Objective: The present study investigates the association between congruence of patients' and therapists' perceptions of the therapeutic bond and symptom improvement. Method: Bond congruence-outcome associations were examined on the within- and between-dyad level for 580 patients (mainly depression and anxiety) receiving cognitive-behavioral therapy. Symptom change was assessed on a session-to-session level as well as from pre- to posttreatment. For the between-dyad analyses, the truth and bias model was applied. For the within-dyad analyses, polynomial regression and response surface analysis were conducted. Results: On the between-dyad level, higher temporal congruence between patients' and therapists' bond ratings (i.e., their correlation) was associated with better treatment outcomes. Additionally, the average discrepancy between therapists' and patients' bond ratings showed a significant quadratic association with treatment outcome. A tendency for therapists to moderately rate the bond lower than their patients' showed lowest posttreatment symptom scores. On the within-dyad level, we found that when patients' and therapists' ratings were in "agreement," higher bond scores were associated with fewer next-session symptoms. For "disagreement," the results showed that if therapists rated the bond as weak, whereas their patients rated it as strong, higher subsequent symptom distress was observed than if patients rated the bond as weak and their therapists rated it as strong. Conclusions: The present study highlights the importance of therapists being vigilant to session-to-session changes in the therapeutic bond to adjust their interventions accordingly.
- Response surface analysis
- Truth and bias model
- Within and between-patient analysis