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 DAW1033B2 in Obstructive Sleep Apnea (NCT03426631), a randomized phase 1, phase 2 study. The primary outcome is Apnea Hypopnea Index (AHI, Events/Hour of Sleep).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03426631 |
| Phase | Phase 1, Phase 2 |
| Condition | Obstructive Sleep Apnea (OSA) |
| Allocation | Randomized |
| Design model | Crossover |
| Masking | Quadruple |
| Enrollment | 13 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Brigham and Women's Hospital |
| Results posted | Yes |
2. Clinical Question
Population
The registry record concerns people with obstructive sleep apnea (OSA). Enrollment was 13. The primary outcome results list 12 observations for each treatment condition; the record does not explain the difference between enrollment and the outcome counts. The outcome description discusses a subgroup characterized by moderate sleep apnea, mild obesity, and specified ventilatory traits, but the reported results do not identify a subgroup analysis.
Intervention
The intervention was a DAW1033B2 oral capsule administered before sleep. The registry describes the trial as phase 1 and phase 2, randomized, quadruple-masked, and crossover. The primary outcome was assessed over one night. Details of the dosing regimen beyond administration before sleep are not reported in the registry record.
Comparator
The comparator was a placebo oral capsule, also described as administered before sleep. The listed outcome results report a median AHI of 33.0 events per hour of sleep for placebo and 20.4 for DAW1033B2. The record does not explain the order in which participants received the two conditions or provide additional details about the crossover schedule.
Primary question
The primary question was whether DAW1033B2 changed apnea-hypopnea index (AHI), measured as events per hour of sleep, compared with placebo. The registry reports a Wilcoxon (Mann-Whitney) analysis of final values, a median difference of -11.7, and p = 0.47. No confidence interval for that treatment contrast is reported in the registry record.
3. Trial Design
The registry describes a randomized, quadruple-masked crossover trial of DAW1033B2 in obstructive sleep apnea. Its phase is listed as phase 1 and phase 2, its status as completed, and its enrollment as 13. Brigham and Women's Hospital is listed as sponsor. The registry lists a start date of 2017-09-15 and a primary completion date of 2019-10-01.
In a crossover design, participants receive more than one treatment condition, so comparisons can be made within participants. That is a general description of the design type, not a claim about the exact sequence or conduct of this trial. The record does not report treatment order, the interval between conditions, washout procedures, or whether all participants completed both conditions. Randomization and quadruple masking are specified, but the identities of the four masked parties are not reported in the registry record.
The intervention and comparator are listed as oral capsules: DAW1033B2 and placebo. Both arm descriptions say the capsule was given before sleep. The primary outcome is AHI over one night. Results list 12 observations under each condition, while enrollment is listed as 13; the reason for this difference is not reported in the registry record. The record does not state how the analysis handled the crossover structure or whether the reported group comparison accounts for pairing within participants.
The trial record includes one outcome measure and one reported analysis. It does not describe an allocation ratio, the method used to generate the random sequence, the masking procedures, or a sample-size justification. These details should not be inferred from the design labels alone.
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.
Overall Study
| Stage | Placebo First, DAW1033B2 Second | DAW1033B2 First, Placebo Second |
|---|---|---|
| STARTED | 7 | 6 |
| COMPLETED | 7 | 5 |
| NOT COMPLETED | 0 | 1 |
Reasons for not completing this period:
| Reason | Placebo First, DAW1033B2 Second | DAW1033B2 First, Placebo Second |
|---|---|---|
| Lost to Follow-up | 0 | 1 |
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 | All Analyzed Participants |
|---|---|---|
| Participants analysed | 12 | |
| Age, Continuous | years | 54 (46.3 to 58.5) |
| Sex: Female, Male: Female | Participants | 3 |
| Sex: Female, Male: Male | Participants | 9 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 |
| Race (NIH/OMB): Asian | Participants | 0 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 |
| Race (NIH/OMB): Black or African American | Participants | 4 |
| Race (NIH/OMB): White | Participants | 8 |
| Race (NIH/OMB): More than one race | Participants | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 |
6. Primary Endpoint
The registry lists the primary outcome as Apnea Hypopnea Index (AHI, Events/Hour of Sleep), measured over 1 night.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | Inter-Quartile Range |
| Unit | events/hour of sleep |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is apnea-hypopnea index (AHI), expressed as events per hour of sleep. AHI counts apnea and hypopnea events relative to time asleep, making it a measure of the frequency of breathing disturbances during sleep. The registry specifies a median and an inter-quartile range as the intended summary, with a timeframe of one night.
AHI is used in sleep-apnea research because it expresses event frequency on a common time scale. A median describes the middle of the observed values and is less sensitive than a mean to unusually high or low observations. The inter-quartile range describes the central spread of a distribution. These are general statistical explanations; the record does not give inter-quartile ranges for the posted arm results.
The results list a median AHI of 33.0 events per hour of sleep for placebo and 20.4 for DAW1033B2. They also list lower and upper values for each condition, but their statistical meaning is not specified in the record. A single night provides a limited snapshot; the registry does not report repeated-night measurements or longer-term AHI results. The endpoint therefore describes the posted one-night measurements, not a broader or longer-term outcome.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted outcome results give a median AHI of 33.0 events per hour of sleep under placebo and 20.4 under DAW1033B2. The reported comparison is a Wilcoxon (Mann-Whitney) analysis of final values, with a median difference of -11.7 and p = 0.47. The record does not provide a confidence interval for this contrast. It also does not explain how the reported median difference was calculated or how the crossover design was represented in the analysis.
The reported p-value does not establish that the treatments are equivalent, nor does it establish that a difference is absent. In general, a p-value describes how compatible the observed data are with a specified null model; it is not the probability that the null hypothesis is true. A value of 0.47 indicates that the reported test did not produce strong evidence against its null hypothesis under the test's assumptions. It does not quantify the size or precision of the treatment effect by itself.
The individual arm summaries are medians, and the record lists lower and upper values of 24.1 and 41.9 for placebo, and 6.4 and 47.9 for DAW1033B2. The registry does not identify what these bounds represent. They should therefore not be interpreted as confidence intervals or as a measure of treatment-effect uncertainty. The IQR is named as the dispersion measure, but no IQR values are reported.
Because this was a crossover trial, within-participant comparisons may be important to the analysis. The reported method is named as Wilcoxon (Mann-Whitney), but the record does not say whether a paired version or another approach was used. The posted contrast therefore should be read as the analysis described in the registry, without assuming it fully accounts for the crossover structure.
The outcome counts are 12 under each condition, compared with enrollment of 13. The record does not explain missing or excluded observations, and it does not report an analysis population. Consequently, the results alone do not show why the counts differ or how missing outcomes may have affected the estimate. No additional secondary outcome results are posted.
Overall, the registry reports the outcome summaries and test result, but the uncertainty around the treatment contrast is not characterized by a confidence interval. A non-significant test is not proof of no effect, while the lower median in the DAW1033B2 condition is not sufficient by itself to establish a treatment effect. Interpretation depends on the analysis details and uncertainty that the registry record does not report.
9. Statistical Methods Explained
The reported inferential method is the Wilcoxon (Mann-Whitney) test, and the outcome is summarized using medians. In general, this rank-based test compares the distributions of observations between conditions using their relative ordering rather than relying on a normal-distribution model for the raw values. It is commonly used when data are skewed or when an analysis based on ranks is preferred. Its interpretation is not automatically a test of medians alone: depending on distributional shapes, it can reflect broader differences in the distributions.
The registry labels the analysis parameter as “Median Difference (Final Values)” and reports -11.7, with p = 0.47. A confidence interval for the difference is not reported. The method description does not specify whether the test was paired, how the median difference was derived, or whether covariates were included. Those details matter in a crossover trial, where measurements from the same participant under different conditions are usually related. The registry record does not explain how that relationship was handled, so the test label should not be taken as evidence of a particular paired-analysis procedure.
In general, a confidence interval gives a range of estimates compatible with the data and a stated statistical procedure. Its width communicates precision: wider intervals indicate less precision, and narrower intervals indicate more. No confidence interval for the treatment contrast is reported in the registry record. The outcome listings include lower and upper values alongside each median, but their meaning is not stated; they cannot be treated as intervals for the treatment effect without that information.
The outcome description names the inter-quartile range as the dispersion measure. In general, the inter-quartile range spans the central half of a distribution and is useful for describing spread when observations are skewed or outliers are present. The posted arm results do not include numerical inter-quartile ranges. As a result, the variability of AHI within each condition cannot be characterized from the stated dispersion measure in the registry results.
Over-dispersion is a general concern when outcome variability exceeds what an assumed statistical model expects. It can make standard errors too small and confidence intervals too narrow if the model does not account for the extra variability. Rank-based procedures do not depend on the same normality assumptions as a simple comparison of means, but they still require appropriate handling of the study design and data dependence. The registry does not report an over-dispersion assessment or a model-based analysis, so no conclusion about that issue can be drawn for this trial.
The crossover structure also raises general design considerations. Period effects, treatment order, and carryover can influence comparisons if they are present and not addressed appropriately. The record does not report the treatment sequence, washout interval, tests for period or carryover effects, or the statistical method used to account for them. It would be inappropriate to assume that these issues were assessed or that they were absent.
In general, randomization aims to reduce systematic differences in treatment assignment, while masking can reduce bias in treatment delivery, outcome assessment, or reporting. The registry identifies the trial as randomized and quadruple-masked, but does not describe the procedures, sequence generation, concealment, or the identities of the masked parties. Those labels do not by themselves establish how successfully these protections were implemented.
The record states enrollment of 13 and lists 12 outcome observations in each condition. It does not report whether the analysis included all randomized participants, only participants with measurements under both conditions, or another population. It also does not explain missing-data handling. Without those details, the analysis population and the implications of incomplete outcome information are uncertain.
The registry lists one outcome measure and one analysis. It does not report a statistical analysis plan, prespecified test details, adjustment for multiple comparisons, interim analyses, or a sample-size justification. Their absence from the record does not establish whether these steps were or were not conducted; they are not reported in the registry record.
In general, a planning calculation based on an observed effect is not a measure of the trial's observed power. The stated planning assumptions describe a two-sided significance level of 0.05, equal allocation, a normal approximation, and an effect set as a fraction of the observed effect. Such illustrations are conditional calculations for planning and should not be interpreted as the original design rationale or as evidence about the probability that this trial detected an effect. The trial's own sample-size justification is not reported in the registry record.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Wilcoxon (Mann-Whitney) | Median Difference (Final Values) -11.7 | not reported | 0.470 |
The registry reports the sponsor's analysis as a Wilcoxon (Mann-Whitney) test of final values, with a median difference of -11.7 and p = 0.47. No confidence interval or additional analysis details are reported. There is no recalculated treatment estimate presented in the registry entries for comparison with the sponsor's result, so no separate recalculation should be substituted for the posted analysis.
In general, a recalculation from summary statistics would be unadjusted and could fail to reproduce a sponsor analysis that uses participant-level data, a different analysis population, or design-specific methods. In a crossover study, handling the relationship between measurements from the same participant can also affect an analysis. The registry does not provide enough detail to establish whether the reported test accounted for that structure or to independently reproduce its result.
There is no stated discrepancy between the sponsor result and a recalculation. The sponsor's reported analysis is therefore the relevant inferential result in the registry. Its p-value should be interpreted as evidence from the named test, not as proof of equivalence or of no treatment effect. The absence of a confidence interval and of fuller method details limits assessment of the estimate's precision and the analysis assumptions.
11. Multiplicity, Missing Data and Analysis Populations
The registry lists one outcome measure and one reported analysis. It does not report a multiplicity adjustment or state whether any outcomes or analyses were prespecified in a statistical analysis plan. With only the posted information, it is not possible to assess whether additional outcomes, subgroups, time points, or analyses were considered but not reported. The record lists zero secondary outcomes, but this does not establish what may have been measured outside the posted registry results.
The record gives enrollment as 13 and reports outcome observations of 12 under placebo and 12 under DAW1033B2. It does not explain why the reported counts differ from enrollment or describe withdrawals, exclusions, missing measurements, or imputation. It also does not identify an intention-to-treat, per-protocol, or other analysis population. These details are not reported in the registry record.
For a crossover study, interpretation of missing data can depend on whether participants have measurements under both conditions and on how incomplete pairs are handled. This is a general consideration, not a claim that missing pairs occurred in this trial. The record does not provide participant-level flow or a missing-data analysis, so the effect of incomplete information cannot be assessed.
No interim analysis, stopping rule, or statistical analysis plan is reported in the registry record. Their absence from the registry should not be interpreted as proof that they did not exist; the record does not provide those details.
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 |
|---|---|---|---|
| Placebo | 0 | 12 | 0.0% |
| DAW1033B2 Oral Capsule | 0 | 12 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Placebo | 0 of 12 |
| DAW1033B2 Oral Capsule | 0 of 12 |
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 |
|---|---|
| Placebo | 0 of 12 |
| DAW1033B2 Oral Capsule | 1 of 12 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Placebo | DAW1033B2 Oral Capsule |
|---|---|---|---|
| Heartburn (gastroesophageal reflux) | Gastrointestinal disorders | 0 of 12 | 1 of 12 |
The registry lists no serious adverse events for either condition: the serious-event count is 0 under placebo and 0 under DAW1033B2. It also lists 0 deaths under each condition. The number at risk for serious events is 12 for each condition. These figures describe the recorded events in the listed safety groups; they do not establish that no other symptoms or events occurred.
For other, non-serious adverse events, the record lists 0 affected participants under placebo and 1 under DAW1033B2, with 12 at risk in each group. No serious adverse-event terms are listed. The registry does not describe event types, severity grading for the non-serious event, relatedness assessments, timing, or how events were collected.
In general, adverse-event counts from a small trial can be imprecise, and the absence of a recorded serious event does not demonstrate absence of risk. These counts also should not be interpreted as a comparative safety test. The registry does not report a formal safety analysis or enough detail to assess differences between treatment conditions.
13. What the Result Does, and Does Not, Mean
The registry results describe one-night AHI measurements under placebo and DAW1033B2. The reported medians are 33.0 and 20.4 events per hour of sleep, respectively. The reported Wilcoxon (Mann-Whitney) comparison gives a median difference of -11.7 and p = 0.47. No confidence interval for the contrast is reported, and the meaning of the lower and upper values attached to the arm summaries is not specified.
These results do not establish that the treatments are equivalent or that a difference is absent. A p-value of 0.47 is not an estimate of the probability that there is no treatment effect. Nor does the lower median in the DAW1033B2 condition, by itself, demonstrate a treatment effect. The uncertainty around the treatment contrast cannot be fully assessed without an interval estimate and clearer analysis details.
The trial is described as a crossover study, but the registry does not report how the analysis handled within-participant measurements, treatment sequence, or potential period and carryover effects. It also does not explain the difference between enrollment of 13 and the 12 observations listed in each treatment condition. These gaps limit assessment of design-specific analysis and missing-data issues.
The findings concern the posted outcome, population, and one-night timeframe. They do not establish longer-term effects, results for an unreported subgroup, or effects on outcomes other than AHI. The record lists no serious adverse events and one non-serious event in the DAW1033B2 condition, but sparse event counts do not establish comparative safety.
In general, statistical evidence must be interpreted together with precision, design, analysis population, and completeness of reporting. The registry provides a test result and descriptive summaries, but not all information needed to evaluate those dimensions. Accordingly, the results are best read as the posted summaries and analysis, with uncertainty about their precision and analysis details.
14. Why This Trial Matters Statistically
This trial is statistically notable because it combines a crossover design with a small listed outcome count and a rank-based comparison. Crossover studies can use within-participant information, but that requires methods that reflect the dependence between measurements from the same person. The registry names a Wilcoxon (Mann-Whitney) analysis but does not explain whether or how the crossover structure was accounted for. That makes the analysis description important to interpretation.
The contrast is reported without a confidence interval. A p-value alone does not show the range of treatment effects compatible with the data, so it gives limited information about precision. The outcome summaries identify medians and a planned inter-quartile range, but do not report numerical IQR values. The lower and upper values listed with the medians are not defined in the record, leaving their statistical interpretation unclear.
Enrollment is 13, while each condition's outcome listing has 12 observations. The record does not explain the counts or specify the analysis population. This illustrates why participant flow and missing-data reporting matter: a reader needs to know which observations informed the estimate and how incomplete measurements were handled.
These points do not invalidate the reported result. They identify limits on what can be learned from the registry entry alone. Clear reporting of the paired analysis, outcome dispersion, confidence interval, analysis population, and missing-data handling would help readers assess the precision and design alignment of the result.
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.
Several trial-specific details needed to interpret the reported comparison are not reported in the registry record. Although the design is labeled crossover, the treatment order, schedule, washout procedures, and method for handling within-participant measurements are not described. The record also does not state whether period or carryover effects were considered.
The reported arm summaries give medians and lower and upper values, but the meaning of those bounds is unspecified. The stated dispersion measure is the inter-quartile range, yet no numerical IQR values are reported. No confidence interval is given for the treatment contrast, limiting assessment of its precision.
Enrollment is listed as 13, while each treatment condition has 12 outcome observations. The registry does not explain the difference, provide participant flow, identify the analysis population, or describe missing-data handling. It does not report the method for generating the random sequence, allocation concealment, masking procedures, or which parties were masked.
The record includes one outcome measure and one analysis but does not report a statistical analysis plan, multiplicity approach, interim analyses, or the trial's sample-size justification. The outcome timeframe is one night, so the posted endpoint does not describe measurements over longer periods. These are limits of the registry information and should not be treated as evidence that unreported procedures were absent.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: DAW1033B2 in Obstructive Sleep Apnea — NCT03426631.
- Registry record last updated 2020-02-19; status last verified 2020-02.
- Results first posted 2020-02-07.
- 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
NCT03426631 is a completed, randomized, quadruple-masked crossover trial of DAW1033B2 in obstructive sleep apnea, listed as phase 1 and phase 2. Enrollment was 13. The intervention and comparator were oral capsules taken before sleep: DAW1033B2 and placebo.
The primary endpoint was median AHI in events per hour of sleep over one night. The registry reports medians of 33.0 under placebo and 20.4 under DAW1033B2, with a Wilcoxon (Mann-Whitney) median difference of -11.7 and p = 0.47. No confidence interval for the contrast is reported. The listed outcome count is 12 for each condition, and the reason it differs from enrollment is not reported.
The safety record lists no serious adverse events or deaths in either condition, and one non-serious adverse event under DAW1033B2. Details of the crossover analysis, missing data, and the statistical meaning of the arm-level lower and upper values are not reported in the registry record.