Homelessness, drug preference predict relapse, treatment failure in opioid use disorder

Homelessness, parole and probation status, drug preference, and other factors that tend to affect tolerability of medication initiation must be considered when matching patients with opioid use disorder admitted to inpatient treatment to either buprenorphine or extended-release naltrexone, suggests a recent study.

The investigators examined 50 demographic and clinical characteristics as moderators of the effects of medication assignment on relapse to regular opioid use and failure to initiate treatment in a multicentre 24-week randomized trial of assignment to buprenorphine-naloxone (n=287) vs extended-release naltrexone (n=283), comprising inpatients planning to start treatment for opioid use disorder.

Logistic regression, with correction for multiple testing, was used to estimate moderator-by-medication interactions.

The intent-to-treat analysis revealed that homeless patients had a lower relapse rate when given extended-release naltrexone compared with buprenorphine-naloxone (51.6 percent vs 70.4 percent; odds ratio [OR], 0.45, 95 percent confidence interval [CI], 0.22–0.90), while those who were not homeless had a higher relapse rate when given extended-release naltrexone compared with buprenorphine-naloxone (70.9 percent vs 53.1 percent; OR, 2.15, 95 percent CI, 1.44–3.21).

Interaction was not significant in the subsample of patients who initiated medication, with a similar pattern of lower relapse with extended-release naltrexone compared with buprenorphine among homeless patients (41.4 percent vs 68.6 percent; OR, 0.32, 95 percent CI, 0.15–0.68), but less difference among those who were not homeless (57.2 percent vs 52.0 percent; OR, 1.24, 95 percent CI, 0.80–1.90).

Moderators for failure to initiate medication were as follows: stated preference for medication (failure was less likely if patients were given their preferred medication), parole and probation status (fewer failures with extended-release naltrexone for those on parole or probation), and presence of pain and timing of randomization (more failure with extended-release naltrexone for those endorsing moderate to severe pain and randomized early while still undergoing medically managed withdrawal).