A nutrition education intervention to improve eating behaviors of children with autism spectrum disorder: Study protocol for a pilot randomized controlled trial

Abstract

Autism spectrum disorder (ASD) is a developmental disorder that affects communication and social behaviors. Children with ASD often experience mealtime behavior challenges and selective eating behaviors. They also tend to consume fewer fruits and vegetables and more high-energy dense foods, compared to neurotypical peers. A nutrition intervention was designed to prevent the development of feeding disorders and the long-term negative health impacts associated with poor dietary intake. This randomized controlled trial will evaluate the feasibility and preliminary efficacy of the nutrition education intervention for children with ASD and their parents through the Early Intervention (EI) services. We will recruit EI providers and parent-child dyads (n = 48) from EI programs, and randomly assign them into Autism Eats intervention (n = 24) or enhance usual care (EUC) comparison group (n = 24). The Autism Eats is 10 weekly sessions delivered individually as part of EI, while the EUC group will receive only 1 nutrition education session and then weekly parent handouts. The Autism Eats integrates ASD-specific feeding strategies and behaviorally-focused intervention strategies such as goal setting. Feasibility indicators include reach/participation, attrition, completion, fidelity, compatibility, and qualitative participant feedback. Outcome measures include dietary intakes and mealtime behaviors of children with ASD using 3-day food records and a validated questionnaire, the Brief Autism Mealtime Behavior Inventory (BAMBI). We will examine whether there are differences in children’s food intakes, variety, diet quality, and mealtime behaviors between Autism Eats and EUC groups at post-intervention and 5-month follow-up assessment. This study will provide critical data to inform a full-scale randomized controlled trial.

1. Introduction

The prevalence of autism spectrum disorder (ASD) has steadily risen over the past few decades, with current estimates at one in 44 children affected in the United States [1]. Children with ASD are more likely to experience food selectivity and problematic mealtime behaviors compared to their neurotypical peers [2–5]. In addition, poor dietary intake in children with ASD is a growing concern as they tend to consume few fruits and vegetables and excess calories from high-energy dense foods such as sugar-sweetened beverages and highly processed snacks [6], which may contribute to developing diet-related non-communicable conditions such as obesity [7–10].

Food selectivity in children with ASD can be categorized into mild, moderate, and severe based on the diversity and quality of foods [11]. Feeding clinics are available for children who are clinically diagnosed with feeding disorders (e.g., avoidant/restrictive food intake disorder, ARFID) or severe food selectivity. However, there is a lack of available programs that address mealtime behaviors and healthy eating for children with ASD who have mild to moderate food selectivity or nutritional issues [12,13]. One of the few interventions is the Autism MEAL Plan, which is a parent training program for caregivers with a three- to eight-year-old child with ASD and moderate food selectivity and showed promising results in improving mealtime behaviors and quantity of food consumed by participants [11,12]. Further research is needed to determine whether nutrition education interventions can improve both mealtime behaviors and diet quality in youth with ASD who do not have feeding disorders or severe food selectivity.

The first few years of life are critical for developing food preferences and dietary habits [14]. Poor dietary habits established in early childhood can persist into adulthood and increase the risk for developing obesity and other chronic diseases [15]. As problematic mealtime behaviors and feeding issues can start early in children with ASD [16], early intervention is paramount to prevent further escalations and to improve dietary behaviors. Early childhood nutrition interventions, especially those using behavioral theories such as social cognitive theory, have shown promise for improving eating behaviors and slowing weight gain in neurotypical children [17,18], but their effectiveness is understudied among children with ASD. Further research is needed to determine whether an early childhood nutrition intervention emphasizing both ASD-specific mealtime behaviors and healthy eating behaviors can mitigate problematic eating behaviors and support the development of healthy dietary habits for children with ASD.

The study protocol describes a pilot randomized controlled trial to evaluate the feasibility and preliminary efficacy of the Autism Eats nutrition education program that integrates nutrition ASD-specific feeding strategies and behaviorally-focused nutrition education for improving dietary intake and mealtime behaviors of children with ASD. Objectives are (1) to test the feasibility of enrollment, implementation, and evaluation of the Autism Eats intervention; and (2) to evaluate dietary intake and problematic mealtime behaviors of children with ASD as primary outcomes. Weight status and parental feeding behaviors are explored as secondary outcomes.

2. Methods

2.1. Study design

This is a pilot randomized controlled trial (RCT) with Autism Eats intervention and “enhanced usual care (EUC)” comparison arms (Fig. 1). By collaborating with the local Early Intervention (EI) services (Part C of Individuals with Disabilities Education Act), we will recruit children with ASD and their parents (48 dyads). After baseline data collection, each dyad will be randomly assigned to either Autism Eats (n = 24) or EUC condition (n = 24). EI providers will implement the Autism Eats intervention or EUC, which will be integrated within the regular sessions of the EI services. Outcome data collection will be conducted at three time points: baseline, post-intervention, and 5-month follow-up. All study materials are available in English and Spanish. The study protocol has been approved by the university Institutional Review Board.

Fig. 1.

Study design.

2.2. Community partner

Children diagnosed with ASD are eligible to be enrolled in the EI program. EI services are offered to eligible children (age birth to 36 months) who are at risk for developmental disabilities or delays, including ASD. We established our partnership with Early Steps, the statewide EI system in Florida, by collaborating with them on our formative research and pilot testing of the intervention. Because Early Steps has no income requirement and does not charge families for services, we anticipate reaching families from diverse socioeconomic backgrounds. An Early Steps liaison will be supported by the project and assist with EI provider and dyad recruitment. The liaison will be one of the service coordinators who is trained in early childhood and has experience in managing Early Steps providers.

2.3. Intervention material refinement

The Autism Eats materials were initially developed and pilot-tested in 2020 [19] and further refined based on EI provider and parent feedback. Nine EI providers and seven parents of children with ASD provided their feedback on the amount of preparation time, clarity of instructions, the amount of time allocated to each lesson, relevance of lesson activities to the program goal, feasibility of completing the lesson activities, satisfaction with the content, and relevance to improving nutrition in children with ASD. Providers and parents rated each component on a 5-point Likert scale with responses from “strongly disagree” to “strongly agree.” They also provided written feedback. The mean of overall scores for all lessons was 4.4 ± 0.5 from the EI providers and 4.1 ± 0.2 from the parents, respectively, indicating that both providers and parents perceived the intervention materials as feasible and acceptable. Based on the written suggestions from both EI providers and parents, the Autism Eats intervention materials have been further revised.

2.4. Participants

2.4.1. Inclusion and exclusion criteria

Eligible providers must be employees of EI service agencies who work with EI enrolled children with ASD at the time of recruitment. Eligible children should be enrolled in EI services and clinically diagnosed with ASD by a licensed practitioner. As the study takes up to 5 months for each family to complete before children graduate from the Early Steps program, children must be 31 months or younger at the time of study enrollment. Parents must be 18 years or older and speak fluent English or Spanish. Children are excluded if they are exclusively breastfed, take medications that may interact with appetite (i.e., Olanzapine, Risperidone, or Aripiprazole), have severe GI conditions such as irritable bowel syndrome, are clinically diagnosed with feeding disorders, have severe food selectivity (accepting five or fewer total food items) [11], or have other serious medical comorbidities such as cancer. Dyads who have previously participated in a similar nutrition intervention study are excluded. As the study requires parent participation, children who are receiving EI services at a daycare setting are also be excluded.

2.4.2. Recruitment

This study includes a two-level plan for recruiting EI providers and parent-child dyads. EI providers will be recruited through the Early Steps programs in Florida. EI providers from Hillsborough and Polk Counties will be recruited first, and then the recruitment will be strategically expanded to other counties until the target sample size is reached. There is no eligibility restriction on location if dyads reside in Florida. The Early Steps liaison will assist with participant recruitment. A recruitment flyer describing the research study will be distributed by email, and research staff will conduct a brief information session (in-person or virtually) for EI providers. At the second level, a parent recruitment flyer will be distributed through the EI services, and EI providers who agree to participate will inform potentially eligible parents about the study and recruit them to contact the research team. Recruitment materials will be distributed repeatedly until the target sample size is met.

2.4.3. Screening, enrollment, and randomization

Children’s ASD diagnosis will be screened by the Early Steps liaison who has access to ASD diagnosis records. The liaison will record ASD diagnosis including who diagnosed ASD (e.g., psychologist, neurologist), date of the diagnosis, level of ASD (i.e., level 1, 2 or 3), autism diagnostic observation schedule (ADOS) score if available, and child’s age at diagnosis on a tracking sheet. The ASD diagnosis tracking sheet will only include the de-identified numeric (ID) numbers without children’s names. Any protected health information (PHI) from those who are not eligible to participate or refuse to participate will be destroyed immediately. Written informed consent will be obtained from EI providers and parents.

After baseline data collection, parent-child dyads (n = 48) will be randomized into either the Autism Eats (n = 24) or the EUC (n = 24) comparison condition. Random assignment will be conducted by the study statistician utilizing a random number generator.

2.4.4. Retention plan

Training sessions will be provided for EI providers and monetary incentives will be distributed. Training session include a 90-min presentation prior to the intervention implementation and a 90-min mid-training session after Week 5. During the implementation phase, trained assistants will be available to support providers by answering questions and reviewing lesson activities together as necessary. This on-going support has been effective for retaining the providers in our formative study [19]. To maximize retention for parent-child dyads, incremental monetary incentives and regular phone/text/email reminders will be used. Social media components will also encourage participant engagement in between post-intervention and follow-up assessment periods. Our research team will contact all parents one-week and 24-h prior to their scheduled assessment(s). To potentially improve retention of dyads in the EUC condition, we will provide an option to receive the Autism Eats intervention materials once they complete the study.

2.5. Intervention

2.5.1. Autism Eats intervention

The Autism Eats intervention consists of ten 25-min weekly lessons plus two 25-min monthly booster sessions (Table 1) [16,19–22]. Parent-child dyads will receive the lessons through their regular EI services implemented by EI providers. The intervention integrates ASD-specific feeding strategies (e.g., repeated exposure, food chaining, and making regular mealtime routines), and behaviorally-focused nutrition content and activities (e.g., goal setting and healthy meal planning). Social cognitive theory is the theoretical basis of the behavioral change strategies [23].

Table 1.

Components of Autism Eats nutrition intervention

Components	Example Activity
Lessons: topics and objectives
L1. Feeding Milestones  Understand feeding milestones for infants and toddlers.	Feeding milestone screening
L2. Sensory Properties of Foods: Taste, Flavors, & Textures  Explore the role of senses as it pertains to taste, flavor, and texture in the development of food preferences.	Exploring senses activity
L3. Introducing New Foods  Understand and utilize strategies for introducing new foods into the child’s diet.	Repeated exposure
L4. Balanced Eating and Nutrition  Examine the benefits of healthy foods and learn how to integrate them in a daily menu.	Creating a meal plan
L5. Food Allergies, Special Diets, and Supplements  Understand food allergies, special diets, and supplements that may affect or be beneficial for their child.	Food sorting activity
L6. Beverages  Identify age-appropriate beverages and examine the benefits of healthy beverage selection.	Setting a family goal
L7. Mealtime Routines and Schedules  Establish a mealtime routine and create a consistent eating schedule.	Mealtime routine
L8. Restructuring Food Environment  Develop strategies to make healthy choices and have a pleasant experience while eating outside the home.	Vegetable and yogurt dip
L9. Hunger and Fullness Cues  Build confidence in recognizing when the child is hungry or full.	Responding to hunger and fullness cues
L10. Maintaining Healthy Nutrition  Create long-term strategies to sustaining healthy eating habits.	Long-term goal setting
Booster L1. Where are We Now?  Review previous lessons and discuss barriers for healthy eating and ways to overcome them.	Overcoming barriers
Booster L2. Celebrating Achievements  Celebrate healthy eating achievements with positive reinforcement.	Positive reinforcement practice
Static website repository Project description and lesson resources	Handouts available online
Social media Private social media (Facebook) group page	Extra motivational resources and parent engagement

L = Lesson

EI providers are accustomed to providing individualized care via a coaching model and this model is applicable for Autism Eats activities (e.g., educating caregivers on the session content and using each child’s favorite and/or culturally appropriate foods for lesson activities). The two-monthly booster sessions are designed to reinforce the progress achieved from previous lessons and to expand food choices. The provider manual also contains instructions for a telehealth setting if the EI service is delivered virtually due to the COVID-19 pandemic restriction or any other reasons. There is no content difference between in-person and telehealth versions. Parents will have access to a static website repository where they can download lesson resources such as handouts, and each parent participant will be invited to a private Facebook group once they complete Lesson 10 to continue active engagement until the end of the study period.

2.5.2. Enhanced usual care

To maintain scientific control for attention threats and increase rigor, EUC group will receive a one-time 25-min nutrition lesson embedded in the regular EI services in the first week after randomization, and then receive 9 weekly parent handouts and two monthly booster handouts that incorporated materials from We Can! program, which is a publicly available program that contains evidence-based nutrition education materials [24]. After the first session, providers will briefly review weekly handouts with parents at the beginning of their regular EI services, and parents will also receive the link to the We Can! Website. Each parent will be invited to a private Facebook EUC group once they receive the Week 10 materials, to continue active engagement and provide comparable intervention dose.

2.6. Measures

2.6.1. Feasibility indicators

2.6.1.1. Reach/participation.

We will track EI providers and parent-child dyads reached by the recruitment methods. The number of providers and dyads reached and those who respond to the invitation will be documented, and the number of people who agree to participate will be recorded, using a master tracking log. Our target participation rate is >60% calculated by the percentage of providers and dyads enrolled in the study divided by the number of those who are eligible and contacted the research team (Table 2).

Table 2.

Measures and benchmarks for feasibility indicators

Indicators	Measures	Benchmarks
Reach/Participation	Tracking log	Number of potential participants reached; >60% of eligible providers and parent-child dyads who contacted the team with research interests agree to participate
Attrition	Tracking log	<20%
Completion	Lesson completion checklist, assessment tracking sheets, and attendance records	>75%
Fidelity	Lesson observations (5-point scale per lesson)	Mean >3.0
Compatibility	Exit survey (5-point scale); provider exit interviews	Mean >3.0; Qualitative coding: positive feedback
Participant feedback	Provider and parent interviews	Qualitative coding: positive feedback

2.6.1.2. Attrition.

Using the participant tracking log, we will monitor and document those who drop out and reasons for dropping out, and clearly indicate whether it was the provider or the parent who decided to drop out. EI providers and parents work closely together within their regular EI services, regardless of their participation status in the study. If either an EI provider or parent wish to withdraw from the study, it is likely that they will discuss that decision together. The primary data to calculate attrition rate will be based on the number of dyads. Our target attrition rate is 20% or below.

2.6.1.3. Completion and fidelity.

We will use an intervention completion checklist adapted from our pilot study [13,25]. Percent completion rate will be calculated for intervention lessons, material distribution, and social media component implementation. EI providers will record lesson completion right after each lesson on the checklist form provided in the provider manual, and assistants will track social media component implementation. In addition, two intervention sessions per family in the Autism Eats intervention group will be observed by an assistant and checked for fidelity using an observation form [13,25].

2.6.1.4. Compatibility.

We will examine whether the interventions fit within the context of EI services. Compatibility of the intervention content and activities for different levels of children’s autism features and symptoms will also be examined with qualitative exit interviews with the EI providers at 5-month follow-up assessment [25,26]. For those who are unavailable to participate in the interview, we will send exit interview questions online via REDCap. There is a 4-question compatibility scale adapted from the validated Appropriateness (defined as the perceived fit, relevance, or compatibility of an intervention for a given practice setting, provider, or consumer) scale [27] and open-ended questions.

2.6.1.5. Participant satisfaction and overall feedback.

General satisfaction and feedback questions will be asked to providers and parents during the semi-structured exit interview. Trained assistants will conduct the interviews virtually or by phone, which will be video/audio-recorded. All files will be converted to audio-only files and transcribed verbatim using an external transcription service.

2.7. Outcomes

2.7.1. Dietary intake

The summary of outcome measures is shown in Table 3. Dietary intake are assessed using 3-day food records, which is a standardized self-report dietary intake assessment and has been used in studies to assess children’s dietary intake, including those with ASD [28–30]. A detailed instruction and a step-by-step training video on how to complete a 3-day food record will be sent to parents via email. The video has detailed instructions on each food record section and how to accurately measure food amounts with portion size guidelines. Parents will record children’s dietary and supplement intake for three days (2 weekdays and 1 weekend day) using a food log, and then a research assistant will schedule an interview with the parents to collect missing information and confirm information recorded on food logs [31,32]. The assistant will then manually enter the data into the Automated Self-administered 24-Hour Dietary Recall (ASA24^®), a dietary assessment tool used to analyze nutrient and food group estimates [33]. We will focus on average daily dietary intakes by food groups (i.e., fruits and vegetables) and daily food variety [22,34,35]. Daily food variety will be assessed by calculating the average number of unique food items consumed over three days by each child [35]. Supplements will be excluded from the food variety score. The ASA24 system generates data on number of unique food items and number of supplements. For all children aged 2 years or older, diet quality will be examined using the Healthy Eating Index total and sub-component scores (adequacy vs. moderation food categories) [36]. Children’s dietary data will be collected at baseline, post-intervention, and at 5-month follow-up.

Table 3.

Measures or expected outcomes and time for efficacy

Outcomes		Measures / Expected Outcomes	Time
Primary	Dietary intake	3-day food records (ASA24™) / Food intakes (e.g., fruit and vegetables ↑), daily food variety ↑, diet quality (e.g., HEI fruit ↑, vegetables ↑, plant protein ↑)	T1, T2, T3
	Mealtime behaviors	BAMBI via REDCap / Problematic behaviors ↓	T1, T2, T3
Secondary	Weight status (height and weight)	A stadiometer/ruler and a scale; birth to 36 months, weight-for-length percentiles based on the CDC growth chart. BMI for parents	T1, T3
	Feeding behaviors	Child Feeding Questionnaire (CFQ)	T1, T2, T3
Participant characteristics	Demographics	Survey via REDCap	T1
	Health conditions	Family history of ASD and comorbid health conditions such as epilepsy, sleep disorders, and anxiety disorders via REDCap	T1

T1: baseline assessment; T2: post-intervention assessment; T3: follow-up assessment at 5 months.

ASA24 = Automated Self-administered 24-Hour Dietary Recall; REDCap = Research Electronic Data Capture; BMI = Body mass index; CDC = the Centers for Diseases and Control Prevention; BAMBI = Brief Autism Mealtime Behavior Inventory; ASD = Autism Spectrum Disorder

2.7.2. Mealtime behaviors

Parents will complete the Brief Autism Mealtime Behavior Inventory (BAMBI) [37], a validated measure to assess problematic mealtime behaviors, via REDCap. The BAMBI contains 18 questions using a 5-point scale for reporting the frequency of a behavior. Scores range from 5 to 90 and higher scores indicate higher problematic mealtime behaviors. Each question also has a yes or no option for parents to indicate if they perceive the behaviors as problematic. The BAMBI subscales are Limited Variety, Food Refusal, and Features of Autism. The survey takes ~5 min to complete and will be used at baseline, post-intervention, and 5-month follow-up.

2.7.3. Child feeding practices

The Child Feeding Questionnaire (CFQ) is a validated self-report instrument that assesses parental feeding practices in terms of feeding beliefs, attitudes, and practices regarding child feeding and obesity proneness [38]. The CFQ has 31 questions using a 5-point Likert scale that assesses 7 factors, including parental control over child feeding (restriction, monitoring, and pressure to eat) and attitudes for child feeding (perceived parent weight, perceived child weight, concerns for child weight, perceived responsibility) [38].

2.7.4. Anthropometric measurements

Research assistants will assess height and weight of parents and children at baseline and at 5-months follow-up with a portable stadiometer (Seca 213) and a digital weight scale (Tanita WB-800AS) following the standardized protocol adapted from the NHANES Anthropometry Procedures Manual [39]. In this study, a standard protocol of height and weight measurement is available for both in person and virtual setting. Virtual participants will receive a metal ruler and a bathroom scale (BalanceFrom LLC. Digital Body Weight Scale) via mail, and an assistant will meet virtually with the family to measure their height and weight. As a standard weight status assessment for birth to 36 months, weight-for-length percentile based on the Center for Disease Control and Prevention (CDC) growth chart will be calculated [40].

2.7.5. Participant characteristics, covariates, and potential moderators

Questions on demographic characteristics and history of other illnesses and comorbid health conditions such as epilepsy, sleep disorders, and anxiety disorders will be asked at baseline via REDCap. Parent BMI will be calculated to be used as a covariate in data analysis. If height and weight information is available for the spouse of the participating parent or the other biological father/mother, it will be self-reported by the parent.

2.8. Data management and security

Data will be collected and maintained in accordance with legal and ethical standards and centrally stored in a master database. REDCap is a secure, HIPAA compliant web application used to collect data. Data files will be stored in a password protected secure database in a locked research office. Only authorized research staff members can have access to and possess the data files to conduct data analyses. Data files will be maintained for 5 years after completion of the research and will be destroyed based on the university data disposal procedure and regulations.

2.9. Data sharing plan

Data will be shared adhering to the National Institute of Mental Health (NIMH) regulations that are required for research with individuals with ASD. De-identified data will be submitted to the NIMH Data Archive (NDA) that allows other authorized researchers to have access to the data for secondary analysis. We will refer any individual data request to the NDA system or make the data and associated documentation available to users under a data-sharing agreement based on the NIH data sharing policy and implementation guidance. Our aggregated data and results will also be available through ClinicalTrials.gov (Registration No. NCT05194345) following their requirement and guidelines.

2.10. Data analytical plan

2.10.1. Sample size and power

Primary outcomes are dietary intake (average daily intake of fruits and vegetables, food variety, and diet quality) and mealtime behaviors. The sample size and power calculation were not formally performed. Pilot RCTs do not have the same objectives as a main trial and using formal power considerations is usually not necessary [41]. An n = 25 pilot RCT sample size (per arm) with a small effect size (0.2) and n = 15 per arm with a medium effect size (0.5) are recommended. Based on previous nutrition education literature and from our own experiences, we conservatively anticipate a small to medium preliminary effect size (approximately 0.3–0.4), and our target sample size of 48 parent-child dyads satisfies pilot RCT sample size recommendations.

2.10.2. Feasibility indicators

Descriptive statistics (e.g., mean, standard deviation, range, percentage) will be used to examine data distributions and frequencies of feasibility indicators including reach/participation, attrition, completion, fidelity, and compatibility. We will further examine potential factors (e.g., demographic characteristics) that may affect the feasibility using crosstabs and Chi-square analysis. Qualitative data from provider and parent interviews will be transcribed verbatim and thematic analysis will be performed to determine patterns in the data.

2.10.3. Outcome evaluation

2.10.3.1. Primary and secondary outcome analysis.

We will examine variance and effect sizes of key outcomes from baseline to post-intervention, as well as baseline to 5-month follow-up. We will first assess whether Autism Eats and EUC groups are balanced through randomization in terms of demographic variables at baseline. For each outcome variable, we will use boxplots and scatterplots with confidence intervals along with descriptive statistics to visually inspect the changes across pre-intervention, post-intervention, and 5-month follow-up by group. In addition, longitudinal analysis will be conducted for each outcome variable with time as a within-subject factor and treatment condition as a between-subject factor to examine the difference in changes between Autism Eats and EUC groups. Demographic factors such as child sex will be controlled as covariates if they are unbalanced between groups.

2.10.3.2. Exploratory analysis.

We will explore whether weight-for-length percentile and weight status of children differ between Autism Eats and EUC groups at 5-month follow-up. Descriptive statistics will be used to summarize weight-for-length percentile and weight status by group, and effect sizes will be computed.

3. Discussion

The current study will provide critical information on the feasibility and preliminary data on efficacy of a nutrition education intervention in improving eating and mealtime behaviors of children with ASD enrolled in the EI services. EI services help children with developmental disabilities learn new skills by providing family training, counseling, and home visits with occupational/physical/speech therapy and other services including nutrition [42]. Nutritional assessment is listed as part of the EI services, yet there is a lack of availability and utilization of nutritional resources and intervention within the EI services. Findings from previous research support the need for interventions early in childhood to increase variety and promote healthy eating among children with ASD [3]. The current study utilizes a rigorous RCT design to address early childhood feeding practices, nutrition, and home environment critical to establishing eating habits that influence healthy growth for children.

There are some potential limitations. As a pilot study, the target sample size is not expected to have adequate power to find statistically significant groups differences. We also recognize that self-report dietary records (parent proxy) may not generate the most reliable data. Considering specific challenges that children with ASD experience daily and their emotional vulnerability, we selected a less invasive nutrition assessment method that has been frequently used in this population [43]. Having both in-person and virtual height and weight assessment options using different measurement devices may also produce less accurate and reliable data. Despite these potential limitations, the current study uses a novel approach to mitigate ASD-specific mealtime problems and to promote development of healthy eating habits among children with ASD integrated into the EI services. As the EI service system is already established as a sustainable intervention delivery channel, it provides a unique opportunity to reach the early childhood age group and their parents.

4. Conclusions

Results from this pilot randomized controlled trial will provide critical information for future nutrition intervention studies for young children with ASD. If shown to be efficacious, integrating a nutrition intervention into the EI services may be a cost-effective way to provide an important service to this population with significant policy and practice implications. Given the high rate of mealtime behavior problems, pediatric obesity, and the overall increased risks for diet-related chronic conditions, the significance and public health relevance of this project is substantial. Further research utilizing a larger scale randomized controlled trial with a longer duration and diverse samples with different socioeconomic and cultural backgrounds is warranted.