The Use of High Frequency Oscillations With NIV in Hypercapnic COPD Participants: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Use of High Frequency Oscillations With NIV in Hypercapnic COPD Participants (NCT02680639), a randomized na study. The primary outcome is Carbon Dioxide Levels.

2. Clinical Question

The clinical motivation in the registry concerns expiratory flow limitation (EFL), which it describes as airway compression when pressure outside the airway exceeds pressure inside it. The study was designed to detect participants with EFL and to examine ventilation settings in participants with hypercapnic COPD. Its primary endpoint is carbon dioxide level, not an outcome such as survival, hospitalization, or long-term symptoms.

The study is described as randomized and factorial, but the record does not provide enough detail to reconstruct a complete factorial structure or identify a prespecified set of pairwise contrasts. It describes randomizing treatment order and allowing carbon dioxide to stabilize between conditions. These details suggest that measurements were intended to be made repeatedly under different settings, but the statistical analysis for those repeated measurements is not reported in the registry record.

3. Trial Design

The registry describes a randomized, unmasked study with a factorial model and an enrollment of 10. The study status is terminated. The record lists chronic obstructive pulmonary disease as the condition and identifies Philips Respironics as the sponsor. The recorded study start is 2015-11-13, and the primary completion date is 2016-08-24. The reason for termination is not reported in the registry record.

Participants were to undergo eligibility procedures at the first visit. At the second visit, the protocol description says a baseline auto-EPAP measurement would be obtained. Treatment order would then be randomized, with at least a 10 minute washout between treatments to allow carbon dioxide to stabilize and return to baseline. The description names automatic determination of optimal EPAP, optimized EPAP with no peak-to-peak oscillations, optimized EPAP with maximum peak-to-peak oscillations, and non-optimized EPAP with no peak-to-peak oscillations. It does not explain how the factorial label maps onto these conditions or give a full allocation scheme.

The listed device is the BiPAP Synchrony ventilator. In the automatic condition, EPAP and IPAP are adjusted to find an EPAP that abolishes EFL, with peak-to-peak oscillations set at 2.5 cmH2O. In the optimized conditions, EPAP and IPAP are set to the pressure determined by auto-EPAP, with peak-to-peak oscillations set at 0 cmH2O or 5 cmH2O. In the non-optimized condition, EPAP is set at 4 cmH2O and IPAP at 10 cmH2O, with oscillations set at 0 cmH2O.

The registry record does not report details such as the sequence-generation method, allocation concealment, or analysis population. No masking was used. The small enrollment and terminated status frame how much can be concluded from the record, particularly because no numerical outcome results are reported.

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.

Screening Period

Reasons for not completing this period:

High Frequency Oscillation (HFO) Therapy

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 Carbon Dioxide Levels, measured over Baseline and 20 minutes.

The primary endpoint is carbon dioxide level, reported as a mean with standard deviation and measured in mmHg. The registry specifies transcutaneous carbon dioxide (TcCO2) monitoring at baseline and 20 minutes. It describes TcCO2 as a non-invasive alternative to arterial blood sampling and says carbon dioxide levels would be measured for each participant as they completed each arm.

Carbon dioxide is relevant to this study because participants have hypercapnic COPD and the study examines ventilation settings. A measurement made during each condition can describe the level observed under that setting. Comparing measurements across conditions could address whether the settings are associated with different short-term carbon dioxide readings, provided the analysis accounts for the study's repeated measurements and timing. The registry record does not report the actual values, estimates of differences, confidence intervals, or statistical tests.

Mean and standard deviation convey different features of a continuous measurement. The mean describes its central level, while the standard deviation describes variability among measurements or participants, depending on how the data are summarized. Neither statistic alone establishes how precisely a treatment-condition contrast has been estimated. No definition of a primary contrast, method for handling the two time points, or clinically important difference is reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a continuous primary outcome summarized by a mean and standard deviation. It does not report a fitted statistical model, a treatment contrast, or an inferential analysis. The concepts below explain how these methods are generally interpreted; they should not be taken as a description of an analysis actually performed for this study.

What does a mean describe?

In general, a mean gives the arithmetic center of a set of quantitative observations. For carbon dioxide measurements, it would describe the average recorded level for a defined condition, time point, and analysis population. The interpretation depends on what observations are included. A mean at baseline is not interchangeable with a mean at 20 minutes, and a mean pooled across conditions does not isolate the effect associated with any one setting. The registry names the outcome and time points but does not provide these summaries.

What does the standard deviation add?

In general, the standard deviation describes how dispersed individual observations are around their mean. It is not the same as the standard error, which describes uncertainty in an estimated quantity such as a mean. A larger standard deviation indicates greater spread in the observed measurements, but does not by itself say whether the estimated treatment contrast is precise. Precision also depends on the amount of usable information and the analysis structure. The registry specifies standard deviation as the dispersion measure but reports no numerical standard deviations.

How should a treatment contrast be interpreted?

In general, a contrast is a defined comparison between conditions, such as the mean outcome under one setting compared with the mean under another. Its interpretation requires clarity about the conditions, time point, and participant population being compared. A contrast estimate describes the magnitude and direction of the difference as calculated; an uncertainty interval, such as a confidence interval, describes a range of values compatible with the data and the analysis assumptions. A confidence interval is not a guarantee that the true value lies within its bounds. No contrast estimate or confidence interval is reported in the registry record.

Because the study description calls for measurements on participants across treatment conditions, a general analysis would need to account for the dependence among measurements from the same participant. Treating all measurements as if they came from unrelated participants can give an inappropriate estimate of uncertainty. Depending on the design and question, analysts may use paired comparisons or a repeated-measures model. This is general methodological context only: the registry record does not report which approach, if any, was used, nor does it describe adjustment for treatment order or period.

Why can over-dispersion matter?

In general, over-dispersion means that the variability in data is greater than a model assumes. It is especially discussed for count outcomes, where an overly restrictive variance assumption can make standard errors too small and uncertainty appear narrower than it is. The primary endpoint in this study is continuous, so the registry's stated summary is a mean and standard deviation. It does not report a model or a variance assumption, and there is no basis in the registry record to claim that over-dispersion was assessed or present. The general lesson is that uncertainty calculations should match the observed variability and data structure.

What do significance tests and planning calculations mean?

In general, a p-value evaluates how compatible the observed data are with a specified null hypothesis under the assumptions of a statistical test. It does not measure the probability that the null hypothesis is true, the size of an effect, or its clinical importance. A p-value without an effect estimate and uncertainty interval gives an incomplete account of results. No p-value or test result is reported for this study.

The record describes planning assumptions for sample-size illustrations: a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and target effects defined as fractions of an observed effect. Such calculations depend on the assumed effect and variability and answer a planning question under those assumptions. They are not observed power and do not show that a completed study had a specified chance of finding an effect. Because the registry does not report a numerical primary-outcome effect or a single primary comparison, the assumptions cannot be used to infer the study's achieved precision or power.

The registry record lists four outcome measures and zero analyses. It does not say which statistical tests were planned, whether any interim analyses occurred, how missing observations were handled, or whether a statistical analysis plan was finalized. Those omissions prevent a reader from judging model fit, multiplicity control, or the sensitivity of the reported conclusions to analytical choices. No inferential result should be inferred from the design label or endpoint definition alone.

10. Multiplicity, Missing Data and Analysis Populations

The registry identifies a primary endpoint and three secondary outcomes: respiratory rate, tidal volume, and minute ventilation for carbon dioxide. It also describes multiple ventilation conditions and measurement at baseline and 20 minutes. Considering more than one outcome, time point, or treatment contrast can create multiplicity: the more hypotheses examined, the greater the chance that at least one apparently notable result arises by chance if no adjustment or prespecification is used. This is a general statistical concern, not a claim that a particular error occurred in this study.

The registry record does not report a multiplicity strategy, a hierarchy of secondary outcomes, or which comparisons were considered confirmatory. It also does not report whether analyses were prespecified or whether any interim analyses were conducted. The record lists zero analyses, so there are no reported test results from which to determine how multiplicity was handled. A reader should not assume either that results were adjusted or that they were not.

Missing-data handling is also not reported in the registry record. The enrollment is 10, but the record does not give the number with measurements at each time point or in each condition, reasons for missing measurements, or rules for including participants in an analysis. The analysis population—such as all enrolled participants, all participants who completed the procedures, or a subset with usable measurements—is not specified. For repeated measurements, missingness at one condition or time can affect which comparisons are available and how representative the analyzed observations are.

The registry does not report a statistical analysis plan, sensitivity analyses, or an approach to deviations from the intended sequence. These details matter for understanding whether the analysis addresses the randomized order and within-participant design as intended. Without them and without posted outcome estimates, the scope and robustness of any statistical inference cannot be assessed.

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 5% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: atrial fibrillation (10.0%).

The registry lists one serious adverse event among 10 participants at risk during the screening period. The serious-event term identified is atrial fibrillation, reported at 10.0%. The screening entry records zero other adverse events and zero deaths. These figures describe the recorded screening-period counts and do not, by themselves, indicate that the event was caused by study procedures or a ventilation setting.

The listed treatment-sequence groups have smaller denominators: 2 participants at risk in one group and 1 in another. Each of those entries records zero serious adverse events, zero other adverse events, and zero deaths. These groups are identified by the order of ventilation conditions. The small numbers at risk mean that zero recorded events in those groups should not be read as evidence that an event could not occur or that the conditions have equivalent safety.

The registry record does not report a causal assessment, event timing in relation to individual treatment conditions, or further details about the atrial fibrillation event. The screening count and the treatment-sequence counts also refer to different periods and denominators, so they should not be combined into a single estimate. They are descriptive registry entries, not a comparative safety analysis.

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

The record establishes that investigators planned a randomized study of ventilation conditions in participants with hypercapnic COPD, with carbon dioxide measured non-invasively at baseline and 20 minutes. It identifies a primary outcome, three secondary outcomes, the device, and several pressure and oscillation settings. It also describes randomizing treatment order and using washout periods to allow carbon dioxide to stabilize before the next condition.

It does not establish the size or direction of an outcome difference. Numerical primary and secondary outcome results are not reported in the registry record, and no analyses are listed. There are no treatment-effect estimates, uncertainty intervals, or p-values from which to judge either the magnitude of a contrast or its precision. The endpoint definition is not a finding.

The study's enrollment of 10 and terminated status are important context, but they do not explain why the study ended or tell how many participants contributed usable measurements. The reason for termination, analysis populations, and missing-data approach are not reported in the registry record. The absence of those details limits interpretation of both the study's conduct and any potential comparison.

The recorded adverse-event figures are descriptive and do not attribute events to a treatment condition. The single serious event recorded during screening and the zero events recorded in the listed treatment groups cannot support a reliable comparative safety conclusion. Likewise, planning assumptions for sample-size illustrations are not evidence of observed power or of an actual treatment effect.

Accordingly, the posted information supports a description of the research question and protocol features, but not a conclusion that one ventilation setting produced a particular carbon dioxide response, was more precise, or was safer than another. Those judgments would require outcome estimates, methods, and participant-level analysis details that are not reported in the registry record.

13. Why This Trial Matters Statistically

This study is statistically interesting because it combines multiple ventilation conditions with repeated carbon dioxide measurements in the same participants. Repeated observations can make within-participant comparisons informative, but the measurements are generally correlated rather than independent. How that dependence is represented affects standard errors, confidence intervals, and tests. The registry gives the broad design description but does not report the analytical approach used to address this structure.

The endpoint is continuous and is specified as a mean with standard deviation, while the procedure includes both baseline and 20-minute measurement. This makes the definition of the treatment contrast important: a comparison of values at a single time point is not the same as a comparison of changes from baseline. The registry record does not state which contrast was primary or how the time points were incorporated into analysis. That distinction is essential to interpretation, not a technical detail that can be inferred from the endpoint title.

The factorial designation also makes clear reporting important. A factorial model can represent effects of more than one intervention component, while the described conditions vary pressure optimization and peak-to-peak oscillations. But the record does not specify the intended factors, interaction terms, or contrast set. Without that information, the label alone cannot determine which effects were meant to be estimated.

Finally, the terminated status, enrollment of 10, and absence of listed analyses make the record useful as an example of the difference between a stated research question and reported evidence. An endpoint and a protocol do not yield an effect estimate on their own. The statistical meaning of the trial depends on actual outcome data, a clear analysis population, and methods that match its repeated-measures structure.

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 trial-specific limitation is the lack of reported primary-outcome values. The registry describes carbon dioxide measurements at baseline and 20 minutes but gives no means, standard deviations, contrasts, confidence intervals, or test results. As a result, neither the size nor the precision of any observed response can be assessed. The record also lists zero analyses, so no statistical findings for the primary or secondary outcomes are available.

The recorded enrollment is 10, but the number completing each condition or contributing data at each time point is not reported. The registry does not describe missingness, protocol deviations, or the analysis population. It is therefore not possible to know whether all enrolled participants contributed to every comparison or how incomplete observations were handled.

The study status is terminated, but the reason is not reported in the registry record. The record also does not provide the statistical analysis plan, explain the factorial structure in detail, name a primary contrast, or describe adjustment for treatment order and repeated measurements. These gaps make it difficult to assess whether any analysis would have matched the intended design. The registry describes a washout between conditions, but does not report data showing whether carbon dioxide returned to baseline or how that criterion was evaluated.

For safety, the listed treatment-sequence groups contain very few participants at risk, and the entries do not provide a causal assessment or detailed event timing. Their counts cannot establish comparative safety. Overall, the record supports a description of the planned comparison and recorded adverse events, but not a complete evaluation of efficacy, precision, or safety.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: The Use of High Frequency Oscillations With NIV in Hypercapnic COPD Participants — NCT02680639.
• Registry record last updated 2019-05-06; status last verified 2019-04.
• Results first posted 2019-04-22.
• 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

NCT02680639 is a randomized, unmasked study of ventilation settings in participants with hypercapnic COPD. It lists Philips Respironics as the sponsor, an enrollment of 10, and a terminated status. The protocol description includes automatic determination of optimal EPAP and other conditions varying EPAP, IPAP, and peak-to-peak oscillations on a BiPAP Synchrony ventilator.

The primary outcome is mean carbon dioxide level, with standard deviation, measured in mmHg by transcutaneous monitoring at baseline and 20 minutes. Respiratory rate, tidal volume, and minute ventilation for carbon dioxide are listed as secondary outcomes. The registry record does not report numerical outcome results or analyses. It records one serious adverse event during screening among 10 participants at risk, with atrial fibrillation listed at 10.0%, and no serious events in the listed treatment-sequence groups.