A hospital-based addiction consultation service (START) roughly doubled in-hospital MOUD initiation and substantially improved 30-day linkage to OUD care compared with usual care.

Background

Despite effective medications for opioid use disorder (MOUD), many hospitalized patients with OUD do not start MOUD in the hospital or connect to treatment after discharge. Hospitalization presents a key opportunity to initiate MOUD and coordinate follow-up care.

Patients

• Population: 325 adults (≥18 years) hospitalized with OUD at 3 US hospitals (Los Angeles, CA; Albuquerque, NM; Springfield, MA), November 2021–September 2023.
• Key demographics: Median age 41 years; 65.5% male at birth; 48.0% Hispanic, 38.5% White, 8.6% American Indian/Alaska Native, 6.5% Black.
• Social factors: 53.8% unhoused in the past year; 50.2% unemployed.
• Exclusions: Already receiving MOUD during the current hospitalization.

Intervention

Substance Use Treatment and Recovery Team (START): An addiction medicine specialist plus a care manager provided motivational interviewing–based engagement, addiction-focused discharge planning, coordination with inpatient/outpatient clinicians, and weekly follow-up calls for 1 month post-discharge. Care included MOUD initiation and linkage support.

Control

Usual care: Standard hospital practices at each site. At one site, a consultation-liaison service (psychiatry/social work) was available; at others, primary teams could initiate MOUD. Patients did not receive the START team intervention.

Outcome

• Primary outcomes:
  • Initiation of MOUD (buprenorphine, methadone, or naltrexone) during hospitalization (EMR-confirmed).
  • Linkage to OUD-related care within 30 days post-discharge (patient self-report).
• Secondary outcomes:
  • OUD-focused discharge plan documented.
  • Initiated or continued MOUD after discharge.
  • Obtained post-discharge outpatient medical care related to OUD.
  • Opioid use days in 30 days after discharge.

Study Design

• Pragmatic, multicenter, parallel-group randomized clinical trial (1:1 START vs usual care).
• Three hospitals in CA, NM, and MA; enrollment November 2021–September 2023; last follow-up December 2023.
• Intention-to-treat analyses; adjusted risk ratios via Poisson regression with robust errors.
• Trial registration: ClinicalTrials.gov NCT05086796.

Level of Evidence

Level I (randomized clinical trial).

Follow up period

In-hospital through discharge for MOUD initiation; 30 days post-discharge for linkage and other secondary outcomes (interviews completed within a 60-day window).

Results

• Primary outcomes
  • In-hospital MOUD initiation: 57.3% (94/164) with START vs 26.7% (43/161) usual care; aRR 2.10 (97.5% CI 1.51–2.91). ARR 30.6%; NNT 4.
  • Linkage to OUD care ≤30 days: 72.0% (90/125) with START vs 48.1% (50/104) usual care; aRR 1.49 (97.5% CI 1.15–1.93). ARR 23.9%; NNT 5.
• Secondary outcomes
  • OUD-focused discharge plan documented: 49.4% vs 27.3%; aRR 1.80 (95% CI 1.36–2.41). ARR 22.1%; NNT 5.
  • Initiated/continued MOUD after discharge: 52.4% vs 30.8%; aRR 1.71 (95% CI 1.23–2.39). ARR 21.6%; NNT 5.
  • Obtained outpatient OUD-related care: 34.4% vs 18.3%; aRR 1.89 (95% CI 1.18–3.03). ARR 16.1%; NNT 7.
  • Opioid use days (30 days post-discharge): Median 0 (IQR 0–10) vs 0 (IQR 0–14); adjusted IRR 1.25 (95% CI 0.64–2.43) — no significant difference.
• Notes: Effects were consistent across sites and prior MOUD exposure strata; longer hospital stay modestly increased likelihood of initiation (no interaction with group).

Limitations

• Linkage outcomes were self-reported (potential reporting bias), although applied equally to both groups.
• Follow-up completion was 70.5% overall, higher in START (possible differential follow-up).
• No blinding; potential variability in “usual care” across sites (including some MOUD by primary teams).
• Short post-discharge follow-up (30 days) and limited power to detect differences in opioid use days.
• Conducted at large, resource-rich hospitals with existing addiction medicine capacity, which may limit generalizability.

Funding

Supported by the National Center for Advancing Translational Sciences and the National Institute on Drug Abuse (U01TR002756-01A1). The funders had no role in study design, data collection/analysis, manuscript preparation, or publication decisions.

Citation

Ober AJ, Murray-Krezan C, Page K, et al. Hospital Addiction Consultation Service and Opioid Use Disorder Treatment: The START Randomized Clinical Trial. JAMA Internal Medicine. 2025;185(6):624-633. Published online April 7, 2025. doi:10.1001/jamainternmed.2024.8586.