The Frankfurt-Hamburg Long Term Therapy Study

Brockmann, J., Schlüter, T., & Eckert, J. (2006). Langzeitwirkungen psychoanalytischer und verhaltenstherapeutischer Langzeitpsychotherapien. Eine vergleichende Studie aus der Praxis niedergelassener Psychotherapeuten {Long-term effects of long-term psychoanalytic and long-term behavior therapy. - A comparative study from general practices of psychotherapists}. Psychotherapeut, 51, 15–25.
Jakobsen, T., Rudolf, G., Brockmann, J., Eckert, J., Huber, D., Klug, G., . . . Leichsenring, F. (2007). Ergebnisse analytischer Langzeitpsychotherapie bei spezifischen psychischen Störungen: Verbesserungen in der Symptomatik und in interpersonellen Beziehungen {Results of psychoanalytic long-term therapy in specific diagnostic groups: Improvement in symptoms and interpersonal relationships}. Zeitschrift für Psychosomatische Medizin und Psychotherapie, 53, 87-110.
Salzer, S., Leibing, E., Jakobsen, T., Rudolf, G., Brockmann, J., Eckert, J., . . . Leichsenring, F. (2010). Patterns of interpersonal problems and their improvement in depressive and anxious patients treated with psychoanalytic therapy. Bulletin of the Menninger Clinic, 74(4), 283-300.

Brief Summary

The prospective study compares 31 patients in long-term behavior therapy (CBT) with 31 patients in long-term psychoanalytic therapy (PA). A naturalistic design was applied within the German health system. All patients underwent a diagnostic interview (SCID) by an external interviewer. Only patients who fulfilled the DSM III-R criteria for a depression or an anxiety disorder were included in the study. Although the diagnosis of the patients undergoing long-term CBT and long-term PA were comparable, we found that they differed in various ways. (We did not randomise the patients.)

The differences arose in a number of characteristics. PA-patients were higher educated, use less psychotropic medication and had a lower strain of symptoms (SCL-90-R GSI: PA= 0.9 vs. BT=1.5. PA-patients also differed in the access to psychotherapy. They introduced themselves more to therapy (vs. recommended by professionals). The average duration of long-term CBT was 2.4 years and 64 sessions. The average duration of long-term PA was 3.6 years and 209 sessions. Symptoms (SCL-90-R) and interpersonal problems (IIP) were examined at the beginning, after 1 year, 2.5 years and 3.5 years and after 7 years.

Both groups showed significant degrees of improvement within the first 3.5 years and remained stabilized in the following 3.5 years regarding the symptomatic aspects. Focussing on the interpersonal problems, group PA showed further improvement after the 3.5 years period. The CBT group however couldn’t show any further improvement after 3.5 years but they stabilized.

After 3.5 years B -patients had a symptom strain GSI = 0.8 (SCL-90-R) with nearly the same results after 7 years. In other words: CBT patients ended up with a symptom strain with which PA-patients started therapy.

Consume of psychotropic medication was different after 7 years (11% of the PA-patients and 23% of the CBT patients). CBT patients had seen the continuing medication not as a failure of therapy.

Relapses (defined as statistic significant changes at 3.5 years that did not longer exist after 7 years) were low (both 19%) compared to short time therapy relapse rates for patients with depression.

31% of the PA patients and 12 % of the CBT patients have looked for further therapy. The difference between the groups is significant. We did not ask for the reasons why and what kind of therapy they had chosen.

Data of the study were used in other studies (Salzer et al. 2010, Jakobsen et al. 2007)


These results point out why comparison studies concerning the matter of therapy using parallelized samples cannot always match up to reality.

Most patients did profit from therapy and felt satisfied by the therapy, but the profit was different.

Different people choose or are recommended by professionals to different treatments. Treatment theory matters (Fonagy & Allison 2014).


Dr. Josef Brockmann

Egenolffstr. 29, Frankfurt/M Germany