T. Herzog et A. Hartmann, PSYCHODYNAMICALLY ORIENTATED TREATMENT IN ANOREXIA-NERVOSA - OVERVIEWAND SYSTEMATIC REVIEW, PPmP. Psychotherapie, Psychosomatik, medizinische Psychologie, 47(9-10), 1997, pp. 299-315
Psychodynamic approaches to treatment in anorexia nervosa (AN) remain
influential. It is unclear, how a psychodynamic orientation influences
outcome with this patient group. To our knowledge, there exists no sy
stematic review (sensu Cochrane) of the subject. Methods: Extensive el
ectronic and hand searches. were conduced (updated in December 1995) t
o identify case reports and all longitudinal and comparison studies us
ing ''psychodynamic'' treatment approaches, published in English, Fren
ch and German. Case reports were examined for treatment ''components''
. Group studies meeting operationalised criteria were independently ra
ted for methodological quality, 12 aspects of treatment and its delive
ry (setting, type, orientation and focus) and therapy dose and duratio
n. Effect sizes were calculated for post treatment and follow-up. Mult
iple regression analysis was used to identify outcome predictors. Find
ings: Six treatment comparison studies and seven follow-up studies met
criteria. in most case reports and studies treatment included differe
nt settings and techniques and an explicit focus on the symptom. In th
e two studies directly comparing undifferentiated psychodynamic treatm
ent and ''treatment as usual'' resp. with psychodynamically based diso
rder specific treatments the latter had better results. Outpatient tre
atment was as successful as inpatient treatment (in somewhat different
populations). It was not possible to identify outcome predictors. Int
erpretation: ''Psychodynamic'' treatment in AN is more pragmatic and d
isorder orientated than generally acknowledged; patients seem to fare
better in specialized treatment programmes delivered by experienced te
ams; less cachectic patients can be treated as outpatients by a specia
lized team. The exact contribution to treatment outcome of the psychod
ynamic orientation remains to be delineated. There is a great need for
well designed and well presented treatment outcome studies.