This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Omega-3 Fatty Acids For Treatment Of Young Children With Autism (OMG) (NCT01248728), a randomized phase 2 study. The primary outcome is Change From Baseline in the Pervasive Developmental Disorder-Behavioral Inventory (PDDBI) - Autism Composite Score.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT01248728 |
| Phase | Phase 2 |
| Condition | Autism Spectrum Disorder |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Enrollment | 38 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Holland Bloorview Kids Rehabilitation Hospital |
| Results posted | Yes |
2. Clinical Question
Population
The trial concerned young children with autism spectrum disorder. The registry describes autism as encompassing autistic disorder, Asperger Syndrome and PDD-Not Otherwise Specified in the protocol’s terminology. The registry enrollment was 38. It does not report a more detailed age range or the eligibility criteria in the record summarized for the primary outcome.
Intervention
The intervention was a liquid omega-3 fatty acid formulation. The arm description specifies Nutra Sea high-EPA and says the target administration was 3.75ml, containing 1.5 gr of EPA+DHA. The starting administration was 1.875ml, containing 0.75 gr of EPA+DHA, with the dose doubled on week 2.
Comparator
The comparator was a liquid placebo. Its starting administration was 1.875ml, with the dose doubled on week 2; the described administration was 3.75ml. The registry does not report additional information about matching or the placebo’s composition.
Primary question
The primary question was whether mean change from baseline to 24 weeks in the PDDBI Autism Composite Score differed between the randomized omega-3 and placebo groups. The registry defines a negative change as improvement. It reports group mean changes and confidence intervals, alongside a sponsor-reported mixed-model p-value. The statistical contrast estimate and its confidence interval are not reported in the registry record.
3. Trial Design
This was a phase 2, randomized, placebo-controlled trial with a parallel-group model. The registry identifies quadruple masking, meaning the record labels the trial as masked at four levels; it does not specify in the available record which roles those levels represent. Holland Bloorview Kids Rehabilitation Hospital is listed as the sponsor. The study status is completed, and the registry enrollment is 38.
The trial compared dietary supplements: omega-3 fatty acids and placebo. The omega-3 arm description gives a liquid formulation and a dose escalation, with the starting dose doubled on week 2. The placebo arm description also specifies a starting dose doubled on week 2. The registry does not report adherence, treatment discontinuation by arm, or how missing outcome measurements were handled beyond noting one early termination that led to insufficient data for one participant in the omega-3 group.
The primary outcome was change from baseline in the PDDBI Autism Composite Score, assessed over baseline and 24 weeks. The arm-level analysis includes 18 participants for omega-3 and 19 for placebo. These are the reported analysis counts for this outcome, not a substitute for a full account of all randomized participants or every analysis population. The registry notes that one participant in the omega-3 group was dropped from the analysis because of insufficient data due to early termination.
The registry also lists three secondary outcomes: responder classification by the Clinical Global Impression - Improvement, change in the Vineland Adaptive Behavioral Scales Adaptive Functioning Composite, and change in the Preschool Language Scale Total Language. The record reports five outcome measures and four analyses overall. More detail on allocation procedures, the masking process, the study protocol, or the rationale for the planned enrollment is 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: Participants were randomized to the two arms in a 1:1 fashion by the pharmacy, based on a randomization schedule produced by the study biostatistician. They were stratified into one of four strata based on the PDDBI score obtained at screening (four quartiles).
Overall Study
| Stage | Omega-3 Fatty Acids | Placebo |
|---|---|---|
| STARTED | 19 | 19 |
| COMPLETED | 15 | 17 |
| NOT COMPLETED | 4 | 2 |
Reasons for not completing this period:
| Reason | Omega-3 Fatty Acids | Placebo |
|---|---|---|
| Adverse Event | 1 | 1 |
| Withdrawal by Subject | 2 | 1 |
| Lack of Efficacy | 1 | 0 |
Analysis population for the primary outcome, as defined in the registry: 1 participant was dropped from the analysis in the Omega-3 group as there was insufficient data due to early termination.
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.
| Characteristic | Unit | Omega-3 Fatty Acids | Placebo | Total |
|---|---|---|---|---|
| Participants analysed | 19 | 19 | 38 | |
| Age, Categorical: <=18 years | Participants | 19 | 19 | 38 |
| Age, Categorical: Between 18 and 65 years | Participants | 0 | 0 | 0 |
| Age, Categorical: >=65 years | Participants | 0 | 0 | 0 |
| Sex: Female, Male: Female | Participants | 4 | 6 | 10 |
| Sex: Female, Male: Male | Participants | 15 | 13 | 28 |
| Region of Enrollment: Canada | participants | 19 | 19 | 38 |
6. Primary Endpoint
The registry lists the primary outcome as Change From Baseline in the Pervasive Developmental Disorder-Behavioral Inventory (PDDBI) - Autism Composite Score, measured over Baseline and 24 Weeks.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | units on a scale |
| Endpoint classification | Continuous Endpoint |
The primary endpoint was mean change from baseline to week 24 in the Pervasive Developmental Disorder-Behavioral Inventory Autism Composite Score. The registry describes the PDDBI as measuring autism symptomology and says the Autism Composite Score was used as the primary measure of autism symptom severity. Its stated scale range is from 10, indicating lower autism symptom severity, to 100, indicating higher autism symptom severity.
The composite is constructed from T-scores across behavioral and communication domains. The registry description says that scores for Sensory/Perceptual Approach Behaviours, Ritualisms/Resistance to Change, Social Pragmatic Problems, and Semantic/Pragmatic Problems are summed, then the sum of Social Approach Behaviours and Expressive Language scores is subtracted; the result is converted into the Autism Composite as a T-score. The score therefore combines several symptom and behavioral domains rather than representing a single observed behavior.
Change, rather than the endpoint score alone, captures how the measure moved from baseline to the 24-week assessment. The registry states that a negative mean change indicates improvement. This direction is important when interpreting the negative arm means: the sign describes within-group change, not by itself the difference between randomized groups. A treatment comparison requires estimating the contrast between groups, with its uncertainty.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
For the primary outcome, the omega-3 group’s mean change was -4.5 units on a scale, with a 95% confidence interval from -8.5 to -0.6. The placebo group’s mean change was -6.4 units on a scale, with a 95% confidence interval from -10.2 to -2.7. Because the registry defines a negative change as improvement, both reported group means are in the direction of improvement from baseline. These are group-level changes and should not be read as the randomized treatment effect.
The sponsor-reported comparison used mixed models analysis and gave p = 0.5. This is the reported statistical test for the comparison, but the record does not provide a treatment contrast estimate or a confidence interval for that contrast. The group-level confidence intervals describe uncertainty around each group’s mean change; they do not directly give the uncertainty around the between-group comparison. The reported p-value likewise does not describe the size or precision of the treatment contrast.
The two group means differ numerically, but that observation alone is not a measure of comparative effect. It is necessary to distinguish change within each arm from the difference attributable to assignment between arms. A group may improve over time for reasons that are not specific to the assigned intervention. The randomized comparison is the relevant basis for assessing whether the groups’ changes differ, and its reported result is the mixed-model p-value.
A p-value is a measure of how compatible the observed data are with a specified null model, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, nor does it measure clinical importance. The registry does not report an alpha level or a prespecified interpretation threshold for this result. The value p = 0.5 does not provide strong evidence against the null model, but the missing contrast estimate and confidence interval prevent direct assessment of the range of treatment differences compatible with the data.
The intervals for the arm-level means are broad relative to the scale of the reported changes, indicating uncertainty in those mean estimates. More importantly for the trial question, precision of the between-group effect cannot be assessed from the reported arm intervals alone. The registry does not report a standard error for the contrast, or a confidence interval around it. The result therefore offers limited information about how large a difference between omega-3 and placebo may be consistent with the data.
The analysis counts were 18 in the omega-3 arm and 19 in the placebo arm. The registry notes one participant in the omega-3 group was dropped from the analysis because early termination left insufficient data. How the mixed model used any other incomplete observations is not reported in the registry record. These details matter when judging how representative the analyzed participants are of everyone randomized and how sensitive the result may be to missing measurements.
9. Statistical Methods Explained
What does a mean change describe?
A continuous outcome can take a range of values, rather than placing each participant into a category. The PDDBI Autism Composite Score is reported as a continuous measure, and the registry presents the mean change in each randomized arm. A mean is a summary of the values in a group. Mean change compares the group’s measured scores at follow-up with its baseline scores; it does not show how much each individual changed, nor whether all participants changed in the same direction.
The registry states that negative change indicates improvement. That direction permits a reader to understand the sign of each group mean, but it does not transform the arm-specific changes into a randomized treatment effect. For a two-arm trial, the comparative question concerns the difference between changes under the two assignments. That comparison is distinct from asking whether either group’s mean changed from baseline.
What does a confidence interval mean?
A confidence interval expresses uncertainty around an estimate, under the statistical procedure used to construct it. The reported 95% confidence intervals are attached to the arm-level mean changes. In general, a narrower interval indicates greater precision for the estimate it accompanies, while a wider interval indicates less precision. These intervals do not state the probability that a particular treatment effect lies within them. Nor should intervals around separate arm means be treated as a direct interval for the between-arm contrast.
The sponsor’s reported p-value concerns a mixed-model comparison, whereas the listed confidence intervals accompany the individual group means. Because the contrast estimate and its interval are not reported in the registry record, the effect’s direction and precision as a between-group quantity cannot be read directly from those intervals. A p-value alone cannot fill that gap: it does not tell the reader the estimated size of the contrast or the range of plausible values.
Why use a mixed model?
Mixed models are a general class of statistical methods that can account for structured observations, such as repeated measurements from the same participant. In general, observations from one participant may be correlated rather than independent, and a model can represent that structure when estimating changes or group comparisons. The registry identifies the sponsor’s method as mixed models analysis, but does not report the model specification, included covariates, covariance structure, or exact handling of incomplete follow-up. It is therefore not possible to determine from the registry record which particular modeling choices were used.
Model-based results depend on assumptions about the outcome and the data structure. In general, the model’s specification and missing-data assumptions influence the estimated contrast and its uncertainty. The method label alone is not enough to evaluate those assumptions or reproduce the analysis. No additional model details should be inferred from the name of the method.
What does the reported p-value communicate?
In general, a p-value quantifies how unusual data at least as incompatible with a null hypothesis would be if that null model and the analysis assumptions held. It is not a measure of effect size, and it does not give the probability that a treatment is effective or ineffective. The registry’s mixed-model comparison reports p = 0.5. Without a reported contrast estimate and confidence interval, the p-value gives only limited information about comparative magnitude and precision.
A non-small p-value does not prove that the groups are equivalent. Demonstrating equivalence requires a prespecified margin and an analysis designed for that question; such details are not reported in the registry record. Likewise, a p-value should not be interpreted as evidence about clinical importance without considering the size and uncertainty of the estimated effect and the meaning of the outcome.
How should missing measurements be considered?
The registry explicitly notes that one participant in the omega-3 group was dropped from the primary analysis because early termination resulted in insufficient data. The arm-level analysis counts are 18 and 19. The record does not describe the timing or reasons for other missing assessments, whether all available repeated measurements were incorporated into the mixed model, or whether sensitivity analyses were performed. In general, missing-data methods rely on assumptions about why values are absent; different assumptions can affect estimates and precision.
The registry’s enrollment count and the endpoint analysis counts serve different purposes. Enrollment records the trial’s stated enrollment, while the endpoint counts describe participants contributing to the reported arm-level outcome. They should not be treated as interchangeable. The record does not provide enough information to reconstruct participant flow or establish whether the primary analysis followed an intention-to-treat principle.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Mixed Models Analysis | not reported | 0.500 |
The sponsor’s own comparison is reported as mixed models analysis with p = 0.5. This is the registry’s reported inferential result for the primary outcome. The arm-level means and their 95% confidence intervals are also reported, but the registry does not provide a treatment contrast estimate or a confidence interval for that contrast.
A simple unadjusted recalculation of the two-arm comparison is not available in the posted results because the continuous-endpoint comparison values are missing or not estimable for that purpose. Consequently, there is no recalculated contrast to set alongside the sponsor’s model-based comparison. The group means should not be used to invent or reconstruct that result: the sponsor’s mixed-model p-value concerns its stated analysis, while the individual confidence intervals describe the arm means.
The registry does not flag a discrepancy between a sponsor result and a recalculated comparison. The absence of a recalculation is not evidence that the two approaches agree; it means a second numerical comparison is not available. The sponsor analysis can be described as reported, but its estimate, precision, and modeling details cannot be independently assessed from the information in the registry record.
11. Multiplicity, Missing Data and Analysis Populations
The registry lists a primary outcome and three secondary outcomes. The secondary outcomes cover responder classification by the Clinical Global Impression - Improvement, change in the Vineland Adaptive Behavioral Scales Adaptive Functioning Composite, and change in the Preschool Language Scale Total Language. The record also states that five outcome measures and four analyses were reported overall. These outcomes represent multiple opportunities to examine results, so the interpretation of any single test depends in part on how the analyses were planned and how multiplicity was addressed.
Whether the protocol specified a hierarchy, adjusted significance thresholds, or another multiplicity strategy is not reported in the registry record. It is also not reported whether the secondary outcomes were confirmatory or exploratory, or how outcomes and analyses were selected for reporting. Without those details, a reader cannot determine whether the reported statistical tests were interpreted within a prespecified family of hypotheses or considered separately.
The registry’s primary-outcome entry reports the arm means, confidence intervals, and mixed-model p-value. The method specification and a treatment contrast confidence interval are not reported in the registry record. The exact analysis population, beyond the arm counts and the noted early termination, is also not established in the record. In particular, it is not reported whether the primary analysis included all randomized participants according to original assignment, how other missing outcome values were handled, or whether sensitivity analyses tested alternative assumptions.
These omissions do not invalidate the posted result, but they limit independent evaluation of its robustness. Multiplicity procedures, prespecified analyses, and missing-data handling can affect the interpretation of statistical evidence. The available record supports describing the reported result and its limitations, but does not support conclusions about whether unreported analyses would have changed it.
12. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Omega-3 Fatty Acids | 0 | 19 | 0.0% |
| Placebo | 0 | 19 | 0.0% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Omega-3 Fatty Acids | 19 of 19 |
| Placebo | 19 of 19 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Omega-3 Fatty Acids | Placebo |
|---|---|---|---|
| Upper Respiratory Infection | Infections and infestations | 16 of 19 | 14 of 19 |
| Rash | Skin and subcutaneous tissue disorders | 6 of 19 | 9 of 19 |
| Diarrhea | Gastrointestinal disorders | 5 of 19 | 8 of 19 |
| Constipation | Gastrointestinal disorders | 6 of 19 | 6 of 19 |
| Increased Irritability/Anger/Aggression | Psychiatric disorders | 7 of 19 | 2 of 19 |
| Initial Insomnia | Nervous system disorders | 5 of 19 | 4 of 19 |
| Decreased Appetite | Metabolism and nutrition disorders | 6 of 19 | 2 of 19 |
| Fever | Infections and infestations | 4 of 19 | 4 of 19 |
| Vomiting | Gastrointestinal disorders | 4 of 19 | 3 of 19 |
| Abdominal Pain | Gastrointestinal disorders | 3 of 19 | 3 of 19 |
The registry reports zero serious adverse events affected in each arm. For the serious-adverse-event figures, 19 participants are listed as at risk in the omega-3 group and 19 in the placebo group. No top serious-event terms are listed. The registry does not report death counts for either arm in the record; those entries are not available as numerical values.
These figures describe the serious adverse events recorded in the registry, not every possible safety outcome. A count of zero affected participants means no serious adverse event was reported for the listed at-risk group during the registry’s safety reporting. It does not establish that no event could occur, and it does not rule out events that were not captured or not entered in the record. The registry entry does not provide event definitions, event-specific follow-up duration, or further detail about ascertainment.
The record also reports 19 participants at risk for other adverse events in each arm and 19 affected in each arm. This is a separate broad adverse-event summary and should not be confused with the serious-adverse-event counts. The registry does not provide further event-level information in the listed serious-event terms. These summaries are descriptive and should be read alongside the trial’s limited size and the level of detail recorded, rather than as a comprehensive account of safety.
13. What the Result Does, and Does Not, Mean
The result describes a randomized comparison of omega-3 fatty acids and placebo on mean change in the PDDBI Autism Composite Score through 24 weeks. The posted group means were -4.5 for omega-3 and -6.4 for placebo, and the registry defines negative change as improvement. The sponsor’s mixed-model comparison reported p = 0.5. These findings characterize the measurements and analysis reported in the registry; they do not by themselves establish a causal difference of a particular size.
Group-level improvement from baseline is not the same as a difference between randomized assignments. Both arms can change over time, and the relevant comparative question is whether their changes differ. The registry’s p-value addresses a model-based comparison, but no contrast estimate or confidence interval is reported. This makes it difficult to assess the size of the between-group effect or the range of values compatible with the data.
The result does not demonstrate equivalence between omega-3 and placebo. A non-small p-value is not proof that two treatments have identical effects; equivalence requires an appropriate design and prespecified margin, which are not reported in the registry record. Nor does the result establish that either intervention is effective or ineffective. The statistical record supports a cautious description of the observed group changes and reported test, not a medical recommendation.
The trial is identified as a pilot, randomized, placebo-controlled study. Its registry enrollment is 38, and the primary-outcome analysis counts are 18 and 19. These figures and the missing contrast uncertainty limit how precisely the comparative result can be characterized. The early termination and insufficient data noted for one omega-3 participant are relevant to understanding the analysis, while further missing-data methods are not reported in the registry record.
More broadly, interpretation depends on the outcome definition, analysis choices, completeness of follow-up, and whether the reported analysis was prespecified. The registry provides only some of these details. Its results are best understood as a record of one trial’s primary outcome analysis rather than a complete account of all evidence about omega-3 fatty acids and autism.
14. Why This Trial Matters Statistically
This trial illustrates why within-group change and between-group effect must be distinguished. Each arm has a mean change, and the outcome definition assigns a direction to that change. However, the randomized question concerns the comparison between arms. Reading the arm means alone could confuse change over time with a difference associated with treatment assignment.
It also shows why the estimate and its uncertainty matter alongside a p-value. The registry reports a mixed-model p-value, but not the treatment contrast or its confidence interval. The arm-level confidence intervals do not directly provide the uncertainty around the comparison. Without those comparative statistics, readers have limited ability to judge the plausible size of the effect, even though a formal test result is available.
The record also makes the analysis population relevant. The primary outcome arm counts are 18 and 19, and the registry notes one omega-3 participant was excluded from the analysis because early termination left insufficient data. The reason for that omission is reported, but the broader handling of missing observations is not. In a trial with limited enrollment, incomplete outcome information can make the assumptions underlying an analysis especially important to assess.
Finally, this record lists several outcomes and analyses while not reporting a multiplicity strategy or detailed statistical plan. The primary result can be described, but readers cannot fully evaluate how it fits within the planned analysis family. Together, these features make the trial useful for statistical interpretation: the endpoint, contrast, precision, missingness, and analysis plan all shape what can responsibly be concluded from posted results.
15. 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 registry record does not report the estimated between-group contrast for the primary outcome or a confidence interval around that contrast. It reports group means with 95% confidence intervals and a mixed-model p-value, but those pieces do not fully describe the magnitude and precision of the treatment comparison. The model specification and the detailed assumptions behind the sponsor’s analysis are also not reported in the registry record.
The record notes that one participant in the omega-3 group was dropped from the primary analysis due to early termination and insufficient data. It does not provide a complete account of missing assessments, reasons for missingness beyond that noted case, or the handling of incomplete observations in the mixed model. The analysis population and whether the analysis followed original randomized assignments are not reported in sufficient detail to establish.
Although the registry lists secondary outcomes and the overall number of outcome measures and analyses, a multiplicity approach, prespecified testing hierarchy, or distinction between confirmatory and exploratory analyses is not reported. The record also does not report the basis for the planned enrollment or a detailed protocol. Those omissions make it harder to assess the study’s operating assumptions and the role of the primary result among other analyses.
The clinical and measurement details are also incomplete in the available record. The exact eligibility age range, full participant flow, and further information about outcome assessment are not reported in the registry record. Safety reporting is limited to the listed affected and at-risk counts, with no serious-event terms listed and no numerical death counts available. These limitations do not alter the reported values, but they constrain how broadly and confidently the trial’s findings can be interpreted.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Omega-3 Fatty Acids For Treatment Of Young Children With Autism (OMG) — NCT01248728.
- Registry record last updated 2025-07-16; status last verified 2025-07.
- Results first posted 2016-05-18.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
19. Record Summary
The registry identifies NCT01248728 as a completed phase 2 randomized, quadruple-masked, parallel-group trial of omega-3 fatty acids versus placebo in young children with autism. The primary endpoint was mean change from baseline to 24 weeks in the PDDBI Autism Composite Score, for which negative change indicates improvement.
The reported mean changes were -4.5 in the omega-3 arm and -6.4 in the placebo arm. The sponsor-reported mixed-model comparison gave p = 0.5. A between-group estimate and confidence interval are not reported in the registry record, and a simple unadjusted recalculation is not available from the posted continuous-outcome results.
The registry notes one omega-3 participant was dropped from the primary analysis because early termination left insufficient data. Serious adverse events affected zero participants in each arm as reported. The multiplicity strategy, detailed model specification, and broader missing-data approach are not reported in the registry record.