The applicability of the Eating Disorder Inventory in pregnancy

PURPOSE: The aim of our study was validating Eating Disorder Inventory (EDI) among pregnant women, who are vulnerable to eating disorders (EDs). METHODS: In 2012–2013, 1146 women (aged 18–47 years) completed a questionnaire including EDI during the first 3 days after delivery. We checked factorial validity of three diagnostic subscales of EDI with confirmative factor analysis and internal validity by Cronbach’s alpha and item-total correlation. We also tested discriminative validity by comparing average of the three subscale of EDI in case of ED and non-ED groups. RESULTS: When applying the EDI to pregnant women, it seems necessary to exclude five items on three diagnostic subscales: on the Drive for Thinness subscale, 4 items remain (out of 7); on the Bulimia subscale, 6 items remain (out of 7); the Body Dissatisfaction subscale decreases from 9 to 8 items. Cronbach’s alpha and item-total correlation values meet the requirements defined by Garner et al. The internal consistency of the EDI has proved to be appropriate, indicating that it is a reliable screening tool. CONCLUSIONS: Thinking, attitudes, and behaviors connected to eating, along with the relation to altering body weight change during pregnancy. Vomiting usually accompanies pregnancy; body weight gain within wide limits is also regarded as normal during pregnancy. These behaviors and changes are not feasible to use for measuring ED symptoms. These aspects cannot be neglected when screening eating disorders in pregnant women. LEVEL OF EVIDENCE: Level IV evidence obtained from multiple time series with or without an intervention.