Continuous opioid substitution treatment over five years: Heroin use trajectories and outcomes.

Abstract

This is the first national study in England of continuous long-term opioid substitution treatment (OST). All adults were admitted to community OST for opioid use disorder (OUD) in 2008/09 with continuous enrolment to 2013/14 (n = 7719). Heroin use trajectories were identified by multilevel Latent Class Growth Analysis. In Year 6 and 7 of follow-up, the outcome measure (analysed by multilevel, multivariable logistic regression) was 'successful completion and no re-presentation' (SCNR) to community treatment within six months. Five heroin use trajectory classes were identified: 'gradual decreasing' (20.9%), 'decreasing then increasing' (21.7%), 'continued low-level' (17.0%), 'rapid decreasing' (25.6%), and 'continued high-level' (14.8%). At the end of Year 7, 4616 people (60.3%) remained in OST. Of those discharged, 28.8% achieved the SCNR follow-up outcome. SCNR was more likely in the 'gradual decreasing' (adjusted odds ratio [AOR] 2.40; 95% confidence interval [CI] 1.77-3.26), 'continued low-level' (AOR 2.46; CI 1.78-3.40), and 'rapid decreasing' (AOR 3.40; CI 2.43-4.37) classes relative to the 'continued high-level' class. SCNR was more likely among patients employed at admission (AOR 1.45; 95% CI 1.15-1.83) and those receiving adjunctive psychosocial interventions (AOR 1.44; 95% CI 1.03 to 2.02). Among English patients in OST for 5 years, heroin use trajectories were clearly delineated with a gradient of response on the study outcome. Successful completion and no re-presentation was achieved by 28.8% of discharged patients. The rapid decreasing trajectory had the greatest likelihood of positive outcome. Adjunctive psychosocial intervention during OST was associated with positive outcome.