When toddlers don’t want to eat properly – a new clinical definition for a long-standing problem

02.2026
Author Dr. Martin Claßen, Bremen

One frequent subject brought up in consultations at paediatric practices is parents’ concern about their children’s problematic eating habits. At first, it can be easy to assume that this is simply a bit of bad behaviour that is ultimately harmless. However, some children who refuse to eat can struggle to gain weight or risk getting insufficient variety in their diet. The previous classification systems, ICD-10 and DSM-4, offered no diagnosis for these children. The new DSM-5 and ICD-11 include the ARFID (avoidant/restrictive food intake disorder) condition in addition to anorexia nervosa and bulimia.

A recent large-scale study in Norway (Norwegian Mother, Father and Child Cohort Study – MoBa) investigated the prevalence, characteristics and genetic factors involved in ARFI in 35,751 children aged between 3 and 8 on the basis of the children’s health insurance and patient registry data, which the researchers were able to access.

The investigation showed that 2,129 (6.0%) of the children had ARFI symptoms at both ages 3 and 8, while 6,338 (17.7%) only had symptoms at the age of 3 and 3,001 (8.4%) only had symptoms at the age of 8. Children with ARFI symptoms at both ages were found to have significantly more frequent problems with development of language and motor skills, emotional disorders, hyperactivity and oppositional behaviour. They were also diagnosed much more frequently with mental health conditions, developmental delays, autism spectrum disorders, ADHD and epilepsy.

The genome-wide association analysis found two separate gene loci associated with ARFI symptoms at both ages. One of the genes, adenylate cyclase 3, localised on chromosome 2, correlated with the clinical diagnoses of ARFID at both ages 3 or 8 (z = 5.42; P = 3.03 × 10−8). This gene codes ADCY3 and also has an influence on depression and obesity.

Comment: Problematic eating habits in children are not exactly a new problem. The symptoms of ARFID (avoidant/restrictive food intake disorder) have now been described and defined as clear lack of interest in food, avoidance of food because of its sensory characteristics and anxiety about adverse consequences of eating, which manifest themselves through persistent failure to meet dietary and/or energy needs, associated with one (or more) of the following outcomes: 1. Significant weight loss (or failure to achieve expected weigh gain or growth disorders in children) 2. Significant nutrient deficiencies 3. Reliance on enteral feeding or oral dietary supplements It tends to affect toddlers, and differs from anorexia nervosa and bulimia in other respects.

Despite the high prevalence of ARFID, which is estimated to be between 2 and 6% here in Germany, little scientific or medical attention has so far been paid to this problem. So we can be grateful that this large-scale Norwegian study has shown that ARFID is more than just bad behaviour. Its associations with developmental problems, autism and ADHD mean it must be treated more seriously as a developmental disorder. The genetic findings, which were reinforced by other studies on twins, are also interesting in this regard.

Children with ARFI symptoms should be examined closely and given comprehensive treatment, including neuropaediatric care, while ARFID must be considered as a potential cause of failure to thrive. Development of diagnostic and treatment concepts for children with ARFID would also be welcome here in Germany.
 

Reference:
Bjørndal LD, Corfield EC, Hannigan LJ et al. Prevalence, Characteristics, and Genetic Architecture of Avoidant/Restrictive Food Intake Phenotypes. JAMA Pediatr. Published online November 24, 2025. DOI:10.1001/jamapediatrics.2025.4786