Client Attachment as a Predictor of Therapy Outcome and Premature Termination

The primary purpose of this study was to examine how clients’ self-reported adult attachment pattern and their attachment to the counselor are associated with working alliance and premature termination. A total of 65 clients at a large southeastern university counseling center were included in data analysis. Clients in this study completed survey packets including the Experiences in Close Relationships Scale (Brennan, Clark, & Shaver, 1998), the Working Alliance Inventory (Horvath & Greenberg, 1989), the Client Attachment to Therapist Scale (Mallinckrodt, Gantt, & Coble, 1995), the Outcome Questionnaire 45 and 30 items (Lambert et al. 1996), and the Therapeutic Distance Inventory (Mallinckrodt, 2011) at four different time points: (a) pretest, (b) after the 3rd session, (c) after the 5th session, and (d) at termination. The Therapeutic Distance scale is composed of four dimensions, Too Close, Too Distant, Growing Engagement, and Growing Autonomy. Results suggested that interactions between adult attachment (anxiety or avoidance) and therapeutic distance were not significantly associated with working alliance or premature termination. However, therapeutic distance subscales were correlated as direct effects with working alliance and premature termination. Other findings suggested adult attachment did not change over the course of therapy. The Client Attachment to Therapist (CATS) subscales at session 5 and at termination were significantly correlated with premature termination. In addition, working alliance at termination was significantly negatively associated with premature termination. Finally, the CATS-Avoidant-Fearful subscale at session 3 was associated with an increase in symptoms, and working alliance at session 3 was associated with a decrease in symptoms. Implications for theory, psychotherapy, and future research are discussed.