Research progress on dietary characteristics and feeding interventions in children with autism spectrum disorder: a narrative review

Abstract

Background and Objective

Children with autism spectrum disorder (ASD) often experience feeding difficulties, including extreme food selectivity (FS), disruptive eating behaviors, and in severe cases, avoidant/restrictive food intake disorder (ARFID). These issues can result in unbalanced nutrient intake, impaired growth and development, worsened gastrointestinal symptoms (GIS), and significant feeding burdens for families. This review aims to summarize the dietary characteristics of children with ASD, examine the underlying causes, and evaluate common treatments. The primary objective is to provide evidence-based guidance for child health physicians to implement scientific, individualized nutritional management strategies in clinical practice.

Methods

We systematically reviewed journal articles published between 2010 and 2025. The analysis focused on dietary characteristics, related mechanisms, and corresponding dietary intervention strategies for children with ASD.

Key Content and Findings

This review summarizes the characteristic dietary profiles of children with ASD and discusses the underlying mechanisms from several perspectives: sensory processing dysfunction, microbiota-gut-brain axis dysregulation, food allergy or intolerance, and behavioral-cognitive abnormalities. It also systematically evaluates the effectiveness of mainstream nutritional and behavioral interventions. The main treatment is applied behavior analysis (ABA). It works best for improving eating behaviors. Other food-related treatments have mixed results: special diets [like gluten-free and casein-free (GFCF)] are not proven to help core autism symptoms for most children. They may be tried for a short period in certain cases. Probiotics/prebiotics can help with stomach problems, but their effect on autism symptoms is unclear. Omega-3 supplements do not clearly help autism symptoms. Vitamin/mineral supplements should only be used if blood tests show a lack of specific nutrients.

Conclusions

We conclude that ABA constitutes the primary behavioral intervention. Nutritional approaches must be implemented following comprehensive evaluation, targeting specific comorbidities—such as GIS or allergies—or confirmed nutrient deficiencies through precise supplementation. This individualized strategy effectively avoids generalized dietary protocols.

Introduction

Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder. Its core features include deficits in social communication and interaction, restricted interests, and repetitive, stereotyped behaviors (1). Globally, the prevalence of ASD continues to rise. The latest data show that approximately 1 in 36 children in the United States is diagnosed with ASD (2). ASD shows high phenotypic and genetic heterogeneity, which poses significant challenges to understanding its etiology, predicting developmental trajectories, and formulating clinical interventions. Currently, there is no fully effective treatment. Feeding and dietary problems are common comorbid symptoms of ASD, with an incidence ranging from 46% to 89%, significantly higher than the incidence rate of 25–45% seen in typically developing (TD) children (3). In 2025, Litman et al. proposed the concept of “broadly impactful ASD” based on clinical manifestations. This term describes subtypes with core social communication deficits plus significant comorbidities, such as feeding problems, gastrointestinal diseases, and attention deficits. It distinguishes these subtypes from those with only core symptoms or developmental delays (4).

Objectives

The dietary problems of children with ASD are not simply “picky eating” or behavioral deviations. They constitute a syndrome resulting from complex interactions. Multiple factors are involved, including neurobiological, sensory-physiological, gastrointestinal, and psychological-behavioral components (5). These multidimensional dietary issues pose direct health risks. Children may develop micronutrient deficiencies and chronic constipation. Indirectly, these issues can also exacerbate core symptoms like stereotyped behaviors and emotional disorders, potentially through mechanisms like the gut-brain axis. In addition, these problems place a significant feeding burden on families and greatly reduce overall quality of life (6). Therefore, systematic exploration of dietary behavioral characteristics in children with ASD is crucial. Analyzing underlying mechanisms and evaluating evidence-based interventions are also essential. These efforts can improve children’s prognosis and reduce family burdens. We present this article in accordance with the Narrative Review reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-1-0162/rc).

Methods

We have conducted a literature review on the dietary characteristics and nutritional intervention strategies for children with ASD. Table 1 outlines our search strategy. This review includes published journal articles from the past 15 years [2010–2025], which analyze and discuss the dietary profiles and intervention approaches for children with ASD. The search was performed using the PubMed database. High-quality articles were defined as randomized controlled trials, systematic reviews, meta-analyses, and peer-reviewed studies with large sample sizes, low risk of bias, and clear outcome assessments. Priority was given to high-quality articles published in the last 5 years and those with high citation counts, while key classic and landmark studies were also included to avoid selection bias. A total of 55 articles were included, among which 49 (89%) were published within the last 10 years and 36 (65%) within the last 5 years. The selected literature demonstrates satisfactory overall quality, clear research objectives, and reflects current advancements in understanding dietary characteristics and nutritional interventions for children with ASD.

Table 1. The search strategy summary.

Items	Specification
Date of search	December 1, 2025
Database searched	PubMed
Search terms used	Details are provided in Table S1
Timeframe	January 1, 2010 to December 1, 2025
Inclusion and exclusion criteria	Inclusion criteria:
	• Studies on patients with ASD aged under 18 years
	• Studies on dietary behavioral characteristics, mechanisms, or interventions
	• Specific study types, namely randomized controlled trials, observational studies, and systematic reviews
	Exclusion criteria:
	• Non-English/Chinese publications, and studies involving other serious comorbidities
Selection process	To identify potentially relevant studies, titles and abstracts were initially screened against the eligibility criteria. Full texts were subsequently reviewed, with priority given to recent, high-impact publications

ASD, autism spectrum disorder.

Dietary behavioral characteristics of children with ASD

Feeding difficulty is an umbrella term that refers to several problems related to food intake and eating behaviors (7). Feeding difficulties are common in children with ASD. Although TD children may experience temporary dietary fluctuations in early life, feeding problems of children with ASD are more severe and persistent. They may even last into adulthood. These problems mainly manifest in three aspects: extreme food selectivity (FS) and food neophobia (FN), disruptive eating behaviors, and avoidant/restrictive food intake disorder (ARFID), all of which have been discussed below.

Extreme FS and FN

FS, characterized by extreme pickiness, is the core dietary feature of children with ASD (8). Children may be picky about food type, texture, color, and smell. They also resist trying new foods, a trait known as FN.

The diet of children with ASD and FS shows significant heterogeneity. Bandini et al. classified it into three specific manifestations (9). These include food refusal, extremely limited food variety, and high-frequency intake of a single food. FN specifically refers to persistent resistance to new foods. TD children gradually accept new foods through repeated exposure, whereas children with ASD face significant obstacles in this process. This selectivity leads to evident dietary structure abnormalities. Children may have stereotyped requirements for specific textures. For example, they may only accept crispy or pureed foods. They may also show excessive preference for sweetness. They may demand fixed food colors or presentation methods. The result is often poor dietary quality. Children may rely more on ultra-processed foods and high-calorie snacks. They tend to have insufficient intake of fresh fruits, vegetables, and high-quality proteins.

Children with ASD also commonly exhibit deficits or developmental delays in feeding skills, which constitute another major contributor to feeding difficulties. Studies have shown that delayed development of oral motor skills impairs basic functions during feeding, such as chewing and swallowing, and may present as difficulties in breastfeeding, delayed introduction of solid foods, and intolerance to specific food textures (10,11). These oral motor challenges frequently co-occur with sensory processing difficulties, leading to restrictive dietary patterns and food refusal in children with ASD. As children with ASD grow older, delays in fine motor skills further contribute to reduced self-feeding ability and difficulties in using utensils, often necessitating comprehensive rehabilitative interventions (12).

Disruptive eating behaviors

Mealtimes for children with ASD are often accompanied by challenging behaviors, such as crying, aggression, self-injury, and leaving the table. Specific manifestations like spitting out food, holding food in the mouth without swallowing, and hitting tableware may also occur (13). These behaviors turn daily mealtimes into a major source of family conflict and stress.

ARFID

When dietary problems become severe enough, they meet clinical diagnostic criteria. These criteria include significant weight loss, growth retardation, nutrient deficiencies, or severe interference with social functioning. At this stage, the condition can be diagnosed as ARFID (14). The main causes of ARFID are not concerns about body shape or weight, but rather sensory sensitivity, lack of appetite, or fear of negative eating consequences (such as choking) (15). When the feeding problems of children with ASD meet DSM-5 diagnostic criteria, they are diagnosed with ARFID. These criteria include effects on nutritional status, weight, and psychosocial functioning. Therefore, ASD is among the most common comorbid conditions associated with ARFID (16). The subtype driven by sensory sensitivity is particularly prominent (17). Statistics show that 8.2–54.8% of individuals with ASD meet ARFID diagnostic criteria (18). Routine screening for feeding problems is necessary for all children diagnosed with ASD to assess whether they meet ARFID criteria.

Nutritional and health consequences

Macronutrient and micronutrient deficiencies

FS and refusal of new foods are key dietary problems in ASD, which may lead to insufficient nutrient intake and significantly impact children’s health. Children with ASD often prefer high-calorie processed foods; however, their total energy intake does not increase significantly (19). Their dietary structure is severely imbalanced. A large case-control study of children aged 3–12 years found that the ASD group had significantly lower intake of key food categories, including vegetables, fruits, dairy products, and meat/fish (13). This imbalanced diet directly reduces dietary quality. A study using the Chinese Children’s Dietary Index found that compared with non-ASD children, children with ASD had significantly lower dietary scores and higher exposure to food additives (e.g., emulsifiers, sweeteners) (20). An unbalanced diet is an important independent factor associated with comorbid constipation and total gastrointestinal symptoms (GIS) in ASD. A case-control study confirmed that children with ASD had poorer diets, consumed fewer vegetables/fruits, fewer food types, and a higher degree of insufficient or imbalanced dietary intake, as well as more severe constipation and total GIS (21).

Children with ASD face a higher risk of micronutrient deficiencies. Multiple meta-analyses have reached consistent conclusions in this regard. Their intakes of vitamin D and calcium are reportedly lower than their non-ASD peers. Their nutritional intake also falls below dietary reference values, posing clear nutritional risks (19,22). A review focused on the nutritional status of American children with ASD between 2014 and 2025 and identified the following most common deficiencies: vitamin D (25.0%), vitamin A (24.1%), B vitamins (18%), calcium (10.8%), and iron (9.6%) (23). These micronutrients are crucial for neurotrophy, synaptic plasticity, and anti-inflammatory processes. Their deficiency may further compromise neurodevelopment (24). In addition, due to low fish intake, omega-3 fatty acid intake is generally insufficient (22).

Abnormal growth and development

The results of studies on the relationship between dietary problems and growth indicators—height, weight, and body mass index (BMI)—in ASD children are inconsistent. Factors like sample heterogeneity, cultural differences, and comorbid medication use may contribute to this inconsistency. Some studies found a higher proportion of overweight or obese ASD children (25,26). Others reported shorter average heights in this group (19). This height difference may be attributable to long-term nutrient deficiencies, which can impair linear growth rate (27). A recent, large case-control study found no significant correlation between the BMI of children with ASD and factors like FS severity, dietary restrictions, and GIS severity (13). This suggests that even with normal or overweight status, children with ASD may face severe specific nutrient deficiencies, primarily resulting from a lack of variation in their dietary structure. Common deficiencies include calcium and various vitamins.

GIS

GIS are common comorbidities in patients with ASD, with an estimated prevalence of 20–80% (28,29). Chronic constipation is the most common symptom (13). Children with ASD often struggle with communication and cannot directly express discomfort such as abdominal pain and bloating. Instead, they may convey it through behavioral changes. These changes include irritability, aggression, specific vocalizations (e.g., groaning), pressing the abdomen, or sleep disturbances (30). Children with ASD experience diarrhea and constipation more frequently than TD children. Gastrointestinal problems are prevalent among those at high risk for FS in both groups of children (13). Insufficient dietary fiber and water intake are key factors. Intestinal microbiota disorders also contribute to constipation and other GIS (31).

Potential mechanisms of dietary problems

Abnormal sensory processing

Sensory sensitivity is the core mechanism driving food refusal in ASD children (32). Children with ASD may have difficulties in sensory processing and may show hypersensitivity or hyposensitivity. This applies to taste, smell, oral touch, and visual perception. Consequently, they may have abnormally heightened or reduced sensitivity to specific food textures, tastes, and smells during eating. A large body of evidence supports a positive correlation between sensory processing difficulties and feeding difficulties (33). Specific manifestations include tactile defensiveness, wherein children may resist foods with specific textures (e.g., slimy, crispy, and juicy). They may also have taste or smell hypersensitivity, with abnormally high sensitivity to odors or bitter and sour tastes. Such hypersensitivities lead to strong resistance. Visual stereotypy is another manifestation, wherein children ask for food to be presented in a fixed shape, color, or arrangement. A study involving 427 children with ASD found sensory feeding challenges to be the most common presentation, with up to 84.5% of children identified as hypersensitive to food texture (34). In addition, children with ASD exhibiting oral tactile sensitivity have poorer food acceptance. Consequently, their caregivers also experience more negative emotions (35).

Microbiota-gut-brain axis

The microbiota-gut-brain axis is a bidirectional communication network connecting the intestinal microbiota, gastrointestinal system, and central nervous system (36). GIS, FS, and gut microbiota form a complex bidirectional interaction rather than a unidirectional causal chain.

Sensory hypersensitivity and behavioral stereotypy in children with ASD often result in restricted dietary intake and insufficient food diversity, which in turn directly disrupt the structure of the gut microbiota, reduce microbial diversity, and induce gut microbiota dysbiosis (37). Numerous studies have shown that children with ASD commonly exhibit gut microbiota dysbiosis, often characterized by decreased microbial diversity and overgrowth of certain potentially harmful bacteria (36), and some microbial changes are associated with the severity of core symptoms (38). Conversely, imbalanced microbiota and altered microbial metabolites such as short-chain fatty acids may impair intestinal barrier integrity, trigger low-grade neuroinflammation, and aggravate gastrointestinal discomfort (39,40).

Some studies have indicated that there is only a weak direct association between gut microbiota and ASD, suggesting that gut microbiota may play a mediating role rather than a primary pathogenic role (41). Specifically, behavioral characteristics of ASD, including stereotyped behaviors and sensory preferences, lead to a monotonous dietary pattern, which in turn reduces gut microbial diversity and induces microbiota dysbiosis. This view highlights the importance of dietary diversity, and improving dietary diversity may represent a fundamental strategy to break the vicious cycle among GIS, FS, and microbiota imbalance.

Food allergy and intolerance

Children with ASD are at a 2.7 times higher risk of food allergies than TD children (42), with a relatively higher incidence of food intolerance reported as well. Common intolerances include gluten and casein (8,43). de Magistris et al. found significantly elevated serum levels of immunoglobulin G (IgG) antibodies against gluten and casein in the children with ASD, which may be attributable to increased intestinal permeability (“leaky gut”) (44). According to a theory, these food antigens likely enter the circulation through the damaged intestinal barrier and subsequently trigger immune and inflammatory responses. These responses then likely affect brain function and behavior via pathways like the vagus nerve (41). However, the clinical significance of elevated food-specific IgG antibodies remains controversial, and it may only be a marker of intestinal exposure and not a disease cause. Therefore, authoritative European and American allergy societies have clear recommendations in this regard, and they do not recommend food-specific IgG testing for diagnosing food intolerance (45).

Behavioral and cognitive factors

Findings on the association between ASD severity and feeding difficulties are inconsistent (13,46-48). The findings are mixed but lean toward a positive correlation. Restricted repetitive behaviors of ASD are associated with feeding difficulties or restrictions (33). ASD children may insist on consuming a few “safe foods” Minor dietary changes of any kind, such as food type, brand, or presentation method, can trigger strong anxiety and resistance. In addition, a lack of focus on eating is common, with frequent distractions during meals. From a behavioral analysis standpoint, this forms a negative reinforcement cycle: the child refuses to eat a new food item. → The parent feels anxious and adopts forced feeding or provides alternative snacks. → The child’s escape behavior is reinforced. → The child avoids disliked foods or obtains favorite snacks. → The child’s anxiety about new foods worsens, and effectively, the child refuses to eat them. Owing to this cycle, problem behaviors persist and deteriorate.

Evidence-based strategies for dietary interventions

Applied behavior analysis (ABA)

ABA, developed based on learning theory, is an effective method to alleviate ASD-related behavioral problems, including feeding issues. It has the highest level of empirical support, supported by a randomized controlled trial combined with a single-subject design. ABA-based interventions can significantly increase the acceptance of new foods by children with ASD (49). Its implementation usually involves several key steps: (I) conduct a functional behavioral assessment: identify the antecedents and consequences of food refusal. (II) Use gradual exposure: begin with having the child touch and smell new foods and then progress to tasting small doses. (III) Apply differential reinforcement strategies: provide positive rewards for behaviors approaching the target (e.g., bringing food close to the lips). (IV) Adopt escape extinction: insist on providing the target food until the child accepts it; this helps avoid escape from the eating task through refusal. (V) Conduct parent training: ensure consistent implementation of intervention measures in the home environment (50,51).

With validation by numerous studies, ABA-based interventions for problematic eating behaviors are considered the optimal approach. However, long-term follow-up studies on their long-term effects and impacts on quality of life are necessary.

Nutritional and dietary approaches

Elimination diets

The gluten-free and casein-free (GFCF) diet, which eliminates all foods containing wheat (gluten) and milk (casein), is the most common elimination diet. Its theoretical basis is the “opioid excess hypothesis”. The hypothesis suggests that peptides from gluten and casein decomposition enter the brain as a result of increased permeability of the intestine. These peptides produce opioid-like effects, leading to ASD behaviors (52).

Evidence on the efficacy of elimination diets is conflicting. Although some studies and reviews have reported potential benefits, such as behavioral improvements, alleviation of GIS, and better quality of life in some children (28,53), particularly in children with comorbid gastrointestinal conditions (54), some high-quality studies, such as a 2019 Cochrane systematic review and another meta-analysis, have reported contradictory findings. Currently, there is insufficient high-quality evidence supporting the implementation of the GFCF diet to alleviate core ASD symptoms. In addition, this diet may lead to nutrient deficiencies, such as calcium and vitamin D deficiencies. It also imposes a burden on family implementation (55,56). Therefore, the current consensus has clear guidelines on not recommending the GFCF diet as a routine intervention. They recommend reserving it for specific individuals and conditions, such as those with clear clinical evidence of GIS or food intolerance (28,57). A short-term, individualized trial should be conducted under a professional dietitian’s supervision.

Nutritional supplements

Probiotics/prebiotics

Based on the gut-brain axis theory, probiotics and prebiotics interventions demonstrate relatively consistent beneficial effects on GIS among children with ASD. Studies have shown that their supplementation consistently improves GIS in children with ASD, but their direct effect on FS remains unclear (58). Some studies found a positive correlation between GIS and behavioral symptom improvements (59). A systematic review found gut microbiota-based interventions like probiotics, synbiotics, and fecal microbiota transplantation (FMT) to have significant health benefits. These interventions produced “small but clinically relevant” improvements in ASD-related behavioral symptoms, with FMT and compound probiotics exerting the most significant effects (60).

However, meta-analyses focusing on core symptoms like social interaction, communication, and stereotyped behaviors have more conservative conclusions. The findings did not clearly support the effectiveness of probiotics for these core symptoms and found existing evidence insufficient. Many studies also have a risk of bias (61,62). Therefore, microbial intervention has potential as an adjuvant therapy but not as a first-line core treatment.

Omega-3 fatty acids

Omega-3 fatty acids mainly include docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA). Supplementing omega-3 can effectively increase its levels in ASD children. However, the vast majority of randomized controlled trials found no significant clinical improvements in ASD core symptoms or in terms of inflammatory markers with omega-3 supplementation (63,64). A recent umbrella review provided specific findings wherein omega-3 supplementation significantly alleviated certain symptoms, such as multilingual jargon and hyperactive symptoms, in children aged 8 years or younger (65); however, it had no benefit on stereotyped behaviors. Similarly, another recent umbrella review that analyzed 24 meta-analyses supported potential benefits of omega-3 fatty acids and vitamin D in attention deficit hyperactivity disorder treatment (58); however, it showed no evidence supporting their efficacy in ASD.

Vitamins/minerals

Supplementation for specific nutrient deficiencies (e.g., vitamin D, B6, iron, and zinc) identified by laboratory testing is necessary for maintaining overall health and metabolism. A meta-analysis reported that supplementing vitamin B6 may enhance emotional regulation and reduce stereotyped behaviors in children with ASD (66). However, this is to resolve the deficiency and not a specific therapy that directly addresses FS.

Conclusions

Taken together, the feeding problems of children with ASD are complex clinical issues with abnormal sensory processing as the core. It is further complicated by multiple intertwined factors, such as microbiota-gut-brain axis dysfunction, increased food allergy/intolerance risk, and inherent behavioral-cognitive characteristics. In terms of intervention strategies, ABA is recognized as an effective approach for addressing feeding difficulties in children with ASD, as it can directly improve feeding behaviors and broaden food acceptance. Although various dietary adjustments and nutritional supplements (e.g., the GFCF diet, probiotics, omega-3, and vitamins) have certain limitations, they may bring additional benefits to children with specific ASD presentations. These include patients with comorbidities like clear GIS, food intolerance, or specific nutrient deficiencies. Notably, their use requires careful consideration, and they must be a part of an individualized comprehensive management plan rather than a standalone intervention. Furthermore, their use should be closely monitored and comprehensively evaluated by professionals. A “one-size-fits-all” approach must be avoided.

For ASD dietary problems, the ideal intervention model involves multidisciplinary team collaboration. Herein, a pediatrician or developmental-behavioral pediatrician serves as the coordinator and multiple professionals participate, with dietitians providing dietary assessment and nutritional support, occupational therapists addressing oral sensory-motor disorders, and behavior analysts or psychologists designing and implementing behavioral intervention plans. Together, they provide comprehensive, individualized, and sustainable support for children with ASD and their families.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.