The objective of this 12-week prospective study of 183 alcohol use disorder patients was to identify patient characteristics that predict engagement and the course in addiction treatment.

At intake, patients' addiction severity, social functioning, psychiatric symptoms and diagnoses were assessed using semi-structured interviews (Addiction Severity Index, SCID-I, and SCID-II), and self-report questionnaires (Barratt's Impulsiveness Scale, Beck Depression Inventory, Beck Anxiety Inventory, and Symptom Checklist 90-R). Patients were reassessed biweekly for 12 weeks to determine alcohol use.

In the first 28 days following intake, 100 patients had slipped or relapsed on alcohol, 75 were abstinent, and 8 were lost to follow-up. Among those patients who had consumed alcohol, the mean time to first slip was 7.3 days, and the mean time to first relapse was 12.1 days. The rate of early drop-out (<28 days) was 15% among relapsing patients vs. 3% for abstinent patients (p=0.003).

There were no differences between relapsing and abstinent patients in terms of their alcohol severity at intake, or their rates of depression or anxiety disorders. However, relapsing patients had more secondary drug use (p<0.001), lower social functioning (p=0.011), higher levels of impulsivity (p=0.005), and greater rates of Cluster B personality disorders (p=0.004) than abstinent patients. The hierarchical regression model accounted for 38% of the variance in abstinence status at 28 days.

Alcoholic patients who presented with more chaotic lives (impulsivity, secondary drug use, lack of social support) and concurrent Cluster B personality disorders were more difficult to engage in addiction treatment.