Identifying and Predicting Patients with Preventable High Utilization

Identifying and Predicting Patients with Preventable High Utilization

Lead Principal Investigator: Rainu Kaushal, MD, MPH

Organization:  Joan & Sanford I. Weill Medical College of Cornell University

Purpose: In collaboration with patients and health systems, we are creating a conceptual model to identify, characterize, and predict preventable high utilizers using claims, clinical, and social determinant data. This will generate actionable evidence for health systems to improve the health of patients with high healthcare use.

Funded through a Patient-Centered Outcomes Research Institute (PCORI) Award (HSD-1604-35187).

Identifying and Predicting Patients with Preventable High Utilization

Lead Principal Investigator: Rainu Kaushal, MD, MPH

Organization:  Joan & Sanford I. Weill Medical College of Cornell University

Purpose: In collaboration with patients and health systems, we are creating a conceptual model to identify, characterize, and predict preventable high utilizers using claims, clinical, and social determinant data. This will generate actionable evidence for health systems to improve the health of patients with high healthcare use.

Funded through a Patient-Centered Outcomes Research Institute (PCORI) Award (HSD-1604-35187).

Overview

We have operational definitions of high utilizer, preventable high utilizer, and seven categories of high utilizer via thorough review of the literature, consultations with experts, and discussions among faculty at NYC-CDRN, OneFlorida, and CAPriCORN as well as patients and health system stakeholders.
We have successfully developed computable phenotypes for the seven categories of high utilizer and used those computable phenotypes to conduct preliminary analyses on four million patients in NYC-CDRN and OneFlorida.
We are running focus groups to solicit input from patients, clinicians, and health system leaders on best ways to characterize, classify, and prioritize definitions of high utilizer and preventable high utilizer. They are helping us with definitions of categories of preventable high users.

who's involved?

sites

3 participating Networks:
NYC-CDRN, CAPriCORN, OneFlorida
Together with an active, engaged, diverse set of stakeholders, including patients, clinicians and health systems, across 20 institutions

hospital

how we're making a difference

We are creating new conceptual models to define preventable high utilization. While a lot of work has been done on high utilization and preventable utilization, the concept of preventable high utilization is relatively unexplored

data-ipad

We are leveraging and extending the PCORnet Common Data Model to integrate clinical data, claims data, and social determinants data into an analysis-ready data set and running the data characterization queries
We are using PCORnet infrastructure and engagement principles to engage health systems, clinicians and patients. These stakeholders are participating in our focus groups, advisory councils and governance board