Anorexia nervosa and social class

(1) To examine social class status of female patients with anorexia nervosa presenting over a 33-year period; (2) to identify any differences in clinical features between the social classes.

Retrospective survey using comprehensive clinical database of patients referred to a national specialist center for the assessment and treatment of anorexia nervosa. Social class was defined using UK Registrar General's classification of father's occupation. Statistical methods included initial univariate analyses and subsequent ordinal logistic regression.

(1) Social class distribution was consistently weighted toward social classes 1/2. (2) Possible clinical indicators examined included low body weight, binge eating, and consequent weight-regulatory behaviors such as vomiting and laxative and diuretic misuse. These clinical features and their distribution proved to be similar across the social groups. Clinically rated quality of family relationships and types of family constellations were also consistent across the social classes. (3) Dieting prodromata and onset of the disorder occurred at younger ages in social classes 1/2. (4) A modest shift in social class distribution over time was apparent, with slightly more patients presenting post-1985 likely to come from lower social classes. Claims that the social class distribution is a product of referral patterns and acceptances are disputed.

We suggest that the social class bias reflects a sociocultural influence; a product of the disorder significantly often arising as an avoidant response to the conflict between social class-related family values/attitudes and adolescent turbulence within that family.