Study to Predict the Benefits of First-line Humidification Use and Comparison of the Effect of ThermoSmart™ and no Humidification on Adherence: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study to Predict the Benefits of First-line Humidification Use and Comparison of the Effect of ThermoSmart™ and no Humidification on Adherence (NCT01742949), a randomized na study. The primary outcome is Adherence.

2. Clinical Question

The clinical question is framed around use of the device, rather than a direct measure of sleep apnea severity. The registry also names a prediction objective: identifying patients who might be more adherent with high levels of humidification. However, no predictor analysis or subgroup results are reported in the registry record. The posted average adherence values therefore address the overall comparison, not whether particular patient characteristics identify a different response.

The crossover design is relevant to how this question should be interpreted. In a crossover trial, participants may contribute measurements under both conditions, making within-person comparisons potentially important. The record gives arm-level means and a comparison, but it does not report enough detail to determine how sequence, period, or within-person correlation were handled in the analysis.

3. Trial Design

This completed study was randomized, double-masked, and used a crossover model. The registry lists enrollment of 40 participants and identifies obstructive sleep apnea as the condition studied. The sponsor was Fisher and Paykel Healthcare. The recorded study start was 2013-10 and primary completion was 2014-08.

The two conditions were heated humidification using ThermoSmart and no humidification. The latter was described as dry CPAP or APAP. The primary outcome was adherence, reported as hours per night and averaged over the stated 8-week timeframe. The device’s internal software was used to measure adherence, with reporting through InfoSmart software, according to the endpoint description.

Because the model is crossover, it is important not to interpret the listed condition counts automatically as separate sets of participants. The registry reports 40 for each condition, but those entries do not establish that 80 distinct people took part. A crossover study generally obtains measurements from a participant in more than one condition, though the exact sequence and implementation must be checked in the trial record. The treatment sequence, period structure, any washout interval, and details of how the crossover was incorporated into the statistical analysis are not reported in the registry record.

The registry labels masking as double but does not identify which groups were masked or explain how masking was maintained for a comparison involving humidification and dry device use. The specific randomization procedure, allocation concealment, and blinding procedures are not reported in the registry record. The study is marked completed, but completion status alone does not provide details about participant flow, withdrawals, adherence to assigned conditions, or deviations from the intended sequence.

In addition to adherence, the record names Epworth Sleepiness Score, the Short Functional Outcomes of Sleep Questionnaire (FOSQ-10), subjective dry nose, subjective dry mouth, subjective sinusitis, and preference as secondary outcomes. Results are posted for three secondary measures. The record lists seven outcome measures overall, but does not provide posted results for every listed secondary measure.

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 Adherence, measured over 8 weeks.

The primary endpoint is adherence with treatment per night, expressed in hours and averaged over the total measured period. The registry specifies an 8-week timeframe and says adherence was measured using internal software on the CPAP device and reported with InfoSmart software. This endpoint captures the extent to which participants used the prescribed device, rather than measuring symptoms or a physiological response directly.

Adherence is used because actual treatment use is a practical component of treatment delivery: a device cannot be used during nights when a participant does not wear it. In this study, the endpoint is intended to assess whether the humidification condition is associated with different nightly use. That general rationale does not mean that an adherence difference by itself establishes a difference in sleepiness, sleep-related functioning, or any other clinical outcome.

The registry reports mean adherence of 4.7 hours per night with ThermoSmart and 5.0 with no humidification, with standard deviations of 2.7 and 2.5, respectively. The comparison is expressed as a mean difference of 0.3 hours per night for no humidification relative to ThermoSmart. The comparison’s confidence interval includes zero, so the posted estimate is not precise enough to distinguish a difference from no difference at conventional levels. The record does not provide a clinically meaningful threshold for adherence, so the clinical importance of a particular difference is not defined in the registry.

7. Results Posted in the Registry

• No Humidification: mean 5.00 versus 4.70 on control; difference 0.30 (95% CI -0.84 to 1.44), standardized effect d = 0.12; z-test p = 0.606.

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.

Epworth Sleepiness Score

• 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:
  • No Humidification: mean 6.00 versus 6.20 on control; difference -0.20 (95% CI -0.51 to 0.11), standardized effect d = -0.28; z-test p = 0.206.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between No Humidification and the control. The interval gives the range of effects that remain compatible with the data.

Short Functional Outcomes of Sleep Questionnaire (FOSQ-10)

• 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:
  • No Humidification: mean 32.00 versus 32.10 on control; difference -0.10 (95% CI -2.69 to 2.49), standardized effect d = -0.02; z-test p = 0.940.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between No Humidification and the control. The interval gives the range of effects that remain compatible with the data.

Subjective Dry Nose

• 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:
  • No Humidification: mean 4.80 versus 3.10 on control; difference 1.70 (95% CI -0.18 to 3.58), standardized effect d = 0.40; z-test p = 0.077.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between No Humidification and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover several domains: sleepiness, sleep-related functioning, subjective dry-nose symptoms, subjective dry-mouth symptoms, subjective sinusitis, and preference. Results are posted for Epworth Sleepiness Score, FOSQ-10, and subjective dry nose. The record does not provide posted results for all of the listed secondary outcomes.

Across the posted secondary comparisons, the confidence intervals include zero, and the unadjusted p-values do not establish a statistically clear difference. The estimates vary in direction and size across measures, which is unsurprising because the outcomes capture different experiences and are measured on different scales. They should not be combined into a single overall effect or treated as interchangeable evidence about adherence.

These outcomes can provide context around the primary endpoint because they address sleepiness, functioning, or symptoms rather than device use alone. However, they do not change the primary adherence comparison, and none resolves the prediction objective. The registry reports no predictor or subgroup analysis that would show whether some participants had a different adherence response from others.

Unadjusted secondary results carry less weight when several outcomes are examined. In general, testing multiple outcomes creates more opportunities for chance findings, so interpretation depends on whether the outcomes and analysis approach were prespecified and whether multiplicity was addressed. The registry record does not report a multiplicity adjustment or a hierarchy for interpreting the secondary outcomes. As a result, the secondary comparisons are best treated as descriptive and exploratory rather than as independent confirmatory evidence.

The registry lists six secondary outcomes, but only three have posted numerical comparisons. The absence of reported results for the remaining outcomes should not be interpreted as showing no effect. Their estimates and analysis status are not reported in the registry record.

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

• Epworth Sleepiness Score
• Short Functional Outcomes of Sleep Questionnaire (FOSQ-10)
• Subjective Dry Nose
• Subjective Dry Mouth
• Subjective Sinusitis
• Preference

9. Statistical Interpretation

10. Statistical Methods Explained

Several statistical ideas help interpret continuous outcomes such as nightly adherence. The registry reports a mean for each condition and a standard deviation. A mean describes the average value in a group or condition. A standard deviation describes how spread out individual measurements are around that mean. It is not the uncertainty of the mean itself, and it should not be read as a confidence interval.

What does the mean difference represent?

The comparison is reported as no humidification minus ThermoSmart. The means are 5.0 and 4.7 hours per night, and the posted mean difference is 0.3 hours per night. The sign of a difference depends on which condition is subtracted from which. A positive value means the reported mean is higher for no humidification under that stated contrast; it does not, by itself, show that the difference is reliable or important.

The comparison also lists a pooled standard deviation of 2.602. Pooling standard deviations is a way to summarize variability across conditions for a standardized comparison under particular assumptions. The registry does not report a sponsor analysis explaining whether this summary or an alternative was used in its primary statistical plan. A standardized effect size can help compare contrasts measured on different scales, but it does not replace the original units or define clinical importance. The posted standardized effect estimate is not needed to interpret the primary result in hours per night.

What does an unadjusted comparison mean?

The posted comparison is an unadjusted re-calculation from the reported summary statistics. An unadjusted comparison does not incorporate adjustments for other variables unless those are explicitly stated. It is useful as a transparent summary of the reported means, but it is not evidence that the same method was used for the sponsor’s planned analysis. No sponsor-reported analyses are listed in the registry record, and the registry does not report a detailed statistical analysis plan.

The reported p-value is 0.606. In general, a p-value describes how unusual data at least as discordant with a specified null hypothesis would be under the assumptions of the statistical test. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. Here, it does not provide evidence of a statistically clear average difference, but it cannot establish that the conditions have identical effects.

Why does crossover structure matter?

In a parallel-group study, each participant typically contributes data to one assigned condition. In a crossover study, participants may receive more than one condition, so observations from the same person are related. A paired analysis can use the within-person differences and may account for this dependence. An analysis that treats measurements as independent can give an inappropriate estimate of uncertainty if the correlation between repeated measurements is ignored.

The registry identifies the study as crossover but does not report the treatment sequence, period definitions, washout arrangements, or the method used to account for repeated measurements. The arm-level means and standard deviations alone do not reveal the within-person covariance needed to reproduce a paired analysis. Consequently, the posted unadjusted comparison should not be mistaken for a documented crossover-model estimate. A crossover trial can also be affected by period or carryover effects; whether these were assessed is not reported in the registry record.

How should uncertainty be understood?

A confidence interval gives a range of values compatible with the observed data under the method’s assumptions. For the primary mean difference, the interval crosses zero. Zero represents no difference in the contrast as defined. This means the data do not pin down the direction of the average difference with high precision. The p-value and confidence interval convey related information, but the interval is more informative about the range and direction of effects compatible with the analysis.

In general, confidence intervals rely on assumptions about the data and statistical model. For a crossover outcome, those assumptions may concern within-person dependence and period structure. Since the registry does not describe the analysis method, the interval should be interpreted as the interval accompanying the posted unadjusted comparison, not as confirmation that all crossover features were modeled.

Why can over-dispersion matter?

General statistical explanation: over-dispersion occurs when observed variability is greater than a model assumes. When that happens, standard errors may be understated if the model’s variance assumptions are too restrictive, making intervals too narrow and tests appear more decisive than warranted. Adherence is a continuous measure, and the posted standard deviations describe observed variability, but the registry does not report an over-dispersion assessment or a model-based diagnostic. It would be inappropriate to claim that over-dispersion was present or absent in this trial.

The record reports no sponsor analysis for the primary endpoint and no reported analyses for the posted secondary outcomes. It does not state which analysis population was used, how missing observations were handled, or whether the crossover design was modeled with participant-level terms. Those omissions limit methodological review; they do not establish that any particular method was or was not used outside the information recorded in the registry.

11. Confidence Intervals and Precision

The primary estimate is a mean difference of 0.3 hours per night for no humidification relative to ThermoSmart. Its confidence interval crosses zero, and the unadjusted p-value is 0.606. The interval therefore includes both no difference and effects in either direction. It does not isolate a single value or direction with enough precision to support a firm conclusion about average adherence.

Precision is different from the size of the point estimate. A small or moderate point estimate can be uncertain, while a larger estimate can also be uncertain if the data are variable. The primary standard deviations are 2.7 hours per night for ThermoSmart and 2.5 for no humidification. These describe variation among measurements, not the precision of the estimated contrast by themselves. The confidence interval combines information about variability and sample size under the calculation’s assumptions.

In general, a confidence interval is a range produced by a statistical procedure that, under its assumptions, has a stated long-run coverage property. It is not a range containing a known proportion of individual participants, nor is it the probability that the true effect lies within the particular interval. Its practical meaning also depends on whether the calculation fits the design. Here, the study is crossover, while the posted comparison is an unadjusted re-calculation from summary statistics; the record does not state how within-person dependence was addressed in a sponsor analysis.

The registry does not report a threshold for a clinically meaningful adherence difference. Therefore, the interval cannot be judged against a stated minimum important effect. The posted results support describing uncertainty around the average contrast, not declaring that any possible effect is clinically important or unimportant. The lack of a reported sponsor analysis and crossover-specific analysis details further limits how precisely the posted interval can be interpreted.

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, not total events. They are planning illustrations based on the observed effect and use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set at the stated fraction of the observed effect: 100%, 75%, or 50%. These are assumptions for the illustration, not reported assumptions from the trial’s original design; the original design’s sample-size assumptions are not reported in the registry record.

The 80% power column describes the number of participants per arm that the planning calculation estimates would be needed to have an 80% chance of detecting the specified effect under the stated assumptions, if that effect were the true one. The 90% power column uses a higher target chance of detection, so its participant requirement is larger for the same effect assumption. These are theoretical planning quantities, not guarantees about what a future study will find.

The rows use progressively smaller fractions of the observed effect. Smaller effects are harder to distinguish from random variation, so the planning calculation requires more participants per arm to detect them at the same target power. The figures are especially sensitive to the observed effect used as their starting point; an uncertain or small observed effect can produce very large planning requirements when the assumed target effect is reduced.

Because the actual study is crossover, these two-arm planning illustrations should not be read as a reproduction of its original design calculation. The stated illustration assumptions describe equal allocation between two arms, whereas the registry identifies a crossover model. The registry does not report the original sample-size justification or the assumptions used to plan enrollment. This is not observed power: it is a planning exercise based on the observed effect and specified assumptions, not a measure of how much power the completed trial actually had.

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 = 0.300; pooled standard deviation σ = 2.602.
2. n = 2 × (1.96 + 0.842)² × 2.602² / 0.300² = 1,180.8, rounded up to 1,181 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 <- 0.3                      # observed difference in means
sigma <- 2.60192236625154                     # 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(1742949)
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 = 1,181 (80% power), 1,581 (90% power)
 75% of observed effect: n per arm = 2,100 (80% power), 2,811 (90% power)
 50% of observed effect: n per arm = 4,724 (80% power), 6,324 (90% power)
power.t.test: 1181.8 
Simulated power at n = 1,181 per arm: 0.798 (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 identifies a primary adherence outcome and six secondary outcomes. It does not report a prespecified hierarchy among secondary outcomes or an adjustment for multiple comparisons. When several outcomes are tested, the chance of at least one apparently notable result by chance can increase. General statistical approaches to this issue include adjusting significance thresholds or defining a primary outcome hierarchy in advance. Which approach, if any, was planned or used for this study is not reported in the registry record.

The record lists no reported analyses for the primary outcome, and the three posted secondary comparisons also have no reported analyses. Thus, the posted summary statistics do not explain whether the sponsor used a particular model, adjusted for baseline values, accounted for repeated measurements, or followed any prespecified analysis plan. The trial’s statistical analysis plan is not reported in the registry record.

Missing-data methods and analysis populations are also not reported. The record does not state whether the analysis included all randomized participants, only participants with measurements in both conditions, or another defined population. It does not describe the amount or pattern of missing observations, how missingness was handled, or whether sensitivity analyses were undertaken. The enrollment and condition-level counts alone do not establish that all measurements were complete.

These omissions matter particularly for a crossover design. A paired comparison generally depends on measurements from the same participants under both conditions, and incomplete pairs can affect which observations contribute to the estimate. The record does not state how incomplete crossover data, period effects, or possible carryover were handled. It is not possible to infer from the posted averages that such issues were absent or that a particular method was used.

The stated objective of predicting which patients benefit most would ordinarily require a reported method for evaluating predictors or differences in response. No such analysis is listed. The registry record therefore does not provide evidence about predictive factors, multiplicity control, or the robustness of subgroup findings.

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.

The registry reports zero serious adverse events affecting participants in the ThermoSmart condition and zero in the no-humidification condition. For each condition, 40 participants are listed as at risk. It also reports zero other adverse events affecting participants in each condition and zero deaths in each condition.

These figures should be read as the adverse-event counts recorded in the registry, not as proof that no adverse experiences occurred outside the categories or reporting process represented there. The record does not provide event descriptions, severity grading beyond the serious-event category, timing, relatedness assessments, or details of how adverse events were collected. It also does not explain whether the condition-level counts refer to unique participants across crossover periods or how repeated exposure was handled.

Accordingly, the posted safety information is a compact count summary. It reports no affected participants in the listed serious or other adverse-event categories and no deaths, but it does not offer enough detail to evaluate event types, reporting completeness, or comparative risk with precision.

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

The posted adherence means are 4.7 hours per night with ThermoSmart and 5.0 with no humidification. The unadjusted mean difference is 0.3 hours per night in the direction of higher average adherence with no humidification. Its confidence interval crosses zero, and the unadjusted p-value is 0.606. These results do not establish a clear average adherence difference between the conditions.

That interpretation is not the same as saying the conditions are equivalent. Equivalence requires a prespecified margin defining how much difference is acceptable and an analysis designed to assess whether the uncertainty lies within that margin. No equivalence margin or equivalence analysis is reported in the registry record. Nor does the result prove that the true effect is exactly zero; the interval remains compatible with a range of possible effects.

The findings also do not answer whether an individual participant might respond differently from the average. The study description includes an aim of predicting who might be more adherent with humidification, but no predictor analysis is reported. An overall mean can obscure variation between participants, and the posted condition averages cannot identify which characteristics are associated with a different response.

Adherence is the primary measure, not a direct demonstration of changes in sleepiness, functioning, or symptoms. Secondary results are posted for some measures, but they do not establish a definitive effect across those domains, particularly because the registry does not report a multiplicity plan or sponsor analyses. The results should therefore be understood as estimates of device use and related posted measures, not as a broad conclusion about clinical outcomes.

Finally, the crossover structure is important. The available summary statistics do not show how the analysis handled within-participant correlation, treatment sequence, or period effects. The record also does not report missing-data methods or analysis populations. These limits mean that the posted unadjusted comparison is a useful description of the recorded averages, but it should not be treated as a complete account of the trial’s planned analysis.

17. Why This Trial Matters Statistically

This study raises a statistical question that is easy to miss when results are presented as two condition means: the design is crossover, so the same participant may contribute measurements under both conditions. For such a design, the within-person contrast can be more relevant than a comparison that treats condition measurements as independent. The registry’s condition-level means and standard deviations do not reveal the within-person relationship.

The study also illustrates why a point estimate and an uncertainty interval should be read together. The mean difference is 0.3 hours per night, but the confidence interval crosses zero and the unadjusted p-value is 0.606. A single estimate without its uncertainty could suggest more certainty than the posted comparison supports. Conversely, a nonsignificant p-value does not establish equivalence or rule out effects in either direction.

There is a second distinction between an overall treatment comparison and a prediction question. A randomized comparison can estimate an average contrast, while identifying who might respond differently requires additional analyses of participant-level variation or predictors. The registry states the prediction aim but does not report such analyses. That gap is relevant to readers interested not only in average adherence but also in variation across patients.

Finally, the planning illustrations are based on the observed effect and stated assumptions, not on the trial’s original design justification. They show how required sample size changes when the target effect is made smaller, but they do not establish what enrollment the original study needed or whether the completed trial was adequately powered. These distinctions help prevent planning calculations, posted descriptive results, and claims about individual response from being conflated.

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.

The main limitation of interpretation is that the registry identifies a crossover design but does not report the sequence, period structure, washout details, or crossover-specific analysis. The posted comparison is an unadjusted re-calculation from summary statistics, and arm-level means and standard deviations do not provide the within-participant covariance needed to reproduce a paired estimate. The reported interval and p-value therefore cannot be verified as accounting for the design’s repeated measurements.

The registry record does not report participant flow, withdrawals, missing-data methods, or analysis populations. Although 40 participants are listed for each condition, the record does not explain whether all participants contributed complete measurements under both conditions. It also does not report baseline characteristics or detailed eligibility criteria, limiting assessment of how the enrolled participants relate to a broader population with obstructive sleep apnea.

The masking label is double, but the record does not describe who was masked or how masking was maintained. Randomization procedures and allocation concealment are not detailed. The timeframe for the primary outcome is stated as 8 weeks, but the record does not clarify the timing or sequence of each condition within the crossover.

The primary comparison has uncertainty that includes no difference, and the registry does not identify a clinically meaningful adherence threshold. There is also no reported predictor analysis despite the stated interest in identifying participants who might benefit most. For secondary outcomes, results are posted for only some listed measures, and multiplicity handling is not reported. Finally, no sponsor-reported analyses or detailed statistical analysis plan are listed, limiting assessment of whether the posted summary comparisons match the intended analysis.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Study to Predict the Benefits of First-line Humidification Use and Comparison of the Effect of ThermoSmart™ and no Humidification on Adherence — NCT01742949.
• Registry record last updated 2019-05-07; status last verified 2019-05.
• Results first posted 2019-05-07.
• 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

This completed randomized, double-masked crossover study compared heated humidification using ThermoSmart with no humidification during CPAP or APAP use in people with obstructive sleep apnea. Enrollment was 40. The primary outcome was average nightly adherence over 8 weeks, measured using device software.

Mean adherence was 4.7 hours per night with ThermoSmart and 5.0 with no humidification. The unadjusted mean difference was 0.3 hours per night for no humidification relative to ThermoSmart; the confidence interval crosses zero and the unadjusted p-value is 0.606. The registry also posts results for three secondary measures, but reports no analyses for the primary or secondary outcomes.

The registry does not report the crossover-specific analysis details, missing-data methods, analysis populations, multiplicity approach, or predictor analyses. The posted results describe an uncertain average adherence contrast and do not resolve whether individual characteristics predict different responses.