Racial and ethnic disparities in medication for opioid use disorder access, use, and treatment outcomes in Medicare

Abstract

Introduction: Overdose deaths are increasing disproportionately for minoritized populations in the United States. Disparities in substance use disorder treatment access and use have been a key contributor to this phenomenon. However, little is known about the magnitude of these disparities and the role of social determinants of health (SDOH) and provider characteristics in driving them. Our study measures the association between race and ethnicity and visits with Medication for Opioid Use Disorder (MOUD) providers, MOUD treatment conditional on a provider visit, and opioid overdose following MOUD treatment in Medicare. We also evaluate the role of social determinants of health and provider characteristics in modifying disparities.

Methods: Using a population of 230,198 US Medicare fee-for-service beneficiaries diagnosed with opioid use disorder (OUD), we estimate logistic regression models to quantify the association between belonging to a racial or ethnic group and the probability of visiting a buprenorphine or naltrexone provider, receiving a prescription or medication administration during or after a visit, and experiencing an opioid overdose after treatment with MOUD. Data included Medicare claims data and the Agency for Health Research and Quality Social Determinants of Health Database files between 2013 and 2017.

Results: Compared to Non-Hispanic White Medicare beneficiaries, Asian/Pacific Islander, American Indian/Alaska Native, Black, Hispanic, and Other/Unknown Race beneficiaries were between 3.0 and 9.3 percentage points less likely to have a visit with a buprenorphine or naltrexone provider. Conditional on having a buprenorphine or naltrexone provider visit, Asian/Pacific Islander, American Indian/Alaska Native, Black, Hispanic, and Other/Unknown Race were between 2.6 and 8.1 percentage points less likely to receive buprenorphine or naltrexone than white beneficiaries. Controlling for provider characteristics and SDOH increased disparities in visits and MOUD treatment for all groups besides American Indians/Alaska Natives. Conditional on treatment, only Black Medicare beneficiaries were at greater associated risk of overdose than non-Hispanic white beneficiaries, although differences became statistically insignificant after controlling for SDOH and including provider fixed effects.

Conclusion: Ongoing equity programming and measurement efforts by CMS should include explicit consideration for disparities in access and use of MOUD. This may help ensure greater MOUD utilization by minoritized Medicare beneficiaries and reduce rising disparities in overdose deaths.