Corresponding Author Information: Jaakko Stenius

Session Abstract: BACKROUND AND PURPOSE: Better therapeutic alliances predict better treatment results, but little knowledge still exists on the patient characteristics that lead to better alliances.  Therefore, the current study investigated how alliance development is predicted by three measures that assess psychological functioning from different vantage points. We hypothesized that better performance in these measures would predict greater alliance development. SUBJECTS: The sample comprised 128 adult outpatients assigned to long-term psychodynamic psychotherapy (the mean duration of 31.3 months) and suffering from depressive or anxiety disorder. METHODS AND MATERIALS: A cohort study design with repeated measurements was used. The alliance development was measured using the Working Alliance Inventory (WAI) at baseline, and after 7-month, 24-month, and 36-month follow up. The three predictors were the Rorschach-based Ego Impairment Index (EII-2), the Wechsler Adult Intelligence Scale-Revised (WAIS-R), and the Defense Style Questionnaire (DSQ). ANALYSES: Primary "intention-to-treat" (ITT) analyses were performed and ignorable dropouts were assumed. Linear mixed models were used in the statistical analysis. Multivariate models used included the variables under examination and confounding factors. RESULTS: All the three measures showed some ability to predict the development of the alliance during long-term therapy. The WAIS-R was found to be the strongest independent predictor, as higher intelligence scores predicted favorable development of both the patient- and therapist-rated alliance. Greater improvement in the patient -rated alliance was predicted by lower DSQ values, indicating less use of immature defenses, and higher EII-2 values, indicating higher ego impairment. CONCLUSIONS: The findings support the value of pretreatment multi-method assessment for understanding psychological capacities and vulnerabilities that may impact therapeutic relationships and inform about patients' individual treatment needs. As patients with severe psychopathology were excluded, the results should not be generalized to these populations.  More multi-method assessment studies are needed to understand how therapeutic relationship evolve over time and the determinants of this evolution.


Jaakko Stenius | Niuvanniemi Hospital, Kuopio, Finland

Leave a Reply