Objective: The aim of this study was to evaluate the predictors of outcome during a 6-month follow-up among heroin dependent patients receiving buprenorphine/naloxone (BN) maintenance treatment.Methods: Three hundred and ninety-two heroin dependent patients, who were consecutively admitted to the clinic (n=106, 27.04%) or accepted as outpatients for BN maintenance treatment (n=286, 72.96%) were included in the study. Patients were investigated with the Bakirkoy Opioid Withdrawal Scale (BOWS), the Substance Craving Scale (SCS), the Drug Use Disorders Identification Test (DUDIT), the Drug Abuse Screening Test (DAST-10) and the Stages of Change Readiness and Treatment Eagerness Scale (SOCRATES) at baseline evaluation.Results: Among 392 heroin dependent patients, 287 (73.21%) were considered to have relapsed to substance use or to have dropped out of treatment, whereas 105 (26.79%) were considered to be compliant to the BN maintenance treatment. Rates of having a first degree relative with substance abuse, being under probation and having a history of suicide attempts were higher in relapsed/dropout group (RDG) when compared with the maintenance group. Other sociodemographic variables did not differ between these two groups. Mean scores on the BOWS and SCS were higher in the RDG than the maintenance group at the first month, whereas the DUDIT, DAST-10, SOCRATES scores and mean dose of BN did not differ between the groups. Among items of the SCS, “severity of craving” predicted a negative outcome. When type of treatment was included in these regression analyses as an independent variable, outpatient treatment predicted negative outcome together with history of suicide attempt and being under probation.Conclusions: Since the dose of BN did not differ between the RDG and those retained in maintenance treatment, the present study suggests that those with severe withdrawal symptoms, particulary those with a high severity of craving, may need a higher dose of BN. A history of suicide attempt and being under probation are high risks for a negative outcome, particularly among those in outpatient treatment. Thus, more observed (supervised) use of BN in the first two weeks, which is more available during inpatient treatment, may improve outcome in outpatient maintenance treatment.