Eating disorders in adolescent girls with insulin-dependent diabetes mellitus: a population-based case-control study

In this study the prevalence of eating disorders in a population-based cohort of 89 female patients with type 1 diabetes, 14-18 y of age, was compared with that in age-matched healthy controls. Of all diabetic girls in the study area, 92% participated in the study. The majority were treated with multiple insulin injections and the mean HbA1c of the participants was 8.4%. On average, diabetic girls were 6.8 kg heavier than the controls. A two-stage design was used. The first consisted of a validated self-report questionnaire, the Eating Disorder Inventory (EDI). Girls who had high scores were then interviewed about eating habits and mental health using a semistructured interview, the BAB-T (Assessment of Anorexia-Bulimia - Teenager version). No cases of anorexia or bulimia nervosa were found, but 15 diabetic patients (16.9%) compared with 2 control girls (2.2%), p<0.01, had disturbed eating behaviour according to the questionnaire. In 6 of these 15 diabetic girls an eating disorder was confirmed at the interview, mainly binge eating and self-induced vomiting. None of the control girls showed an eating disorder. Overweight diabetic girls scored higher on EDI than non-overweight diabetic girls (chi2 = 4.9; p = 0.038). No relationships were found between EDI scores and metabolic control (HbA1c), dose of insulin, frequency of hypoglycaemia or diabetic ketoacidosis.