Continuous Positive Airway Pressure (PAP) Titration and Treatment Versus Auto-adjusting PAP Treatment for Sleep Apnea: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Continuous Positive Airway Pressure (PAP) Titration and Treatment Versus Auto-adjusting PAP Treatment for Sleep Apnea (NCT00988351), a randomized na study. The primary outcome is Positive Airway Pressure Adherence (Nightly Use of Treatment).

2. Clinical Question

The trial asks whether initiating PAP through an auto-adjusting pathway produces a different pattern of treatment adherence from a pathway using attended titration followed by CPAP. It also evaluates subjective sleepiness, reaction time, quality of life, satisfaction and other treatment measures in its stated aims. The posted results include the primary adherence outcome and several secondary measures; reaction-time and satisfaction results are not reported in the registry record.

The comparison is about two ways to start PAP treatment, not about PAP versus no treatment. As a result, the reported difference estimates how the two initiation strategies compare on the measured outcomes. The registry labels its primary analysis as non-inferiority or equivalence, but a margin defining what difference would count as acceptable is not reported in the registry record. Without that margin, the result can be described as an estimated difference with uncertainty, but not as a demonstrated equivalence or non-inferiority conclusion.

3. Trial Design

This was a randomized, prospective, parallel-group study of patients diagnosed with obstructive sleep apnea by portable monitoring. The registry records an enrollment of 156 and identifies the study as completed. The recorded start was 2010-02 and primary completion was 2012-12. The phase is listed as not applicable.

One group followed a pathway of attended CPAP titration using polysomnography and then CPAP treatment. The other received auto-adjusting PAP treatment without a titration. The two groups therefore differed in the way PAP was initiated, while both pathways involved PAP treatment. The registry identifies the allocation as randomized and the model as parallel.

Masking is recorded as none. Participants and clinical staff therefore were not masked according to the registry record. The record does not describe the randomization sequence, allocation concealment, or operational procedures used to assign participants. Those details are not reported in the registry record.

The primary endpoint was average nightly hours of PAP use at 6 weeks after starting treatment. Its description says that nights without use were counted as zero, and that the population was patients using CPAP or APAP at the clinic visit. Arm-level results are reported for 59 participants in the CPAP-titration pathway and 66 in the auto-adjusting pathway. The registry enrollment and the endpoint analysis counts are different figures; the reasons for that difference are not reported in the registry record.

The registered secondary outcomes include changes in Epworth Sleepiness Scale and Functional Outcomes of Sleep Questionnaire scores, residual apnea-hypopnea index, and treatment pressure. The study aims also mention reaction time and PAP satisfaction. Results for all the aims are not posted in the registry record.

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:

Analysis population for the primary outcome, as defined in the registry: patients using cpap or apap at clinic visit

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: Apnea+hypopnea index \>= 10/hour

6. Primary Endpoint

The registry lists the primary outcome as Positive Airway Pressure Adherence (Nightly Use of Treatment), measured over 6 weeks after starting treatment.

The primary endpoint is mean nightly use of positive airway pressure, measured in hours, at 6 weeks after treatment started. The registry describes the measure as average nightly hours of use and explicitly includes zero for nights when treatment was not used. Counting those nights gives a measure of use across nights, rather than describing use only on nights when a device was worn. This is relevant because adherence includes both how much treatment is used and whether treatment is used at all on a given night.

The registry reports the endpoint as a mean with a standard deviation. The CPAP-titration group had a mean of 3.99 hours and a standard deviation of 2.3; the auto-adjusting group had a mean of 4.45 hours and a standard deviation of 2.3. The reported contrast is the auto-adjusting group minus the CPAP-titration group, a mean difference of 0.46 hours.

Average use is useful for comparing typical group-level treatment exposure, but it does not show the full distribution of individual use. A mean can be influenced by participants with higher or lower use, and it does not reveal how many participants did not use treatment on any night. The registry’s definition does incorporate zero-use nights into the average, but separate counts or distributions of use are not reported in the registry record. Nor does the endpoint, by itself, measure symptom improvement or quality of life.

7. Results Posted in the Registry

• Auto-Adjusting Positive Airway Pressure: mean 4.45 versus 3.99 on control; difference 0.46 (95% CI -0.35 to 1.27), standardized effect d = 0.20; z-test p = 0.264.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change in Epworth Sleepiness Scale

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Auto-Adjusting Positive Airway Pressure: mean -4.10 versus -3.70 on control; difference -0.40 (95% CI -2.07 to 1.27), standardized effect d = -0.08; z-test p = 0.639.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Auto-Adjusting Positive Airway Pressure and the control. The interval gives the range of effects that remain compatible with the data.

Change in Functional Outcomes of Sleep Questionnaire

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Auto-Adjusting Positive Airway Pressure: mean 2.60 versus 2.20 on control; difference 0.40 (95% CI -0.87 to 1.67), standardized effect d = 0.11; z-test p = 0.536.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Auto-Adjusting Positive Airway Pressure and the control. The interval gives the range of effects that remain compatible with the data.

Residual Apnea-hypopnea Index

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Auto-Adjusting Positive Airway Pressure: mean 5.50 versus 4.90 on control; difference 0.60 (95% CI -1.09 to 2.29), standardized effect d = 0.13; z-test p = 0.486.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Auto-Adjusting Positive Airway Pressure and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary results cover changes in sleepiness and quality of life, along with residual apnea-hypopnea index. The registry also lists treatment pressure as a secondary outcome, but its results are not reported in the registry record. The study aims mention reaction time and PAP satisfaction; results for those outcomes are likewise not reported in the registry record.

Across the secondary results that are posted, the estimated contrasts are modest relative to their uncertainty. The group-level patterns do not point uniformly in one direction across all measures: changes in sleepiness and quality of life have different measurement scales and interpretation, while the residual index is a distinct physiological measure. The reported confidence intervals for these posted comparisons include zero. This leaves uncertainty about the size and direction of the between-group differences, rather than establishing that the strategies are identical on these outcomes.

The overall pattern is broadly consistent with the primary result in one important sense: the posted comparisons do not provide a precise, conclusive separation between the treatment pathways. They do not, however, establish equivalence across outcomes. Each endpoint measures a different aspect of treatment, and similar-looking estimates across endpoints do not replace endpoint-specific uncertainty assessment.

Secondary outcomes usually carry less confirmatory weight than a primary endpoint, particularly when several outcomes are tested. Testing many outcomes increases the chance that at least one result appears notable by chance. The registry record does not report a multiplicity adjustment or a prespecified hierarchy for interpreting secondary findings. Their p-values and confidence intervals should therefore be read as exploratory or supportive evidence, not as independent confirmation of a treatment difference. The unreported outcomes also mean the full set of aims cannot be assessed from the posted results.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Change in Epworth Sleepiness Scale
• Change in Functional Outcomes of Sleep Questionnaire
• Residual Apnea-hypopnea Index
• Treatment Pressure (Level of CPAP or 90th Percentile APAP)

9. Statistical Interpretation

10. Statistical Methods Explained

What does a mean difference represent?

The endpoint is continuous: nightly use is measured in hours, and the registry reports a mean for each treatment group. The mean difference expresses the contrast between group averages. In the reported direction, the auto-adjusting group’s mean is compared with the CPAP-titration group’s mean. A positive estimate indicates a higher observed mean in the auto-adjusting group; it does not by itself establish that the difference is reliable or important.

The standard deviation describes variation among individual observations around the group mean. Both groups have a reported standard deviation of 2.3 hours. This variation is distinct from uncertainty in the estimated mean difference. Uncertainty in the contrast depends on the variability of the outcome and the number of observations contributing to each group estimate.

How should the confidence interval be read?

A confidence interval gives a range of values compatible with the data under the statistical model and its assumptions. The recalculated 95% interval for the primary difference runs from -0.348 to 1.268 hours. Its inclusion of zero means that a zero difference is among the values compatible with the data; the interval also includes positive and negative differences. The interval should not be read as a range containing a fixed proportion of individual participants’ use values, nor as proof that every value in it is equally plausible.

Confidence intervals reflect sampling uncertainty, not every possible source of bias. They do not account automatically for missing-data problems, deviations from the planned analysis, measurement limitations, or choices made after seeing the data. How those issues were handled is not reported in the registry record.

What does the p-value add?

The registry reports a two-sided t-test for the primary mean difference, with p = 0.27 in the sponsor analysis. The unadjusted recalculation gives p = 0.264. A two-sided test assesses evidence against a null difference in either direction. The p-value is calculated under the null hypothesis and the assumptions of the test; it is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of the difference.

The p-value and confidence interval address related questions. A p-value does not show which differences are plausible in the original measurement scale as directly as the confidence interval does. In this study, both summaries indicate uncertainty around the observed difference rather than a sharply determined contrast.

Why can variability matter?

When outcome values vary substantially among participants, group means are estimated less precisely unless there is enough information to offset that variability. The standard deviations describe the spread of nightly-use measurements. More generally, statistical models also need to account for features such as skewness, repeated measurements, or over-dispersion when those features are present. Over-dispersion means that observed variability exceeds what a model assumes; if ignored, uncertainty can be understated. The registry record does not report a model assessment for these distributional features. The posted primary comparison is a t-test, so the interpretation rests on the appropriateness of that comparison for the recorded data.

What is the status of the recalculation?

The recalculated comparison is an unadjusted calculation from the posted group means, standard deviations and analysis counts. It is not a replacement for the sponsor’s analysis. For the primary endpoint, the recalculated mean difference and interval closely correspond to the sponsor-reported values, with small rounding differences in the displayed results. The discrepancy flag is false. The sponsor’s own analysis therefore remains the principal reported analysis, and the recalculation serves as a transparent cross-check using the posted summary statistics.

How does the analysis label affect interpretation?

The registry identifies the primary analysis as non-inferiority or equivalence, while also reporting a two-sided t-test and a confidence interval for the mean difference. For a non-inferiority interpretation, the key question is whether the confidence interval excludes differences beyond a prespecified unacceptable margin in the relevant direction. For equivalence, the interval must lie within prespecified bounds on both sides. Those bounds are not reported in the registry record. A conventional test of whether the difference is zero cannot substitute for a margin-based assessment. The available values describe the estimated difference and its uncertainty, but do not establish a non-inferiority or equivalence conclusion.

The registry does not report a detailed statistical analysis plan, handling of missing observations, or whether any covariate adjustment was used. The sponsor’s listed method for this endpoint is a two-sided t-test on final values. It is not possible to infer from the registry record whether other analyses were prespecified or performed.

11. Confidence Intervals and Precision

The primary estimated mean difference is 0.46 hours, with a recalculated 95% confidence interval from -0.348 to 1.268 hours. The interval crosses zero, so the data remain compatible with no average difference as well as with differences in either direction. It is also wide enough to include values on both sides of the estimate, which signals that the precise size of the contrast is uncertain.

The interval is an estimate of uncertainty in the group contrast, not a prediction interval for an individual participant’s nightly use. Individual observations vary: the reported standard deviation is 2.3 hours in each group. The interval around the mean difference and the spread among individual measurements answer different questions. A narrow interval would indicate a more precise estimate of the average contrast; it would not imply that individual use values are tightly clustered.

The registry’s sponsor-reported 95% confidence interval is -.35 to 1.26, which is close to the unadjusted recalculation. The minor displayed differences are consistent with rounding and do not alter the broad reading: the estimate is uncertain and the interval spans zero. The registry record does not report a non-inferiority margin, so the interval cannot be judged against a criterion for an acceptable loss of adherence.

Precision also depends on the analysis population and data completeness. The registry does not report missing-data handling or the reasons endpoint counts differ from the recorded enrollment. Without those details, the confidence interval captures model-based sampling uncertainty but cannot resolve uncertainty about how the analyzed observations relate to all enrolled participants.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm for a continuous-outcome comparison. They are not total event counts. The illustration assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It uses the observed effect as a reference and considers planning for 100% of the observed effect, 75% of the observed effect, or 50% of the observed effect.

The 80% power column gives the planned participants per arm for a design with an 80% chance of detecting the selected effect under the stated assumptions, if that effect is the true effect and the model assumptions hold. The 90% power column expresses the corresponding planning target for a 90% chance. These are design probabilities under assumptions, not guarantees about a future study’s result.

For the primary outcome, planning for 100% of the observed effect gives 393 participants per arm for 80% power and 526 participants per arm for 90% power. Planning for 75% of the observed effect gives 698 participants per arm for 80% power and 934 participants per arm for 90% power. Planning for 50% of the observed effect gives 1,570 participants per arm for 80% power and 2,102 participants per arm for 90% power. The pattern illustrates why smaller effects require more participants: they are harder to distinguish from ordinary sampling variation.

These values are planning illustrations based on the observed estimate, not the original trial’s design assumptions. The original sample-size justification is not reported in the registry record. This is not observed power: it does not describe the probability that the completed trial detected an effect after the results were known. The planning values should be read as conditional illustrations, not as evidence that a particular design would necessarily produce a particular result.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 0.460; pooled standard deviation σ = 2.300.
2. n = 2 × (1.96 + 0.842)² × 2.300² / 0.460² = 392.4, rounded up to 393 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- 0.46                      # observed difference in means
sigma <- 2.3                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(988351)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 393 (80% power), 526 (90% power)
 75% of observed effect: n per arm = 698 (80% power), 934 (90% power)
 50% of observed effect: n per arm = 1,570 (80% power), 2,102 (90% power)
power.t.test: 393.4 
Simulated power at n = 393 per arm: 0.799 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The sponsor-reported primary analysis is a two-sided t-test of the mean difference in final values. It reports a difference of 0.46 hours, a 95% confidence interval from -.35 to 1.26, and p = 0.27. The unadjusted recalculation from the posted summary statistics gives a difference of 0.46 hours, a 95% confidence interval from -0.348 to 1.268, and p = 0.264.

The estimates agree closely in direction and magnitude. The small differences in the interval endpoints and p-value are consistent with the distinct rounding shown in the registry record. The discrepancy flag is false, so there is no indicated conflict between the recalculation and the posted analysis. The sponsor’s analysis remains the main reported result; the recalculation is a check based on summary values, not a substitute for the analysis of the underlying participant-level data.

The registry’s non-inferiority or equivalence label still requires caution. The sponsor-reported analysis does not include a margin in the information posted, and the registry record does not report the decision criterion. The numerical agreement between the sponsor result and the recalculation does not answer whether a non-inferiority or equivalence threshold was met. That conclusion cannot be evaluated from the posted information alone.

15. Multiplicity, Missing Data and Analysis Populations

Several outcomes are named in the study aims and registry record. When many hypotheses are examined, the chance of at least one apparently unusual result can increase, even when there are no true differences. Statistical approaches can control this issue by adjusting significance thresholds, specifying a testing sequence, or identifying a limited set of confirmatory comparisons in advance. The registry record does not report whether any such multiplicity strategy was used. A prespecified hierarchy or a clear distinction between confirmatory and exploratory analyses is also not reported in the registry record.

This uncertainty matters most when interpreting secondary endpoints. Their individual results may be useful for describing patterns, but they should not be treated as a collection of independent confirmations without knowing how the analyses were planned and whether multiplicity was addressed. The registry lists more secondary outcomes than are represented among the posted secondary results, and the unposted results cannot be assessed.

Missing-data handling is not reported in the registry record. In particular, the record does not state how participants without a recorded 6-week outcome were handled, whether any imputation was performed, or whether sensitivity analyses assessed alternative assumptions. Missing outcomes can affect both the estimated group means and their uncertainty, especially if the reasons for missingness differ between groups or relate to treatment use.

The analysis population is also not reported in the registry record. The primary outcome description refers to patients using CPAP or APAP at the clinic visit, while the endpoint results report analysis counts by arm. The registry does not clarify whether the analysis followed an intention-to-treat population, a per-protocol population, or another rule for inclusion. Those approaches answer somewhat different questions, particularly in a non-inferiority or equivalence context.

The registry identifies randomization and a two-sided t-test, but it does not provide a complete statistical analysis plan. Details such as baseline adjustment, handling of protocol deviations, and any planned sensitivity analyses are not reported in the registry record. These gaps do not erase the posted estimates, but they limit assessment of how robustly the analysis addresses missingness, multiplicity, and population definition.

16. 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 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse events for 78 participants at risk in each group and records 0 participants affected in each group. No serious adverse event terms are listed. The serious-event figures therefore describe no affected participants in the posted registry entries; they do not establish that the risk of serious events is zero.

For other adverse events, the record reports 19 participants affected among 78 at risk in the CPAP-titration group and 12 affected among 78 at risk in the auto-adjusting group. These are listed separately from serious adverse events. The registry does not provide event terms or additional detail in the posted information, so the nature, severity and timing of those other events cannot be assessed from these figures.

Deaths are not reported in the registry record. Adverse-event counts should be interpreted in light of the study’s size and follow-up, and should not be taken as a complete characterization of safety. These descriptive counts do not establish whether one pathway has a different safety profile from the other.

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

The primary data show that average nightly PAP use at 6 weeks was 3.99 hours in the CPAP-titration group and 4.45 hours in the auto-adjusting group. The estimated mean difference was 0.46 hours in favor of higher observed average use in the auto-adjusting group. The confidence interval includes zero and allows differences in either direction, so the estimate does not determine a clear between-group difference.

That result is not the same as proving that the approaches are equivalent. Equivalence requires prespecified bounds defining differences small enough to count as practically equivalent, and the confidence interval must be assessed against those bounds. Non-inferiority similarly requires a prespecified margin and an appropriate decision rule. Neither margin is reported in the registry record. The registry’s analysis label cannot replace those missing criteria.

The result also does not establish that either treatment strategy works or does not work. The comparison was between two PAP initiation pathways, not between PAP and no treatment. A difference in average use is only one outcome and does not, by itself, establish a difference in symptoms, quality of life, reaction time, satisfaction, or longer-term outcomes. Some secondary results are posted, but other outcomes named in the study aims are not reported in the registry record.

The follow-up for the primary endpoint is 6 weeks after treatment started. The finding therefore concerns a short-term measure and should not be extended to longer-term adherence. The mean summarizes group behavior and does not describe every participant’s experience or the distribution of individual use. The registry record also does not report missing-data procedures or the analysis population, which limits assessment of how complete and representative the endpoint results are.

18. Why This Trial Matters Statistically

This trial illustrates why an estimate, its uncertainty and the study’s decision framework need to be read together. The observed primary difference is positive, but the confidence interval extends across zero and includes differences in both directions. Reporting only the point estimate could make the direction seem more certain than the data warrant; reporting only a p-value would conceal the range of values compatible with the results.

It also highlights a common issue in non-inferiority and equivalence trials. Those questions are not answered simply by failing to find a statistically clear difference. They require margins chosen in advance, together with an analysis that evaluates the interval against those margins. Although the registry labels the primary result as non-inferiority or equivalence, the relevant margin is not reported. This prevents a complete assessment of the intended claim.

The trial compares two active treatment-initiation pathways and includes adherence alongside patient-reported and physiological outcomes. These measures address related but non-identical questions. A strategy may produce one pattern for device use and another for symptoms or residual respiratory events, so each outcome needs a clear definition and appropriately cautious interpretation.

Finally, the registry record illustrates the importance of complete reporting. Information about multiplicity, missing data, analysis populations and some intended outcomes is absent. These details shape how confidently results can be interpreted and compared with future studies. Clear reporting of estimates, confidence intervals, analysis decisions and decision margins makes clinical trial evidence more useful without treating statistical uncertainty as a binary verdict.

19. 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 comparison was unmasked, as the registry records masking as none. Awareness of treatment pathway could influence behavior, treatment use, clinic interactions or responses to subjective measures. The registry record does not report procedures intended to limit such effects.

The primary endpoint is measured 6 weeks after treatment starts, so it does not establish whether group differences persist over longer follow-up. The endpoint is a mean of nightly use, including zero-use nights, but the registry does not provide the distribution of use or separate counts of nights without use. A group average can conceal substantial variation among participants.

The registry reports enrollment of 156, while the primary endpoint arm counts are 59 and 66. The reasons for the difference are not reported in the registry record. Missing-data methods and the analysis population are also not reported, making it difficult to determine how participants without endpoint measurements were handled or how the analysis relates to all enrolled participants.

The registry describes diagnosis by portable monitoring, but the record does not report further diagnostic eligibility details or the representativeness of the enrolled population. The primary outcome description refers to patients using CPAP or APAP at the clinic visit. The relationship between that wording and the analysis population is not clarified.

The non-inferiority or equivalence label cannot be fully evaluated because the margin and decision rule are not reported in the registry record. Multiplicity procedures, detailed analysis plans, and results for some outcomes named in the study aims are also not reported. These limitations constrain interpretation of the intended claims and the breadth of conclusions that can be drawn.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Continuous Positive Airway Pressure (PAP) Titration and Treatment Versus Auto-adjusting PAP Treatment for Sleep Apnea — NCT00988351.
• Registry record last updated 2014-11-03; status last verified 2014-10.
• Results first posted 2014-11-03.
• 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.

23. Record Summary

NCT00988351 is a completed randomized, prospective, parallel-group study comparing attended CPAP titration followed by CPAP treatment with auto-adjusting PAP treatment without titration in obstructive sleep apnea diagnosed by portable monitoring. The registry records an enrollment of 156 and no masking.

The primary outcome was mean nightly PAP use, including nights with zero use, at 6 weeks. The group means were 3.99 hours and 4.45 hours, respectively. The estimated difference was 0.46 hours, with a recalculated 95% confidence interval from -0.348 to 1.268 and an unadjusted p-value of 0.264. The sponsor-reported analysis is closely aligned.

The interval spans zero, and the non-inferiority or equivalence margin is not reported in the registry record. Secondary results are posted for several measures, but treatment pressure and some outcomes named in the study aims have no results reported in the registry record. Missing-data methods, multiplicity procedures and the analysis population are not reported.