Determining the influence of type 1 diabetes on two common eating disorder questionnaires

This research evaluated the level of influence that having type 1 diabetes (T1DM) has on responses to questions about food choices, eating concerns, dietary restraint, and others that are included on two widely used, validated eating disorder (ED) questionnaires and examined responses to these two questionnaires from patients with T1DM and an eating disorder (ED-T1DM) and an ED-no-diabetes.

An expert panel rated each item on the Eating Disorders Examination Questionnaire (EDE-Q) and Eating Disorders Inventory, version 3 (EDI-3) regarding T1DM level of influence on item interpretation. These questionnaires were completed by 2 matched samples (ED-T1DM, n = 48 and ED-no-diabetes, n = 96); responses were compared between the samples with particular attention to items of high T1DM influence.

The expert panel identified that 50% (19/38) of the items on the EDE-Q and 6.6% (6/91) on the EDI-3 could be highly influenced by having T1DM. Before Bonferroni correction, the 2 groups responded statistically different on 9 out of 38 items on the EDE-Q and 27 out of 91 items on the EDI-3; generally responses were healthier for those with ED-T1DM than ED-no-diabetes. Of these items, on the EDE-Q, 5 were rated high T1DM influence and on the EDI-3, 3 were rated high.

Having T1DM influences responses on ED questionnaires developed for the general population. This influence may be greater when questionnaires focus on eating, weight, and shape and result in misinterpretation of total and subscale scores by even well-trained clinicians. A careful review of individual item responses by the treatment team is warranted.