Development and validation of a claim-based provider-level measurement of use of medications to treat opioid use disorder

Abstract

ObjectiveMedication for opioid use disorder (MOUD) is effective but underused. Measuring the percentage of a provider's patients with an opioid use disorder (OUD) who receive MOUD may drive quality improvement and stimulate greater use of medications. This study introduces and tests a provider-level measure of MOUD receipt. MethodsThe study used claims and enrollment data from 32 states in the 2014 Medicaid Analytic Extract to measure the proportion of a provider's patients who received MOUD within 30 days of their OUD diagnosis. The research team assessed measure reliability with several tests to establish the effect of provider on MOUD receipt; and assessed the validity by correlation with a measure of emergency department visits or hospitalizations related to substance use. ResultsThe sample included 434,484 individuals treated for OUD by one or more of 9398 providers. The mean provider score was 38 %, indicating that 38 % of the average provider's patients received an MOUD within 30 days of an OUD diagnosis (44 % for clinicians [N = 5344] and 31 % for facilities [N = 4054]). Provider performance varied considerably. The interquartile ranges were 11 %–79 % and 9 %–45 % among clinicians and facilities, respectively. The measure reliably distinguished between lower- and higher-performing providers and demonstrated convergent validity, as indicated by a significant and moderately sized negative correlation between MOUD receipt and substance use–related hospitalizations or emergency department visits. ConclusionsThe measure may help to improve access to MOUD and OUD outcomes by identifying providers who could benefit from technical assistance, quality improvement initiatives, and resources to expand MOUD prescribing.