In Home Assessment of Three Anti-Snoring Devices, a Cross Over Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for In Home Assessment of Three Anti-Snoring Devices, a Cross Over Study (NCT03333876), a non-randomized na study. The primary outcome is Bed Partners' Rating of Sleep Disturbance Due to Partner Snoring.

2. Clinical Question

3. Trial Design

This was a completed, nonrandomized crossover study of three anti-snoring devices. The registry lists an enrollment of 58, the condition as snoring, and Philips Respironics as the sponsor. The recorded start date is 2017-10-09 and the primary completion date is 2017-12-03. Masking is listed as none. The study is not presented as a randomized comparison, so the device groups should not be treated as if they were randomized, like-for-like control groups.

The interventions were a nasal dilator, mandibular advancement, and positional therapy. The study description names Mute, myTAP V, and the Sleep Positional Trainer as examples of the devices, alongside the Silent Night mobile application. The record describes the app as including an intake questionnaire, guidance to one of the device approaches, overnight snoring recording, and a snore score. The primary outcome timeframe is listed as 5 weeks, with feedback evaluated at the end of each period. The duration of each period and the sequence or timing of device use are not reported in the registry record.

A crossover design generally evaluates more than one intervention within participants, allowing comparisons that can account for stable differences between people. That is a general feature of the design, not a description of the analysis performed in this study. Without the assignment procedure and the order and timing of device use, it is not possible to assess how the study addressed order effects, changes over time, or carryover from one device period to another. The registry record does not report those design details or the statistical model used for the primary comparison.

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: Of the 29 couples, 2 participants screened out. Therefore 27 participants started the study.

Screening Period

Reasons for not completing this period:

Nasal Dilator

Reasons for not completing this period:

Mandibular Advancement

Reasons for not completing this period:

Positional Therapy

Analysis population for the primary outcome, as defined in the registry: This includes all bed partners that completed the subjective feedback survey.

5. Baseline Characteristics

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

Population: This is all participants that were consented and eligible to participant in the study, all baseline demographics include snorer and bed partner with the exception of neck circumference and Body Mass Index (BMI).

6. Primary Endpoint

The registry lists the primary outcome as Bed Partners' Rating of Sleep Disturbance Due to Partner Snoring, measured over 5 weeks.

The primary endpoint is bed partners’ rating of sleep disturbance due to their partner’s snoring. It is described as a subjective response to the question of how much the snoring disturbed sleep the previous night, recorded on a 0 to 10 scale. The registry defines 0 as the worst and 10 as the best. Ratings were evaluated at the end of each period, and the endpoint is summarized as a mean with a standard deviation. The timeframe is listed as 5 weeks.

This measure focuses on the experience of the bed partner rather than a device’s effect on an objective measure of snoring. A rating scale can capture a participant’s or bed partner’s experience in a form that can be summarized across respondents. As a general statistical point, a mean describes the average recorded score, while a standard deviation describes how much scores vary around that average. Neither quantity, by itself, establishes whether a difference between devices is important or precise.

The registry record does not report the endpoint scores, means, standard deviations, number of completed surveys, or comparisons among devices. It also does not report whether the same bed partner submitted a rating for each period or how incomplete ratings were handled. Therefore, the endpoint definition is available, but its observed distribution and results are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical methods explain how observations are turned into estimates and how uncertainty is represented. The registry record identifies a continuous primary outcome summarized as a mean and standard deviation. It does not report a primary analysis model, an estimate, a confidence interval, or a test result. The explanations below describe general concepts and do not imply that a particular method was used for this study.

What does a mean and standard deviation describe?

For a continuous measure, the mean is the arithmetic average of recorded values. The standard deviation describes the spread of individual values around that average. A mean alone can conceal substantial variation: participants or bed partners can report different experiences even when the average score is similar. The standard deviation is descriptive and is not the same as the uncertainty around the mean. A standard error or confidence interval is used to express uncertainty in an estimate, with the calculation depending on the analysis and data structure.

Here, the endpoint was a bed partner’s rating on a 0 to 10 scale, with 0 identified as the worst and 10 as the best. The record does not give the observed mean or standard deviation for any device period. It also does not say whether the primary comparison used the raw ratings, a change in ratings, or another summary. Those details are not reported in the registry record.

Why does the crossover structure matter?

As a general concept, a crossover study collects observations under different interventions across periods. When the same participant contributes observations under multiple interventions, an analysis can account for the fact that observations from one person are related rather than independent. This within-participant structure can be useful because stable personal characteristics do not vary across that participant’s periods. However, a valid analysis also has to consider whether outcomes change with time, whether the order of interventions matters, and whether the effect of one intervention persists into a later period.

The registry record labels the study as a crossover, but does not report the device sequence, period-specific schedule, washout procedure, or statistical model. It therefore cannot establish whether a paired analysis was used or how period, sequence, and possible carryover effects were handled. These are questions about the analysis plan, not evidence that a particular problem occurred.

What can uncertainty measures tell readers?

In general, a confidence interval gives a range of values compatible with the data under the method’s assumptions. A narrow interval indicates greater precision than a wide interval, but precision does not by itself establish clinical importance. A p-value is meaningful only in relation to a defined hypothesis and analysis. Neither measure is reported for the primary endpoint in the registry record. Without an estimate and its uncertainty, readers cannot assess how well the data distinguish among possible differences in bed-partner ratings.

How should planning illustrations be read?

The registry description of the sample-size illustration identifies a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It also states that the effect to detect is a stated fraction of the observed effect. For time-to-event outcomes, the illustration is based on a total number of events; for other outcomes, it is expressed as participants per arm. These are planning assumptions, not evidence that the trial was designed or analyzed that way. The record does not report the trial’s own sample-size justification.

A planning illustration based on an observed effect is not observed power and does not establish the probability that a study has detected a real effect. It is conditional on the assumptions and effect size used for the calculation. In this study, the primary outcome is continuous, while the planning description refers to a two-arm illustration. The study itself is described as a nonrandomized crossover of three device approaches. The illustration should therefore not be mistaken for a description of the trial’s allocation or primary analysis. The record does not provide a numerical planning result in the information available for this analysis.

More generally, statistical interpretation depends on both the estimate and the design. A numerical difference may be vulnerable to confounding when allocation is not randomized. In a crossover study, a difference may also reflect period or order effects if those are not adequately handled. Statistical methods can address some sources of variation under assumptions, but the method and its assumptions need to be reported for readers to evaluate them. The registry record does not report these details for the primary outcome.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Nasal Dilator", "Mandibular Advancement", "Positional Therapy")
n   <- c(24, 25, 27)
m   <- c(5, 3.8, 4.4)
sp  <- c(1.6, 2.2, 2.5)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Nasal Dilator  24  5.000  1.600  4.324  5.676
CARM  Mandibular Advancement  25  3.800  2.200  2.892  4.708
CARM  Positional Therapy  27  4.400  2.500  3.411  5.389

11. Multiplicity, Missing Data and Analysis Populations

The registry lists the bed-partner rating as the primary outcome and four secondary outcomes: users’ acceptance of each solution, bed partners’ acceptance of each solution, overall satisfaction of the bed partner for each solution, and the total number of audio recordings. Considering multiple outcomes or multiple comparisons can increase the chance of finding an apparently notable result by chance. As a general statistical approach, a study may define one primary comparison in advance or use an adjustment when several tests are treated as confirmatory. The registry record does not report a multiplicity strategy, a hierarchy among comparisons, or which comparisons were treated as confirmatory.

The record identifies one analysis, but does not describe its specification or explain how analyses for the primary and secondary outcomes were handled. The primary population is described as all bed partners who completed the subjective feedback survey. The number who completed it, the number of missing ratings, and reasons for missing data are not reported in the registry record. Methods for handling missing observations, including any imputation or sensitivity analysis, are also not reported.

The record does not identify an intention-to-treat population, a per-protocol population, or another analysis population for the primary endpoint. Nor does it explain whether all enrolled participants contributed data to each device period. Because the study is a crossover, the amount of data available for each person and period matters for interpreting the analysis. Without completion counts and analysis-population definitions, the reader cannot determine how the endpoint population relates to the enrollment of 58 or how missing observations might affect the reported results. No primary-outcome results are reported in the registry record.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry lists serious adverse events for each device group, with 27 at risk in each group, zero serious events affecting participants in each group, and zero deaths in each group. For other adverse events, the record lists zero affected in the nasal dilator group, 3 affected in the mandibular advancement group, and zero affected in the positional therapy group. These are the posted counts; the registry record does not provide descriptions of the other events in this information.

These figures should be read as recorded counts, not as proof that no serious events occurred outside the recorded study information or that the devices have equivalent safety. The record does not report event definitions, severity details for the other events, timing, relatedness assessments, or whether a participant could be counted in more than one group. Because this is a crossover study and the listed groups each have 27 at risk, the counts should not be interpreted as three independent randomized arms. The enrollment of 58 does not establish that the safety groups represent separate sets of participants.

The registry record does not list serious-event terms. Zero recorded serious events and zero deaths are therefore descriptive entries, not a comparative statistical analysis. No significance test or estimate of uncertainty for safety differences is reported.

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

The record describes a study designed to assess three anti-snoring device approaches in a crossover framework, with bed-partner ratings of sleep disturbance as the primary outcome. It also records that the study was completed and lists safety counts for the three approaches. These details describe the study’s scope and the outcomes it set out to collect.

The record does not report the primary ratings, means, standard deviations, or comparisons among devices. It therefore does not show whether ratings differed, how large any difference was, or how precisely a difference was estimated. It also does not report confidence intervals or p-values for the primary outcome. The direction and magnitude of the primary outcome result are not reported in the registry record.

Even if comparative estimates were available, interpretation would need to account for the nonrandomized allocation and the crossover structure. In general, nonrandomized assignment can make device groups differ in ways that influence outcomes, while crossover observations may be affected by period or sequence. The registry record does not report how these concerns were addressed. This is a limitation of what can be interpreted from the record, not evidence that a bias necessarily occurred.

The safety entries show zero recorded serious events and zero deaths for each listed group, and 3 participants affected by other adverse events in the mandibular advancement group. These counts do not establish comparative safety or explain the nature of those events. Overall, the registry information describes the trial and selected safety counts, but does not provide enough outcome data to assess the primary comparative result.

14. Why This Trial Matters Statistically

This trial raises several statistical issues that are common in clinical research. First, crossover studies collect information across periods, so observations may be related within a participant. Analyses need to reflect that structure rather than assume all observations are independent. The registry record identifies a crossover design but does not report the primary analysis method or how period and sequence were handled.

Second, allocation is listed as nonrandomized. Randomization is one way to reduce systematic differences between treatment groups, while nonrandomized allocation may leave potential confounding. In a crossover design, each participant may contribute observations under more than one approach, but that feature does not automatically resolve all sources of bias. Clear reporting of allocation and analysis is important for judging what comparisons can support.

Third, the primary endpoint is a subjective rating from bed partners. Such measures can address an experience that an objective recording alone may not capture, but their interpretation depends on clear measurement and complete reporting. The registry defines the scale and timing, yet does not report the observed scores or the number of completed surveys. Without those details, the endpoint cannot be interpreted comparatively.

Finally, transparent reporting distinguishes a study’s design from its results. The record gives enrollment, intervention categories, endpoint definition, and selected safety counts, but does not report the primary outcome estimates or their uncertainty. This makes the study statistically instructive as an example of why design labels, analysis methods, outcome data, and missing-data information all matter when evaluating a clinical comparison.

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 most direct limitation for interpretation is that the registry record does not report the primary endpoint results. Device-specific ratings, means, standard deviations, comparisons, and measures of uncertainty are absent. The record also does not report the number of bed partners who completed the survey, although the primary population is defined by survey completion. This prevents assessment of how much primary-outcome information was available.

Important design and analysis details are not reported in the registry record. These include the device-use sequence, the duration of each period, any washout procedure, the assignment process, and the statistical model for the crossover data. The record also does not state how period effects, order effects, or possible carryover were evaluated. These omissions limit assessment of the validity and precision of comparisons, without establishing that any specific bias occurred.

The registry record does not report the sample-size justification for the trial, a detailed analysis plan, a multiplicity strategy, or methods for handling missing outcome data. The primary analysis population beyond bed partners who completed the survey is not described. For safety, the record lists counts but does not provide event descriptions or relatedness assessments for other adverse events. The nonrandomized allocation and lack of masking are additional design features to consider when interpreting subjective ratings. No primary-outcome conclusion can be drawn from the registry information alone.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: In Home Assessment of Three Anti-Snoring Devices, a Cross Over Study — NCT03333876.
• Registry record last updated 2019-02-15; status last verified 2019-02.
• Results first posted 2019-02-15.
• 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.

19. Record Summary

The registry record describes a completed, nonrandomized crossover study of a nasal dilator, mandibular advancement, and positional therapy for snoring. Enrollment was 58, masking was none, and the sponsor was Philips Respironics. The primary outcome was bed partners’ rating of sleep disturbance due to snoring on a 0 to 10 scale, assessed at the end of each period over a timeframe listed as 5 weeks.

Primary-outcome results, including ratings and comparisons among devices, are not reported in the registry record. Safety entries list zero serious adverse events and zero deaths in each group; other adverse events affected 3 participants in the mandibular advancement group and none in the other listed groups. The record does not report the primary analysis method, the completed survey count, or details of missing-data handling.