LAUSR

BACKGROUND AND OBJECTIVES This paper examines the discrepancy between implicit and explicit negative self-associations (NSA) after cognitive psychotherapy for depression as a predictor of long-term outcome. METHODS One hundred and twenty patients completed an Implicit-Association Test relating the self with depressive attributes and a self-report questionnaire with identical item content, at the end of time-limited outpatient depression psychotherapy. At post-treatment and at 3-, 6-, and 12-month follow-up, patients completed the BDI-II. We used different strategies to operationalized implicit and explicit NSA discrepancies and three-level Hierarchical linear models to analyze the effects. RESULTS We found significant interactive effects of discrepancy between implicit and explicit NSA and the direction of the discrepancy on long-term outcome. In patients with a greater explicit than implicit NSA (a damaged self-esteem pattern) a greater absolute discrepancy was associated with worse long-term outcome in terms of BDI scores at the end of follow-up and rate of change during follow-up. Consistently, with an alternative method, we found that damaged self-esteem discrepancies were associated with worse estimated BDI-II scores at the end of follow-up. LIMITATIONS The inclusion in the sample of only treatment completers limits the generalizability of the results. Furthermore, the follow-up period captured only the first 12 months after treatment. CONCLUSIONS Our results support the notion that a discrepancy between implicit and explicit negative self-associations may pose a risk factor for deterioration after psychotherapy for depression.