Information from eating disorder clinics across five continents suggests that anorexia nervosa is becoming an increasing problem in children and young adolescents. There is some indication that anxiety disorders in childhood may be a major risk factor for the development of anorexia nervosa. Early recognition and family treatment for this disorder are essential to prevent chronic impairment.
Anorexia nervosa developing in early adolescence was well documented in the case of Princess Margaret of Hungary, who lived and died in the 13th century  She was the daughter of King Bcla IV, who had her enter a Dominican convent during her early childhood. Her history comes from a complete copy of depositions by witnesses who gave evidence in the process of her beatification, which began less than 5 years after her death. Her eating behaviors were indistinguishable from those of young anorexia nervosa patients of today. Although there is documentation of fasting female saints in the middle ages,  the fasting did not appear to occur during childhood.
Obtaining the precise information to answer the question as to whether anorexia nervosa is an increasing problem in children and adolescents requires population-based interview data ascertaining the prevalence of anorexia nervosa, with age-of-onset distribution for different time cohorts. This data is simply not available. Studies of changing rates of anorexia nervosa published in the recent literature are limited to specific populations, have small sample sizes, or are based on questionnaires rather than personal interviews. Age of onset is presented as a mean statistic, rather than the number of cases with a specific age of onset. Table I summarizes the more recent published rates of anorexia nervosa. It should be noted that the studies from England  and Brazil  reported the greatest incidence and prevalence in females from age 10 through 19 or 10 through 13, respectively. In Singapore,  there was an increase in adolescents with anorexia nervosa admitted to a clinic over the years 1994 to 2002. Another study conducted in New South Wales, Australia  concluded that there was an increasing prevalence of anorexia nervosa in a younger age group. A questionnaire study carried out in South Australia concluded that there was a decrease in strict dieting between the years of 1995 and 2005 in the age group of 15 through 65.  A Finnish twin study of birth cohorts between 1975 and 1979 found a rather low incidence of anorexia nervosa (0.27%) for ages 15 to 19. 
A more specific documentation of pre- and early adolescent cases of anorexia nervosa admitted to an eating disorder treatment program (Halmi et al, unpublished data) is presented in Table II. Overall, it seems reasonable to form the opinion from these studies across four continents that anorexia nervosa is an increasing problem in children and adolescents.
Prepubertal and early adolescent onset of anorexia nervosa may be increasing; however, there are not sufficient cases with adequate samples to assess common risk factors. There is a suggestion that childhood anxiety may be a liability for developing anorexia nervosa. In a genetic study of over 600 women, 39% of women with a diagnosis of anorexia nervosa reported a history of overanxious disorder of childhood, and of those 94% met criteria for this disorder before meeting criteria for anorexia nervosa.  Although overanxious disorder of childhood is no longer a DSM-IV diagnosis, it was not only associated with the development of anorexia nervosa in this study, but also associated with the presence of additional anxiety disorders, such as generalized anxiety disorder, obsessive-compulsive disorder, specific phobia, social phobia, and panic disorder. 'Ihc authors concluded that childhood overanxious disorder is pernicious and associated with greater severity and longer duration of anorexia nervosa. 
Growth chart trajectories of co-occurring symptomatology were examined in a large community sample of adolescent females ranging in age from 12 to 15 years, with annual assessments over a 5-year period. In this study, initial depression predicted increases in eating and substance abuse symptoms, and initial eating disorder symptoms predicted increases in substance abuse problems.  Ihis study showed that depressive, eating, antisocial, and substance abuse symptoms operated differently as risk factors for one another, and thus the authors suggested that there may be reliable temporal sequencing of cooccurring forms of psychopathology. Therefore, co-occurrence of these symptoms may be due partially to the fact that over time certain symptom domains increase the risk of symptom growth in other domains.
There is substantial evidence that dieting is a major risk factor in the development of anorexia nervosa.  Dieting practices are now an aid to self-presentation, because consumerism and the mass market have blurred the exterior marks of social distinction (status) and personal difference (identity), according to the sociologist Turner.  This effect may be extending to 9- and 10-year-old children.
The recognition of pre- and early adolescent anorexia nervosa has directed a focus on family therapy for treatment of this disorder. Nonetheless, the more seriously ill anorexic patients continue to need a period of hospitalization. Over the past two decades, hospital treatment for eating disorders has changed from a long-term treatment, of the disorder to stabilization of acute episodes.
A specific example from the Westchester Division of the New York Presbyterian Hospital is shown in
An assessment during the next decade of the effect of these readmissions on the more seriously ill anorexia nervosa patients is crucial. It is very likely that there will be an increase in morbidity and mortality rates for pre-and early adolescent, onset patients with anorexia nervosa. This may be prevented with adequate length of hospitalizations; ie, discharge at BMI >19, and early diagnosis with specific family therapy for anorexia nervosa.