When Food Avoidance Meets The Pelvic Floor: ARFID, The Microbiome, And Constipation In Pediatric Pelvic Health

For pediatric pelvic health practitioners, the child who “won’t poop” and the child who “won’t eat” are often the same. You tend to meet these kids in the context of withholding, soiling, or refractory constipation, and you build your plan of care around the pelvic floor and the gut.

You often focus on the bowel symptoms because that’s what brings families through your doors. But sometimes the constipation story is actually part of a much bigger picture. You start asking a few more questions and realize the kids barely eat anything beyond a handful of “safe” foods.

Increasingly, the eating side of the story deserves a name and a place in your clinical reasoning: Avoidant/Restrictive Food Intake Disorder (ARFID).

ARFID sits at a fascinating crossroads of feeding, the gut microbiome, bowel function, and the pelvic floor. It is one of the clearest examples of why pediatric pelvic health cannot be practiced in a silo. This blog post walks through these connections and how you can assess, treat, and refer.

ARFID became a formal diagnosis in the DSM-5 in 2013, replacing the older, narrower concept of “selective eating disorder.” Unlike anorexia or bulimia, children with ARFID aren’t restricting food because they’re worried about weight, body shape, or appearance.

Instead, food avoidance is typically driven by one or more of the following:

• Lack of interest in eating
• Avoidance based on sensory sensitivities to textures, smells, temperatures, or colors
• Fear of negative consequences such as choking, vomiting, abdominal pain, or GI discomfort

The impact can be significant. Children may experience nutritional deficiencies, weight loss or poor growth, dependence on supplements or enteral feeding, and social limitations.

ARFID frequently co-occurs with:

• Autism spectrum disorder (ASD)
• Attention-deficit/hyperactivity disorder (ADHD)
• Anxiety disorders
• Gastrointestinal conditions, including:
  • Reflux
  • Functional GI disorders
  • Food allergies
  • Oral-motor difficulties
  • Constipation

Constipation is one of the most common reasons these children may be referred to you, so let’s take a look at how this may present in the clinic.

As a pelvic health practitioner, one of the most important things to understand is that the relationship between ARFID and constipation is bidirectional.

On one side, a restrictive diet shapes the gut. Children with ARFID often gravitate toward a limited carbohydrate-heavy “beige” diet that is low in fiber and fluid variety. That dietary pattern is itself a recognized contributor to functional constipation.

On the other hand, constipation can drive food avoidance. A child who experiences early satiety, abdominal pain, bloating, or painful defecation learns to eat less. Stomach and digestive problems are recognized triggers of ARFID because food can come to feel like a source of misery.

In other words, the constipation you treat may be both a consequence of restrictive eating and a maintaining factor. This bidirectional loop is exactly the kind of self-perpetuating cycle that you are trained to interrupt.

An exciting recent development is the emerging picture of the gut microbiome in ARFID, and it ties the feeding and bowel stories together at a mechanistic level.

We know that dietary variety plays an important role in establishing a diverse gut microbiome during childhood. Long-term food restriction alters both the composition and function of the microbiome.

In the first observational study looking at the microbiome in children with ARFID, investigators compared 102 children with ARFID to 33 healthy controls and found meaningful differences in both microbial diversity and composition. They also identified differences in microbial genes involved in carbohydrate metabolism, which may reflect the high-carbohydrate dietary patterns commonly seen in ARFID (Ye et al., 2023).

Building on this, a 2024 conceptual review proposed a microbiota-gut-brain axis model of ARFID, in which a homogeneous, nutrient-depleted diet alters the microbiome, which in turn influences gastrointestinal symptoms, gut-brain signaling, and the very appetite and sensory-processing pathways that perpetuate avoidant eating (Schneider et al., 2024). The model is still theoretical, and the evidence base is young, but it gives us a coherent narrative linking what a child eats, how their gut feels, and how their bowel and brain behave.

For pelvic health, the takeaway is that the microbiome is not a separate topic from constipation. Microbial shifts affect stool consistency, motility, and visceral sensation. And the uncomfortable gut that results may further reinforce both food avoidance and withholding behaviors.

Now connect all this back to the muscles you treat.

Functional constipation affects an estimated 9-12% of children. In the withholding cycle, a painful or frightening bowel movement leads a child to retain stool voluntarily. Retained stool grows harder and larger. That rectal distention dulls the urge to defecate, and the next attempt is even more painful, reinforcing the avoidance.

Children accomplish this withholding by actively contracting the pelvic floor and squeezing the gluteal muscles. You recognize this retentive posturing in any child.

Over time, this learned guarding can evolve into pelvic floor dyssynergia: the paradoxical contraction or failure to relax the pelvic floor muscles and anal sphincter during attempted defecation (StatPearls, 2024). The child’s body is, in effect, holding the door shut at the exact moment it needs to open.

Now layer ARFID in, and the clinical picture becomes even clearer.

A child with a fear-of-aversive-consequences ARFID profile may have started avoiding food because defecation became painful. A child with sensory-driven ARFID may eat a fiber-poor diet that keeps the stool hard and the cycle alive. And a child with constipation-related early satiety may be eating less because their rectum is full.

In each scenario, treating the pelvic floor without addressing the feeding concerns leaves part of the cycle intact. Likewise, treating the feeding concerns without addressing bowel function and pelvic floor coordination will limit progress.

That’s why these kids often require you to think beyond any single system.

There is one more structure that quietly governs whether the pelvic floor can do its job: the diaphragm.

The diaphragm and the pelvic floor function as the lid and the base of a single intra-abdominal pressure canister, with the deep abdominal wall and spinal stabilizers forming the sides. When a child breathes diaphragmatically, the two move as a coordinated piston. Dynamic MRI work has shown that the diaphragm and pelvic floor move in phase-locked synchrony: both descend caudally on inhalation and recoil cranially on exhalation (Talasz et al., 2011).

That rhythmic descent matters because every quiet breath delivers to the pelvic floor a small, repetitive lengthening cue. This keeps the pelvic floor mobile and teaches it to yield to downward pressure rather than brace against it. This is the same mechanism a child must access to relax and let the pelvic floor descend for a bowel movement.

Apical breathing interrupts that system. When a child breathes high into the chest (recruiting the accessory neck and shoulder muscles instead of descending the diaphragm), the diaphragm never makes its full downward excursion, and the pelvic floor loses its rhythmic, breath-driven descent.

The pressure canister becomes disorganized: pressure is generated and managed at the top of the chest rather than distributed and modulated through a mobile diaphragm and a responsive pelvic floor. A pelvic floor that is no longer cued to lengthen with each breath tends to settle into a guarded and shortened posture, which is precisely the cause for the dyssynergia described above. The child then cannot relax and descend the pelvic floor on demand to evacuate, no matter how hard they push.

Additionally, the zone of apposition decreases, and rib flares become present. Now we are dealing with a poorly activated and inefficient core as well.

This is where the ARFID picture closes another loop. Many of these children are anxious. Anxiety is itself a powerful driver of shallow, apical breathing and chronic pelvic floor holding. Food restriction, gut discomfort, and fear of eating raise the arousal level:

1. The breath rises into the chest.
2. The pelvic floor braces.
3. The stool is retained.
4. The discomfort grows.
5. And the avoidance is reinforced.

The diaphragm is one of the levers that determines whether the whole pressure system works for the child or against them.

Clinically, this is why diaphragmatic breath retraining belongs at the foundation of pelvic floor training and effective defecation mechanics. Restoring a true diaphragmatic breath re-establishes the piston, giving the pelvic floor permission to lengthen and descend. This supports a “blow, don’t push” evacuation strategy rather than a breath-held strain that drives the pelvic floor in the wrong direction.

For these kids, teaching the breath can be as central to bowel recovery as the laxative protocol or the dietary expansion.

ARFID is a condition no single discipline can fully treat. The complexity of its presentation—nutritional, gastrointestinal, behavioral, sensory, and musculoskeletal—is why the literature repeatedly calls for coordinated, multidisciplinary care (University of Gothenburg, n.d.).

A well-rounded team often includes:

• The pediatrician or pediatric GI provider: to evaluate for organic causes, manage disimpaction and maintenance laxative therapy, and monitor growth and nutritional status.
• The registered dietitian: to assess intake, address deficiencies, and expand dietary variety in a way that supports both nutrition and the microbiome.
• The feeding therapist (OT or SLP): to address sensory aversions, oral-motor skills, and the mealtime behaviors at the heart of the avoidance.
• The mental health provider: to treat the anxiety, trauma history, or rigidity that frequently underlies ARFID, often through CBT-AR.
• The pediatric pelvic health PT or OT: to assess and retrain pelvic floor coordination, restore diaphragmatic breathing and the pressure system, teach effective defecation dynamics and positioning, and address withholding and retentive posturing.

The pelvic health practitioner’s role here is bigger than biofeedback and toileting mechanics. You are frequently the provider who screens for and addresses the constipation that nobody else has connected to the child’s eating. Because, as you know, families often arrive for one concern and never mention the bowel and bladder issues at all.

You are well-positioned to ask the questions that reveal the full picture, and to make referrals that pull the rest of the team together.

Practical Takeaways For The Clinic

• Screen the bowel in every restrictive eater, and screen eating in every withholder. The two so often co-occur that finding one should prompt you to ask about the other.
• Treat constipation as a possible cause of food avoidance, not just a result of it. A full rectum, early satiety, and painful defecation can all suppress appetite and intake.
• Frame dietary variety as gut-health work, not just nutrition. Expanding the diet supports the microbiome, which improves motility and stool consistency.
• Refer early and refer wide. If you suspect ARFID rather than ordinary picky eating (faltering growth, nutritional concern, supplement dependence, or functional impairment), loop in GI, dietetics, feeding therapy, and mental health rather than treating the pelvic floor in isolation.

In my online course, Pediatrics Level 1 – Treatment of Bowel and Bladder Disorders, you will learn more about how restrictive eating ties into pelvic floor function. If you want to deepen your understanding and confidently navigate the treatment framework, sign up today!

References

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). https://doi.org/10.1176/appi.books.9780890425596

Children’s Health. (n.d.). Pediatric avoidant/restrictive food intake disorder (ARFID). Retrieved June 4, 2026, from https://www.childrens.com/specialties-services/conditions/avoidant-restrictive-food-intake-disorder

Dinkler, L., et al. (2024). ARFID InitiativE Sweden (ARIES): Study protocol for a large-scale genetic and registry-linked cohort study on avoidant/restrictive food intake disorder. BMJ Open. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12007039/

Hall, C., Illingworth, V., Sims, A., Whitlingum, G., & Cryer, J. Should children under 5 and those with constipation be overlooked from ARFID research? eClinicalMedicine, 2024; 73. https://doi.org/10.1016/j.eclinm.2024.102667

Nationwide Children’s Hospital. (n.d.). Biofeedback for constipation and pelvic floor dysfunction. Retrieved June 4, 2026, from https://www.nationwidechildrens.org/family-resources-education/health-wellness-and-safety-resources/helping-hands/biofeedback-for-constipation-and-pelvic-floor-dysfunction

Schneider, E., Schmidt, R., Cryan, J.F. and Hilbert, A. (2024), A Role for the Microbiota-Gut-Brain Axis in Avoidant/Restrictive Food Intake Disorder: A New Conceptual Model. Int J Eat Disord, 57: 2321-2328. https://doi.org/10.1002/eat.24326

StatPearls. (2024). Pediatric functional constipation. NCBI Bookshelf. https://www.ncbi.nlm.nih.gov/books/NBK537037/

Talasz, H., Kremser, C., Kofler, M. et al. Phase-locked parallel movement of diaphragm and pelvic floor during breathing and coughing—a dynamic MRI investigation in healthy females. Int Urogynecol J 22, 61–68 (2011). https://doi.org/10.1007/s00192-010-1240-z

University of Gothenburg. (n.d.). Avoidant/restrictive food intake disorder (ARFID). Retrieved June 4, 2026, from https://www.gu.se/en/gnc/avoidantrestrictive-food-intake-disorder-arfid

Ye, Q., Sun, S., Deng, J., Chen, X., Zhang, J., Lin, S., Du, H., Gao, J., Zou, X., Lin, X., Cai, Y., & Lu, Z. (2023). Using 16S rDNA and metagenomic sequencing technology to analyze the fecal microbiome of children with avoidant/restrictive food intake disorder. Scientific reports, 13(1), 20253. https://doi.org/10.1038/s41598-023-47760-y