Objective: Since there are discrepant findings in the literature, we studie
d the longer-term course in a large sample of 103 DSM-IV anorexia nervosa (
AN) patients. Method: Assessments were made at four points of time: beginni
ng of therapy, end of therapy, 2-year follow-up, and 6-year follow-up. Self
-rating scales as well as expert-rating interview data were used. Eating di
sorder-specific and general psychopathology were assessed. These data were
also compared with data on the 6-year course of patients with bulimia nervo
sa and binge earing disorder, respectively, who were treated at the same in
stitution at about the same time. Results: The participation rate at the tw
o follow-ups was high (97.9% of those alive). The general pattern of result
s over time of those alive at 6-year follow-up was as follows: substantial
improvement during therapy, moderate tin many instances nonsignificant) dec
line during the first 2 years posttreatment, and further improvement from 3
to 6 years posttreatment. At the time of the 6-year follow-up, 26.8% had A
N, 9.9% had bulimia nervosa-purging type (BN-P), 2.0% were classified as ea
ting disorder not otherwise specified (ED-NOS) all diagnosed according to D
SM-IV criteria; more than one half (55.4%) showed no major DSM-IV eating di
sorder. Based on an operationalized global outcome score at 6-year followup
, 34.7% had a good outcome, 38.6% an intermediate outcome, 20.8% a poor out
come, and 6 of 101 persons (5.9%) were deceased. Body mass index was 17.9 /- 2.8 at the 6-year follow-up,. amenorrhea was still found in 23.9%. Discu
ssion: In comparison to samples with bulimia nervosa or binge eating disord
er, the 6-year course of anorexia nervosa was less favorable. Mortality was
rather high and symptomatic recovery protracted; predictors of unfavorable
6-year course were the presence of binges during 4 weeks before index trea
tment, psychiatric comorbidity, and low body weight at discharge from index
treatment. (C) 1999 by John Wiley & Sons, Inc.