Abstract
IMPORTANCE: Nearly 727 000 individuals in the US died of opioid overdoses between 1999 and 2022. The current workforce of addiction medicine specialists is inadequate to address the scale of this crisis, and primary care clinicians (PCCs) do not feel sufficiently supported to treat opioid use disorder (OUD). OBJECTIVE: To evaluate whether an electronic health record-integrated clinical decision support system (CDSS) increases OUD diagnosis and treatment in primary care. DESIGN, SETTING, AND PARTICIPANTS: This pragmatic cluster randomized clinical trial was conducted from April 2021 to December 2023. Primary care clinics in 3 health systems in 4 US states were randomized to receive or not receive an electronic health record-integrated CDSS aimed at improving OUD diagnosis and treatment. Eligible patients were aged 18 to 75 years, visited a randomized clinic, and had an OUD diagnosis in the last 2 years, opioid overdose in the last 6 months, or risk score indicating high risk of OUD or opioid overdose. Data were analyzed from September 2023 to October 2024. INTERVENTIONS: The OUD CDSS provided personalized treatment recommendations to patients and PCCs in intervention clinics. MAIN OUTCOMES AND MEASURES: Primary outcomes were likelihood to receive (1) an OUD diagnosis (among high-risk patients without a baseline OUD diagnosis), (2) a naloxone prescription, or (3) a prescription of a medication for OUD (MOUD) or specialty referral, all within 30 days of first eligible (index) visit, and (4) days covered by a MOUD prescription in the 90 days after index. RESULTS: Among 10 891 patients meeting eligibility criteria, 5918 (54.3%) were female, and the mean (SD) age was 48.0 (13.9) years. There was no difference in OUD diagnoses within 30 days between groups. Patients in the intervention group had more naloxone orders (80 of 5538 [1.4%] vs 40 of 5353 [0.7%]; odds ratio, 1.76; 95% CI, 1.14-2.72) and orders for MOUDs or treatment referral (775 of 5538 [14.0%] vs 503 of 5353 [9.4%]; odds ratio, 1.48; 95% CI, 1.05-2.08) within 30 days. There were no differences in median (IQR) days covered by MOUD over 90 days postindex between intervention (84 [55-90] days) and usual care (83 [55-90] days; rate ratio, 1.00; 95% CI, 0.93-1.08) or in overdose or death rates during the intervention period. CONCLUSIONS AND RELEVANCE: In this cluster randomized clinical trial, the intervention improved rates of naloxone orders and OUD treatment in primary care but did not affect days covered by a MOUD over 90 days postindex or overdose or death rates. These findings demonstrate an OUD CDSS can help increase access to OUD treatment in primary care. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT04198428.