The Evaluation of the Toffee Nasal Pillows Mask for the Treatment of Obstructive Sleep Apnea: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Evaluation of the Toffee Nasal Pillows Mask for the Treatment of Obstructive Sleep Apnea (NCT03242148), a na na study. The primary outcome is Toffee Mask Usability.

2. Clinical Question

The clinical question concerns how participants with obstructive sleep apnea rated the usability of the Toffee Nasal Pillows Mask after using it at home. The record describes a prospective, non-randomized, non-blinded investigation of the mask's performance, comfort and ease of use. Its primary outcome is titled “Toffee Mask Usability” and is based on a participant questionnaire administered during the second visit.

This question is about reported usability rather than a randomized comparison of treatments. The registry also names objective treatment performance as a secondary outcome, but does not give its result in the record. The primary outcome description does not report how many participants selected each response category, or any summary of those responses. As a result, the record establishes what the investigators intended to assess, but does not allow the usability ratings or their precision to be evaluated.

3. Trial Design

The registry describes a prospective, non-randomized, non-blinded, single-group device investigation. The intervention is the Toffee Nasal Pillows Mask, and no separate control arm is listed. Allocation is recorded as NA and masking as none. Accordingly, the design is not a randomized comparison and cannot separate changes associated with mask use from other explanations by comparing outcomes against a concurrent control group.

The record lists enrollment as 38. Its brief summary says that up to 45 people with obstructive sleep apnea were to be recruited from the Ohio Sleep Medical Institute. The record does not explain the difference between the planned recruitment ceiling and the listed enrollment, or state how many participants contributed data to the primary outcome. These should not be treated as interchangeable measures of analyzed participants.

The mask-use period is described as 14 ± 5 days from visit 2, with the primary outcome timeframe stated as 14 ± 5 days in-home. The primary questionnaire was administered during the second visit and asked about overall usability. The study is marked completed; its recorded start date is 2017-09-16 and its primary completion date is 2017-11-03.

The registry lists five outcome measures and no analyses. It does not report a statistical analysis plan, the recruitment or follow-up flow, the number of participants completing the questionnaire, or an account of deviations from the intended use period. Completion status alone does not establish that every planned assessment was obtained or analyzed.

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

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 Toffee Mask Usability, measured over 14 ± 5 days in-Home.

The primary endpoint is “Toffee Mask Usability,” measured as a count of participants according to the endpoint description. The accompanying narrative says participants rated overall mask usability on a five-point Likert-type scale: very easy, easy, neither, difficult or very difficult. The questionnaire was administered during the second visit, with the timeframe recorded as 14 ± 5 days in-home.

A usability questionnaire is used to capture participants' subjective experience, which is distinct from an objective performance measure. The stated response options give a range from favorable ease ratings through a neutral response to difficulty. In general, such ratings can describe how users experience a device, but they do not by themselves establish objective treatment performance or a causal effect compared with another device or no device.

The registry record does not report the response counts, a summary score, or a rule for converting the five response categories into the stated count of participants. It therefore does not show the distribution of usability ratings or how that endpoint was analyzed. The record also does not state whether all enrolled participants were expected to complete the questionnaire or how incomplete responses were handled.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical information should be read in light of what the registry reports and what it leaves unspecified. The study is listed as a single-group investigation, with no random allocation and no masking. The primary outcome description is a subjective questionnaire, while its parameter is described as a count of participants. The registry record does not provide a statistical analysis plan that reconciles those descriptions or explains the form of the primary analysis.

What does a single-group design estimate?

In general, a single-group study can describe outcomes observed among participants who received the intervention. It does not create a concurrent comparison with a different treatment or control condition. Without such a comparison, differences that might have appeared under another condition cannot be estimated directly from the design. This is a design limitation, not a statement about whether the observed experience was favorable or unfavorable.

Non-randomized allocation also means that assignment was not determined by randomization, and the record does not describe a comparison assignment in any case. Non-blinding means participants were not masked to the intervention. For a subjective usability questionnaire, respondents know which device they are rating. In general, responses can reflect a combination of the experience of use and expectations, preferences or other circumstances. The registry does not report methods to assess or account for these influences.

How should the response scale be understood?

The listed questionnaire choices are ordered categories, from very easy to very difficult, with neither between those ends. In general, an ordered response scale conveys rank or direction, but the spacing between adjacent categories is not necessarily equal. A count for each response category would preserve the response distribution. A combined count or score could answer a different question, depending on how categories were grouped or coded. The registry record does not say whether any such grouping or scoring was used.

The endpoint description calls the outcome a count of participants, but does not identify which response or responses qualify for that count. It does not state whether the reported outcome would distinguish all response categories or collapse them into a binary outcome. No numerical primary result is reported, and the registry lists no analyses. Therefore, there is no basis in the record for reconstructing the primary analysis or assessing its uncertainty.

What do confidence intervals and significance tests communicate?

In general, a confidence interval expresses the statistical uncertainty around an estimate under the assumptions of the analysis. A narrow interval indicates greater precision than a wide interval, while neither width nor location establishes that an estimate is free from bias. A significance test assesses compatibility between data and a specified statistical model or null hypothesis. Neither a confidence interval nor a p-value can correct weaknesses in study design or unclear outcome definitions.

The registry record does not report a confidence interval, p-value, or analysis of the primary usability outcome. It also does not report whether any hypothesis test was planned. The sample-size assumptions describe planning illustrations based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms and a normal approximation. Those are characteristics of the illustrations, not evidence that this single-group trial used those assumptions in its own design or analysis. The record does not report a trial-specific sample-size justification.

Why does over-dispersion matter in general?

For count outcomes, an analysis may rely on assumptions about how much observations vary around an average. Over-dispersion occurs when the observed variability exceeds what a chosen model expects. If it is ignored, estimated uncertainty can be too small and inference can appear more precise than warranted. This is a general consideration for count analyses; the registry does not report an analysis model, dispersion estimate, or over-dispersion assessment for this outcome.

The record's planning assumptions also state that time-to-event illustrations use a total event count based on the Schoenfeld approach, while other illustrations give participants per arm. These are generic planning conventions described in the record. They do not turn this single-group device investigation into a two-arm trial, and the record does not identify a time-to-event primary endpoint.

What can be concluded from the available reporting?

Statistical interpretation requires an estimate, its denominator, and enough information about the method to understand what the estimate represents. For the primary outcome, the registry provides the questionnaire description and timeframe but no outcome values, analyzed population, missing-data account or method. As a result, precision and statistical evidence cannot be evaluated from the registry record. The listed safety counts are available, but they do not supply the missing primary outcome analysis.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists five outcome measures and identifies Toffee Mask Usability as the primary outcome. It does not report a statistical analysis plan, a hierarchy for testing outcomes, or an approach to multiplicity. Multiplicity refers in general to the increased chance of a misleading positive finding when several hypotheses or outcomes are examined. Whether that issue applied to analyses in this study cannot be determined from the registry record because no analyses are listed and no results for the outcome measures are reported.

The record does not say whether the five-point usability responses were analyzed separately, combined into a summary, or converted into a count using a prespecified rule. It also does not report whether the secondary objective performance outcome was analyzed, or how any conclusions across outcomes would be interpreted. No multiplicity adjustment or rationale for not using one is reported in the registry record.

Missing-data handling is not reported. The registry gives enrollment as 38, but does not state the number who completed the primary questionnaire, the number of incomplete questionnaires, or the reasons for any missing responses. It does not describe imputation, exclusion, or sensitivity analyses. Without that information, it is not possible to assess whether the available responses represent the enrolled group or whether missingness could affect interpretation.

The record does not define an analysis population, such as all enrolled participants or those who completed the in-home use period and questionnaire. It does not report whether participants were included according to an intention-to-treat principle, a per-protocol rule, or another approach. The relevant population and denominator for the primary endpoint therefore remain unclear.

11. 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 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry lists the Toffee Nasal Pillows Mask group with 38 participants at risk for serious adverse events and zero affected by serious adverse events. It also lists 38 at risk for other adverse events and eight affected. The death count is zero. These are registry counts, not a comparison with a control group.

Serious adverse events and other adverse events are separate categories in the record. A count of zero in the serious category means that no serious events are recorded for this group in the posted data; it does not mean that no adverse events of any kind occurred. The eight participants listed with other adverse events should not be relabeled as eight events: the record characterizes the figures as counts of participants with an event.

The registry does not report the terms for the other adverse events, their timing, duration, severity, or whether investigators considered them related to the mask. It also does not provide an account of adverse-event collection or follow-up completeness. Consequently, the figures describe the recorded counts but do not provide a detailed characterization of safety or a comparative estimate.

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

This record documents a completed prospective investigation of a nasal pillows mask among participants with obstructive sleep apnea. It specifies a single-group, non-randomized, non-blinded design, an in-home use period of 14 ± 5 days, and a primary questionnaire about overall usability. Those features clarify what the study set out to assess and the context in which participants were asked to rate the mask.

The record does not report the primary outcome results. It gives no category counts, summary of the usability ratings, estimate, confidence interval or p-value. It also does not report how the five response categories were translated into the endpoint described as a participant count. Therefore, the posted information does not establish what participants' ratings were or how precisely any usability estimate was measured.

The design also sets limits on what an observed result could mean. A single-group study has no concurrent comparator listed, and it is not randomized or blinded. In general, such a study can describe participant experience during the observed period, but it cannot by itself establish how that experience would compare with another mask or a different condition. The registry record does not report adjustment methods that would address such a comparison.

The safety figures list zero serious adverse events, eight participants with other adverse events, and zero deaths. They are descriptive counts without event terms or a control-group comparison. They do not resolve the absence of primary usability results or show whether other adverse events were related to the mask.

Thus, the registry establishes the study's stated purpose, endpoint description, exposure timeframe, design and recorded adverse-event counts. It does not provide enough outcome or analysis information to judge the primary usability findings, their precision, or comparative effects.

13. Why This Trial Matters Statistically

This trial is statistically notable because the endpoint is a subjective ordered response, while the registry describes its parameter as a count of participants. That combination makes clear outcome coding important. Counts for each response category, a prespecified grouping, or a summary score would represent different estimands. The registry record does not explain which representation was used, so a reader cannot connect the endpoint label to a specific analysis.

The study also illustrates how design determines the scope of inference. The single-group, non-randomized and non-blinded structure can provide a record of participants' reported experience, but it does not offer a randomized control comparison. Statistical methods cannot supply the missing counterfactual comparison: more elaborate modeling would not, by itself, remove the design limitation or establish what would have happened under another condition.

Transparent reporting of denominators and missing responses is particularly important for a questionnaire endpoint. Enrollment is listed, but the record does not state how many participants posted a usable response. Without that information, even category counts would be difficult to interpret in relation to the study group. Similarly, a stated endpoint needs an analysis description and uncertainty estimate to show how much the data support an interpretation.

The record's planning illustrations also underscore the distinction between a study's design and generic sample-size calculations. The stated assumptions include equal allocation between two arms, whereas the trial is recorded as a single-group study with no like-for-like control comparison. Planning illustrations can show how design assumptions affect required sample sizes in general, but they are not observed trial results or evidence of observed power.

These features make the record useful for understanding the difference between a prespecified question, a study design, and analyzable outcome evidence. Here, the first two are described, while the primary outcome analysis is not reported.

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.

Several trial-specific limitations arise from the registry information. The primary outcome description identifies a five-point usability scale, but the endpoint is also described as a count of participants. The record does not explain how responses were coded, whether categories were combined, or what the count represents. This leaves the intended estimand and interpretation uncertain.

The registry lists enrollment as 38 and says up to 45 participants were to be recruited. It does not explain the relation between those figures or report the number who completed the in-home period or questionnaire. The primary outcome result, its denominator, and missing-response handling are all absent. The number of analyses is listed as zero, and no primary analysis is described.

The absence of a comparator is a further limitation for comparative claims. The study is single-group, non-randomized and non-blinded. Participant-reported usability can be described, but the design does not provide a randomized estimate of how ratings differ under another condition. The registry does not report methods for accounting for expectations or other influences on subjective ratings.

Although objective treatment performance is named as a secondary outcome, no result is reported. The registry also does not give details about participant characteristics, adherence to the intended mask-use period, outcome collection procedures, or protocol deviations. These omissions limit evaluation of the context and completeness of the observations.

For safety, the record gives counts but no event terms, timing, severity details for other adverse events, attribution, or comparator data. The figures therefore cannot provide a detailed account of the types or context of reported events.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: The Evaluation of the Toffee Nasal Pillows Mask for the Treatment of Obstructive Sleep Apnea — NCT03242148.
• Registry record last updated 2021-09-22; status last verified 2021-08.
• Results first posted 2021-06-01.
• 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.

18. Record Summary

The registry record describes a completed, prospective, non-randomized, non-blinded, single-group evaluation of the Toffee Nasal Pillows Mask for obstructive sleep apnea. Enrollment is listed as 38, and the brief summary states that up to 45 participants were to be recruited. The recorded dates are 2017-09-16 for study start and 2017-11-03 for primary completion.

The primary outcome is a usability questionnaire administered during the second visit after an in-home period of 14 ± 5 days. Participants rated overall usability on a five-point scale. The registry lists five outcome measures and no analyses, and does not report primary outcome results, the analyzed population, or missing-data handling. Objective treatment performance is listed as a secondary outcome, without a reported result.

The safety record lists zero serious adverse events, eight participants with other adverse events, and zero deaths. It does not identify the other event terms. No separate comparator group is listed.