Eating Disorders 227The change in incidence and prevalence since its descriptionBulimia nervosa was first described in 1979. This raises the question “Has
there been an increase in the incidence of bulimia nervosa from before to after
the time of its description?” Two studies carried out since this time enable this
question to be answered. The first was by Kendler et al. (1991), who inter-
viewed 2,163 female subjects from a twin register in the US. Using DSM
criteria they allocated a definite or probable diagnosis of bulimia to 60 indi-
viduals to enable the calculation of the proportion of individuals who could
have received a diagnosis at any time in their life – the lifetime prevalence.
They then examined this prevalence according to the age of the patients.
This analysis showed that those born before 1950 were less likely to have
had bulimia than those born after 1960. These data also indicated that the
prevalence by the age of 25 for those born after 1959 was 4.6 times higher
than for those born before 1950. The older patients may have been more
likely to have forgotten their bulimia or to have been less aware of it due to
the absence of any publicity about it at the time. Alternatively, these results
could suggest that bulimia was very rare prior to its description in the late
1970s. This second explanation is supported by a study by Lucas and Soundy
(1993). This study involved a retrospective analysis of cases that fulfilled
the criteria for bulimia nervosa as recorded in the medical database for the
entire population of Rochester, Minnesota, in the US (about 60,000 in 1985).
The analysis revealed only a very small number of cases prior to 1979 but a
very sharp increase in cases after the description of bulimia nervosa became
formal. Russell (1995) reviewed the evidence for the change in incidence
of bulimia before and after his original description (see figure 10.3). He con-
cluded that although there may have been cases prior to this time, they were
very rare. He also stated that the increase in prevalence following this time could
be due to the recognition of clinical characteristics by health professionals.
Furthermore, Russell suggested that “the description of the disorder...
encouraged the appearance of the key behaviors in vulnerable young women
who consequently acquired the illness as if by contagion” (1995, p. 15).
The question also arises, “Has there been an increase in the prevalence
of bulimia since it was first described?” One study carried out in the US
examined the prevalence of bulimia in 1983 and 1986 and reported a three-
fold increase from 1 percent to 3.2 percent among female students when using
a strict criteria, involving bulimic episodes at least weekly, and self-induced
vomiting or laxative abuse (Pyle et al., 1983; Pyle and Mitchell, 1986). In
contrast to these results, Cooper and colleagues replicated their own 1983 study