Evaluation of a National Care Coordination Program to Reduce Utilization Among High-cost, High-need Medicaid Beneficiaries With Diabetes.

Abstract

Medical, behavioral, and social determinants of health are each associated with high levels of emergency department (ED) visits and hospitalizations. The objective of this study was to evaluate a care coordination program designed to provide combined "whole-person care," integrating medical, behavioral, and social support for high-cost, high-need Medicaid beneficiaries by targeting access barriers and social determinants. Individual-level interrupted time series with a comparator group, using person-month as the unit of analysis. A total of 42,214 UnitedHealthcare Medicaid beneficiaries (194,834 person-months) age 21 years or above with diabetes, with Temporary Assistance to Needy Families, Medicaid expansion, Supplemental Security Income without Medicare, or dual Medicaid/Medicare. Our outcome measures were any hospitalizations and any ED visits in a given month. Covariates of interest included an indicator for intervention versus comparator group and indicator and spline variables measuring changes in an outcome's time trend after program enrollment. Overall, 6 of the 8 examined comparisons were not statistically significant. Among Supplemental Security Income beneficiaries, we observed a larger projected decrease in ED visit risk among the intervention sample versus the comparator sample at 12 months postenrollment (difference-in-difference: -6.6%; 95% confidence interval: -11.2%, -2.1%). Among expansion beneficiaries, we observed a greater decrease in hospitalization risk among the intervention sample versus the comparator sample at 12 months postenrollment (difference-in-difference: -5.8%; 95% confidence interval: -11.4%, -0.2%). A care coordination program designed to reduce utilization among high-cost, high-need Medicaid beneficiaries was associated with fewer ED visits and hospitalizations for patients with diabetes in selected Medicaid programs but not others.