| A significant body of research has highlighted the importance of the therapeutic alliance to treatment outcome in psychotherapy. In addition, the therapeutic rupture has been the focus of intense study as a potential mediator to positive treatment outcome. This study examines whether certain structural properties of alliance rupture and repairs are related to treatment outcome, specifically (1) frequency of rupture and repair events, (2) temporal proximity of rupture and repair events to each other, and (3) location of rupture and repair events (early, middle, or late in treatment). Participants were 63 patients who underwent a brief (30 session) psychotherapy in one of two treatment conditions (cognitive-behavioral therapy or brief relational therapy). Two ancillary aims contrasted the primary cohort of treatment "completers" with a smaller cohort of dropout patients (N= 20), and any differences between treatment groups. Ruptures and repairs were measured by patient-rated working alliance scores (WAI), and clinical outcome was determined through patient and therapist-rated Target Complaints (TC) and Inventory of Interpersonal Problems (IIP), patient-rated Symptom Checklist, Brief Symptom Inventory (SCL-BSI), and therapist-rated Global Assessment Scale (GAS). Ruptures and repairs were cataloged by a method of control charting that was used to determine WAI scores that crossed a threshold of statistical significance.;Bivariate correlations revealed that "prolonged ruptures," which grouped consecutive rupture sessions together, were negatively correlated with therapist-rated symptoms at termination. The same relationship was found with increases in repair frequencies when the repair occurred less than 2 sessions from the respective rupture; the association did not hold for repairs occurring 3 or more sessions out. ANOVA tests did not reveal any difference between rupture or repair locations. ANOVA tests did not detect any difference in outcome between top-performing completers and dropouts. ANOVA testing revealed that BRT patients were rated higher in symptom endorsement from therapists when contrasted to patients receiving CBT. Chi-square testing did not reveal any differences in independent variables with regard to treatment group. Implications for clinical practice are discussed, as well as directions for future research. |
