Introduction: Redefining Restrictive Eating in Pediatrics
Avoidant/Restrictive Food Intake Disorder (ARFID) represents a significant shift in how clinicians view restrictive eating behaviors. Unlike traditional eating disorders such as anorexia nervosa or bulimia nervosa, ARFID is characterized by an avoidance or restriction of food intake that is not driven by body image distortion or a desire for weight loss. Instead, it is typically motivated by sensory sensitivity, a lack of interest in eating, or fear of aversive consequences like choking. Despite its inclusion in the DSM-5, our understanding of its prevalence in the general population, its developmental trajectory, and its underlying biological mechanisms has remained limited. A groundbreaking study by Bjørndal and colleagues, published in JAMA Pediatrics, provides the most comprehensive look to date at the prevalence, characteristics, and genetic architecture of avoidant/restrictive food intake (ARFI) phenotypes.
The MoBa Study: Mapping the ARFI Landscape
This research utilized data from the Norwegian Mother, Father, and Child Cohort Study (MoBa), a massive population-based project following children born between 1999 and 2009. The researchers analyzed a sample of 35,751 children, using mother-reported data at ages 3 and 8 years, and linked this information with diagnostic data from national health registries.
To capture the full spectrum of the condition, the study defined two main tiers of the phenotype:
1. ARFI-broad: Children exhibiting a narrow range of food consumption or restricted eating patterns.
2. ARFI-clinical: A subset of the broad group who also displayed markers of clinical significance, such as nutritional deficiency, interference with psychosocial functioning, or the need for nutritional supplements.
Furthermore, the researchers categorized children based on the persistence of symptoms: transient (present only at age 3), emergent (present only at age 8), and persistent (present at both ages). This longitudinal approach allowed for a nuanced understanding of how restrictive eating evolves during early childhood.
Prevalence and Developmental Trajectories
The findings reveal that ARFI is far more common in the general pediatric population than previously estimated. Among the 35,751 children studied, the prevalence of the ‘ARFI-broad’ phenotype was substantial: 17.7% were transient, 8.4% were emergent, and 6.0% were persistent.
When applying stricter clinical criteria, the prevalence of ‘ARFI-clinical’ was 6.3% overall. Specifically, 3.2% were transient, 1.4% were emergent, and 1.8% were persistent. These figures suggest that while many children may experience phases of picky eating, a significant minority—nearly 2% of the population—struggles with a persistent, clinically significant form of restrictive eating that lasts throughout early and middle childhood.
Unmasking the Genetic Architecture
One of the study’s most significant contributions is its exploration of the genetic underpinnings of ARFI. Using genome-wide association analyses, the researchers estimated the single-nucleotide variant heritability (SNV-h2) of ARFI phenotypes to be between 8% and 16%. This confirms that restrictive eating behaviors are not merely products of the environment or parenting styles but have a clear biological basis.
Table 2. Avoidant/Restrictive Food Intake (ARFI) Symptoms and Clinical Significance Indicators in Children With Persistent, Transient, and Emergent ARFI.
| Symptom | No. (%) | ||
|---|---|---|---|
| ARFI–broad persistent (n = 2129 [5.96%]) | ARFI–broad transient (n = 6338 [17.73%]) | ARFI–broad emergent (n = 3001[8.39%]) | |
| ARFI symptoms: age 3 y | |||
| Does not eat well | 608 (28.56) | 1070 (16.88) | NA |
| Not happy eating food | 214 (10.05) | 400 (6.31) | NA |
| Fussy | 1281 (60.17) | 2402 (37.90) | NA |
| Careful to make sure child eats enough | 572 (26.87) | 1306 (20.61) | NA |
| If child says not hungry, try to get him/her to eat | 767 (36.03) | 3069 (48.42) | NA |
| Child needs guidance or regulation to eat enough | 758 (35.60) | 1597 (25.20) | NA |
| ARFI symptoms: age 8 y | |||
| Does not enjoy tasting new foods | 600 (28.18) | NA | 422 (14.06) |
| Gets full easily | 432 (20.29) | NA | 470 (15.66) |
| Eats slowly | 615 (28.89) | NA | 813 (27.09) |
| Takes more than 30 min to finish meal | 88 (4.13) | NA | 63 (2.10) |
| Gets full before finished meal | 311 (14.61) | NA | 448 (14.93) |
| Does not enjoy a variety of foods | 656 (30.81) | NA | 461 (15.36) |
| Is not interested in tasting new food | 811 (38.09) | NA | 689 (22.96) |
| Eats less when upset | 190 (8.92) | NA | 456 (15.19) |
| Leaves food on plate at the end of a meal | 297 (13.95) | NA | 335 (11.16) |
| Eats less when angry | 231 (10.85) | NA | 482 (16.06) |
| Clinical significance indicators | |||
| Any clinical indicator (ARFI-clinical) | 624 (29.31) | 1157 (18.25) | 484 (16.13) |
| Type of clinical indicator | |||
| Weight loss/failure to growa | 481 (22.59) | 943 (14.88) | 363 (12.10) |
| Nutritional deficiencyb | 138 (6.48) | 232 (3.66) | 110 (3.67) |
| Clinical diagnosis of eating disorderc | 152 (7.14) | 105 (1.66) | 61 (2.03) |
The analysis identified two independent genome-wide significant loci. Most notably, a strong association was found with the ADCY3 (Adenylate Cyclase 3) locus (z = 5.42; P = 3.03 × 10−8) for the ARFI-clinical phenotype. ADCY3 is known to be involved in the regulation of appetite and energy homeostasis, and variants in this gene have previously been linked to obesity and type 2 diabetes. Its identification in the context of ARFI suggests a shared biological pathway between restrictive eating and broader metabolic regulation.
Figure 1. Developmental Characteristics for Children With and Without Avoidant/Restrictive Food Intake (ARFI) Phenotypes (N = 35 751).
Furthermore, the study quantified genetic correlations between ARFI and other phenotypes. Small to moderate genetic correlations were found with mental health conditions (such as ADHD and autism spectrum disorder), cognitive/educational outcomes, anthropometric traits (BMI), and gastrointestinal disorders. This suggests that the genetic risk for ARFI overlaps with the genetic architecture of neurodevelopmental and metabolic health.
Neurodevelopmental and Clinical Correlates
The longitudinal data allowed the researchers to examine how ARFI relates to a child’s broader development. Children in the persistent ARFI-broad group exhibited significantly more developmental difficulties compared to their peers without ARFI. These difficulties were evident as early as 6 months of age and persisted through 14 years.
Affected children were more likely to show delays or challenges in motor skills, social communication, and cognitive functioning. The strong association with neurodevelopmental traits suggests that for many children, ARFI is not an isolated eating behavior but rather one manifestation of a broader neurodivergent profile. This aligns with clinical observations that ARFID is highly comorbid with autism spectrum disorder (ASD) and attention-deficit/hyperactivity disorder (ADHD).
Expert Commentary: Clinical Implications
The implications of this study for pediatricians and mental health professionals are profound. First, the 6.3% prevalence rate of clinical ARFI underscores the need for routine screening in primary care settings. Clinicians should be mindful that persistent restrictive eating is often associated with broader developmental risks.
Second, the discovery of the ADCY3 association and the heritability of the trait should help reduce the stigma often felt by parents. Rather than viewing restrictive eating solely as a behavioral or parenting issue, it should be recognized as a complex phenotype with a significant genetic component. This biological perspective may pave the way for more targeted pharmacological or behavioral interventions in the future.
Finally, the study highlights the importance of ‘persistence’ as a clinical red flag. While transient picky eating is common, children whose symptoms persist from age 3 to age 8 are at the highest risk for long-term developmental difficulties and should be prioritized for comprehensive multidisciplinary assessment.
Conclusions and Future Directions
This landmark study clarifies that ARFI is a prevalent and clinically significant condition within the pediatric population. By identifying the genetic architecture and developmental trajectories of ARFI, the researchers have provided a foundation for future precision medicine approaches. The findings suggest that interventions should not only focus on expanding a child’s diet but also provide broad support for associated neurodevelopmental and gastrointestinal challenges. Future research should continue to explore the functional role of ADCY3 in restrictive eating and investigate whether early interventions can alter the developmental trajectory of children with persistent ARFI phenotypes.
Funding and ClinicalTrials.gov
This study was supported by grants from the Research Council of Norway and the South-Eastern Norway Regional Health Authority. Data originated from the Norwegian Mother, Father, and Child Cohort Study (MoBa). The study was preregistered on the Open Science Framework.
References
1. Bjørndal LD, Corfield EC, Hannigan LJ, et al. Prevalence, Characteristics, and Genetic Architecture of Avoidant/Restrictive Food Intake Phenotypes. JAMA Pediatr. 2026;180(1):45-55. doi:10.1001/jamapediatrics.2025.4786 IF: 18.0 Q1 .2. Bryant-Waugh R. Avoidant restrictive food intake disorder: an illustrative case example. Clin Child Psychol Psychiatry. 2013;18(4):489-496.
3. Grilo CM. Avoidant/restrictive food intake disorder (ARFID) at age 7 years in a nationwide cohort: prevalence, co-occurrence, and risk factors. Z Kinder Jugendpsychiatr Psychother. 2021;49(4):284-290.
