Opioid relapse and its predictors among methadone maintenance patients: a multicenter, cross-sectional study in Vietnam

Abstract

Background: Opioid relapse, one of the common and severe problems during methadone maintenance treatment, can give rise to poor treatment outcomes. This study measured the opioid relapse rate and its associated factors among methadone maintenance patients in Vietnam.

Methods: Information about the demographic characteristics and social support of 655 patients was collected through direct interviews. Medical records were used to gather data on treatment characteristics. Relapse was determined via urine opioid test results.

Results: The overall relapse rate of patients during treatment was 13.1%. According to the multivariate logistic regression model, living in mountainous areas (adjusted odds ratio (aOR) = 3.63, 95% CI 1.90-7.46) and long duration of drug use in the past (aOR = 1.06, 95% CI 1.03-1.09) were associated with an increase in the odds of opioid relapse. By contrast, living with many family members (aOR = 0.69, 95% CI 0.55-0.85), having longer treatment time (aOR = 0.80, 95% CI 0.73-0.87), and completely adhering to treatment (aOR = 0.38, 95% CI 0.23-0.64) were protective for opioid relapse. As per the univariate analyses, the odds of opioid relapse declined by 25% for each increase of one close friend or relative (OR = 0.75, 95% CI 0.66-0.86). Regarding social support (range score: 0-100), each additional increase of one score was associated with a 1% decrease in the odds of opioid relapse (OR = 0.99, 95% CI 0.98-0.99). Patient sex, education level, occupation type, patient's monthly income, family's monthly income, the number of previous treatments, daily methadone dose, comorbidity, and received antiretroviral therapy were not associated with opioid relapse among patients (p > 0.05).

Conclusions: Residence, the role of family and social support, and treatment adherence should be paid more attention to guarantee and enhance the success of methadone maintenance treatment.

Keywords: Associated factor; Concurrent drug use; Methadone maintenance treatment; Opioid relapse; Social support.

Fig. 1

The receiver operating characteristic (ROC) curve analysis for the multivariate logistic regression model. Model: Opioid relapse ~ residence (metropolitan/mountainous) + the number of family members living with MMT patients + duration of drug use in the past (year) + treatment time (year) + treatment adherence (yes/no). The Hosmer–Lemeshow goodness of fit test results demonstrated that the multivariate logistic regression model could adequately fit the data (χ² = 4.742, df = 8, p = 0.785). The AUC of this model was 0.789 (95% CI 0.741–0.839)