New perspectives on the therapeutic alliance. Therapist personality and the moderating role of attachment representations

A growing body of research highlights that certain therapists’ personal characteristics could positively or negatively influence the quality of the therapeutic alliance (Ackerman & Hilsenroth, 2001, 2003). Preliminary evidences suggest that more securely attached clinicians form stronger alliances with their patients, but more systematic studies are needed (Degnan et al., 2016). Furthermore, the therapists’ personality remains an under-researched area despite its clinical relevance. This study aimed to investigate the relationship between therapists’ personality characteristics and the bond, goals, and tasks of the working alliance (Bordin, 1979), and to explore the moderating role of therapists’ attachment in the relationship between personality functioning and overall therapeutic alliance. Fifty-five psychodynamic therapists were interviewed with the Adult Attachment Interview (George et al., 1985) followed by the Clinical Diagnostic Interview (Westen & Muderrisoglu, 2003) to assess their personality with the Shedler-Westen Assessment Procedure-200 (Westen & Shedler, 1999). They were also asked to complete the Working Alliance Inventory-T (Horvath & Greenberg, 1989) on a patient currently in treatment. In order to compensate the underrepresentation of specific AAI classifications, a well-established dimensional approach (Waters et al., 2007) has been employed. Findings showed that the bond, goal, and task components were positively associated with therapists’ healthy personality functioning, and negatively related to SWAP-200 scales characterized by emotional dysregulation or interpersonal detachment. Furthermore, the relationship between therapists’ personality functioning and overall working alliance was moderated by higher level of attachment security. These findings promote a better understanding of the “therapist effects”, one of the most important emerging topics in psychotherapy research (Baldwin & Imel, 2013).