348-OR: Impact of Endocrinology ECHO on Utilization in Medicaid Patients with Complex Diabetes

Background: Poorly controlled diabetes can be difficult to manage, particularly in high-risk patients and in settings without an endocrinologist. Project ECHO is a case-based learning and tele-mentoring model that connects primary care providers (PCPs) with specialists. We evaluated the impact of the ECHO model on healthcare utilization and care for Medicaid patients with diabetes in New Mexico (NM). Methods: Between January 2015 and April 2017, patients with complex diabetes from 10 federally qualified health centers in NM were recruited to receive diabetes care from a PCP and community health worker upskilled via weekly Endo ECHO sessions. We included patients with 6 months of continuous enrollment in the NM Medicaid claims database and performed 5:1 propensity matching of participants to comparison Medicaid beneficiaries. We used a difference-in-difference (DID) approach to compare utilization and processes of care between study participants and controls. Results: Of 539 Medicaid patients enrolled in Endo ECHO, 235 met inclusion criteria and were successfully matched to comparison patients. Outpatient visits increased for study participants in the pre- versus post-enrollment period (801 vs. 1056 per 1,000 patients/month) but not for comparators (785 vs. 680 per 1000 patients/month, DID p<0.01). We observed no difference in inpatient (131 vs. 134 per 1,000 patients/month) or ED utilization (24 vs. 26 per 1,000) in the pre- versus post-enrollment period for study participants. The intervention was associated with increased metformin utilization (DID 15.2%, p=0.01), but had no effect on expenditures, statin utilization, or ACE inhibitor/ARB utilization for hypertension. Discussion: Patient enrollment in Endo ECHO was associated with increased outpatient utilization and increased compliance with some, but not all, process measures. No impact was observed on hospitalizations or ED visits. Ongoing evaluation will determine the impact of the ECHO model on clinical outcomes. Disclosure S. Blecker: Advisory Panel; Self; Medtronic. E. Lemieux: None. M. Paul: None. C. Berry: None. M.F. Bouchonville: None. B. Hager: None. S. Arora: None. J. Billings: None. Funding The Leona M. and Harry B. Helmsley Charitable Trust