226 Eating Disorders
It is clear from this that the definition of bulimia nervosa is more ambigu-
ous than that for anorexia and more open to interpretation based on an
understanding of social norms. For example, what constitutes a binge to one
person may not to another if they have a different sense of what is normal.
Who has it?Many individuals who would fulfill the criteria for bulimia nervosa con-
duct their bulimic behavior in private and maintain apparently normal and
functioning lives. They therefore do not come into contact with health pro-
fessionals. Further, the course of bulimia is varied, with different symptoms
peaking at different times. In addition there are no clear cutoff points for
the severity of different symptoms, and definitions of the core clinical
symptoms are dependent upon what is considered normal. Estimating the
prevalence of bulimia nervosa is, therefore, problematic. Some popular lit-
erature claims that the majority of middle-class women are bulimic (Wolf,
1990). Similarly, studies using self-report measures among female student
populations have reported the prevalence of bulimic episodes as being as
high as 60 percent (Halmi, Falk, and Schwartz, 1981; Hart and Ollendick,
1985). Prevalence rates of bulimia nervosa as defined by DSM criteria have
also been high. For example, Johnson, Lewis, and Hagman (1984) reported
a rate of 8 percent among female American high school students; Pope,
Hudson, and Yurgelun-Todd (1984) reported a rate of 14 percent; Hart and
Ollendick (1985) reported a rate of 17 percent; and Halmi, Falk, and Schwartz
(1981) reported a rate of 19 percent. More conservative estimates, how-
ever, are shown by community-based studies. For example, Pyle et al. (1983)
used a questionnaire based upon DSM-III criteria and concluded that
4.5 percent of their large group of US female college students fulfilled the
criteria for bulimia nervosa. In a similar vein, Cooper and Fairburn (1983)
carried out a questionnaire survey of women attending a family planning
clinic in southern England. They reported that 20.9 percent reported bulimic
episodes as defined as episodes of “uncontrollable excessive eating” but that
only 1.9 percent fulfilled the criteria for “probable bulimia nervosa” as defined
as bulimic episodes, self-induced vomiting, and a morbid fear of fatness.
This study was replicated in 1987 (Cooper, Charnock, and Taylor, 1987)
and reported a comparable rate of 1.8 percent. In general, the prevalence
of bulimia nervosa is accepted as being between 1 and 2 percent of the female
population, with the peak age of onset being between 15 and 19 years. It
is therefore twice as common as anorexia nervosa.