Self-efficacy has been identified as one of the most consistent variables that predict the outcome of alcohol treatment. However, many previous studies in this field failed to control for other important predictors (e.g., dependences severity, psychiatric symptoms, and treatment goal). Our study's first goal was to evaluate the predictive value of self-efficacy when most other relevant variables were statistically controlled. The second goal was to compare the predictive values of self-efficacy assessed with the Situational Confidence Questionnaire (SCQ), and general self-efficacy assessed with a single question.
Four hundred and fifteen patients with alcohol dependence from 12 residential alcohol use disorder (AUD) treatment programs were assessed at treatment admission, discharge, and the 1-year follow-up. A stepwise logistic regression for abstinence was calculated using all predictors. For those predictors that were significant, a Cox survival regression analysis was performed to predict the time to the first drink after discharge.
Only abstinence as treatment goal, alcohol use during treatment, and general self-efficacy as measured by 1 question were revealed to be significant predictors in the stepwise regression, whereas all other variables, including self-efficacy as measured by the SCQ, were not significantly associated with abstinence at the 1-year follow-up. Cox survival regression analysis showed a significant difference in the time to first alcohol use between patients with maximum general self-efficacy and those with lower general self-efficacy, when the other 2 significant variables were controlled for.
General self-efficacy, that is, the patients' own prognosis of his success in remaining abstinent, was a central variable in predicting residential alcohol dependence treatment outcome. Self-efficacy showed a more accurate prognosis of outcome when it was assessed with just 1 question, than when assessed with the SCQ. With this simplified assessment, knowledge of the prognostic value of self-efficacy could be made applicable for everyday practice.