/2024
Behind the Curtain: Comparing Predictive Models Performance in 2 Publicly Insured Populations
11/28/2022
Predicting Hospital Readmissions in a Commercially Insured Population over Varying Time Horizons
07/01/2022
Hilltop Pre-AH Model™ Infographic
11/01/2020
Predictive Modeling Suite: How We Do It
06/29/2020
Maryland Primary Care Program (MDPCP) Pre-AH Risk Score Specifications and Codebook, Version 3
01/01/2020