ScanBrit Dietary Intervention in Autism: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for ScanBrit Dietary Intervention in Autism (NCT00614198), a randomized phase 2 study. The primary outcome is Change in Diet Groups Scores on One of Several Measures Used Against Pre-defined Statistical Thresholds as Evidence of Improvement..

2. Clinical Question

3. Trial Design

ScanBrit is a completed, randomized, parallel-group, phase 2 study with an enrollment of 72. The registry identifies the study as single-blind and describes it in the summary as a matched-pair adaptive trial. Who was blinded, how matching was performed, and the details of the allocation procedure are not reported in the registry record.

The study began in 2006-04, with primary completion in 2008-10. During stage 1, one group followed a gluten- and casein-free diet while the comparison group continued with no special dietary intervention. The planned first-stage duration was 8 or 12 months. The registry describes an interim decision based on whether the dietary group surpassed statistical thresholds. If it showed significant improvements, both groups proceeded to a further 12 months on the diet. This means that the later follow-up did not preserve the original diet-versus-no-diet contrast.

The description links the decision to progress with improvements on outcome measures, but the exact thresholds, decision rules, and analysis used for that decision are not reported in the registry record. The record also does not report how matched pairs were handled in the statistical analysis, whether analyses accounted for the adaptive design, or how the single-blind procedure was implemented. Those details matter when interpreting estimates from a trial whose treatment assignment and comparator changed over time.

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: 92 participants were initially identified for possible study recruitment. 82 participants were available for possible study allocation; 10 participants dropped out prior to randomisation for various reasons, leaving 72 participants.

Stage 1

Stage 2

Analysis population for the primary outcome, as defined in the registry: Per protocol analysis. Analysis was conducted on n=26 (Gluten- and casein-free dietary intervention) and n=29 (No gluten- and casein-free dietary intervention) at 12 months. Analysis was conducted on n=18 and n=17 at 24 months (both on gluten- and casein-free dietary intervention).

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.

6. Primary Endpoint

The registry lists the primary outcome as Change in Diet Groups Scores on One of Several Measures Used Against Pre-defined Statistical Thresholds as Evidence of Improvement., measured over Baseline - 8 months - 12 months - 24 months.

The primary outcome is described as change in diet-group scores on several behavioral and developmental measures, assessed against pre-defined statistical thresholds as evidence of improvement. It is not a single scale or a single time-point comparison. The listed measures are ADOS, GARS, VABS, and ADHD-IV, with the timeframe given as baseline, 8 months, 12 months, and 24 months.

The registry gives instrument-specific cutoffs and categories. ADOS cutoffs distinguish autism and autism-spectrum classifications across modules; GARS ranges describe probability categories; VABS ranges describe adaptive ability categories; and ADHD-IV records symptom indications. These measures address different aspects of behavior, development, or functioning. Using them can provide a broader description than relying on a single measure, but interpretation depends on knowing which scores and thresholds were specified for each analysis and how results across measures were handled.

The outcome description explains the intended role of the measures, but the registry record does not report the observed changes, group estimates, uncertainty intervals, or which thresholds were met. A secondary outcome concerns changes to the appearance of multiple compounds in urine samples.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean score represent?

The primary outcome uses the mean of a continuous measure. A mean summarizes the average score in a group, but it does not describe every participant's score or the spread of individual outcomes. In this study, several instruments are named, so a mean from one measure would represent only that instrument and the participants included in its analysis. The registry record does not report those means or the corresponding group comparisons.

Score interpretation also depends on the scale. The registry lists cutoffs and categories for ADOS, GARS, VABS, and ADHD-IV, but the measures do not all describe the same construct. A score change should therefore be interpreted using the relevant instrument's scoring rules, the direction of its scale, and the planned threshold. The registry provides some threshold descriptions, but does not report which specific score changes occurred or how the measures were combined in the primary analysis.

How should standard errors and confidence intervals be read?

The primary outcome specifies standard error as its dispersion measure. In general, a standard error describes the sampling uncertainty of an estimated mean; it is not a description of the full range of participant scores. A confidence interval uses an estimate and its uncertainty to express a range of values compatible with the data under the method's assumptions. Narrower intervals generally indicate greater precision, while wider intervals indicate less precision. The registry record does not report standard errors or confidence intervals for the primary outcome, so precision cannot be evaluated from the posted results.

What would a group comparison establish?

In general, a comparison between randomized groups estimates the difference associated with assignment under the trial's design and analysis assumptions. Random allocation can support a fair comparison at the point of assignment, but the usefulness of a particular estimate also depends on follow-up, outcome measurement, adherence, and the analysis population. Here, the reported primary-outcome population is per protocol, and the record does not state whether the analysis accounted for matching or describe a full analysis plan. It is therefore not possible to determine how the reported design features were incorporated into a primary estimate.

The comparator also changes after stage 1 if the stated progression condition is met: both groups are then assigned to the diet. A comparison at a later time point would not be a continuing comparison of randomized diet and no-diet groups. The precise interim decision procedure and the analysis intended for the later stage are not reported in the registry record.

Why do multiple measures and time points matter?

When a study assesses multiple measures or time points, it creates more than one opportunity to observe a result that appears notable. In general, multiplicity can increase the chance of false-positive findings if each result is interpreted independently without an appropriate strategy. Statistical approaches may specify a primary measure, adjust for multiple comparisons, or define a decision rule in advance. The registry describes several measures, pre-defined statistical thresholds, and an interim decision, but does not report the thresholds in full or whether a multiplicity adjustment was used. No specific adjustment should be assumed.

What do the planning assumptions mean?

The stated sample-size assumptions describe a planning illustration based on the observed effect, using a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is described as a stated fraction of the observed effect. Such planning figures are hypothetical illustrations, not observed power and not proof that the trial was adequately sized for its outcomes. The registry record does not report a sample-size justification from the trial's own design, and no numerical planning estimates are stated in the record.

In general, over-dispersion means that variation in the observed data is greater than a model assumes. When relevant, ignoring it can make uncertainty estimates too small and tests too optimistic. The registry does not report an over-dispersion assessment or identify a model for the continuous outcomes, so there is no basis to say whether this issue arose or how it was addressed. More generally, the numerical estimates and statistical methods needed to reproduce or fully assess the primary analysis are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The primary outcome encompasses several measures and multiple time points, and the title describes changes in scores on one of several measures assessed against pre-defined statistical thresholds. The registry also describes an interim decision at 8 or 12 months that could determine whether the study proceeded to a further stage. These features raise questions about how the primary comparison was selected and how repeated opportunities for a positive finding were handled.

The record does not report a complete multiplicity strategy, whether thresholds were adjusted across measures or time points, or the exact rules used at the interim decision. It is not appropriate to infer that a correction was or was not used. The number of outcome measures is listed as 2 and the number of analyses as 1, but those registry counts do not resolve how the several named instruments and assessment times were treated in the analysis.

The primary-outcome population is described as per protocol. The record gives analysis counts of n=26 and n=29 at 12 months, and n=18 and n=17 at 24 months, when both groups were on the diet. It does not report the reasons for the differing counts, a missing-data strategy, or whether participants with unavailable outcomes were included in any other analysis. The record does not specify an intention-to-treat analysis or describe imputation or sensitivity analyses. Without those details, the potential effect of missing observations and exclusions cannot be assessed.

The registry does not report a statistical analysis plan or enough detail to determine how the adaptive feature, matched-pair description, and per-protocol population were handled jointly. These omissions limit assessment of both the false-positive risk and the robustness of any outcome estimate.

11. Safety Results

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

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 two groups for adverse events. In the gluten- and casein-free dietary intervention group, 38 participants were at risk for serious adverse events and 0 were affected; 38 were at risk for other adverse events and 0 were affected. In the group with no gluten- and casein-free dietary intervention, 34 participants were at risk for serious adverse events and 0 were affected; 34 were at risk for other adverse events and 0 were affected. No serious adverse-event terms are listed.

These figures mean that the registry reports zero affected participants in the listed serious and other adverse-event categories for both groups. They do not show that no event of any kind occurred, establish that all events were captured, or address outcomes outside the recorded categories. The registry record does not report death counts; the death fields are not reported in the registry record. The at-risk counts are not, by themselves, a substitute for details about the duration or completeness of safety follow-up.

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

The record establishes that ScanBrit was a completed randomized study of a gluten- and casein-free dietary intervention in children with autism spectrum disorders, with a no-special-diet comparison during stage 1. It identifies the primary-outcome instruments and assessment times and describes a conditional progression to a later stage in which both groups were assigned to the diet.

It does not establish the size or direction of any primary-outcome difference. Numerical outcome estimates, standard errors, confidence intervals, and p-values are not reported in the registry record. The description of pre-defined thresholds does not reveal whether they were met, and the registry does not give the exact analysis needed to evaluate a threshold-based decision. The record therefore cannot support a conclusion about how precisely the intervention group differed from the comparison group during stage 1.

At the later follow-up, both groups were on the diet according to the study description. Those observations cannot be interpreted as a continuing randomized diet-versus-no-diet comparison. The per-protocol counts indicate the size of the analyzed populations at specified times, but do not show why counts changed or how missing outcomes were handled.

The listed safety data report no affected participants in the serious or other adverse-event categories, while death counts are not reported in the registry record. This is a statement about the recorded figures, not a general determination about safety. The distinction between a missing result and a result showing no effect is important: the registry record does not report the primary outcome estimates, so their absence cannot be interpreted as evidence of no difference.

13. Why This Trial Matters Statistically

Statistically, this trial is notable for combining random allocation with a matched-pair and adaptive design description. The intervention and comparator were distinct during stage 1, but the registry says both groups could proceed to the diet after an interim decision. This makes the timing and definition of each comparison central to interpretation. Later measurements do not preserve the initial diet-versus-no-diet contrast.

The primary outcome also spans multiple behavioral and developmental instruments and several follow-up times. That breadth reflects an attempt to assess different domains, but it makes the analysis strategy important: readers need to know which measures drove conclusions, how thresholds were applied, and how multiplicity was handled. The registry record does not report those details or the numerical outcome estimates.

The per-protocol populations vary across follow-up, and the record does not explain the changing analysis counts or describe missing-data methods. These details affect how confidently readers can understand the analyzed population and the precision or potential bias of an estimate. The safety figures are also best read narrowly as registry-reported counts for named adverse-event categories, rather than as a complete assessment of every possible safety outcome.

The study therefore illustrates why design descriptions and endpoint definitions are not enough for a full statistical interpretation. Transparent reporting of estimates, uncertainty, analysis populations, decision rules, and missing-data handling is needed to assess what the trial's results show.

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 main limitation for interpreting the registered results is that numerical primary-outcome findings are not reported in the registry record. Without group estimates and measures of uncertainty, the magnitude and precision of any changes cannot be assessed. The record also does not identify a single primary score comparison that captures the several instruments and time points.

The adaptive progression is described in broad terms, but its exact statistical thresholds and decision procedure are not reported. The registry does not explain how the matched-pair feature was implemented or whether the analysis accounted for it. It also does not report the full statistical analysis plan, multiplicity handling, or the rationale for the sample size. The planning assumptions described in the record are illustrations and do not supply the trial's own design justification.

The reported primary analysis is per protocol. The registry provides counts at 12 and 24 months but not the reasons for changes in the analyzed populations, a missing-data approach, or sensitivity analyses. That leaves uncertainty about the impact of exclusions and unavailable measurements. Because both groups were assigned to the diet for the later stage, comparisons at 24 months cannot be read as a continuing randomized contrast between diet and no special diet.

The safety record reports zero affected participants in the listed serious and other adverse-event categories, but does not provide death counts or a detailed account of safety ascertainment. These limitations concern what can be concluded from the registry record; they do not establish that particular analyses or events were absent from other study materials.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: ScanBrit Dietary Intervention in Autism — NCT00614198.
• Registry record last updated 2010-10-01; status last verified 2010-05.
• Results first posted 2010-10-01.
• 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

NCT00614198, ScanBrit Dietary Intervention in Autism, is a completed phase 2 randomized parallel-group study with enrollment of 72. It compared a gluten- and casein-free diet with no special dietary intervention during stage 1. The registry describes the study as single-blind and matched-pair, with an adaptive later stage in which both groups could receive the diet.

The primary outcome concerns changes in scores across several behavioral and developmental measures from baseline through 24 months. Per-protocol analysis counts are reported at 12 and 24 months, but numerical primary-outcome results are not reported in the registry record. The record also does not report the complete statistical analysis plan, missing-data methods, or multiplicity strategy. Safety listings show zero participants affected in the serious and other adverse-event categories in both groups; death counts are not reported in the registry record.