P3.002: Are Adult Screeners for Eating Disorders Useful for Pediatrics?

Background: Clinical practice guidelines for pediatricians recommend routine screening for eating disorders. The AAP recommends screening for eating disorders during the annual visit or sports exams. However, there are no validated tools for the pediatric population. There are also no transdiagnostic tools which include avoidant restrictive food intake disorder (ARFID), the most common eating disorder in childhood. Because of a lack of a validated tool, screening measures developed in older populations are often used. We examined the performance of eating disorder screening surveys developed for use in adult populations in a pediatric sample.

Methods: Participants were recruited via CloudResearch Connect, an online platform that connects researchers with potential participants. Adult caregivers with a child between the ages of 6-12 were eligible. The study included completing an online survey and a virtual diagnostic interview to determine the presence or absence of a child eating disorder diagnosis. Survey completers were contacted approximately1-week later with the opportunity to complete the interview. Eating disorder diagnosis was determined using the Diagnostic Interview for Anxiety, Mood, and OCD and Related Neuropsychiatric Disorders- Child and Adolescent Version (DIAMOND-KID).18 As part of the survey battery, caregivers completed the SCOFF, which consists of five questions addressing the core features of anorexia and bulimia nervosa; a score of 2 indicates a likely diagnosis. The Eating Disorder Screen for Primary Care (ESP) was also completed, which is a five-item screener developed for adults in primary care. Two or more items endorsed indicate a possible eating disorder. Items were modified to change self-report to “my child.” This study was reviewed and approved by the BRANY IRB.

Results: N=1,760 caregivers completed the survey battery and 255 completed the interview. Parents/caregivers who completed the interview were, on average, age 39.4 (SD = 6.5), primarily White (72%), cisgender women (66%) and married (75%). Nearly a third of children whose caregivers completed the interview were diagnosed with an eating disorder on the DIAMOND-KID (n = 82; 32%). The SCOFF performed poorly, flagging less than 10% of those children identified as having an eating disorder. The ESP performed better, flagging 62% of children with an eating disorder; however, had a 46% false positive rate.

Conclusion: There is a critical need for a pediatrics eating disorder screening measure. We amplify this need by showing that other commonly used screeners developed in adults are not appropriate for use in children. A pediatric screening measure is especially important for young children given ARFID is the most common diagnosis at this age, yet no screeners exist that are transdiagnostic across all eating disorders.