CPAP vs ASV for Insomnia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for CPAP vs ASV for Insomnia (NCT02365064), a randomized na study. The primary outcome is Insomnia Severity Index.

2. Clinical Question

The comparison is between two positive airway pressure modes rather than between treatment and no treatment. That matters when interpreting the estimate: it describes the difference between the posted ASV and CPAP group results, not the effect of either mode against an untreated group. The registry labels the study randomized, single masked, and parallel. It does not identify who was masked or give further details about the allocation process.

3. Trial Design

The registry lists this as a completed, randomized, single-masked, parallel-group trial. Enrollment is listed as 78. The record gives a start date of 2015-02 and a primary completion date of 2017-08. It identifies ResMed as the sponsor. The phase is recorded as NA.

Participants were assigned to a CPAP active-comparator arm or an ASV experimental arm. CPAP is described as a fixed pressure for inspiration and expiration. ASV is described as using different inspiratory and expiratory pressures and adjusting pressure support to meet a target minute ventilation. Those descriptions identify the treatment modes, but they do not establish the exact device settings, duration of use, adherence expectations, or other treatment procedures; these details are not reported in the registry record.

The primary outcome is listed as the Insomnia Severity Index, assessed at baseline and 4 months. The posted arm summaries use 21 participants for CPAP and 19 for ASV. These outcome-specific counts should not be read as the enrollment count or as a complete account of participant flow. Reasons for the difference between enrollment and the outcome counts, withdrawals, and missing outcome data are not reported in the registry record.

The record names five secondary outcomes: Sleep Quality Rating, Global Morning Rating, Sleep Onset Latency, Wake After Sleep Onset (WASO), and Sleep Efficiency. Numerical results are posted for three of those outcomes. The record does not report the masking target, allocation concealment, stratification, treatment adherence, or a detailed statistical analysis plan. A study protocol or original sample-size justification is not reported 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:

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: Baseline information available for ITT

6. Primary Endpoint

The registry lists the primary outcome as Insomnia Severity Index, measured over Baseline and 4 months.

The primary endpoint is change in insomnia severity from baseline to 4 months, measured with the Insomnia Severity Index (ISI). The registry describes the ISI as a subjective questionnaire with seven questions whose answers are added to produce a total score. Scores range from 0 - 28, and a higher score indicates more severe insomnia. The measure therefore gives a common numerical scale for comparing reported insomnia severity between the randomized groups.

The registry lists the measure as a mean with standard deviation. For CPAP, the posted mean is 9.29 with a standard deviation of 6.57 among 21 participants. For ASV, the posted mean is 13.21 with a standard deviation of 5.17 among 19 participants. The comparison is reported as an ASV-minus-CPAP mean difference of 3.92. Because a higher ISI score indicates more severe insomnia, the positive direction of this difference corresponds to a higher posted score in the ASV group. These summaries describe the posted endpoint; they do not by themselves explain how missing observations, baseline values, or any covariates were handled.

The ISI is used because the trial’s stated question includes insomnia severity and the scale directly records participants’ subjective symptoms. A questionnaire score is not the same as a measure of sleep-breathing events, even though the trial summary also refers to those events. The registry’s primary endpoint is the change in insomnia severity, so conclusions about the posted primary comparison should remain focused on that outcome.

7. Results Posted in the Registry

• Adaptive Servo-Ventilation: mean 13.21 versus 9.29 on control; difference 3.92 (95% CI 0.27 to 7.57), standardized effect d = 0.66; z-test p = 0.035.

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.

Sleep Quality Rating

• 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:
  • Adaptive Servo-Ventilation: mean 1.65 versus 1.06 on control; difference 0.59 (95% CI -0.07 to 1.25), standardized effect d = 0.55; z-test p = 0.082.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Adaptive Servo-Ventilation and the control. The interval gives the range of effects that remain compatible with the data.

Global Morning Rating

• 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:
  • Adaptive Servo-Ventilation: mean 24.33 versus 15.39 on control; difference 8.94 (95% CI -1.47 to 19.35), standardized effect d = 0.52; z-test p = 0.092.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Adaptive Servo-Ventilation and the control. The interval gives the range of effects that remain compatible with the data.

Sleep Onset Latency

• 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:
  • Adaptive Servo-Ventilation: mean 7.17 versus -6.09 on control; difference 13.26 (95% CI -9.25 to 35.77), standardized effect d = 0.36; z-test p = 0.248.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Adaptive Servo-Ventilation and the control. The interval gives the range of effects that remain compatible with the data.

The registry names five secondary outcomes, while numerical results are posted for Sleep Quality Rating, Global Morning Rating, and Sleep Onset Latency. For each of these posted outcomes, the point estimate is in the direction of a higher mean in the ASV group than in the CPAP group. The uncertainty is substantial: the unadjusted intervals for these secondary contrasts include zero, and their unadjusted p-values do not give strong evidence against a null difference on their own. Sleep Efficiency and WASO are named as secondary outcomes, but numerical results for them are not reported in the registry record.

This broad pattern is directionally similar to the primary mean contrast, which also has a higher posted score in the ASV group. Similar directions across outcomes can be informative as a descriptive pattern, but they do not establish that the outcomes measure the same construct or that one result confirms another. Sleep quality, morning ratings, sleep onset latency, and insomnia severity are distinct measures, and their scales and timeframes differ. The secondary findings should not be combined into a single overall effect without a prespecified method.

Unadjusted secondary results carry less weight when several outcomes are examined because the chance of obtaining an apparently notable result by chance can increase as more comparisons are made. The registry record does not report a multiplicity strategy, a hierarchy of outcomes, or adjusted secondary analyses. The estimates and intervals are therefore more useful as exploratory descriptions than as independent confirmatory findings.

The primary result remains the prespecified focus identified in the registry, while secondary outcomes can help describe other aspects of sleep experience. However, no analysis details are reported for the secondary comparisons, and the registry does not provide enough information to determine whether the same participants contributed to every outcome or how missing observations were handled. These gaps reinforce the need to interpret the secondary pattern cautiously rather than treating it as a set of definitive tests.

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

• Sleep Quality Rating
• Global Morning Rating
• Sleep Onset Latency
• Wake After Sleep Onset (WASO)
• Sleep Efficiency

9. Statistical Interpretation

10. Statistical Methods Explained

The registry classifies the primary outcome as continuous and reports a mean and standard deviation for each arm. A mean is the arithmetic average of the measured values in a group. A standard deviation describes how dispersed the values are around that average. It is not the same as the uncertainty in the mean: uncertainty also depends on the number of observations and on the method used to analyze them.

What does the mean difference represent?

The reported comparison is the ASV mean minus the CPAP mean. The estimate of 3.92 therefore expresses the difference between the two posted group means on the ISI scale. A positive estimate means the ASV mean is higher in this subtraction order. Since the registry says higher ISI scores indicate more severe insomnia, the direction can be described without implying that the difference is causal, clinically important, or applicable to every participant.

The comparison is labeled a mean difference and is a simple unadjusted recalculation from the posted summary statistics. An unadjusted comparison uses the group summaries without accounting for other variables. Randomization is intended to support a fair comparison of assigned groups, but the registry’s brief record does not establish how the result was analyzed or whether all randomized participants were included. It would not be appropriate to describe the displayed calculation as an adjusted model, a repeated-measures model, or an analysis of covariance.

How should the standard deviations be read?

The posted standard deviations are 6.57 for CPAP and 5.17 for ASV. They convey within-group variation in the outcome values, while the mean difference describes the separation between group averages. When variation is substantial relative to the contrast, participant-level values can overlap even when group means differ. The group summaries alone do not show the shape of either distribution, influential observations, or whether the data meet assumptions of a particular model.

In general, a standard error describes the estimated sampling uncertainty of a statistic, such as a mean or mean difference. A confidence interval combines an estimate with an indication of that uncertainty. The interval is not a range containing a fixed proportion of individual participant scores. Nor does it express the probability that the true effect lies in the particular interval after the data have been observed. These are general explanations of the statistical concepts, not claims about an unreported analysis procedure.

What does the p-value tell the reader?

The posted unadjusted p-value for the primary contrast is 0.035. In general, a p-value is calculated under a null model and summarizes how unusual the observed statistic, or a more extreme one, would be if that model were true. It does not tell the reader the probability that the null model is true. It also does not indicate whether an observed difference is important to patients, whether it will replicate, or whether the model’s assumptions are suitable.

The registry record lists no reported analyses. The comparison is therefore best described as an unadjusted recalculation from the posted arm summaries, rather than as a reported sponsor inferential result. The record does not report the statistical model, handling of baseline values, missing-data method, analysis population, or any multiplicity adjustment. Those omissions limit how specifically the p-value can be interpreted. No discrepancy is flagged between a sponsor analysis and the recalculation.

Why do analysis details matter?

Different analysis choices can target different quantities. Comparing follow-up scores, comparing changes, and adjusting follow-up scores for baseline can yield different estimates, especially if baseline values differ or observations are missing. The endpoint description refers to change from baseline to 4 months, but the registry record does not report enough detail to identify the precise calculation behind the posted values. Statistical interpretation should not fill that gap by assuming a preferred method.

General statistical practice also distinguishes an analysis population from the number of observations shown for an outcome. The posted arm counts identify the number used in those summaries, but the registry does not say whether they represent all randomized participants, an intention-to-treat population, a per-protocol population, or another set. It also does not state how unavailable scores were handled. Without those details, it is not possible to assess the influence of missingness or exclusions from the record alone.

What is the role of the planning calculations?

The sample-size illustrations are planning calculations based on the observed effect, not the trial’s original design. They use the assumptions listed in the registry material: a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set to the stated fraction of the observed effect. They are not evidence that the trial had a particular probability of detecting an effect. The original design assumptions and sample-size justification are not reported in the registry record.

In general, planning estimates depend on the effect size being targeted, outcome variability, allocation, significance level, and desired power. They can help illustrate how a future study might be planned under stated assumptions, but they cannot repair uncertainty in the current trial’s outcome reporting. Sample-size planning should be interpreted separately from the observed group results and from the p-value.

11. Confidence Intervals and Precision

The confidence interval for the primary comparison describes uncertainty around the estimated mean difference. The posted interval has a positive lower bound and extends to 7.567; the lower endpoint is reported with more than three decimal places, so it is not reproduced. The interval’s position is consistent with the positive point estimate, but the registry record does not provide enough information about the analysis to establish how the interval was constructed.

In general, a confidence interval gives a range of values compatible with the data under the statistical procedure used. A narrower interval indicates greater precision for the target estimate, while a wider interval indicates more uncertainty. It is not a range for individual scores, and it does not by itself determine whether an effect is clinically meaningful. The ISI’s score direction is known from the registry, but an important-difference threshold is not reported in the registry record.

The posted outcome counts are 21 for CPAP and 19 for ASV. The standard deviations are 6.57 and 5.17, respectively. These summaries make clear that the estimate comes from a limited set of outcome observations with variability within each arm. The confidence interval is therefore important context: the point estimate alone would overstate the precision with which the group contrast is known.

Precision also depends on the analysis population, missing-data handling, and model. Those details are not reported in the registry record. The interval should not be read as incorporating uncertainty about all possible analytic choices, multiplicity, or bias. It quantifies uncertainty under its calculation, not uncertainty from every limitation of the study record.

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 sample-size calculations are planning illustrations based on the observed effect, not the trial’s original design assumptions. The entries are participants per arm, not total event counts. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and a target effect equal to the stated fraction of the observed effect. The trial’s own sample-size justification and original planning assumptions are not reported in the registry record.

The 80% power column indicates the number of participants per arm that the calculation estimates would be needed to have an 80% chance of detecting the specified effect under those assumptions. The 90% power column uses a higher target chance of detection, so its planning number is larger for the same effect assumption. These are prospective planning quantities: they describe a hypothetical design under the calculation’s model, not the probability that this completed trial detected an effect.

The rows compare planning for 100%, 75%, and 50% of the observed effect. For example, planning for 75% of the observed primary effect gives 65 participants per arm for 80% power and 87 participants per arm for 90% power. As the target effect gets smaller, more participants are needed to distinguish it from random variation under otherwise unchanged assumptions. The curve visualizes that relationship under the same planning framework; it does not add new trial results or replace the stated assumptions.

This is not observed power. Observed power is not a useful substitute for the estimate, confidence interval, and limitations of the actual results. The planning illustrations should instead be read as examples of how the assumed target effect and desired power affect a future sample-size calculation.

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 = 3.920; pooled standard deviation σ = 5.948.
2. n = 2 × (1.96 + 0.842)² × 5.948² / 3.920² = 36.1, rounded up to 37 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 <- 3.92                      # observed difference in means
sigma <- 5.94806071890574                     # 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(2365064)
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 = 37 (80% power), 49 (90% power)
 75% of observed effect: n per arm = 65 (80% power), 87 (90% power)
 50% of observed effect: n per arm = 145 (80% power), 194 (90% power)
power.t.test: 37.1 
Simulated power at n = 37 per arm: 0.800 (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. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary outcome and five secondary outcomes, but the record does not report how multiplicity was handled. Multiplicity arises when several hypotheses or outcomes are examined: without an adjustment or a prespecified testing hierarchy, the chance of at least one apparently unusual result can increase. No adjustment method, gatekeeping procedure, or confirmatory status for secondary outcomes is reported in the registry record. The unadjusted secondary p-values should therefore be treated as exploratory signals, not as a collection of independent confirmatory tests.

The registry reports no analyses. It does not identify an intention-to-treat, per-protocol, or other analysis population, nor does it state whether the primary comparison included all randomized participants with available data. The outcome-specific counts are 21 for CPAP and 19 for ASV, but the record does not explain why these counts differ from the listed enrollment or describe participant flow. Reasons for missing outcomes or exclusions are not reported in the registry record.

The method used for missing data is also not reported in the registry record. It is not stated whether missing scores were left out, imputed, or addressed through a model, nor whether the analysis assumed that missingness was related to observed or unobserved outcomes. These choices can affect both the estimate and its uncertainty. The available arm summaries cannot determine whether missingness could have biased the comparison.

The registry describes the endpoint as change from baseline to 4 months, but does not provide the analysis specification. It is not reported whether the calculation compared change scores directly, adjusted follow-up scores for baseline, or included other covariates. It is also not reported whether an interim analysis was planned or conducted, or whether any outcomes were added, removed, or changed. Such details should not be inferred from the completed status or from the posted estimates.

15. 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’s serious adverse event figures list zero affected participants in the CPAP group, with 21 at risk, and zero affected participants in the ASV group, with 19 at risk. The deaths field is zero in each group. The record lists no serious adverse event terms. These figures describe what was posted for the registry’s serious-event reporting; they do not establish that no other safety outcomes occurred.

For other adverse events, the record lists 20 affected among 21 at risk in the CPAP group and 16 affected among 19 at risk in the ASV group. These are counts, not estimates of comparative risk, and no causal attribution or event definitions are included in the record. The registry does not report event severity, duration, timing, specific non-serious event terms, or a statistical comparison between arms.

Zero serious adverse events in the posted groups should not be interpreted as proof that serious events are impossible or that the treatments have equivalent safety. The numbers at risk are limited, and the record’s adverse-event summary does not supply the detail needed to evaluate event types or ascertainment. These data should be read as a registry snapshot rather than a complete safety assessment.

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

The registry’s primary result describes an average difference in the posted ISI outcome between two randomized treatment arms. The ASV group mean is higher than the CPAP group mean, and the difference is reported as 3.92 points in the ASV-minus-CPAP direction. Because higher ISI scores indicate more severe insomnia, this is a direction-of-score description. It is not, by itself, proof that either treatment caused a particular outcome for an individual participant.

The posted p-value and confidence interval summarize statistical uncertainty under the unadjusted calculation. They do not answer whether the size of the difference matters to patients, whether the result would persist in another study, or whether the result applies to people outside the trial population. The registry does not report a threshold for clinically important change, detailed participant characteristics, treatment adherence, or the analysis population. Those omissions limit interpretation beyond the observed group summaries.

The result also does not compare either treatment with no therapy. It is a comparison of ASV and CPAP as described in the registry. The trial summary mentions sleep-breathing events as part of its objective, but the posted primary endpoint is the ISI. Conclusions about sleep-breathing events cannot be drawn from the primary outcome alone.

Nor do the secondary findings provide an independent confirmation of the primary contrast. They concern different outcomes, have uncertainty, and are unadjusted; numerical results for two named secondary outcomes are not reported in the registry record. The adverse-event figures likewise do not provide a complete account of safety or a formal comparison of risks.

The most defensible reading is specific: the posted summaries show a higher average ISI score in the ASV arm than in the CPAP arm at the endpoint described as change from baseline to 4 months, with an unadjusted recalculated contrast and associated uncertainty. The registry record does not support a broader claim about clinical benefit, harm, mechanism, or treatment choice.

17. Why This Trial Matters Statistically

This trial is statistically instructive because it contrasts two active pressure modes and reports a patient-reported continuous endpoint. The primary result can be expressed as a mean difference on the ISI scale, while its standard deviations and confidence interval help distinguish the estimated group contrast from the spread of individual outcomes and uncertainty around the estimate. That distinction is central to reading clinical trial results: an estimate, its precision, and its clinical meaning are related but not interchangeable.

The record also illustrates why a p-value should not stand alone. The unadjusted p-value accompanies a modest outcome dataset and a comparison reconstructed from posted summaries, while details of the sponsor’s analysis are absent. Without the analysis population, missing-data approach, baseline handling, and multiplicity strategy, readers cannot fully evaluate the inferential framework. Registry summaries can make a result accessible, but they do not replace a complete statistical analysis plan and report.

The secondary outcomes show another important issue: a set of estimates can have a broadly similar direction while remaining imprecise and subject to multiplicity. A pattern can be worth noting without treating each contrast as confirmatory. The absence of numerical results for Sleep Efficiency and WASO also demonstrates that a listed outcome is not necessarily accompanied by a posted result.

Finally, the planning illustrations separate future study design from the trial’s observed evidence. They show how the number planned per arm changes with the target effect and desired power under stated assumptions. They are not a retroactive measure of whether this trial was adequately powered, and the original design rationale is not reported in the registry record. Keeping these distinct interpretations separate is essential for sound statistical reading.

18. 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 limitations specific to this registry record affect interpretation. The primary outcome description refers to change from baseline to 4 months, but the record does not explain how the posted group means or comparison were calculated. It is not reported whether the analysis adjusted for baseline ISI or used another model. The displayed contrast is a simple unadjusted recalculation from posted summary statistics, and no sponsor analysis is listed.

The outcome counts are smaller than the enrollment figure, but participant flow and reasons for missing outcome data are not reported in the registry record. The analysis population, handling of missing values, and any exclusions are also not reported. Without this information, the possibility and direction of bias from missingness cannot be assessed.

The record labels the design single masked but does not identify who was masked. It does not report allocation concealment, eligibility criteria, detailed participant characteristics, adherence, treatment duration, or device-setting procedures beyond the brief mode descriptions. Those gaps limit evaluation of implementation and generalizability.

Although five secondary outcomes are named, numerical results are posted for only three. The registry does not report a multiplicity approach or a prespecified hierarchy for interpreting them. The adverse-event summary gives counts but not event definitions, non-serious event terms, timing, severity, or attribution. Finally, no clinical-importance threshold for the ISI comparison is reported in the registry record. These limitations are reasons for caution, not evidence that the observed result is invalid.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: CPAP vs ASV for Insomnia — NCT02365064.
• Registry record last updated 2021-09-09; status last verified 2021-05.
• Results first posted 2021-09-09.
• 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.

22. Record Summary

NCT02365064 is listed as a completed randomized, single-masked, parallel-group trial sponsored by ResMed. It compares CPAP, described as the active comparator, with ASV, described as the experimental intervention, in a study concerning insomnia and sleep-disordered breathing. Enrollment is listed as 78, and the primary endpoint is change in Insomnia Severity Index from baseline to 4 months.

The posted primary summaries give an ISI mean of 9.29 for CPAP and 13.21 for ASV, with a simple unadjusted ASV-minus-CPAP mean difference of 3.92 and p = 0.035. The outcome-specific counts are 21 and 19, respectively. Three secondary outcomes have numerical results posted; two other named secondary outcomes do not. The record lists no reported analyses and does not describe the analysis population, missing-data handling, multiplicity approach, or original sample-size justification.