How common is ARFID in preschool children and how can we identify it early?

[Posted on 12th February, 2025 by Lisa Dinkler]

Avoidant Restrictive Food Intake Disorder (ARFID) often begins in early childhood, yet little is known about how common it is or how it manifests in very young children, and well-validated, parent-reported screening tools are lacking. We recently completed a large study addressing these issues by screening for ARFID among preschoolers in the Fyrbodal region of Western Sweden and want to share our findings with you here. Many GNC-affiliated researchers were involved in this study, which started in 2019 and was funded by both the Swedish Research Council (VR) and the Swedish Brain Foundation (Hjärnfonden).

What is ARFID?

Most parents have likely encountered picky eating phases, where their children refuse to touch anything green or insist on eating the same meal every day. For most families, these phases pass without much concern. But what happens when picky eating becomes more than a phase, and when it begins to affect a child’s health, growth, and daily life? This is when ARFID should be considered.

ARFID is a feeding and eating disorder that goes beyond the typical picky eater. It’s not about body image or weight concerns, as in other eating disorders, but stems from low interest in food, sensory aversions (extreme dislike of certain textures, tastes, or smells), or fear of unpleasant consequences of eating, like choking or vomiting. Left unaddressed, ARFID can lead to serious nutritional, medical, and psychosocial challenges and interfere with daily life. Meals become battlegrounds, social eating situations are avoided, and stress escalates for the entire family. 

Although awareness of ARFID has grown in recent years, it’s still a relatively new diagnosis (introduced in 2013), and much remains unknown. Learn the basics of ARFID in our earlier blog post.

ARFID in preschool children

Despite ARFID often starting in early childhood, little is known about prevalence and clinical characteristics of ARFID in preschoolers. This knowledge gap makes it challenging to plan healthcare services and allocate resources effectively. Routine screening for ARFID is not currently part of primary care, such as at child health centre visits, leaving many families without the support they need.

Preschool age is a critical period for identifying and addressing ARFID. Eating patterns established during this time can have a lasting impact on a child’s health and well-being. Early detection and intervention could prevent eating challenges from developing into more severe nutritional, growth, or health problems later in life.

However, distinguishing ARFID from typical picky eating in young children is particularly challenging. It’s normal for children under six to go through phases of food refusal, often referred to as "selective" or "picky" eating. These behaviours usually resolve on their own by age six or seven. The overlap between typical development and ARFID symptoms can lead to two key risks:

• Overdiagnosis, where typical picky eating is mislabelled as ARFID, placing unnecessary strain on healthcare systems.
• Underdiagnosis, where children with significant eating difficulties are dismissed as being "just picky," missing the opportunity for early treatment. 

Parents who express concerns about their child’s eating habits are often reassured that the issue is temporary and will resolve with time. While this is true for many children, some may in fact have ARFID or will go on to develop ARFID—a condition that can have serious and long-lasting effects if not addressed.

Aims of the study

How common is ARFID in preschoolers and how can we identify it early?

To address the lack of validated screening instruments for ARFID in children, we previously developed the ARFID-Brief Screener, a parent-reported questionnaire designed specifically to identify ARFID symptoms in children.

In the current study, we aimed to test how well the ARFID-Brief Screener performs in identifying children with ARFID. Our ultimate goal is to provide a validated screening instrument than can be used routinely in healthcare settings such as a child health centre visit.

We also wanted to better understand how common ARFID is in Swedish preschool children and the symptoms and co-existing problems these children present with. 

What data did we collect?

Between November 2020 and June 2022, parents of children attending 2.5-year and 4-year routine check-ups at 21 child health services centres in the Fyrbodal region north of Gothenburg were invited to participate in the study. Parents of 645 children (50.5% male) completed the ARFID-Brief Screener. They also filled in the ESSENCE-Q, a screener for neurodevelopmental conditions developed at the GNC. 

To confirm diagnoses and learn more about the ARFID symptoms of these children, we then conducted in-depth diagnostic interviews via phone with parents of children who screened positive for ARFID, as well as with a sample of children who screened negative. We also collected information from child health centre records, including growth curves and neurodevelopmental diagnoses. 

Key findings

ARFID-Brief Screener may be a valuable tool in Swedish healthcare

Overall, the ARFID-Brief Screener demonstrated strong psychometric properties when compared to the clinical interview PARDI in this population of preschool children.

In total, 42 of 645 (6.5%) children screened positive for ARFID. Of these 42 children, 29 were followed up via interview, and an ARFID diagnosis was confirmed in 21 children. This means that most children (72%) who screen positive did have had ARFID, underscoring the screener’s utility in identifying true cases. The other 8 children had eating problems that were not severe enough to meet the diagnostic criteria for ARFID.

Of the 603 children screening negative for ARFID, 33 were interviewed, and in all but 2 children, it was confirmed that they didn’t have ARFID. This means a negative screening result was almost always correct (in 94% of cases). This may be particularly reassuring for parents and healthcare providers, as it indicates that children who screen negative are unlikely to have ARFID.

ARFID is not rare among preschoolers

Taking all these data into consideration, the best estimate of the prevalence of ARFID in our sample was 5.9%, showing that ARFID is not rare among preschoolers. However, our study did have some methodological challenges, mainly because data collection was conducted during  the COVID-19 pandemic. Because of this, our prevalence estimate is likely somewhat too high. Still, our results show that ARFID is common among preschoolers and should be considered if eating problems are brought to attention by parents.

ARFID in preschoolers is characterised by psychosocial impairment

The majority of children with ARFID in our study did not exhibit significant weight or growth issues, nutritional deficiencies, or dependence on nutritional supplements. These are potential physical consequences of restrictive eating behaviours based on which criteria for ARFID may be met. For instance, the children with ARFID in our sample did not have significantly lower body mass indexes (BMI) compared to those without ARFID.

Instead, most children with ARFID met diagnostic criteria through the criterion called “marked interference with psychosocial functioning”. Many cases were borderline, just meeting the threshold for this criterion. Commonly reported difficulties included

• Frequent mealtime conflicts
• Anxiety and tantrums during meals
• Eating alone
• Avoidance of eating situations outside the home

This finding aligns with previous research suggesting that psychosocial impairment may serve as an early marker of ARFID, while nutritional and weight-related issues are more likely to emerge later. As children grow older, reduced parental control over their eating habits and increased energy demands during puberty may amplify these physical consequences.

ARFID in preschoolers is characterised by sensory-based avoidance and low interest

All children with ARFID in our study displayed both sensory-based avoidance and low interest for eating or food. This supports previous findings that these two profiles of ARFID typically have an early onset. Our findings also align with prior research showing that children with the Sensory profile or a combined Sensory and Low interest profile typically have BMIs within the normal weight range—consistent with what we observed in our study.

Neurodevelopmental problems are very common in ARFID

Lastly, we found that children with ARFID were at significantly higher risk of co-occurring neurodevelopmental problems compared to those without ARFID. Among children with ARFID, 43.5% scored above cut-off (≥5) on the ESSENCE-Q compared to 28.2% in the non-ARFID group. Additionally, 17-22% of children with ARFID had diagnosed autism and/or ADHD. These findings are consistent with previous research. Studies in clinical samples (i.e. children treated for ARFID) have reported that up to half of them had suspected or diagnosed autism.

Children with ARFID in our study were also nearly three times more likely to fail the structured speech and language test routinely administered at the 2.5-year child heath centre visit. We also found that a higher burden of neurodevelopmental problems was associated with greater ARFID severity and higher scores on the Sensory profile. This aligns with prior research, particularly regarding autism, and supports the idea that eating problems in children with autism and other neurodevelopmental conditions may be more challenging to treat.

Take-home message for parents and clinicians

Image

1. ARFID may be more common than you think. Our study found that 5.9% of preschoolers met criteria for ARFID. While this number may be somewhat overestimated, it highlights that ARFID is at least as common as many other neurodevelopmental conditions in preschool-aged children. It should be considered if eating problems are brought to attention by parents.

2. Incorporate ARFID screening into routine check-ups. Tools like the ARFID-Brief Screener can be integrated into standard health visits, particularly for children with feeding challenges or neurodevelopmental concerns. Our findings suggest that this parent-reported tool is easy to use and effective in identifying children at risk. For parents, it provides an initial indication of whether their child’s eating behaviours warrant further attention. For clinicians, it supports early identification and facilitates timely referrals for intervention. While the screener is not a diagnostic tool, it helps identify children who need further evaluation. Follow-up assessments are necessary to confirm a diagnosis.

3. Weight and growth may not always be affected - focus on psychosocial impacts. Only 13% of children with ARFID in our study showed nutritional or weight-related issues, while the majority (87%) experienced significant psychosocial challenges. These included extremely stressful mealtimes, eating alone, and avoiding social eating situations. This underscores the importance of looking beyond physical health metrics to assess how eating behaviours impact the child’s social interactions, daily functioning, and family dynamics.

4. Neurodevelopmental problems may signal increased ARFID risk. ARFID frequently co-occurs with neurodevelopmental conditions such as autism or ADHD. In our study, nearly half of the children with ARFID showed signs of neurodevelopmental difficulties. These challenges may, in fact, be stronger predictors of ARFID than typical early feeding issues, emphasising their importance in supporting early detection and intervention.

Study participation

If you want to participate in ARFID research, please visit www.ariesstudy.se for 6-14-year-old children) and www.edgi.se for adults.