214 Eating Disorders
different periods in time. The surveys he selected were in south Sweden,
northeast Scotland, Switzerland, and Monroe County, New York, and
Rochester, Minnesota, in the US (Theander, 1970; Kendell et al., 1973; Jones
et al., 1980; Willi and Grossman, 1983; Szmukler et al., 1986; Willi, Giaco-
metti, and Limacher, 1990; Lucas et al., 1991). These results showed that in
Sweden, Scotland, Switzerland, and Monroe County from periods ranging
from the 1930s to the 1980s, the incidence of anorexia nervosa had changed
from less than one in a million per year in the 1930s to around 4 per 100,000
per year in the 1980s. Russell (1995) also considered the data from the
Rochester study (Lucas et al., 1991), which involved a painstaking examina-
tion of 40 years of records of cases presenting to hospital services with a
range of diagnoses with clinical similarities to anorexia nervosa. These results
showed no increase in the incidence of anorexia nervosa, and are particu-
larly important as the study covered the period before 1960, when incidence
rates were thought to be low, and included patients from a range of services.
However, when the data for females aged 15 to 24 were examined in isola-
tion, these results also showed an increased incidence between the periods
1935–1939 and 1980–1984 which was particularly marked after 1950.
Following this analysis Russell (1995) concluded that “there is persuasive
evidence in support of a rising incidence of anorexia nervosa from the 1950s
until the 1980s” (p. 11). This would seem to be particularly apparent in
the 15-to-24 age band. However, since this time the numbers of new cases
with AN appears to have stabilized, with evidence indicating no additional
change since the 1990s (Zachrisson et al., 2008).
The increased incidence in anorexia has not gone unchallenged and has
been called the “medical myth” by Williams and King (1987). They argued
that data illustrating an increase are problematic due to factors such as
deficiencies in the psychiatric case registers, changes in the referral rates,
differences in case definition, demographic changes within a given popula-
tion, and a rise in readmission rates. These factors would result in an increase
in new cases appearing in hospital units and old cases being recounted as
new. This challenge to the purported rise in anorexia nervosa has also been
supported by studies based on local and national case registers (Nielsen,
1990; Willi, Giacometti, and Limacher, 1990). Therefore, while much research
points to an increase in the occurrence of anorexia, this conclusion should be
treated with caution given the potential for many methodological problems.
The incidence of anorexia nervosa can be considered in terms of demo-
graphic factors including age, gender, social class, ethnicity, and sexuality.
The stereotype of the anorexic often involves an emaciated woman looking