Background. A possible association between childhood feeding problems and maternal eating disorder has been suggested by a clinic-based self-report questionnaire study. A community study was conducted, using standardized psychiatric interviews, to investigate the strength and specificity of this putative association.

Methods. Four-year-old children were screened using a self-report version of the Behaviour Screening Questionnaire, completed by mothers, and the Pre-School Behaviour Checklist, completed by teachers. Three groups of children were identified for follow-up: children with feeding problems (N = 42), children with a non-feeding form of disturbance (i.e. shyness, fearfulness or behavioural disturbance; N = 79), and a random sample of children with no disturbance (N = 29). The presence of feeding problems was confirmed by assessment of a filmed family meal, with ratings made blind to child group and maternal mental state. Maternal current and past affective disorder and current and past eating disorder were systematically assessed, blind to child status, using the Anxiety Disorders Interview Schedule and the Eating Disorder Examination respectively.

Results. Compared with the mothers of the two comparison groups of children, the mothers of the children with feeding problems had no raised rate of any affective disorder, either current or past, but they did have a markedly raised rate of both current and past DSM-IV eating disorder. The odds ratio of maternal eating disorder for the children with feeding problems was significantly raised at 11·1 (CI 1·4–91·8).

Conclusion. There is a strong and specific association between childhood feeding problems and maternal eating disorder.