Background
Continuous abstinence and retention in treatment for alcohol and drug use disorders are central challenges for the treatment providers. The literature has failed to show consistent, strong predictors of retention. Predictors and treatment structure may differ across treatment modalities. In this study the structure was reinforced by the addition of supervised urine samples three times a week and mandatory daily work/structured education activities as a prerequisite of inclusion in the program.
Methods
Of 128 patients consecutively admitted to buprenorphine maintenance treatment five patients dropped out within the first week. Of the remaining 123 demographic data and psychiatric assessment were used to predict involuntary discharge from treatment and corresponding cumulative abstinence probability. All subjects were administered the Structured Clinical Interview for DSM-IV-TR, and the Symptom Checklist 90 (SCL-90), the Alcohol Use Disorder Identification Test (AUDIT), the Swedish universities Scales of Personality (SSP) and the Sense of Coherence Scale (SOC), all self-report measures. Some measures were repeated every third month in addition to interviews.
Results
Of 123 patients admitted, 86 (70%) remained in treatment after six months and 61 (50%) remained in treatment after 12 months. Of those discharged involuntarily, 34/62 individuals were readmitted after a suspension period of three months. Younger age at intake, poly-substance abuse at intake (number of drugs in urine), and number of conduct disorder criteria on the SCID Screen were independently associated with an increased risk of involuntary discharge. There were no significant differences between dropouts and completers on SCL-90, SSP, SOC or AUDIT.
Conclusion
Of the patients admitted to the programme 50% stayed for the first 12 months with continuous abstinence and daily work. Poly-substance use before intake into treatment, high levels of conduct disorder on SCID screen and younger age at intake had a negative impact on retention and abstinence.