Classification of Children and Adolescents With Avoidant/Restrictive Food Intake Disorder

Abstract

BACKGROUND AND OBJECTIVES: Evidence suggests that children and adolescents with avoidant/restrictive food intake disorder (ARFID) have heterogeneous clinical presentations. To use latent class analysis (LCA) and determine the frequency of various classes in pediatric patients with ARFID drawn from a 2-year surveillance study. METHODS: Cases were ascertained using the Canadian Pediatric Surveillance Program methodology from January 1, 2016, to December 31, 2017. An exploratory LCA was undertaken with latent class models ranging from 1 to 5 classes. RESULTS: Based on fit statistics and class interpretability, a 3-class model had the best fit: Acute Medical (AM), Lack of Appetite (LOA), and Sensory (S). The probability of being classified as AM, LOA, and S was 52%, 40.7%, and 6.9%, respectively. The AM class was distinct for increased likelihood of weight loss (92%), a shorter length of illness (<12 months) (66%), medical hospitalization (56%), and heart rate <60 beats per minute (31%). The LOA class was distinct for failure to gain weight (97%) and faltering growth (68%). The S class was distinct for avoiding certain foods (100%) and refusing to eat because of sensory characteristics of the food (100%). Using posterior probability assignments, a mixed group AM/LOA (n = 30; 14.5%) had characteristics of both AM and LOA classes. CONCLUSIONS: This LCA suggests that ARFID is a heterogeneous diagnosis with 3 distinct classes corresponding to the 3 subtypes described in the literature: AM, LOA, and S. The AM/LOA group had a mixed clinical presentation. Clinicians need to be aware of these different ARFID presentations because clinical and treatment needs will vary.

Subtypes of avoidant/restrictive food intake disorder in children and adolescents: a latent class analysis

“…Cases were reported through national surveillance units which survey most of the consultant paediatricians and child and adolescent psychiatrists in the UK and ROI, rather than from specialty eating disorder services 6 , 7 , 10 , 15 or paediatricians alone. 18 …”

“…However, the first two classes had similar patterns of response to the variable indicators used to define classes and a considerable number of individuals were assigned to more than one class with similar probability. The authors 18 proposed that these individuals belonged to a combination of those subgroups, suggesting the existence of a fourth subgroup which was not presented as a class. 18 Our study used six simple binary indicators of proposed drivers of food avoidance to empirically derive ARFID subgroups, which we suggest offers a more robust methodology.…”

“… 17 Katzman et al. 18 used LCA in a sample of children and adolescents with ARFID ascertained through active paediatric surveillance in Canada using a combination of indicator variables based on i) ARFID diagnostic criteria, ii) sign and symptoms, iii) hospitalizations, and iv) autism spectrum disorder (ASD) comorbidity. Results identified three subgroups: lack of appetite sensory sensitivity, and ‘acute medical’, characterised by medical hospitalisation.…”

Clinical features of adult patients with avoidant/restrictive food intake disorder presenting for medical stabilization: A descriptive study

“…Also, our study may not be applicable to adult patients who meet the ARFID criteria through supplement dependence alone as they may be able to maintain a normal body weight and be unbothered by restrictive eating patterns. Furthermore, most of the cohort were FOAC; therefore, findings may not generalize to other ARFID presentations, especially as LOI/SC have been reported to involve less weight/nutritional disruption in pediatric ARFID (Katzman et al, 2022).…”

Avoidant restrictive food intake disorder and pediatric feeding disorder: the pediatric gastroenterology perspective

“…The definition, while including sequelae of malnutrition, does not hold nutrition as a major element and tends to focus on the disorder as a behavioral problem that, in turn, may result in malnutrition and possible growth failure. Expert panels have proposed subtypes of ARFID, including those based on fear/anxiety, absent interest/appetite, and sensory problems; some publications have included those with more severe nutritional sequelae as a separate subcategory [6,7]. It is notable that these subtypes do not acknowledge the contribution from complex medical (e.g., gastroschisis, leukemia, sequelae of prematurity) or skill-based deficits (e.g., congenital airway anomalies, Down syndrome or other hypotonia, sequelae of cerebrovascular accident) to the feeding dysfunction and understandably view the feeding problem as a behaviorally driven dysfunction.…”