Effect of Health Information Exchange Plus a Care Transitions Intervention on Post-Hospital Outcomes Among VA Primary Care Patients: a Randomized Clinical Trial

Abstract

Background Health information exchange (HIE) notifications when patients experience cross-system acute care encounters offer an opportunity to provide timely transitions interventions to improve care across systems. Objective To compare HIE notification followed by a post-hospital care transitions intervention (CTI) with HIE notification alone. Design Cluster-randomized controlled trial with group assignment by primary care team. Patients Veterans 65 or older who received primary care at 2 VA facilities who consented to HIE and had a non-VA hospital admission or emergency department visit between 2016 and 2019. Interventions For all subjects, real-time HIE notification of the non-VA acute care encounter was sent to the VA primary care provider. Subjects assigned to HIE plus CTI received home visits and telephone calls from a VA social worker for 30 days after arrival home, focused on patient activation, medication and condition knowledge, patient-centered record-keeping, and follow-up. Measures Primary outcome: 90-day hospital admission or readmission. Secondary outcomes: emergency department visits, timely VA primary care team telephone and in-person follow-up, patients’ understanding of their condition(s) and medication(s) using the Care Transitions Measure, and high-risk medication discrepancies. Key Results A total of 347 non-VA acute care encounters were included and assigned: 159 to HIE plus CTI and 188 to HIE alone. Veterans were 76.9 years old on average, 98.5% male, 67.8% White, 17.1% Black, and 15.1% other (including Hispanic). There was no difference in 90-day hospital admission or readmission between the HIE-plus-CTI and HIE-alone groups (25.8% vs. 20.2%, respectively; risk diff 5.6%; 95% CI − 3.3 to 14.5%, p = .25). There was also no difference in secondary outcomes. Conclusions A care transitions intervention did not improve outcomes for veterans after a non-VA acute care encounter, as compared with HIE notification alone. Additional research is warranted to identify transitions services across systems that are implementable and could improve outcomes.