Eating habits and attitudes among mothers of children with feeding disorders

BMJ 1995; 310 doi: (Published 28 January 1995) Cite this as: BMJ 1995;310:228
  1. Alan Stein, senior research fellowa,
  2. Jennifer Stein, visiting graduate studenta,
  3. Elizabeth A Walters, consultantb,
  4. Christopher G Fairburn, Wellcome Trust senior lecturerc
  1. a Section of Child and Adolescent Psychiatry, University of Oxford Park Hospital for Children, Oxford OX3 7LQ
  2. b Park Hospital for Children, Oxford OX3 7LQ
  3. c Department of Psychiatry, University of Oxford Warneford Hospital, Oxford OX3 7JX
  1. Correspondence to: Professor Alan Stein, Leopold Muller Department of Child and Family Mental Health, Royal Free Hospital School of Medicine and Tavistock Clinic, London NW3 2PF.
  • Accepted 25 November 1994

Feeding disturbances in young children are common.1 Furthermore, eating disorders are a substantial source of psychiatric morbidity among women of childbearing age.2 There has been no published research examining whether feeding disturbances of clinical severity in children are related to abnormal eating habits and attitudes among their mothers—though there is evidence that parents with eating disorders can adversely influence their young children's development, including feeding.3 If there were such a relation it would have important implications for the management of such children.

Subjects, methods, and results

Over 12 months the mothers of all children with ICD-10 (International Classification of Diseases, 10th revision)4 diagnoses of a feeding disorder who were referred to local child psychiatric clinics covering one geographical area were identified (n=32). ICD-10 criteria require that a feeding disorder in infancy or childhood “generally involves refusal of food and extreme faddiness.”4 One mother declined to participate and one general practitioner requested that a family should not be approached. Thirty index families were therefore available for study. Each of the 30 children was matched with the next child of the same sex and age referred to the clinic with a behavioural disorder but not a primary eating or feeding problem (n=30). The children's ages ranged from 2 to 12 years (feeding disorder group mean 7.4 (SD 3.6) years; behavioural disorder group mean 7.3 (3.0) years). There were 18 boys and 12 girls in both the index and control groups. All mothers were asked to complete the eating disorder examination-questionnaire.5 This is a standardised self report questionnaire based on a widely used interview and generates five subscales: dietary restraint, bulimia, concern about eating, concern about body shape, and concern about weight.5 Wilcoxon signed rank tests for matched pairs were carried out to compare the two groups (see table for means).

Mothers' eating disorder examination-questionnaire subscale scores (values are means (SD))

View this table:

The index group of mothers scored significantly higher on all five subscales of the questionnaire: dietary restraint (z=2.71; P<0.01), bulimia (z=2.93; P<0.01), concern about eating (z=3.11; P<0.01), concern about body shape (z=2.61; P<0.01), and concern about weight (z=2.92; P<0.01). A further question was raised about whether the mothers of clinically referred children differed from community samples. In order to examine this, data were extracted from a study in which the eating disorder examination-questionnaire was administered to a community sample of 243 women of childbearing age.5 The 108 mothers in that sample were included for this analysis. One way analysis of variance was conducted across the three groups, followed by a Tukey test to examine which groups were different (table). There were no significant differences between the mothers of the behaviourally disturbed children and the community mothers on any measure. In contrast, the mothers of children with feeding disorders scored significantly higher than community mothers on all subscales except concern about body shape.


These results provide evidence that feeding disturbances in children are specifically associated with distrubed eating habits and attitudes among mothers: mothers of the children with feeding disturbances showed significant disturbances in eating habits and attitudes whereas the mothers of behaviourally disturbed children were not different from community controls. Though causal relations cannot be determined from a cross sectional study, it is possible that abnormal attitudes to body shape and weight and disturbed eating habits among mothers may play a role in feeding disturbances in children. Further research is needed. But in the mean time, when children present with feeding disturbances there should be careful and sensitive questioning about the parents' own eating habits and attitudes.

We are grateful to the families who participated; to Sandra Cooper, Marianne O'Connor, and Sarah Beglin for help; to Richard Hooper for statistical advice; and to Peter Cooper for helpful suggestions. The research was funded by the Wellcome Trust (grant No 035035). AS was supported by the Wellcome Trust (grant No 035035) and the Tedworth Trust, JS held a Rotary Foundation scholarship, and CF is a Wellcome Trust senior lecturer (grant No 013123).


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract