Background. Twin studies have concluded that there is a substantial genetic contribution to the
aetiology of eating disorders. The aim of the present study was to estimate the genetic contribution
to the aetiology of self-reported eating disorders in a sample of representative twins.
Method. A population cohort of 34142 young Danish twins was screened for eating disorders by
a mailed questionnaire.
Results. Concordance rates differed significantly across monozygotic and dizygotic twin pairs for
broadly defined self-reported anorexia nervosa and bulimia nervosa. Heritability estimates of 0·48,
0·52 and 0·61 respectively were estimated for narrow and broad definitions of self-reported anorexia
nervosa and for self-reported bulimia nervosa.
Conclusions. There is a genetic contribution to the aetiology of self-reported eating disorders in the
general population. The relationship between self-reported and clinical eating disorder remains to