The Evaluation of a Nasal Pillows Mask for the Treatment of Obstructive Sleep Apnea (OSA) in NZ: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Evaluation of a Nasal Pillows Mask for the Treatment of Obstructive Sleep Apnea (OSA) in NZ (NCT02741635), a na na study. The primary outcome is Participants Choose the Toffee Mask Based on First Impressions Over 3 Market Leading Masks.

2. Clinical Question

3. Trial Design

The registry describes a completed, single-group device investigation sponsored by Fisher & Paykel Healthcare. It lists an enrollment of 15 and characterizes the study as having no allocation and no masking. There is no randomized assignment or separate control arm, so the design does not provide a like-for-like randomized comparison between the trial mask and another mask.

Participants currently using a nasal pillows mask were to use the new mask for one overnight in-lab polysomnography session. They were to retain their usual CPAP or APAP device and settings during the study. The stated purpose was to assess mask performance, including leak and comfort, participants’ overall acceptance, and feedback on design and first impressions. The registry identifies the primary outcome as a participant count based on choosing the Toffee mask over three market-leading masks. It lists comfort and ease of fitting or adjustment relative to participants’ current masks as secondary outcomes.

The record lists the study as starting in 2016-04 and reaching primary completion in 2016-05, with status completed. These status and enrollment fields do not substitute for outcome data. The number of participants included in each outcome analysis, the actual outcome values, and the statistical analysis methods are not reported in the registry record. A completed study can still have an incomplete public account of its results.

The population and overnight setting make the investigation relevant to immediate user impressions in a sleep-laboratory context. They do not, by themselves, establish how the device performs across other settings or over longer periods. The registry record does not report longer-term follow-up, a broader clinical endpoint, or a protocol for evaluating clinical effectiveness.

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

Reasons for not completing this period:

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Participants Choose the Toffee Mask Based on First Impressions Over 3 Market Leading Masks, measured over 12 hours.

The primary endpoint is described as the count of participants who choose the Toffee mask based on first impressions over three market-leading masks. The listed timeframe is 12 hours. As a count, the endpoint is intended to convey how many participants made the specified choice, rather than the magnitude of a continuous measurement such as leak or comfort on a numeric scale. However, the record does not report the observed count, the number assessed for this outcome, or additional details about how a choice was elicited or defined.

A participant-choice endpoint can be useful in an early device evaluation because it records an immediate preference or first impression. It is distinct from an endpoint that directly measures a clinical outcome, and it should not be interpreted as evidence of clinical benefit. The registry describes broader study interests in performance, comfort, acceptance, and design feedback, but the primary outcome title concerns preference.

Interpreting a count normally requires knowing the relevant analysis population and denominator: the count alone does not show how many participants had an opportunity to make a choice. A proportion could provide that context, but no such value should be inferred when the outcome result and denominator are not reported in the registry record. The outcome’s response options and handling of participants who did not choose a mask are also not reported. Without those details and the result itself, the endpoint’s observed meaning and precision cannot be assessed.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a participant count represent?

The primary outcome is classified as a count of participants with an event, where the event is choosing the Toffee mask according to the outcome title. A count is a direct tally. It is not automatically a percentage, a rate, or a measure of the size of a treatment effect. To interpret a count in context, readers generally need the number of participants whose outcome was assessed, the rule for deciding whether the event occurred, and a clear description of any missing responses. Those details and the observed count are not reported in the registry record.

For a binary outcome, each assessed participant is generally classified as having or not having the event under a stated definition. A count can then be described alongside its denominator, or expressed as a proportion if the denominator is known and reported. No proportion should be inferred for this investigation: the registry does not provide the outcome count or the number assessed. The primary outcome’s timeframe is 12 hours, but the record does not further explain the timing or procedure for eliciting the choice.

Why does the design affect interpretation?

This is a single-group study with no allocation to a separate control arm. The study plan describes participants using the new mask during an overnight laboratory session and retaining their usual CPAP or APAP device and setting. The secondary outcome titles refer to participants’ usual masks, but that is not the same as a randomized parallel comparison. A within-person report or preference can be informative about an individual’s experience, yet it can be influenced by the order and setting of assessments, expectations, or recall. These are general considerations; the registry does not report whether or how they were addressed.

Masking is listed as none. In general, lack of masking can matter when an outcome involves subjective judgment, because participants know which device they are using and may be aware of the study’s focus. This does not establish that responses were biased. It means that interpretation should distinguish subjective impressions from outcomes assessed independently of participant judgment. The registry does not report a blinded assessment or a method for reducing expectation effects.

How should uncertainty be assessed?

A confidence interval is a statistical range used to communicate the precision of an estimated quantity under a specified model and method. It does not guarantee that the true value lies in that interval, and its interpretation depends on the analysis assumptions. For a participant count or a proportion, an interval would require the result and the relevant analysis denominator. The registry record reports neither outcome results nor confidence intervals, so the precision of the primary or secondary outcomes cannot be assessed.

With a limited enrollment such as the 15 listed in the record, individual responses can have substantial influence on a summary. This is a general feature of small studies, not a calculated finding about this trial. A lack of reported statistical results also means that no hypothesis test, p-value, or adjusted estimate can be attributed to the study. The registry lists zero analyses, and the sponsor’s analysis methods are not reported in the registry record.

What do the planning assumptions mean?

The record describes sample-size assumptions as a planning illustration computed from the observed effect, rather than as the study’s own design. The assumptions specify a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the illustration considers fractions of the observed effect. These are planning assumptions, not results from the study and not observed power. Equal allocation between two arms does not describe the registered single-group design. The record does not report the trial’s original sample-size justification, and these assumptions should not be treated as that justification.

In general, a sample-size calculation asks how many participants a proposed design might require under specified assumptions about the effect, variability, significance level, and desired power. If the assumed effect is based on an observed estimate, uncertainty in that estimate can make the illustration unstable. A planning illustration cannot demonstrate that a study had adequate power after completion, nor can it replace a reported analysis. No numerical planning results are discussed because the registry record does not give those values.

Why can over-dispersion matter?

For count outcomes, a model may assume a particular relationship between the expected count and its variability. Over-dispersion means the observed variability is greater than that assumed by a model; ignoring it can make uncertainty appear too small. This is a general statistical issue, especially for counts accumulated across differing exposure times. Here the primary measure is described as a participant count, not a time-based event rate, and no count model or dispersion estimate is reported. It would therefore be inappropriate to claim that over-dispersion was present or that a particular method was used.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Toffee Nasal Pillows Mask")
n      <- c(14)
events <- c(4)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Toffee Nasal Pillows Mask  14  4  28.6  8.4  58.1  11.7  54.6  11.8  55.1

11. Multiplicity, Missing Data and Analysis Populations

The record lists a primary outcome and two secondary outcome descriptions: participant choice based on first impressions, comfort relative to the usual mask, and ease of fitting and adjustment relative to the current mask. When several outcomes are examined, the chance of at least one apparently positive finding can increase if each is tested separately without accounting for multiplicity. This is a general statistical concern; it does not show that any finding in this study was significant or spurious.

A multiplicity strategy could specify which outcome is primary, how secondary outcomes will be interpreted, whether tests are adjusted, or whether results are descriptive. Such a strategy and any testing hierarchy are not reported in the registry record. The record lists zero analyses, and outcome results, p-values, and confidence intervals are not reported. The number of outcome measures is three, but this does not reveal how many tests, if any, were planned or performed.

Missing-data information is also absent. The record lists an enrollment of 15 but does not report how many participants completed the overnight assessment, contributed to each outcome, or had missing responses. It does not describe imputation, exclusions, or sensitivity analyses. Analysis populations—such as all enrolled participants or only those with an observed outcome—are not reported in the registry record. These gaps prevent assessment of whether missingness or analytic choices could affect the findings. No assumptions about missing responses or participant exclusions should be made.

12. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry lists one group, Toffee Nasal Pillows Mask, with 15 participants at risk for serious adverse events and zero affected. It also lists 15 at risk for other adverse events and zero affected. These are the reported event counts for the group; they do not establish that every possible safety event was actively assessed or that no event occurred outside the reported categories.

The deaths field is not populated, so deaths are not reported in the registry record. The list of leading serious-event terms is empty. An empty term list does not provide detail beyond the absence of listed terms. The registry does not report event definitions, collection procedures, follow-up beyond the stated study context, or whether all participants contributed safety information.

Safety counts should be read in light of the number assessed, the observation period, and the completeness of event reporting. The record’s risk denominators and affected counts offer limited descriptive information, but no comparative safety estimate is available from this single-group design. These figures do not establish a broader safety profile or predict event frequency in other populations or settings.

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

The record documents an evaluation focused on initial impressions and overnight use of a new nasal pillows mask among participants with obstructive sleep apnea who already use this type of mask. Its primary outcome concerns participant choice, and its secondary outcomes concern comfort and ease of fitting or adjustment relative to a usual mask. These aims are relevant to understanding user experience, but the registry does not report the outcome values.

Without reported results, it is not possible to say how many participants preferred the Toffee mask, how participants rated comfort, or whether they found fitting or adjustment easier. The enrollment of 15 is not itself an outcome and cannot answer those questions. The record also does not report estimates of leak, objective sleep measures, or statistical uncertainty. No conclusion about the direction or size of any effect can be drawn.

Even if preference results were available, a first-impression choice would not by itself answer whether a mask improves clinical outcomes or performs better over longer-term use. The registry describes a one-night in-lab assessment and does not report longer follow-up. The single-group design also lacks a randomized control arm for separating device-related differences from other influences. These are limits on what this design could establish, not claims about what participants experienced.

The appropriate conclusion is narrow: the registry describes a completed investigation intended to collect overnight performance and user feedback, but it does not provide results sufficient to characterize those responses. The record can inform readers about the study question and its design, while leaving the primary and secondary findings unresolved.

14. Why This Trial Matters Statistically

This investigation illustrates a distinction between a study’s stated purpose and the evidence that can be interpreted from its registry record. The purpose includes user experience and device performance, but a clear outcome title alone does not establish what participants reported. Without the observed values, denominators, and analysis results, readers cannot quantify the responses or their uncertainty.

It also highlights why endpoint definition matters. A count of participants choosing a mask is a different type of outcome from a measurement of leak, a comfort score, or a clinical endpoint. Each answers a different question and requires its own definition and reporting. The registry identifies choice as the primary outcome and describes comfort and fitting-related secondary outcomes, but it does not provide the response rules or outcome values needed to interpret them fully.

The single-group design and unmasked assessment are especially relevant for subjective outcomes. In general, a preference may reflect the user’s experience in the study setting as well as the device itself. A randomized comparison and clearly specified outcome procedures can help address some interpretive questions, but the registry does not report such a comparison or detailed procedures. This study should therefore be read as an investigation of immediate impressions, not as a randomized efficacy comparison.

Finally, planning assumptions must be kept separate from the completed study’s evidence. A hypothetical two-arm calculation does not describe the registered single-group design and is not observed power. The record is a useful reminder that transparent reporting of results, analysis populations, uncertainty, and missing data is needed for readers to assess what a small device investigation establishes.

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 results account is incomplete for statistical interpretation. The registry lists the primary and secondary outcomes but does not report their observed values, analysis denominators, confidence intervals, or statistical tests. It also does not report how participants made their choice, what response options were available, or how comfort and fitting responses were recorded. These omissions prevent assessment of the outcome definitions and precision.

The enrollment field lists 15, but the number contributing to each outcome is not reported. There is no account of missing responses, withdrawals, exclusions, or analysis populations. The record does not report the original sample-size justification, interim analyses, or a prespecified statistical analysis plan. Its planning illustration uses assumptions for two equally allocated arms, whereas the registered design is single-group; the illustration should not be interpreted as the trial’s design or as observed power.

The investigation was unmasked and had no randomized control arm. Because the outcomes include subjective impressions and comparisons with usual masks, the absence of a parallel randomized comparison limits causal interpretation. The overnight laboratory setting and the described one-night use also constrain what can be inferred about experiences in other settings or over longer periods. The registry does not report enough detail to assess how broadly the recruited participants represent other people with OSA.

Safety reporting is limited to the listed group counts. The deaths field is not populated, and event definitions and collection details are not reported in the registry record. These limitations do not establish that unreported events occurred; they mark the boundaries of what can be concluded from the record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: The Evaluation of a Nasal Pillows Mask for the Treatment of Obstructive Sleep Apnea (OSA) in NZ — NCT02741635.
• Registry record last updated 2024-09-19; status last verified 2024-08.
• Results first posted 2020-05-15.
• 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 describes a completed, single-group evaluation of a new nasal pillows mask in participants with obstructive sleep apnea. It lists an enrollment of 15, no allocation, and no masking. Participants were to use the mask for one overnight in-lab polysomnography session while using their usual CPAP or APAP device and setting.

The primary outcome concerns the number choosing the Toffee mask based on first impressions over three market-leading masks, with a listed timeframe of 12 hours. Secondary outcomes concern comfort and ease of fitting or adjustment relative to usual masks. The registry does not report outcome results, analysis populations, or statistical analyses. It lists zero participants affected by serious adverse events and zero affected by other adverse events in the mask group; deaths are not reported in the registry record.