Negative attentional bias for positive recovery-related words as a predictor of treatment success among individuals with an alcohol use disorder

This study assessed relationships between clients' attentional bias (AB) for different types of stimuli and their treatment outcomes. Alcohol AB during detoxification has previously been shown to predict relapse, but further research was needed to clarify this relationship. The current study determined whether AB for recovery-related words would also predict treatment outcome. Participants were 45 clients undergoing alcohol detoxification, and a control group of 36 staff members. They rated words for personal relevance in four categories (alcohol-related, neutral, positive change-related, and negative change-related). Participants completed an individualized Stroop task containing their chosen words. They were also assessed on readiness-to-change, difficulties with emotion regulation, drinking problems, anxiety, and depression. Clients were interviewed at a three-month follow-up to determine their treatment outcome. As predicted, questionnaire measures did not predict clients' treatment outcome (p > .05). A logistic regression model indicated that the best predictor of treatment outcome was AB for positive change-related words (p = .048), with successful individuals having less AB for these words than for the other word categories. Although this finding was unexpected, it was supported by significant relationships between positive change-related interference scores and continuous measures of drinking at follow-up [i.e. number of units drunk (p = .039) and number of drinking days (p = .018)]. The results suggest that positive change-related words are a better predictor of treatment outcome than are either alcohol-related words or negative change-related words. •Treatment outcome was predicted by implicit, not explicit measures.•An attentional bias for positive change-related words predicts treatment outcome.•Positive change-related words are a better predictor than alcohol-related words.•A logistic regression model correctly identified 85% of successful outcomes.