Outpatient alcoholism treatment: Predictors of outcome after 3 years

Aims: This prospective study investigated predictors for relapse 3 years after completion of an intensive outpatient treatment programme for alcoholism. Design: As previous studies mainly revealed that severity of alcohol dependence, and comorbid psychopathology were predictive for subsequent relapses, the impact of these and other pre-treatment variables on the 36-month outcome was evaluated in a logistic regression analysis. A structured interview was used to assess the variables. Patients were personally interviewed at entry to, and the end of, an outpatient treatment programme, and 6, 12, 24 and 36 months after the end of treatment. One hundred and three alcohol-dependent participants who were taking part in an outpatient treatment were consecutively recruited. Results: Seventy-four patients completed the treatment programme. At the follow-up after 36 months, 2 patients had died (after heavy alcohol relapse) and 88 (88%) of the remaining patients could be located and personally re-interviewed. Forty-four (43%) patients were abstinent, 46 (45%) had relapsed and 12 (12%) were classified as improved for the total follow-up period according to the classification proposed by Feuerlein and Kufner. Based on a logistic regression analysis, significant variables for prediction of relapse were treatment drop-outs, female sex and sum of positive life events prior to treatment (relapsers had significantly fewer positive life events). Conclusions: In contrast to previous studies we could not confirm the importance of determinants known as risk factors for relapse like severity of alcohol dependence. The strongest predictor for relapsing after treatment is treatment drop-out. Since women were at an increased risk for relapse gender-specific treatment approaches should be considered. In summary, the effectiveness of the studied intensive outpatient treatment programme, with an abstinence rate of 43% for the total follow-up period of 3 years, is favourable although selection criteria of must be taken into account. (c) 2005 Elsevier Ireland Ltd. All rights reserved.