Abstract
BACKGROUND: Primary care settings tailored for persons with experience of homelessness (PEH) could enhance opioid use disorder treatment delivery, but evidence is lacking. OBJECTIVE: To examine medication treatment for opioid use disorder (MOUD) among PEH who received homeless-tailored primary care in the Department of Veterans Affairs (VA). METHODS: Receipt of MOUD was assessed from electronic health records among VA primary care patients with experience of homelessness and OUD in 2016-2020. We estimated the proportion who received MOUD over time and applied mixed effect Poisson models with entropy balance weighting to estimate differences in MOUD by primary care type (homeless-tailored vs. mainstream). Secondary analyses examined the consistency of findings across 18VA service regions. FINDINGS: The percentage of eligible PEH (n = 45,864) receiving any MOUD rose from 42.1 % to 51.0 % over time. Half (51.3 %) of those initiating MOUD received > 30 days MOUD over one year. In unadjusted models, the proportion receiving MOUD was slightly higher in homeless-tailored primary care, compared to mainstream primary care (48.8 % vs 46.4 %, Unadjusted Incidence Rate Ratio=1.09, 95 % CI=1.03-1.16). After covariate adjustment, there was no statistical difference between groups (Adjusted IRR=0.97, CI=0.92-1.02). This proved broadly consistent across VA service regions. CONCLUSIONS: Concomitant with a national VA initiative to tailor primary care services for PEH, half of diagnosed patients received MOUD. Yet evidence of durable treatment was low, and the homeless-tailored clinics did not outperform mainstream clinics. Efforts to tailor primary care for PEH may require specialized addiction staffing and implementation support to improve MOUD care in these settings.