Full personal, weight, and eating histories were taken of 92 women who entered a treatment study of bulimia nervosa. These findings are reported and compared with those described in other research projects and in large-scale postal surveys.
The results of a controlled trial of different therapies on bulimia nervosa have been described elsewhere (Freeman, Barry, Turnbull, & Henderson 1988). In this paper the characteristics of the subjects in that study are compared with those given in three other experimental studies, three descriptive reports, and two postal surveys. These were chosen on the basis of being conducted on a reasonable number of subjects (n > 15) with well-defined inclusion criteria and sufficient data for comparison.
As Fairburn and Garner (1986) have indicated, discrepancies in research studies of bulimia have been complicated by the use of different sets of diagnostic criteria. Diagnoses of bulimia (APA, 1980) and Bulimia Nervosa (Russell, 1979) embrace persons with a broadly similar eating disorder, but the latter diagnosis necessitates a previous history of actual or cryptic anorexia nervosa. In the absence of any firm agreement as to diagnostic criteria, most researchers have used DSM-TI1 [American Psychiatric Association (APA), 19801 with additional restrictions designed to tighten up what is widely acknowledged to be a rather loose definition. The proposed DSM-III-R criteria for Bulimia (APA, 1985) may resolve this difficulty.