Patient Sex as Moderator of Effects of Transference Interpretation in a Randomized Controlled Study of Dynamic Psychotherapy

Article excerpt

Objective: To examine whether men and women respond differentially to brief dynamic psychotherapy, with or without transference interpretations.

Method: Data from the First Experimental Study of Transference Interpretation were used. Patients (n = 100) were randomized to receive 2 different dynamic psychotherapies during 1 year, with either a moderate level of transference interpretations or no transference interpretations. We used the following outcome measures: the Psychodynamic Functioning Scales, Inventory of Interpersonal Problems-Circumplex Version, Global Assessment of Functioning (GAF), and total mean score of Symptom Checklist-90-Revised (Global Severity Index [GSI]). Change was assessed using linear-mixed models.

Results: On average, men and women responded equally across treatments. The moderator analyses, using the 2 secondary outcome measures, GAF and GSI, demonstrated that women responded better to therapy with transference interpretations, compared with men, whereas men responded better to therapy without transference interpretations, compared with women. When the moderator sex was combined with the moderator quality of object relations (QOR), a strong effect emerged: men with high QOR showed a large negative effect of transference interpretations, and women with low QOR showed a large positive effect.

Conclusions: In terms of symptom change, women responded better to therapy with transference interpretations, while men responded better to therapy without transference interpretations. Patient sex showed moderator effects over and above the moderator effects of QOR.

Clinical Trial Registration Number: NCT00423462

Can J Psychiatry. 2009;54(2)78-86.

Clinical Implications

* This study indicates that women and men responded differentially to transference interpretations in brief dynamic psychotherapy.

* Women responded significantly better than men to transference interpretations in dynamic psychotherapy.

* This study may contribute in the development of more targeted treatment interventions for men and women.


* We do not know whether our results can be generalized to long-term psychotherapy or are sustained after psychotherapy.

* The study was not large enough for precise estimates of effect size.

* Specifically trained psychotherapists are good for providing internal validity, but less than optimal for generalizing to standard practice.

Key Words: sex, gender, dynamic psychotherapy, transference

Interpretation of transference is one of the key components of dynamic psychotherapy. However, which patients are best suited for this technique has been debated. Mainstream clinical thinking has maintained that transference interpretations are anxiety-provoking and should therefore be used only for less disturbed and suitable patients in brief dynamic psychotherapy,1-2 although empirical research on this is contradictory.3 In the FEST study,4 patients (n = 100) were randomized to receive brief dynamic psychotherapy with either a moderate level of transference interpretations or without transference interpretations. There was no difference in the effectiveness of the 2 treatments; for the typical (average patient) me 2 treatments were equally effective on all 4 outcome measures. However, in patients with a history of less mature object relations, brief dynamic psychotherapy with transference interpretations was more beneficial than therapy without such interpretations. For patients with mature object relations, there were small negative effects of transference interpretations.4

Patient variables that influence treatment outcome are either predictors or moderators of outcome.5-8 Both predictors and moderators are pre-treatment variables that affect the strength or direction of a treatment response. Predictors do so regardless of treatment condition. Moderators, however, differentially influence outcome depending on treatment condition. …


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.