The real‐world impact of dosing of methadone and buprenorphine in retention on opioid agonist therapies in Ukraine

BACKGROUND AND AIMS Ukraine's HIV epidemic remains concentrated among opioid-dependent people who inject drugs (PWID) where opioid agonist therapies (OAT) like methadone (MMT) and buprenorphine (BMT) maintenance treatments are the most cost-effective HIV prevention strategies, but remain under-scaled. This study aimed to measure the association between dose and type of OAT prescribed and treatment retention. DESIGN Observational longitudinal cohort study. PARTICIPANTS AND SETTING Patients (n = 15 290) prescribed OAT throughout Ukraine from 2004 through 2016. MEASUREMENTS Data were analyzed using time-event strategies to estimate cumulative treatment retention, defined as time to OAT discontinuation. Cumulative retention proportions at 1, 12 and 36 months were assessed for outcomes. Cox regression with log-rank likelihood assessed independent predictors of treatment discontinuation. FINDINGS The proportion prescribed high (MMT: > 85 mg; BMT: ≥ 16 mg), medium (MMT: > 40-85 mg; BMT: > 6-15 mg) and low (MMT: ≤ 40 mg; BMT: ≤ 6 mg) dosages was 25, 43 and 32%, respectively. Retention was significantly higher for BMT than MMT both at 12 (89 versus 75%) and 36 months (80 versus 56%). Although dosing levels for BMT did not influence retention, increasing dosages for MMT were significantly associated with higher retention rates at 1 (90, 96, 99%), 12 (59, 78, 91%) and 36 (34, 59, 79%) months, respectively. Independent predictors associated with 12-month OAT discontinuation were medium [adjusted hazard ratio (aHR) = 2.23; 95% confidence limit (CL) = 1.95-2.54] and low (aHR = 4.96; 95% CL = 4.37-5.63) OAT dosage relative to high dosage, male sex (aHR = 1.27; 95% CL = 1.14-1.41), MMT relative to BMT prescription (aHR = 1.57; 95% CL = 1.32-1.87) and receiving OAT in general (aHR = 1.22; 95% CL = 1.02-1.46) or tuberculosis (aHR = 1.43; 95% CL = 1.10-1.85) hospitals, relative to specialty addiction treatment and AIDS center settings. Lower dosages contributed more to dropout especially at 1 month (aHR 3.12; 95% CL = 2.21-4.41 and aHR 7.71; 95% CL = 5.51-10.79 for medium and low dosages, respectively). Younger age was significantly associated with OAT discontinuation only at 36 months (aHR = 1.08; 95% CI = 1.02-1.15). CONCLUSIONS Higher dosages of opioid agonist therapies, especially for methadone maintenance treatment patients, appear to be associated with higher levels of treatment retention in Ukraine.