Effect of Liraglutide in Obese Subjects With Moderate or Severe Obstructive Sleep Apnoea: SCALE™ - Sleep Apnoea: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effect of Liraglutide in Obese Subjects With Moderate or Severe Obstructive Sleep Apnoea: SCALE™ - Sleep Apnoea (NCT01557166), a randomized phase 3 study. The primary outcome is Change From Baseline in Apnoea-hypopnoea Index (AHI).

2. Clinical Question

The registry record identifies the trial as phase 3 and states that it was conducted in North America. The central statistical comparison is the difference in mean change in AHI between the randomized groups. A negative change represents a reduction in the measure from baseline; the posted group means are therefore read as average changes, not as the groups’ AHI levels at the end of the trial. The record also lists changes in body weight, fasting plasma glucose and HbA1c as secondary outcomes. It does not report the trial’s eligibility details beyond the stated population description, so further characteristics of participants should not be inferred.

3. Trial Design

The registry record describes a phase 3, randomized, double-blind, parallel-group trial of liraglutide 3.0 mg and placebo. Its official title describes a 32-week, placebo-controlled, multicentre and multinational trial; the brief summary states that the trial was conducted in North America. The record lists an enrollment of 359 and a completed status. The start date is 2012-06-07 and the primary completion date is 2013-06-01.

The primary outcome was change from baseline in AHI at Week 32, with the outcome timeframe given as Week 0 and Week 32. The primary population is described as the full analysis set, including all randomized subjects, with 334 subjects contributing to the statistical analysis. The primary endpoint results list 168 participants in the liraglutide group and 166 in the placebo group. These endpoint-specific counts should be read as the participants included in the reported outcome summaries, rather than as a replacement for the registry’s total enrollment figure.

Three secondary outcomes are posted: change from baseline in body weight, fasting plasma glucose and glycosylated haemoglobin (HbA1c). Their reported group counts vary by outcome. The registry record does not explain the differing counts or describe how missing measurements were handled. It also does not report the trial’s original sample-size justification, planned interim analyses, or detailed statistical analysis plan. The record identifies ANCOVA as the method for the reported primary comparison, but the details of that model are 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.

Before assignment: The trial consisted of a 2-week screening period before randomisation.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set (FAS) - included all randomised subjects. 334 subjects contributed to the statistical analysis.

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 Change From Baseline in Apnoea-hypopnoea Index (AHI), measured over Week 0, Week 32.

The primary endpoint is the mean change from baseline in the apnoea-hypopnoea index (AHI), measured in events per hour of sleep. The registry record describes AHI as the number of apnoea and hypopnoea events per hour and identifies it as a measure used to diagnose and classify obstructive sleep apnoea severity. It gives severity categories of none at 4.9 or fewer events/hour, mild from 5.0 to 14.9, moderate from 15.0 to 29.9, and severe at 30.0 or more.

Change from baseline expresses how the group average moved over the stated treatment period rather than reporting only an end-of-treatment value. A negative mean change indicates that the average AHI was lower than at baseline. The endpoint is relevant to the stated clinical question because it measures the frequency of breathing events used to characterize the condition under study. However, a group mean change does not show how every participant’s AHI changed, nor does it by itself describe movement between severity categories. The registry record posts the average changes and their variability, along with an adjusted treatment comparison. It does not report individual-level transitions between AHI categories or other measures of how participants experienced the condition.

7. Results Posted in the Registry

• Liraglutide 3.0 mg: mean -12.22 versus -6.08 on control; difference -6.14 (95% CI -11.43 to -0.85), standardized effect d = -0.25; z-test p = 0.023.

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 From Baseline in Body Weight (kg)

• 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:
  • Liraglutide 3.0 mg: mean -6.73 versus -1.87 on control; difference -4.86 (95% CI -6.12 to -3.60), standardized effect d = -0.80; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Liraglutide 3.0 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Fasting Plasma Glucose

• 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:
  • Liraglutide 3.0 mg: mean -0.15 versus 0.17 on control; difference -0.32 (95% CI -0.49 to -0.15), standardized effect d = -0.38; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Liraglutide 3.0 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Glycosylated Haemoglobin (HbA1c) (%)

• 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:
  • Liraglutide 3.0 mg: mean -0.36 versus -0.17 on control; difference -0.19 (95% CI -0.25 to -0.13), standardized effect d = -0.64; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Liraglutide 3.0 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Across the three posted secondary outcomes, the group mean changes are more negative in the liraglutide group than in the placebo group. The reported unadjusted comparisons also have negative mean differences, with p-values listed as <0.001 for each. This gives a consistent direction across outcomes involving body weight, fasting plasma glucose and HbA1c, alongside the negative primary AHI comparison. These outcomes describe different measures and scales, so their numerical estimates should not be compared directly with one another.

The secondary results add context to the primary finding by showing the pattern of other measured changes during the same treatment period. They do not establish the mechanism behind the primary result, and they do not replace the primary endpoint analysis. The registry record lists these as secondary outcomes but does not report a multiplicity adjustment or a complete statistical plan for their interpretation.

When several outcomes are tested, the chance of obtaining one or more small p-values by chance can increase. General explanation: adjustment procedures can account for this multiplicity, while unadjusted p-values do not make that correction. Because the status of any adjustment is not reported in the registry record, the posted unadjusted secondary comparisons carry less inferential weight than a prespecified, multiplicity-controlled analysis would. They are informative descriptions of the posted results, but should not be treated as independent confirmatory evidence without information about the testing plan.

Group counts differ across the secondary outcomes. The registry record does not explain those variations or report how missing measurements were handled. The results therefore need to be read with the reported outcome-specific populations in mind, rather than assumed to represent identical participant sets for every measure.

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

• Change From Baseline in Body Weight (kg)
• Change From Baseline in Fasting Plasma Glucose
• Change From Baseline in Glycosylated Haemoglobin (HbA1c) (%)

9. Statistical Interpretation

10. Statistical Methods Explained

The main endpoint is continuous: each participant’s AHI change is a numerical measurement in events/hour, and the registry reports group means and standard deviations. The registry record also reports an adjusted ANCOVA treatment estimate. These summaries answer related but not identical questions. The group means describe the average observed change in each arm; the ANCOVA estimate is the sponsor’s model-based comparison.

What does a mean difference represent?

In general, a difference between group means describes how far apart the two groups’ average outcomes are, using the stated comparison direction. For the primary outcome, the posted liraglutide mean change is -12.22 and the placebo mean change is -6.08 events/hour. The simple unadjusted re-calculation is -6.14 events/hour. Because the endpoint is change from baseline, a negative comparison indicates a more negative average change in the liraglutide group. It does not imply that each individual experienced a decrease, or that the mean difference is the outcome for a typical participant.

Standard deviation describes the spread of individual observations around a group mean. The posted primary endpoint standard deviations are 23.34 for liraglutide and 25.9 for placebo. In general, a standard deviation is not the same as the uncertainty in the estimated mean: it describes variation among observations, while a standard error or confidence interval describes uncertainty in an estimate. The registry record identifies standard deviation as the dispersion measure for the primary outcome.

Why are there both an unadjusted re-calculation and an ANCOVA result?

The unadjusted re-calculation uses the posted group means and their sample summaries to form a direct comparison. It is useful for interpreting the raw separation between the reported averages, but it cannot account for covariates or other modeling choices unless those are explicitly incorporated. The registry record reports the sponsor’s primary analysis as ANCOVA, or analysis of covariance, and gives an estimated treatment difference of -6.10, a 95% confidence interval from -11.0 to -1.19, and p = 0.015. In general, ANCOVA estimates a group contrast while accounting for specified covariates. Which covariates were used and the other model details are not reported in the registry record.

The simple unadjusted re-calculation gives p = 0.023. Its p-value and the sponsor’s p-value are not interchangeable: they come from different analyses. The unadjusted calculation does not supersede the sponsor’s reported model-based analysis. The record does not identify a discrepancy between the posted summary statistics and the sponsor’s analysis.

What does the confidence interval communicate?

In general, a confidence interval gives a range of estimates compatible with the observed data under the analysis and its assumptions. A narrower interval indicates greater precision, while a wider interval indicates less precision. The reported primary 95% interval extends from -11.0 to -1.19 events/hour. It describes uncertainty in the estimated group contrast, not a range containing most individual participant outcomes. Its limits should be interpreted in the scale of the outcome, events/hour.

A p-value describes how unusual data at least as discordant with a specified null hypothesis would be under the test’s assumptions. It is not the probability that the null hypothesis is true, and it does not measure clinical importance. The sponsor’s p = 0.015 is the result from the reported ANCOVA analysis; the unadjusted re-calculation gives p = 0.023. Neither number alone describes the size or precision of the effect, so the estimate and interval are important alongside the p-value.

Why can over-dispersion matter?

General explanation: when count outcomes vary more than a simple statistical model expects, a model that underestimates that variability can produce confidence intervals that are too narrow and tests that are too optimistic. The primary endpoint is a count rate of events per hour, but the registry record reports it as a continuous measure summarized by means and standard deviations and analyzed using ANCOVA. It does not report a count model or an over-dispersion assessment. Therefore, over-dispersion should not be assumed to have been present or absent in the trial’s analysis.

Overall, the posted summaries support a transparent unadjusted comparison, while the sponsor’s ANCOVA result is the reported trial analysis. They share the same general direction but are not identical estimands or methods. The registry record does not provide enough model detail to recreate the ANCOVA or determine exactly why its estimate and uncertainty differ from the simple re-calculation.

11. Confidence Intervals and Precision

The sponsor’s primary estimate is -6.10 events/hour, with a 95% confidence interval from -11.0 to -1.19. The interval’s full range is on the negative side of zero, and its width conveys that the data are compatible with a range of average differences rather than one exact value. In general, a confidence interval reflects sampling uncertainty under the analysis model and its assumptions; it is not a guarantee that the true value lies within the interval, nor does it describe the range of individual participants’ outcomes.

The interval should be considered together with the endpoint scale. It expresses uncertainty in the difference in mean AHI change, measured in events/hour. It is not an interval for the mean change within either arm and should not be interpreted as a range of individual AHI changes. The reported group standard deviations, 23.34 for liraglutide and 25.9 for placebo, describe variability in observations and are distinct from the confidence interval around the treatment estimate.

The registry record reports the adjusted ANCOVA interval, while a simple unadjusted re-calculation gives a mean difference of -6.14 and p = 0.023. The adjusted analysis reports p = 0.015. The interval for the adjusted estimate is the appropriate uncertainty statement for that reported model; the unadjusted calculation is not a substitute for it. Details needed to reproduce the model-based interval, including the covariates and model specification, are not reported in the registry record.

Precision is not the same as clinical importance. Whether the estimated difference is large enough to matter clinically is not reported in the registry record. The confidence interval helps describe statistical uncertainty, but it cannot answer that separate question by itself.

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 figures are planning illustrations expressed as participants per arm, not total events. They use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each row considers an effect to detect equal to the stated fraction of the observed effect: 100%, 75% or 50%. These are planning assumptions for the illustration; the trial’s original sample-size assumptions and justification are not reported in the registry record.

The 80% power column means the illustrated number of participants per arm for a design that would have an 80% chance of detecting the specified effect under the stated assumptions, if that effect were the true effect and the model assumptions held. The 90% column gives the corresponding planning number for a 90% chance. These are design probabilities, not probabilities that the treatment effect is real. The displayed numbers are participants per arm; they are not totals across both arms.

Smaller effects require more participants to detect with the same power under otherwise unchanged assumptions. Accordingly, planning for 75% or 50% of the observed effect yields larger per-arm sample-size figures than planning for 100% of the observed effect. The curve is read as a visual counterpart to these planning values: as the assumed effect becomes smaller, the required sample size increases. These figures are not observed power. They are prospective planning illustrations based on the observed effect, not evidence about the probability that the completed trial detected an effect.

Because the observed effect is used to define the planning scenarios, the estimates are sensitive to how well that observed value reflects the effect that a future study should target. The registry record does not report whether these illustrations resemble the original trial’s planning process; the listed assumptions belong to the illustration, not necessarily to the trial design.

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 = -6.140; pooled standard deviation σ = 24.646.
2. n = 2 × (1.96 + 0.842)² × 24.646² / -6.140² = 252.9, rounded up to 253 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 <- -6.14                      # observed difference in means
sigma <- 24.6455496605835                     # 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(1557166)
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 = 253 (80% power), 339 (90% power)
 75% of observed effect: n per arm = 450 (80% power), 602 (90% power)
 50% of observed effect: n per arm = 1,012 (80% power), 1,355 (90% power)
power.t.test: 253.9 
Simulated power at n = 253 per arm: 0.804 (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’s reported primary analysis used ANCOVA and gave an estimated treatment difference of -6.10 events/hour, with a 95% confidence interval from -11.0 to -1.19 and p = 0.015. This is the registry record’s reported inferential analysis for the primary comparison.

The simple unadjusted re-calculation from the posted group means is -6.14 events/hour, with p = 0.023. The re-calculation is based on the summary statistics and does not reproduce ANCOVA. The two estimates are close in value and point in the same direction, but their p-values and methods differ. The ANCOVA result is the sponsor’s own analysis; it should be interpreted as the trial’s reported analysis rather than replaced with an unadjusted reconstruction.

The registry record does not report the covariates, full model specification or other details needed to reproduce the ANCOVA. It also does not report an explanation for the numerical differences between the adjusted result and the simple re-calculation. The record does not flag a discrepancy. The re-calculation is therefore best treated as a transparent supplementary comparison of the posted group summaries, not as a correction to the sponsor’s analysis.

15. Multiplicity, Missing Data and Analysis Populations

The registry record reports one primary endpoint and three secondary outcomes. It gives an ANCOVA result for the primary endpoint and unadjusted comparisons for the secondary outcomes. The record does not report whether the primary analysis was part of a hierarchical testing strategy, whether secondary tests were adjusted for multiplicity, or what significance threshold was planned for the trial. Those details are not reported in the registry record.

General explanation: testing multiple outcomes creates more opportunities for a nominally small p-value to occur by chance. A multiplicity strategy can limit that risk by defining an order of testing or adjusting the inference across outcomes. Without a reported strategy, the secondary p-values should be read as unadjusted evidence and not as a set of fully controlled confirmatory tests. The registry record’s primary ANCOVA p-value is reported, but the associated prespecification and multiplicity context are not reported in the registry record.

The primary population is described as the full analysis set, including all randomized subjects, and 334 subjects contributed to the statistical analysis. The primary endpoint group summaries include 168 subjects assigned to liraglutide and 166 assigned to placebo. Secondary outcome counts vary by measure. The registry record does not state why counts vary, whether all randomized participants had an observed measure, or what methods were used for missing values. It does not report whether imputation, a likelihood-based approach, or another missing-data method was used; none should be assumed.

The record also does not report sensitivity analyses for missing data, an analysis of departures from the intended treatment, or additional analysis populations beyond the stated full analysis set description. Such information would help readers assess how robust the estimates are to incomplete outcome data and analysis choices. The reported counts and estimates can be described as posted, but the missing-data assumptions behind them cannot be reconstructed from the registry record.

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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: angina pectoris (1.1%), sinus arrest (0.6%), cholelithiasis (0.6%), pneumonia (0.6%), procedural pain (0.6%).

The safety summary lists serious adverse events for 6 of 176 participants at risk in the liraglutide group and 6 of 179 participants at risk in the placebo group. These are participant counts in the registry’s safety summary. The record also lists other adverse events for 117 of 176 participants at risk with liraglutide and 84 of 179 at risk with placebo. These categories should be read as reported, and the registry record does not provide enough detail to characterize event duration, severity beyond the serious-event classification, or causal attribution.

The listed serious-event terms include angina pectoris at 1.1%, and sinus arrest, cholelithiasis, pneumonia, procedural pain and spinal fracture at 0.6% each. Those percentages are the registry’s posted values. They describe reported terms and do not by themselves establish that a treatment caused an event or provide a complete comparative safety assessment.

Death counts are not reported in the registry record. The safety figures should be interpreted as descriptive counts over the recorded safety population, not as a complete account of all possible harms or a basis for individual medical decisions.

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

The posted primary results show a larger average reduction in AHI from baseline in the liraglutide group than in the placebo group at Week 32. The sponsor’s ANCOVA estimate is -6.10 events/hour, with a 95% confidence interval from -11.0 to -1.19 and p = 0.015. The simple unadjusted re-calculation from the arm means is -6.14 events/hour. These statistics describe the average difference between groups in the analyzed trial population.

They do not establish that every participant had a reduction, that participants moved into a different AHI severity category, or that the average difference is clinically important. The registry record does not report a threshold for clinical importance or individual transitions between severity categories. A group mean can conceal a broad range of individual outcomes, and the posted result should not be interpreted as an individual prediction.

The result applies to the stated endpoint and 32-week timeframe. It does not describe effects beyond that period. The registry record also does not provide the detailed statistical model specification or missing-data methods needed to reproduce the sponsor’s estimate fully. The primary analysis population is identified as the full analysis set, but the reasons for outcome-specific participant counts are not reported.

Secondary outcomes show a consistent direction of average change across the measures posted, but their comparisons are unadjusted and multiplicity handling is not reported. These results are therefore contextual rather than a substitute for the primary analysis. Safety counts are descriptive; they do not establish causation or provide a complete assessment of risks.

In general, a trial’s randomized comparison can estimate an average difference between assigned groups under the trial’s design and analysis assumptions. It does not guarantee an outcome for a particular person, and it does not answer questions beyond the populations, outcomes and timeframes actually reported. Further clinical interpretation requires context not contained in the registry record.

18. Why This Trial Matters Statistically

This trial illustrates why a clinical endpoint should be read in its measurement units and alongside the analysis method. AHI is reported as events/hour, and its change from baseline provides a direct comparison of average changes over the stated treatment period. The sponsor’s adjusted ANCOVA estimate and a simple re-calculation from group means provide related views of the same comparison, but they are not interchangeable analyses.

The reported 95% confidence interval adds information that a p-value alone cannot provide: it communicates the range of estimates compatible with the model and data, while keeping the interpretation on the AHI scale. At the same time, statistical precision and clinical importance are distinct. The registry record does not state a clinical-importance threshold, so the statistical result should not be converted into a claim about the practical importance of the estimated difference.

The record also offers a useful example of the limits of registry summaries. It identifies the analysis population, outcome measures, sponsor’s ANCOVA result and secondary comparisons, but does not report the full analysis plan, multiplicity procedure or missing-data methods. Those omissions matter when assessing how much inferential weight to place on multiple outcomes and how reproducible the estimates are.

Finally, the secondary results and safety counts must be interpreted according to their role and limitations. Consistency across secondary measures can provide context, while unadjusted comparisons remain less conclusive when multiplicity handling is unknown. Descriptive adverse-event counts support transparent reporting but do not establish causation. Together, these features make the trial a useful case for separating the observed data, the statistical model, and the broader conclusions those results can support.

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 registry record gives limited detail about the primary ANCOVA. The covariates, model specification and assumptions are not reported, so the sponsor’s adjusted estimate cannot be independently reproduced from the record. The simple unadjusted re-calculation is not a substitute for those missing model details.

The record identifies the full analysis set as including all randomized subjects and states that 334 subjects contributed to the primary statistical analysis, but does not explain the relationship between that count, the overall enrollment of 359, and the outcome-specific group counts. Counts also vary among secondary outcomes. Reasons for the variation, missing-data procedures and sensitivity analyses are not reported in the registry record.

Multiplicity handling for the secondary outcomes is not reported. Their p-values are posted as unadjusted comparisons, so the evidence from those tests should be interpreted cautiously when considering several outcomes together. The record does not provide the trial’s original sample-size justification or planned interim analyses; the displayed sample-size planning figures are illustrations based on the observed effect and should not be confused with original design assumptions.

The primary outcome is a group average change in AHI, and no individual-level distribution or severity-category transition results are reported. The clinical-importance threshold for the mean difference is not reported. The registry record also does not provide follow-up results beyond the stated Week 32 timeframe. Safety reporting is descriptive; death counts are not reported, and the posted serious-event terms and counts do not by themselves establish causality.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Effect of Liraglutide in Obese Subjects With Moderate or Severe Obstructive Sleep Apnoea: SCALE™ - Sleep Apnoea — NCT01557166.
• Registry record last updated 2017-11-01; status last verified 2017-09.
• Results first posted 2015-02-09.
• 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.

23. Record Summary

NCT01557166 is a completed phase 3 randomized, double-blind, parallel-group trial comparing liraglutide 3.0 mg with placebo in obese subjects with moderate or severe obstructive sleep apnoea. The registry record lists an enrollment of 359 and describes a 32-week trial conducted in North America.

The primary endpoint was change from baseline in AHI at Week 32. The sponsor reported an ANCOVA estimated treatment difference of -6.10 events/hour, with a 95% confidence interval from -11.0 to -1.19 and p = 0.015. The simple unadjusted re-calculation from the posted group means is -6.14 events/hour. Three secondary outcomes are also posted, with unadjusted comparisons.

The record does not report the detailed analysis plan, multiplicity handling, missing-data methods, or the original sample-size justification. Safety summaries include serious and other adverse-event counts; death counts are not reported in the registry record.