Childhood sexual abuse and adolescent eating disorder : findings from the Victorian Adolescent Health Cohort study

Objective: To examine the relationship between childhood sexual abuse (CSA) before the age of 16 years and later onset of bulimia and anorexia nervosa symptoms in females.

Design: A longitudinal cohort study of adolescents observed from August 1992 to March 2003. The cohort was defined in a 2-stage cluster sample using 44 Australian schools in Victoria.

Setting: Population based.

Participants: A total of 1936 persons participated at least once and survived to the age of 24 years, including 999 females. The mean (SD) age of females at the start of follow- up was 14.91 (0.39) years; and at completion, 24.03 (0.55) years.

Main Exposure: Self-reported CSA before the age of 16 years was ascertained retrospectively at the age of 24 years.

Outcome Measures: Incident Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition)–defined partial syndromes of anorexia and bulimia nervosa were identified between waves 4 (mean age, 16.3 years) and 6 (mean age, 17.4 years) using the Branched Eating Disorder Test.

Results: The incidence of bulimic syndrome during adolescence was 2.5 (95% confidence interval, 0.80-8.0) times higher among those who reported 1 episode of CSA and 4.9 (95% confidence interval, 1.9-12.7) times higher among those who reported 2 or more episodes of CSA, compared with females reporting no episodes, adjusted for age and background factors. The association persisted after adjusting for possible confounders or mediators measured 6 months earlier, including psychiatric morbidity and dieting behavior. There was little evidence of an association between CSA and partial syndromes of incident anorexia nervosa.

Conclusion: Childhood sexual abuse seems to be a risk factor for the development of bulimic syndromes, not necessarily mediated by psychiatric morbidity or severe dieting.