TY - JOUR TI - Patient-therapist working alliance rating discrepancies predicting alcohol treatment participation and clinical outcomes DO - https://doi.org/doi:10.7282/t3-4mzb-9m88 PY - 2020 AB - Client and therapist perceptions regarding their therapeutic working alliance have been shown to be independently predictive of treatment outcomes. The extent to which patients and therapists agree or disagree about the quality of the therapeutic relationship may be a distinct variable from either individual’s perspective of the quality of the relationship. Recent studies of alliance have explored patient and therapist congruence and discrepancy regarding measurements of the alliance and the degree to which they are predictive of treatment outcomes. This study centers on an investigation of patient-therapist discrepancy in evaluations of the therapeutic relationship as a predictor of treatment participation and outcomes (i.e., alcohol use, alcohol-related problems). To achieve study aims, secondary data analyses were conducted using 741 participants from the outpatient sample of Project MATCH. To examine the extent to which patient-therapist Working Alliance Inventory (WAI) rating discrepancies predicted posttreatment outcomes and treatment participation, a difference score was computed and multiple regression analyses conducted. To further examine patient-therapist WAI discrepancy as a predictor of posttreatment outcomes and treatment participation, both patient and therapist WAI scores were categorized as high or low alliance based on a median split, yielding a 2 by 2 study design (therapist score vs. patient score; low vs. high). Study hypotheses based on this 2 by 2 study design were tested using four analyses of covariance (ANCOVAs). Working alliance discrepancy was not found to be a significant predictor of posttreatment alcohol use behavior, alcohol-related negative consequences, or treatment participation when controlling for baseline measures. No significant interaction effect was found for the combination of patient and therapist alliance levels on posttreatment outcomes when controlling for baseline measures, although significant main effects (p < .05) were detected. The results of the current study do not support the hypothesis that alliance discrepancy accounts for differences in treatment outcomes and treatment participation. Limitations of the study and future directions for research examining alliance discrepancy are discussed. KW - Alcoholism -- Treatment KW - Alcohol treatment KW - Clinical Psychology LA - English ER -