Post-Discharge Treatment Engagement Among Patients with an Opioid-Use Disorder

Opioid misuse is a growing public health problem, and estimates show a 150% increase in opioid-related hospital stays over the last two decades. This study examined factors associated with substance use treatment engagement following a hospitalization for opioid use disorder or overdose.

This study analyzed the Truven Health Analytics MarketScan® Commercial Claims and Encounters (CCAE) database for 2010 through 2014 to study post-hospitalization substance use disorder (SUD) treatment of individuals aged 18-64 who had an inpatient admission for an opioid-use disorder or opioid overdose. Engagement in post-discharge SUD treatment was defined as having at least two unique outpatient visits within 30 days of a hospitalization. Generalized estimating equations (GEEs) with a binomial link were used to determine the factors associated with SUD treatment engagement.

Only 17% of patients engaged in SUD treatment within 30 days of hospital discharge. A behavioral health outpatient visit prior to the SUD admission increased the odds of engaging in SUD treatment by 1.34 (CI: 1.25-1.45), an antidepressant prescription drug fill prior to the SUD admission increased the odds by 1.14 (CI: 1.07-1.21), a benzodiazepine fill prior to the SUD admission increased the odds by 1.14 (CI: 1.07-1.21), a principal diagnosis for an SUD at index admission increased the odds by 2.13 (CI: 1.97-2.30), an alcohol-related disorder diagnosis at index admission increased the odds by 3.13 (CI: 2.87-3.42), and an additional SUD diagnosis at the index admission increased the odds by 2.72 (CI: 2.48-2.98).

We found low rates of SUD treatment engagement following hospitalizations for opioid use disorders and overdoses. Patients with prior engagements with behavioral health providers were more likely to engage in follow-up care; therefore, providers may need to focus additional efforts on patients admitted to the hospital with opioid-use disorders who do not have an existing provider relationship.