Nibbles
The debate concerning the diagnostic classi®cation of eating disorders had an upswing some years ago (Beumont et al., 1994;Waller, 1993). During the last couple of years, however, it seems to have calmed down. Perhaps due to the DSM±IV coming into general use? A recent paper by van der Ham and colleagues (1997) now throws fresh wood on the fading ®re. In a longitudinal design and by way of sophisticated multivariate statistics they show that the only dimension actually differentiating among a group of adolescents categorized into four DSM±III±R eating disorder diagnoses (AN, AN/BN, BN hx AN, and ED NOS) is the bulimic behaviours (binge eating, vomiting, purging). Do I recall having heard this before? What's more, they ®nd that, as time passes by, the diagnostic groups get more and more alike in this dimension too, and after 4 years there are virtually no differences at all.
Another ®nding that immediately gives me a sense of recollection is the outcome ®gures for bulimia nervosa reported by Keel and Mitchell (1997). In an ambitious review of outcome studies for bulimia nervosa they ®nd that, at intermediate followup (5 to 10 years), about 50 per cent of the patients have recovered fully, while approximately 20 per cent continue to meet the full criteria for bulimia nervosa. In addition, there seems to be no reliable predictors for outcome. Treatment has a positive in¯uence on the course and outcome in the shorter perspective (6 months to 4 years), but subsequently it does not seem to have any substantial effect. Strikingly similar outcome ®ndings have repeatedly been reported for anorexia nervosa. Does this mean that the two disorders have similar natural courses across 5 to 10 years?
McKisack and Waller (1997) review the literature on group therapy for bulimia nervosa (including ®ve previous reviews). Their general ®ndings are neither surprising, nor encouraging. Group therapy is found to be generally but moderately effective, and more so if the duration or intensity of treatment is increased or if other concurrent treatments are added. It is also concluded that no speci®c theoretical orientation is clearly preferable, and that some patients are more suited for group therapy than others. What is discouraging about McKisack's and Waller's conclusions to me is that they seem to re¯ect the lack of speci®c knowledge in the entire ®eld of eating disorders treatment. One of the points they emphasize is the need to identify which patients will bene®t from group therapy (and its speci®c forms) and which patients are likely to drop out. It seems to me that this kind of knowledge (what type of treatment to offer to what patient under what circumstances) is one of the most needed if we are to improve the effectiveness not only of group therapy for bulimics, but of eating disorders treatment in general.
If I add together the conclusions of the three papers above they seem to indicate problems with our way of describing and subdividing eating disordered patients. Since this is an old interest of mine (see e.g.