Effectiveness of psychoanalytic psychotherapy for children and adolescents with severe anxiety, depressive, and disruptive psychopathology in a naturalistic treatment setting
Weitkamp, K., Daniels, J. K., Hofmann, H., Timmermann, H., Romer, G., & Wiegand-Grefe, S. (2014). Psychoanalytic psychotherapy for children and adolescents with severe depressive psychopathology – preliminary results of an effectiveness trial. Psychotherapy, 51, 138-147.
This partly waitlist-controlled field study aimed to evaluate the effectiveness of psychoanalytic short and long term psychotherapy for children and adolescents employing a prospective design.
231 children and adolescents (aged 4 to 21 years) and their parents who entered psychoanalytic therapy in private practices in northern Germany participated in this ongoing study (154 intervention group, 23 wait-list control and intervention group, and 54 wait-list control group). Data was collected from therapists, parents, and from the patients (aged 10 years and older) at the beginning and the end of treatment, as well as up to 5 points in time during therapy. Follow-up took place at 6 and 12 months after therapy. Amongst other measures, depressive pathology was measured with the CDI, anxiety pathology with the SCARED, disruptive pathology with the external symptom score of the CBCL/YSR, and quality of life with the KIDSCREEN.
The patients received individual psychoanalytic psychotherapy which was predominantly child-focused, complemented by parent sessions usually on a ratio of 4:1. The interventions were based on Anna Freud (1949/1980) and object-relations theory as set out by Winnicott (1958/1988). The actual applied practice of psychoanalytic psychotherapy in children and adolescents was written down in a field manual (Baumeister-Duru, Hofmann, Timmermann & Wulf, 2013). Adherence to this code of practice was checked with a retrospective treatment fidelity checklist filled out by the therapists at the end of treatment for each patient.
Data analyses were carried out using intention-to-treat (ITT) analysis. Missing values were analysed and imputed with expectation maximation (EM). In addition to ANOVAs with repeated measures, mixed linear models were utilised to take into account the nested structure of the data, e. g. therapists treating more than one patient.
Overall, patients showed pronounced impairments at the commencement of outpatient therapy.
Depressive group 50 patients were included in the depressive group. Patients received, on average, 97 sessions of therapy (range: 25-205). At the end of therapy, there was a significant reduction in depression in the treatment group (parent report: d=.88; patient report d=.68). The wait-list control group, which received minimal treatment, displayed a slight, but not statistically significant, symptom improvement in the patient report (d=.07), but a significant improvement in the parent report (d=.49). 66% of the patients could be rated as recovered or improved. At the time of the publication follow-up was still being collected and hence, could not be published (Weitkamp, K., Daniels, J. K., Hofmann, H., Timmermann, H., Romer, G. & Wiegand-Grefe, S.,2014). Analyses with the completed data-set indicate stable results comparable to the anxiety and disruptive pathology.
Anxiety group The 76 anxiety patients received on average 94 therapy sessions (range: 8-300). Both, parents and patients in the intervention group reported moderate symptom improvements at the end of therapy (parent: d=.58; patient: d=.57), which are stable at the 1-year follow-up and increase from the patient perspective (parent: d=.37; patient: d=.80). When comparing the first therapy interval with the (minimal treatment) wait-list control group, both groups improved significantly with small effect sizes and no significant group differences. 69% of the patients could be rated as recovered or improved (Weitkamp, K., Daniels, J. K., Baumeister-Duru, A., Wulf, A., Romer, G., & Wiegand-Grefe, S., in prep.).
Disruptive group On average, the 65 intervention patients received 94.8 sessions (range: 19-300). Both, parents and patients in the intervention group reported moderate improvement of disruptive pathology at the end of therapy (parent: d=.69; patient: d=.63), which are stable at the 1-year follow-up (parent: d=.77; patient: d=.68). When comparing the first therapy interval with the (minimal treatment) wait-list control group, both groups improved significantly in the parent view on disruptive pathology with no significant group differences. The patients noted no significant differences neither in the first therapy interval nor the wait-list group (Weitkamp, K., Daniels, J. K., Daubmann, A., Romer, G., & Wiegand-Grefe, S., in prep.).
The results suggest that psychoanalytic therapy is successful in alleviating different kinds of pathology and improving quality of life for children and adolescents. These effects remain stable across one year follow-up. This naturalistic study is high on external validity with some limitations attached: the control interval had a much shorter duration than most therapies and these wait-list patientes received more often than not some supporting sessions. The therapy duration showed large variance. Lastly, the three symptom groups were not mutually exclusive, in a number of cases patients reported comorbid pathology.
Prof. Dr. Silke Wiegand-Grefe
Medical School Hamburg
Prof. Dr. Georg Romer