Editorial: Limitations on the Predictive Validity of the ARFID Diagnosis

Abstract

Avoidant/restrictive food intake disorder (ARFID) first entered the psychiatric nosology with the 2013 publication of DSM-5. Unlike binge eating disorder (BED), which was also new to DSM-5 but which had first been described by Stunkard in 1959,1,2 ARFID had never been described in the psychiatric literature as a single diagnostic entity. The new diagnosis encompassed clinical constructs that were previously proposed and studied but not described in DSM (ie, causes of "non-organic failure to thrive" including infantile anorexia and post-traumatic feeding disorder,3 and extreme food selectivity in children with autism spectrum disorder4) and the DSM-IV Feeding Disorder of Infancy and Early Childhood (FDIEC).5 The ARFID diagnosis supplanted FDIEC and incorporated earlier descriptions of pediatric feeding problems into a lifespan diagnosis for patients with restrictive eating characterized by food selectivity, poor appetite/lack of interest in eating, or fear of aversive consequences of eating that led to significant weight loss or failure to grow, nutritional deficiency, supplement dependence, and/or psychosocial impairment. Because the diagnosis was so new when DSM-5 was published, the ARFID criteria were not yet supported by descriptive psychopathology research in population-based or clinical samples. Kambanis etal.6 have made an important contribution to the descriptive psychopathology of ARFID by describing the naturalistic illness course over 2 years in a well-characterized adolescent and adult sample.6 In addition to providing novel information about the course of ARFID, findings from Kambanis etal.6 highlight and illustrate 3 limitations of the predictive validity of the current DSM-5-TR ARFID criteria.