The interplay between agency and therapeutic bond in predicting symptom severity in long-term psychotherapy.

Abstract

This study investigated the interplay between agency and therapeutic bond in predicting patient symptoms in outpatient psychotherapy. A total of N = 731 patients provided measurements of agency (Therapeutic Agency Inventory; TAI), therapeutic bond (bond subscale of Working Alliance Inventory-Short Form Revised [WAI-SR]), and symptoms (Symptom Checklist Short Form [SCL-K11]) every fifth session of long-term treatment for up to 60 sessions. When investigated in separate models, both more agency and a stronger therapeutic bond predicted symptom improvement. However, within-person changes and betweenperson differences in agency predicted symptom improvement over and above the effects of therapeutic bond. Multilevel response surface analysis was used to further investigate the interplay between agency and therapeutic bond. When both agency and therapeutic bond levels were high, symptoms improved the most. When agency and therapeutic bond ratings differed, symptom ratings were significantly lower when agency exceeded therapeutic bond levels than when therapeutic bond ratings exceeded agency. Findings suggest that both agency and therapeutic bond are important treatment factors, but outcome could be improved when a strong therapeutic bond is combined with an equally strong sense of agency that empowers patients to pursue changes in their lives. When a strong therapeutic bond is present, but the patient feels less agentic, therapists may want to foster agency to improve outcomes. Public Significance StatementThis study suggests that both agency and therapeutic bond are relevant predictors of psychotherapy outcome. Ideally, patients should both experience a strong bond with their therapist and also feel empowered to influence their own therapeutic process as needed. When agency and therapeutic bond levels are vastly different, therapists should intervene especially in cases of lower agency.

Therapist and client perceptions of the working alliance: Codevelopment, linear growth, variability, and client functioning.

“…However, 45 sessions are not unusual for therapy conducted in some European countries. For example, a recent article published in the Journal of Counseling Psychology (Jennissen et al, 2022) had an average of 44.9 ( SD = 15.9) sessions. Many sessions per client can be considered a strength because DSEM analyses perform better with a larger number of time points (sessions).…”

Changes in meaning in life, working alliance, and outcome in psychodynamic psychotherapy: What leads to what?

“…If MIL and client outcome were assessed every session, as was the working alliance, powerful statistical models like dynamic structural equation modeling could be used to examine session-to-session relationships among these variables. We note that other peer-reviewed researchers examining the relationship between working alliance and outcome have also assessed these variables periodically (e.g., Jennissen et al, 2022 assessed alliance and outcome every 5th session).…”