An Integrated Nutrition Intervention to Promote Healthy Eating Habits for Children With ASD: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for An Integrated Nutrition Intervention to Promote Healthy Eating Habits for Children With ASD (NCT05194345), a randomized na study. The primary outcome is Change in Fruit and Vegetable Intake From Baseline, as Measured by the Daily Amount of Fruit and Vegetables Consumed by Children.

2. Clinical Question

The study addresses nutrition and feeding challenges among children with autism spectrum disorder. The registry describes an intervention that combines autism-specific feeding strategies with behaviorally focused nutrition content. The comparison is between that structured program and enhanced usual care materials, not between the intervention and an absence of all nutrition information.

The principal outcome is a change from baseline in children’s daily fruit and vegetable intake. The intended comparison is between groups at post-intervention, 10 weeks from baseline, and at 5-month follow-up. The registry record does not report the numerical group estimates or the between-group contrasts, so it does not establish how large any change was or how precisely the groups differed.

3. Trial Design

The registry describes a randomized, single-masked, parallel-group trial with an enrollment of 132. The study is marked completed. Its start date is 2022-01-27, and its primary completion date is 2024-05-24. The phase is listed as NA. The sponsor is the University of South Florida.

Participants were assigned to Autism Eats nutrition intervention or We Can! enhanced usual care control. Autism Eats is described as 10 lessons plus 2 booster sessions. Its content includes autism-specific feeding strategies such as repeated exposures, food chaining, and regular mealtime routines, alongside behaviorally focused nutrition activities. The enhanced usual care group received materials from an established online resource, described as available in English and Spanish. Further detail about implementation is not reported in the registry record.

“Single” masking indicates that the registry identifies the design as single-masked, but it does not specify which participants, staff, or assessors were masked. The method used to generate the random allocation sequence, the procedures used to conceal assignments, and any steps to limit contamination are not reported in the registry record. These details matter when judging the risk of bias, particularly for a behavioral intervention in which participants may know which program they receive.

The primary outcome was assessed at baseline, post-intervention (10 weeks from baseline), and 5-month follow-up. Other registered outcomes include change in child weight-for-length or BMI percentile and change in child feeding practices measured using the Child Feeding Questionnaire. The record lists 7 outcome measures overall. The full list of measures and their results is not reported in the registry record.

The enrollment figure describes the trial record; it should not be treated as the number with complete primary outcome data. The numbers included in the primary analysis and the extent of follow-up completion are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: One family withdrew due to the participant's own medical reason. Providers were trained and implemented the interventions, but outcomes of the study are assessed mainly for children and some for parents. The target number of providers (n=30 targeted) was met (n=32 enrolled).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Outcome data were only assessed for children. Mean changes of daily amount of fruits and vegetables were calculated. Daily mean intake was used because the data were collected from 3-day food records. Linear mixed-effects models (LMMs) were conducted using restricted maximum likelihood (REML) estimation to evaluate the effects of the intervention and time interaction on the outcome, adjusting for child age, sex, race, and ethnicity.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: 25 parent-dyads (50 individuals) per condition: 25 children per condition and 25 parents per condition. 32 providers implemented the intervention or the enhanced usual care (EUC) control. 14 providers only implemented Autism Eats, 12 providers only implemented EUC, and 6 providers were trained in both programs and implemented the programs based on the random assignments. Outcomes of the providers were not assessed.

Only the first 6 of 8 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Change in Fruit and Vegetable Intake From Baseline, as Measured by the Daily Amount of Fruit and Vegetables Consumed by Children, measured over baseline, post-intervention (10 weeks from baseline), and 5-month follow-up.

The primary endpoint is change in children’s daily fruit and vegetable intake from baseline, measured in cup equivalents. Parents reported intake using 3-day food records. The registry describes calculating a mean daily intake from those records, then comparing change from baseline between Autism Eats and enhanced usual care at post-intervention and at 5-month follow-up.

Food records provide a way to estimate intake across more than one day rather than relying on a single day’s report. Averaging the recorded daily intake is intended to represent usual intake more stably than an isolated measurement. In general, repeated measurements can also describe whether change is maintained beyond the immediate post-intervention assessment. However, the accuracy of a parent-reported food record depends on what is recorded and how well intake is captured; information about validation, adherence, or data quality is not reported in the registry record.

The endpoint is a continuous measure, and the registry identifies the reported summary as a least-squares mean with a 95% confidence interval. A least-squares mean is a model-based estimate; it is not necessarily the same as a simple average of all recorded observations. The exact model specification, including any adjustment variables, is not reported in the registry record.

Fruit and vegetable intake is relevant to the intervention’s stated nutrition goals, but the registry record does not define a minimum clinically important change. Without numerical estimates, the size, direction, and precision of the observed changes cannot be described.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: daily fruit and vegetable intake in cup equivalents, summarized as change from baseline. The registry describes estimating mean changes and analyzing the data with linear mixed-effects models using restricted maximum likelihood (REML). It identifies the intended summary as a least-squares mean with a 95% confidence interval. The numerical estimates themselves are not reported in the registry record.

What does a mixed-effects model do?

In general, a mixed-effects model can represent both average patterns across a study group and the dependence among repeated measurements from the same participant. Measurements from one child are often related to one another, so treating every observation as independent may misstate uncertainty. A mixed model can account for this structure, provided the model is specified appropriately. The exact random-effects structure, fixed effects, covariance assumptions, and any adjustment variables used in this trial are not reported in the registry record.

The registry describes measurements at baseline, post-intervention, and 5-month follow-up. A model can use repeated measurements to estimate change over time and compare patterns between randomized groups. The scientific contrast of interest is the difference between groups in change from baseline at the follow-up assessments. The registry’s description states that those changes will be compared, but the exact contrast definitions and model terms are not reported in the registry record. Therefore, the results cannot be independently reconstructed from the posted information.

Why use restricted maximum likelihood?

In general, REML is an approach to estimating variance components in mixed models. It accounts for the estimation of fixed effects when estimating those variance components and is commonly used for this model class. Its use does not, by itself, guarantee that the model is appropriate: the repeated-measure structure, distributional assumptions, and missing-data assumptions also matter. Diagnostic checks and the trial’s rationale for the chosen model are not reported in the registry record.

What does a least-squares mean represent?

In general, a least-squares mean is an estimated mean from a statistical model, often evaluated at specified levels of model variables or averaged over them according to a defined procedure. It can differ from a raw arithmetic mean, particularly when the model adjusts for other factors or when observations are unbalanced. The adjustment variables and calculation details for this trial’s least-squares means are not reported in the registry record. The label alone does not reveal which covariates were included or how incomplete observations contributed.

How should the confidence interval be read?

In general, a 95% confidence interval gives a range of estimates compatible with the data and the statistical procedure under repeated use of its assumptions. It expresses uncertainty about the estimated mean or contrast; it is not a range containing 95% of individual children’s intake values. Its width reflects precision, while its position relative to a null value can inform a test of a contrast. The interval bounds and the tested contrast are not reported in the registry record, so neither precision nor statistical significance can be evaluated for the primary outcome.

What can be learned from a planning illustration?

The registry describes a sample-size planning illustration based on the observed effect, using a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. It describes considering effects expressed as fractions of the observed effect. Such calculations are planning illustrations, not observed power and not evidence that a particular effect occurred. The numerical planning outputs are not reported in the registry record. They also should not be confused with the trial’s original design justification, which is not reported in the registry record.

In general, planning calculations depend on assumptions about the effect to detect, variability, allocation, significance level, and method of analysis. A calculation based on an observed effect is especially dependent on that estimate and does not substitute for a prospective design rationale. The registry does not report the original sample-size justification or the assumptions used when setting the enrollment target.

The registry lists zero analysis records. This means the posted record does not supply a detailed analysis entry from which to assess the model implementation, planned contrasts, or result reporting. It does not justify filling those gaps by assuming common practices. The stated LMM and REML approach is the available description; additional analytical details are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 7 outcome measures. The primary outcome is evaluated at post-intervention and 5-month follow-up, in addition to baseline measurement. Examining multiple outcomes or time points can increase the chance of at least one apparently unusual result when many statistical tests are conducted. In general, multiplicity procedures can control or otherwise address that issue, but the appropriate approach depends on which comparisons were prespecified and how they were prioritized.

The registry record does not report a multiplicity adjustment, a hierarchy of outcomes, or a rule identifying a single primary time-point contrast. It also lists zero analysis records, and numerical outcome results are not reported. Accordingly, it is not possible to determine whether statistical testing was conducted, which comparisons were considered confirmatory, or how any multiplicity was handled.

The record describes a linear mixed-effects analysis using REML, but does not describe the handling of missing outcome measurements. Missing-data assumptions, imputation methods, sensitivity analyses, and reasons for incomplete follow-up are not reported in the registry record. A mixed model can use incomplete repeated measurements under particular assumptions, but the use of a mixed model alone does not establish how missingness was addressed or whether its assumptions are plausible.

The record says that the primary outcome data were assessed for children. It does not report the numbers of children contributing measurements at baseline, post-intervention, or follow-up, or whether all randomized children were included in the analysis. Intention-to-treat, per-protocol, and other analysis populations are not reported in the registry record. The registry’s total enrollment should not be mistaken for the number analyzed at each time point.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry lists serious adverse events, other adverse events, and deaths separately for children, parents, and providers in the intervention and control groups, with an additional provider group trained for both intervention and enhanced usual care control. The listed serious adverse event counts are 0 in each group, and the listed deaths are 0 in each group. The other adverse event counts are also 0 in each group.

The at-risk counts differ across the provider groups: 14 for Autism Eats providers, 12 for enhanced usual care providers, and 6 for providers trained for both. The child and parent groups each list 25 at risk in each arm. These are the denominators attached to the registry’s adverse-event entries, not the trial’s total enrollment. The record does not explain how these safety-group counts relate to enrollment or to the primary outcome analysis population.

Zero events in the listed groups means that no events of those categories are recorded for the stated at-risk counts in the registry record. It does not establish that no events occurred outside those listed groups or that the intervention has no safety concerns. Further detail about event definitions, collection procedures, follow-up duration for safety, and the relationship between the safety groups and all enrolled participants is not reported in the registry record.

12. What the Result Does, and Does Not, Mean

The registry establishes that a randomized, parallel-group study was conducted to compare Autism Eats with enhanced usual care, and it defines a primary outcome based on children’s daily fruit and vegetable intake. It also reports a modeling approach and gives safety-event counts for specified groups. These are useful descriptions of the study’s aims and record, but they are not a substitute for numerical primary outcome results.

The record does not report the estimated change in intake in either arm, the estimated difference between arms, confidence interval limits, or p-values. It is therefore not possible to say whether the posted results show an increase, a decrease, or a difference between the groups, or whether any estimate is precise. A completed trial status and a planned analysis method do not supply those missing results.

Even if an estimate were available, statistical evidence and practical importance would be different questions. In general, a small effect can be estimated precisely, while a potentially meaningful effect can remain uncertain if the interval is wide. The registry record does not report a threshold for an important change in intake, so clinical or practical importance cannot be judged from the record.

The trial’s findings, if reported elsewhere, would also need to be interpreted in light of the population and implementation described in the registry. The target population is children with autism spectrum disorder in Part C Early Intervention services in Florida; parents and providers are also represented in safety listings. The extent to which results would apply to other populations or service settings cannot be established from the registry record.

The appropriate conclusion from the posted record is limited: it documents the study design, outcome definition, and stated analysis method, but does not provide the numerical primary outcome findings needed to evaluate the between-group result.

13. Why This Trial Matters Statistically

This trial presents a statistically meaningful question because the intervention combines feeding approaches tailored to autism spectrum disorder with behaviorally focused nutrition activities. Its randomized parallel-group design is intended to compare that package with enhanced usual care, while the primary endpoint focuses on daily fruit and vegetable intake. The registry therefore links an intervention question to a measurable continuous outcome and follow-up over time.

The repeated assessments create an opportunity to distinguish change at post-intervention from change at 5-month follow-up. Those comparisons are not interchangeable: one concerns the measurement at 10 weeks from baseline, while the other concerns a later assessment. Clearly reported estimates and uncertainty at both times would help readers understand the timing and persistence of any between-group differences. The numerical estimates are not reported in the registry record.

The analysis description also raises useful methodological questions for clinical researchers. Linear mixed-effects models can address repeated observations, but interpretation depends on the model specification, missing-data assumptions, and clearly defined contrasts. The registry names LMM and REML but does not report enough detail to evaluate those choices fully. This illustrates why a method label alone is not a complete statistical report.

The record also lists multiple outcome measures and two post-baseline assessment times for the primary outcome. Transparent reporting of outcome priority, multiplicity, analysis populations, and missing data would help distinguish prespecified evidence from exploratory findings. Those details are not reported in the registry record.

Finally, the safety listings separate children, parents, and providers, with differing at-risk counts across groups. Reading those denominators carefully avoids treating them as the trial’s total enrollment or as the primary analysis sample. The record is informative about design and measurement, but numerical outcome reporting is necessary for assessing the principal comparison.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The most consequential limitation of the registry record for statistical interpretation is the absence of numerical results for the primary outcome. Estimates by group, between-group contrasts, confidence interval limits, and p-values are not reported in the registry record. Without them, direction, magnitude, and precision cannot be assessed.

The outcome relies on parent-reported 3-day food records. The registry explains the measurement approach but does not report information about data completeness, recording adherence, validation, or procedures for handling implausible or incomplete records. These factors could affect the reliability of the estimated daily intake, but their impact cannot be assessed from the record.

The model description is limited to linear mixed-effects models using REML. The fixed and random effects, covariance structure, covariate adjustments, diagnostic checks, and exact comparisons are not reported in the registry record. Missing-data methods and the analysis population are also not reported. The presence of repeated measurements makes these details important for evaluating how the estimates were obtained.

The record includes multiple outcome measures and several primary-outcome assessment times, but does not report how outcomes and contrasts were prioritized or whether multiplicity was addressed. It also does not report the original sample-size justification. The planning illustration described in the record is based on the observed effect and is not a substitute for that prospective rationale.

Masking is labeled single, but the masked party is not identified. The allocation-sequence method and concealment procedures are also not reported in the registry record. These gaps limit assessment of potential bias and the reproducibility of the analysis. They do not show that bias occurred; they identify information unavailable for judging it.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: An Integrated Nutrition Intervention to Promote Healthy Eating Habits for Children With ASD — NCT05194345.
• Registry record last updated 2025-11-19; status last verified 2025-11.
• Results first posted 2025-11-19.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

NCT05194345 is a completed, randomized, single-masked, parallel-group study sponsored by the University of South Florida. The record lists an enrollment of 132 and compares Autism Eats nutrition intervention with We Can! enhanced usual care. The primary outcome is change in children’s daily fruit and vegetable intake, measured from parent-reported 3-day food records at baseline, post-intervention, and 5-month follow-up.

The registry describes linear mixed-effects models using REML and a least-squares mean with a 95% confidence interval. Numerical primary outcome estimates and detailed analysis records are not reported in the registry record. Safety listings show 0 serious adverse events, 0 other adverse events, and 0 deaths in each listed group. Important details about missing data, analysis populations, multiplicity, and the original sample-size rationale are not reported in the registry record.