Session-to-session effects of alliance ruptures in outpatient CBT: Within- and between-patient associations

Objective: The concept of alliance ruptures has had a large impact on contemporary studies of the alliance. Although this debate mainly focuses on the effect of ruptures and repairs on subsequent alliance and outcome levels within patients, to date no study has actually examined these within-patient effects. The present study fills this lacuna by examining the effect of alliance ruptures and in-session repairs on alliance ratings and symptom impairment in the subsequent session, accounting for the temporal sequence of ruptures and symptom impairment. Method: Ruptures and in-session repairs were rated by 1,210 patients and 147 therapists using a postsession questionnaire (PSQ-P/-T). Alliance was assessed with the Bern postsession reports and symptomatic impairment with the Hopkins symptom checklist, short form. Results: Patient- and therapist-reported ruptures were significant predictors of subsequent alliance and symptom impairment. While sessions in which both patient and therapist perceived a rupture were especially detrimental for next session symptom distress, they were less damaging to next session alliance levels than sessions in which either only the patient or the therapist experienced the rupture. Neither the intensity of the rupture nor the perceived level of in-session repair were associated with next-session fluctuations in symptom or alliance levels within-patient. Conclusion: The findings demonstrate the importance of disentangling between- and within-patient rupture and in-session repair processes to better understand the roles of rupture and repair in treatment.