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 Pilot Study of the Effect of Hyperbaric Oxygen Treatment on Behavioral and Biomarker Measures in Children With Autism (NCT00584480), a na phase 3 study. The primary outcome is Number of Participants With the Given Clinical Global Impression Scale - Improvement (CGI-I) Score.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00584480 |
| Phase | Phase 3 |
| Condition | Autism |
| Allocation | Not applicable (single group) |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 10 |
| Arms in the primary outcome | n/a |
| Lead sponsor | University of California, San Francisco |
| Results posted | Yes |
2. Clinical Question
Population
Children with autism. The registry lists autism as the condition and describes the study population as children; further eligibility details are not reported in the registry record.
Intervention
Active Hyperbaric Oxygen Treatment (HBOT). The registry describes the intervention as an other intervention and states that the study aimed to examine behavioral and biomarker measures, as well as safety and tolerability.
Comparator
There is no separate comparison group in this single-group study. The registry identifies no randomized allocation and no masking. Consequently, the record does not support a direct comparison of HBOT with a control treatment or usual care.
Primary question
How many participants were assigned each score on the Clinical Global Impression Scale - Improvement (CGI-I) at baseline, 8 Weeks from baseline, and 20 Weeks from baseline? The scale was intended to capture a clinician’s assessment of global changes in autistic symptoms, informed by parent interview and direct clinical observation. The registry’s stated goals also include assessing safety and tolerability, but the primary endpoint is the CGI-I score count.
The clinical question is therefore descriptive rather than a controlled estimate of treatment effect. A change in CGI-I scores over the listed assessment times could describe how participants were rated during the study, but without a control group it cannot separate any change associated with the treatment period from changes that might occur for other reasons. The registry record does not report the CGI-I score counts, participant-level changes, or a statistical comparison across time. It also does not report biomarker outcome results. The posted record thus establishes what the investigators intended to measure, but does not provide the primary outcome data needed to describe the observed behavioral ratings.
3. Trial Design
The registry identifies NCT00584480 as a completed study titled “Pilot Study of the Effect of Hyperbaric Oxygen Treatment on Behavioral and Biomarker Measures in Children With Autism.” The sponsor was the University of California, San Francisco. The record lists a Phase 3 designation, while the title describes the work as a pilot study. These are both registry descriptions; the record does not explain their relationship.
This was a single-group study with active Hyperbaric Oxygen Treatment (HBOT). The listed enrollment was 10. Allocation is recorded as NA, and masking as NONE. The record therefore does not describe random assignment, a blinded assessment process, or a separate control arm. Its own note cautions that this is not a randomized comparison and that the arms are not a like-for-like control comparison. For this single-group design, the key interpretive consequence is that observed measurements can describe participants over time but cannot by themselves establish what would have happened without HBOT.
The registry lists a start of 2007-10 and primary completion of 2009-09, and gives the study status as completed. Its brief summary says the main goals were to demonstrate safety and tolerability in autistic children and to measure the effect of HBOT on specific blood chemicals that may play a role in behavior. The listed primary outcome, however, is a count of participants with each CGI-I score, assessed at baseline, 8 Weeks from baseline, and 20 Weeks from baseline.
Further design details—including eligibility criteria, treatment schedule, dose, and the timing or methods for biomarker measurements—are not reported in the registry record. The record also does not explain the Phase 3 designation, describe a comparator, or provide a statistical analysis plan. Those omissions limit how fully the study’s conduct and intended inferential approach can be reconstructed from the registry.
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.
First 8 Weeks (40 Treatments)
| Stage | Hyperbaric Oxygen Treatment (HBOT) |
|---|---|
| STARTED | 10 |
| COMPLETED | 10 |
| NOT COMPLETED | 0 |
4-Week Break From Treatment
| Stage | Hyperbaric Oxygen Treatment (HBOT) |
|---|---|
| STARTED | 10 |
| COMPLETED | 10 |
| NOT COMPLETED | 0 |
Second 8 Weeks (40 Treatments)
| Stage | Hyperbaric Oxygen Treatment (HBOT) |
|---|---|
| STARTED | 10 |
| COMPLETED | 10 |
| NOT COMPLETED | 0 |
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 | Hyperbaric Oxygen Treatment (HBOT) |
|---|---|---|
| Participants analysed | 10 | |
| Age, Categorical: <=18 years | Participants | 10 |
| Age, Categorical: Between 18 and 65 years | Participants | 0 |
| Age, Categorical: >=65 years | Participants | 0 |
| Age Continuous | years | 5.3 (SD 1.8) |
| Sex: Female, Male: Female | Participants | 0 |
| Sex: Female, Male: Male | Participants | 10 |
| Region of Enrollment: United States | participants | 10 |
6. Primary Endpoint
The registry lists the primary outcome as Number of Participants With the Given Clinical Global Impression Scale - Improvement (CGI-I) Score, measured over Baseline, 8 Weeks from baseline, and 20 Weeks from baseline.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | not applicable |
| Unit | participants |
| Endpoint classification | Binary Endpoint |
The primary endpoint is the number of participants with each score on the Clinical Global Impression Scale - Improvement (CGI-I). The registry describes the scale as an assessment of global changes in severity of autistic symptoms. A clinician formulates the score using a parent interview about changes in the child’s behavior and direct clinical observation. The listed score meanings are 0 for no improvement, 1 for minimally improved, 2 for much improved, and 3 for very much improved.
This kind of global rating can combine observations from more than one source into a clinician’s overall judgment. It is used to summarize a broad clinical impression rather than a single narrowly defined behavior. Reporting counts at baseline, 8 Weeks from baseline, and 20 Weeks from baseline could describe how ratings are distributed at those assessment times. It would not, on its own, show which individuals changed scores unless participant-level transitions were also reported.
The registry gives the endpoint title and assessment description but does not report the actual CGI-I counts or any statistical analysis of them. It also does not explain how the scale’s improvement categories are to be interpreted at baseline, when no study-period change has yet occurred. Thus the measure’s intended content is described, but the posted record does not permit evaluation of the observed primary outcome.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record identifies a completed study and defines a primary behavioral endpoint, but it does not report the primary endpoint results. In particular, it gives no counts for participants at each CGI-I score at baseline, 8 Weeks from baseline, or 20 Weeks from baseline. There is no reported estimate of change, uncertainty interval, hypothesis test, or analysis population for this outcome. As a result, the record cannot establish what pattern of clinician ratings was observed or how precisely any change was estimated.
The design also constrains interpretation. This was a single-group study, not a randomized comparison. Even if before-and-after scores were available, changes over time would not isolate the effect of HBOT from other possible explanations. General statistical point: a concurrent control group can help distinguish treatment-associated change from changes that occur over time for other reasons. A single-group description cannot make that contrast. The registry’s own note flags the lack of a randomized comparison.
The registry’s safety information is more concrete but remains limited. For the HBOT group, 1 participant is recorded as having a serious adverse event, among 10 at risk. The record lists Seizure as a top serious term, with 10.0%. It also records 5 participants with other adverse events among 10 at risk. These counts describe what the registry reports; they do not provide event-level detail, timing, severity beyond the serious classification, or a comparison group. Deaths are not reported in the registry record. These entries should not be interpreted as proof that no deaths occurred.
The study summary states an interest in blood chemicals related to behavior, but biomarker results are not reported in the registry record. Nor are the detailed methods needed to evaluate the planned behavioral or biomarker analyses. The registry provides no analysis count, and the number of outcome measures listed is 1. Those record fields do not substitute for a full statistical analysis plan or results report.
Overall, the posted information supports a narrow account: a small, completed, single-group study was registered, a clinician-rated outcome was specified at three assessment times, and adverse-event counts were entered. It does not support an estimate of behavioral benefit, a causal conclusion, or a complete assessment of safety. The absence of primary outcome results is especially important because the named endpoint is the main basis for answering the study’s behavioral question.
9. Statistical Methods Explained
Interpreting clinical trial statistics begins with identifying the design and the question the analysis can answer. This study is recorded as a single-group study with no randomized allocation and no masking. That structure allows descriptive reporting of observations within the treated group. It does not create a randomized contrast between HBOT and another treatment or no treatment. General statistical point: random assignment is valuable because it can make treatment groups comparable, on average, with respect to measured and unmeasured factors. Without that comparison, temporal change may have multiple explanations.
What does a participant count show?
The primary outcome is expressed as the number of participants with each CGI-I score. Counts communicate how many participants fall into each category at a given assessment. The categories are ordered from no improvement through levels of improvement, but the registry presents the endpoint as category counts rather than as a numerical average or a change score. General statistical point: ordered categories carry a ranking, but the distances between neighboring categories are not necessarily equal. Treating such ratings as if each step had the same quantitative meaning requires justification that is not reported in the registry record.
Repeated assessment times can describe the course of ratings, but a set of separate counts does not necessarily show individual trajectories. For example, aggregate counts at baseline and later assessments do not identify which participants moved between categories. Participant-level transitions or a stated repeated-measures analysis would be needed to describe those within-person patterns. Neither is reported in the registry record. The primary endpoint’s assessment schedule is listed, but its observed counts and analysis are not.
Why does the missing comparator matter?
A before-and-after comparison can reveal whether measurements differ across assessment times among the observed participants. General statistical point: that difference does not automatically estimate a causal treatment effect. Natural variation, changes in circumstances, expectations, measurement variability, and other time-related influences can contribute. A concurrent comparison group helps assess these influences, particularly when participants are assigned to groups in a way that reduces systematic differences. Here, no randomized comparator is described, so a causal interpretation cannot be recovered from a within-group trend alone.
Masking is also relevant to subjective ratings. The endpoint is formulated by a clinician using parent interview and direct observation. General statistical point: when observers or participants know the treatment assignment, expectations can influence reports or ratings. The registry records masking as NONE. It does not describe separate blinded outcome assessment, so the record does not support an assumption that the CGI-I ratings were protected from knowledge of treatment.
How should uncertainty be understood?
A confidence interval is a range of values compatible with the data under a specified statistical model and procedure. General statistical point: a wider interval indicates less precision, while a narrower interval indicates greater precision; neither tells whether the result is free from bias or whether a causal interpretation is warranted. No confidence interval or other uncertainty estimate for the primary endpoint is reported in the registry record. Without the outcome counts and a defined analysis, readers cannot assess the precision of any behavioral estimate.
The registered enrollment is 10. In a study with a small enrollment, descriptive counts can be sensitive to the outcomes of individual participants. General statistical point: sparse data can yield unstable estimates and limited ability to distinguish a pattern from random variation. No formal inferential estimate is posted for the primary endpoint, so there is no reported precision measure to weigh against this concern.
What do the planning assumptions mean?
The record describes sample-size planning illustrations based on a fraction of an observed effect. The stated conventions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. These are planning assumptions, not a description of the single-group design used in this study. The record also says that the illustrations are not the trial’s own design. They should therefore not be read as evidence that this trial had a particular power, or that an effect was observed or confirmed.
General statistical point: a sample-size calculation depends on an assumed effect, variability, error rate, allocation, and analysis method. Planning calculations can show how design choices affect the number of participants needed under those assumptions. They do not establish the probability that a completed study has detected a real effect, and they cannot replace reporting the observed results with uncertainty estimates. The record does not provide the trial’s sample-size justification, and the planning illustration should not be treated as one.
The study is labeled Phase 3, but the registry does not provide an analysis plan explaining the planned models, hypothesis tests, handling of repeated measurements, or treatment of missing observations. General statistical point: model choice should match the outcome type, design, and data structure. Because the primary measure is categorical and assessed over time, an analysis would need to clarify whether the goal is to compare category distributions, assess within-participant changes, or answer another question. The record does not specify such an approach. No recalculated result is presented, and the sponsor’s analysis cannot be assessed from the posted outcome information because it is not reported.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists 1 outcome measure and 0 analyses. The primary outcome is the count of participants in each CGI-I score category, assessed at baseline, 8 Weeks from baseline, and 20 Weeks from baseline. The brief summary also refers to blood chemicals, but biomarker results and their analysis are not reported in the registry record. It is therefore unclear whether those measures were analyzed, treated as secondary outcomes, or reported elsewhere.
Multiplicity arises when many outcomes, time points, groups, or statistical tests are examined. General statistical point: conducting many tests can increase the chance of finding an apparently notable result by chance, so an analysis plan may identify a primary test or describe an adjustment strategy. The registry does not report a multiplicity strategy, planned comparisons, or a detailed statistical analysis plan. Since the primary outcome results themselves are absent, the number and nature of any actual tests cannot be established from the record.
Missing-data handling is also not reported. The listed enrollment does not establish how many participants contributed data at each assessment, whether any assessments were missed, or how missing observations were treated. General statistical point: missingness can affect estimates when the likelihood of an observation being absent is related to participant characteristics or outcomes. Different handling methods rely on different assumptions, so transparent reporting is important.
The registry does not identify an analysis population, such as whether analyses included all enrolled participants or only those with particular follow-up information. It also does not describe exclusions or deviations from the intended assessment schedule. These omissions prevent evaluation of whether the reported results, if available elsewhere, would represent all enrolled participants or a selected subset.
11. 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 |
|---|---|---|---|
| Hyperbaric Oxygen Treatment (HBOT) | 1 | 10 | 10.0% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Hyperbaric Oxygen Treatment (HBOT) | 5 of 10 |
Most frequent serious adverse event terms
| Event term | Organ system | Hyperbaric Oxygen Treatment (HBOT) |
|---|---|---|
| Seizure | Nervous system disorders | 1 of 10 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Hyperbaric Oxygen Treatment (HBOT) |
|---|---|---|
| Ear Discomfort | Ear and labyrinth disorders | 3 of 10 |
| Ear Infection | Ear and labyrinth disorders | 2 of 10 |
Serious events with the highest arm-level rate: seizure (10.0%).
The registry records adverse-event information for the Hyperbaric Oxygen Treatment (HBOT) group. It lists 1 participant with a serious adverse event among 10 at risk, and 5 participants with other adverse events among 10 at risk. The serious-event term Seizure is listed with 10.0%. The record does not provide event narratives, timing, duration, severity details beyond the serious classification, or an account of how events were judged related to treatment.
These figures are counts recorded for the study group, not a comparison of risks against a control group. With no separate comparator, the entries do not show whether adverse events were more or less common than they would have been without HBOT. General statistical point: safety counts are easier to interpret when the event definitions, observation period, exposure, and method of collection are clear. Those details are not reported in the registry record.
Deaths are listed as not reported in the registry record; a missing death entry should not be treated as a reported count of zero. The serious and other adverse-event counts also should not be added together, because the record does not state whether participants could appear in both categories. The available information supports only a limited description of the entered counts, not a comprehensive safety assessment.
12. What the Result Does, and Does Not, Mean
The posted results and design information establish that the study was completed, involved active HBOT in a single group, and specified a clinician-rated CGI-I endpoint at baseline, 8 Weeks from baseline, and 20 Weeks from baseline. The record also includes adverse-event counts for the HBOT group. These registry entries describe the registered study and the limited outcome information entered in the registry.
They do not show the observed distribution of CGI-I scores. Because the primary outcome counts are not reported in the registry record, no conclusion about the direction, size, or precision of behavioral change can be drawn from this record. The study summary’s interest in blood chemicals likewise does not establish that any biomarker changed; biomarker results are not reported.
Even if score changes were available, this single-group design would not provide a randomized estimate of what HBOT changed compared with a control condition. A before-and-after pattern could be descriptive, but it would remain open to explanations other than treatment. The registry’s Phase 3 label does not resolve this limitation, and the record does not provide a statistical analysis plan that could clarify the intended inferential approach.
The adverse-event entries show that the registry recorded 1 participant with a serious adverse event and 5 participants with other adverse events, each among 10 at risk. Seizure is listed as a serious-event term with 10.0%. These entries warrant careful description, but they do not establish comparative safety, event causality, or the complete clinical context. The absence of a reported death count cannot be interpreted as evidence that no deaths occurred.
In short, the registry record is informative about the study’s stated aim, design, endpoint definition, and selected safety counts. It does not provide the primary behavioral findings needed to answer the central outcome question, nor the comparator and analysis details needed for a causal estimate. Any interpretation should remain within those limits.
13. Why This Trial Matters Statistically
This study illustrates why a registry entry should be read as a record of design and posted results, rather than as a complete trial report. The entry names a primary outcome and assessment times, but the outcome counts themselves are not reported in the registry record. Knowing how an endpoint was intended to be measured is useful, yet it is not a substitute for seeing the observed data and the analysis used to interpret them.
The single-group design is also statistically important. It can support a description of what was recorded among participants receiving HBOT, but it does not supply a randomized counterfactual. This distinction matters whenever the objective is to attribute change to an intervention. General statistical point: a treatment-associated time trend and a causal treatment effect are not interchangeable claims.
The registry’s juxtaposition of a Phase 3 designation with a title calling the study a pilot is another reason to distinguish labels from methods. The record does not explain that designation. Statistical interpretation should therefore rely on the documented design and outcome information rather than infer features from a phase label.
The adverse-event entries offer a limited example of how registry counts can inform, but not settle, a safety question. Counts need context about ascertainment, timing, definitions, and comparison. Those details are not reported. In addition, absent primary outcome results make it difficult to place the safety information alongside the study’s intended behavioral findings.
For clinical researchers, the central statistical lesson is that transparent reporting requires more than naming an endpoint. Outcome distributions, analysis populations, missing-data handling, uncertainty estimates, and the analysis plan are all needed to judge what the data support. Here, several of those elements are not reported in the registry record. The study therefore matters as an example of the boundary between a registered research question and an interpretable estimate of its answer.
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 of the available record is that it does not report the primary CGI-I outcome counts or an analysis of them. Although the assessment times and score descriptions are given, the observed behavioral results cannot be evaluated. Biomarker results are also not reported in the registry record, despite the study summary’s stated interest in blood chemicals.
The single-group design limits causal interpretation. Allocation is recorded as NA, and the record identifies no like-for-like randomized control comparison. Masking is listed as NONE. The record does not explain whether outcome assessors were otherwise blinded, so no such protection should be assumed. The CGI-I depends on clinician judgment informed by parent interview and observation, making clear assessment procedures important.
Further methods details are absent, including eligibility criteria, the HBOT schedule, treatment exposure, details of outcome collection, and the statistical analysis plan. Missing-data procedures, analysis populations, protocol deviations, and any multiplicity strategy are not reported. These gaps restrict assessment of potential bias, completeness of follow-up, and the appropriateness of any statistical conclusions.
Safety information is limited to entered counts and a listed serious-event term. The record does not provide event narratives, timing, attribution, or a comparison group, and deaths are not reported in the registry record. Consequently, the adverse-event entries cannot provide a comprehensive assessment of safety or comparative risk. The listed enrollment is small, and the record does not report uncertainty estimates for safety or efficacy outcomes.
Finally, the Phase 3 designation and pilot-study title appear together without an explanation of how the designation relates to the study’s design. The registry’s planning illustrations are not the trial’s own design, and no sample-size justification is reported. These limitations concern what can be concluded from the registry record; they do not establish what may be available in other documents.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Pilot Study of the Effect of Hyperbaric Oxygen Treatment on Behavioral and Biomarker Measures in Children With Autism — NCT00584480.
- Registry record last updated 2012-08-07; status last verified 2012-08.
- Results first posted 2012-04-13.
- 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.
18. Record Summary
NCT00584480 is a completed single-group study sponsored by the University of California, San Francisco, with a registry-listed enrollment of 10. The record lists a Phase 3 designation and describes the study as a pilot of Hyperbaric Oxygen Treatment (HBOT) in children with autism. Allocation is NA and masking is NONE.
The primary outcome is the number of participants with each CGI-I score at baseline, 8 Weeks from baseline, and 20 Weeks from baseline. The registry describes clinician ratings informed by parent interview and direct observation, but does not report the score counts or an analysis. Biomarker results are also not reported in the registry record.
For the HBOT group, the record lists 1 participant with a serious adverse event among 10 at risk and 5 with other adverse events among 10 at risk. Seizure is listed as a serious-event term with 10.0%. Deaths are not reported in the registry record. The available information supports a limited description of study design and entered safety counts, but not an estimate of behavioral effect or a controlled comparison.