Pergamon 0005-7967(94)00070-0 Behav. Res. Ther. Vol. 33, No. 4, pp. 363 367, 1995 Copyright ~-~ 1995 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0005-7967/95 $9.50 + 0.00 A PILOT STUDY OF A RANDOMISED TRIAL OF COGNITIVE ANALYTICAL THERAPY VS EDUCATIONAL BEHAVIORAL THERAPY FOR ADULT ANOREXIA NERVOSA JANET TREASURE, *j GILL TODD, I MARTIN BROLLY, t JANE TILLER, t ANNIE NEHMED 2 and FRANCESCA DENMAN 2 ~Eating Disorders Clinic, Maudsley Hospital and Institute of Psychiatry, De Crespigny Park, London SE5 8AF and 2Guys Hospital, St. Thomas St, London SE1 9RT, England (Received 29 June 1994) Summary--The aim of this study was to compare two forms of outpatient treatment, educational behavioural treatment and cognitive analytical therapy for adult anorexia nervosa. Thirty patients were randomly allocated to the two treatments. At one year, the group had gained 6.8 kg, 19/30 (63%) had a good or intermediate recovery in terms of nutritional outcome. The group given cognitive analytical treatment reported significantly greater subjective improvement but there were no differences in other outcome parameters. In conclusion outpatient treatment of adult onset anorexia nervosa leads to an improvement in two thirds of cases. Larger studies will be needed to determine the most effective form of treatment in this group. INTRODUCTION Great advances have been made in the treatment of bulimia nervosa (Freeman, Barry, Dunkeld-Turnbull & Henderson, 1988; Fairburn, Jones, Peveler, Carr, Solomon, O'Connor, Burton & Hope, 1991). In contrast the development and evaluation of psychological treatments for anorexia nervosa is still in its infancy (Fairburn, 1990). The traditional form of treatment for anorexia nervosa has involved skilled nursing in specialised inpatient units (Russell, 1970; Garfinkel, 1986). Unfortunately the relapse rate after discharge is high; up to a third fail to show any improvement at follow-up (Garfinkel & Garner, 1982). Relapse can be minimised by specific psychotherapeutic interventions after discharge (Russell, Szmukler, Dare & Eisler, 1987; Dare, 1990). In 1983 Morgan and colleagues provocatively questioned the place of inpatient treatment: "whilst admission to hospital might make the situation safe for a while, especially when weight is very low or there is a suicidal risk, it can also involve considerable disruption in the patients management: it may represent counterproductive retreat from confron- tation with certain life difficulties and signify confirmation of the sick role in the eyes of relatives who then dissociate themselves from active participation in therapy. It is not always a major therapeutic step forward to admit a patient to a hospital ward, and our findings suggest that criteria for hospital admission in anorexia nervosa should always be scrutinized carefully." The study of Crisp, Norton, Gowers, Halek, Bowyer, Yeldham, Levett & Bhat (1991) was designed to test Morgan's challenging statement. It compared inpatient and outpatient treatments with a no-treatment option. The design of the study posed practical difficulties particularly with compliance and recruitment (Gowers, Norton, Yeldham, Bowyer, Levett, Heavey, Bhat & Crisp, 1989). In the short term, after one year, no differences were found between the treatments. In particular inpatient treatment did not appear to result in advantages. Outpatient treatment has proved to be effective in several studies (Hall & Crisp, 1987; Channon, De Silva, Hemsley & Perkins, 1989; Le Grange, Eizler, Dare & Russell, 1992). Most of these have involved young patients with a short duration of illness, the group with the best prognosis. It is uncertain whether outpatient treatment can be as effective in older, more chronic patients. These *Author for correspondence. 363