Risk Factors for Full- and Partial-Syndrome Early Adolescent Eating Disorders: A Population-Based Pregnancy Cohort Study

Objective: To identify prospective predictors of eating disorders in a population-based sample of 14-year-old boys and girls, using previously collected antenatal, biomedical, familial, demographic, and psychosocial data. Method: Participants (N = 1,597) were drawn from the Western Australian Pregnancy Cohort (Raine) Study. Data were collected during pregnancy, at birth, and when children were aged 1, 2, 5, 8, 10, and 14 years. An adapted version of the Eating Disorder Examination Questionnaire was used to assess eating disorder symptoms at age 14 years. Logistic regression was used to identify prospective predictors of eating disorder caseness, relative to general control and psychiatric control groups. Results: At age 14 years, 6% of the sample met full or partial criteria for a DSM-IV eating disorder. Being female and being perceived as overweight by one's parent were the strongest predictors of eating disorder caseness in the final multivariate models, relative to both control groups. Maternal body mass index, social problems, low social-related self-efficacy, and neurocognitive difficulties were also predictive of eating disorder caseness relative to the general control group only. Conclusions: The results suggest that parent's perceptions of their child's weight are more powerful than objective child body weight in predicting the development of eating disorders. Parent-perceived child overweight was also a specific risk factor for eating disorders, whereas elevated maternal weight and childhood psychosocial difficulties seem to be associated with increased risk for psychiatric disturbance more generally. These results have implications for the prevention of eating disorders, particularly in light of recent increases in the prevalence of childhood obesity. J. Am. Acad. Child Adolesc. Psychiatry, 2009;48(8):800-809.