A Pilot Study of the Effects on Sleep Disordered Breathing (SDB) When Using the Drug Liraglutide for 4 Weeks: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Pilot Study of the Effects on Sleep Disordered Breathing (SDB) When Using the Drug Liraglutide for 4 Weeks (NCT01832532), a non-randomized phase 1, phase 2 study. The primary outcome is Change in Apnea Hypopnea Index (AHI) From Baseline.

2. Clinical Question

The clinical question is about a short-term change in a sleep-study measure, not about longer-term outcomes. The study also included breathing tests, fasting blood samples, weekly sleepiness questionnaires, and blood-sugar checks for the liraglutide group, as described in the registry summary. Those study procedures do not substitute for the primary outcome result. No result for the primary endpoint, uncertainty estimate, or statistical test is reported in the registry record.

3. Trial Design

This completed pilot study is listed as Phase 1 and Phase 2, with a parallel-group, nonrandomized allocation and no masking. Its registry enrollment is 34. The record describes an open-label controlled study, so participants and study personnel were not masked to assigned treatment. Because allocation was not randomized, the groups may have differed at baseline in ways that affect the outcome; the record does not describe an adjustment strategy to address such differences.

The registry summary describes a liraglutide treatment group and a control group. The treatment group used liraglutide daily for four weeks, with dose titration from 0.6 mg/day to 1.8 mg/day or the highest tolerated dose. The control group did not use liraglutide. Both groups were to continue standard care for sleep apnea. The summary describes the intended groups as 20 and 10 adults, respectively; the adverse-event record lists 18 and 9 participants at risk in the corresponding groups. The record does not explain the relationship between these group counts and the overall enrollment.

The listed study start is 2013-04, and the listed primary completion is 2015-12. The primary outcome timeframe is baseline and four weeks. Participants were to have study visits and an overnight research sleep study; those without a recent sleep study could have an additional research study to establish baseline severity. Weekly phone visits included a sleepiness questionnaire. The record does not report the recruitment process, reasons participants may not have contributed outcome data, or details of the analysis population.

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: Enrollment fell short of goal for each cohort

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.

6. Primary Endpoint

The registry lists the primary outcome as Change in Apnea Hypopnea Index (AHI) From Baseline, measured over Baseline and 4 weeks.

The primary endpoint is change in the apnea hypopnea index (AHI) from baseline to the end of the study at four weeks. AHI counts apnea and hypopnea events per hour of sleep, so it expresses the frequency of these breathing events while accounting for the amount of time spent asleep. The registry lists the measure in apnea/hypopnea events per hour of sleep and specifies a mean with standard deviation.

In general, a change-from-baseline measure describes how the outcome at follow-up differs from its starting value for the analysis population. Comparing changes between groups can help distinguish a pattern over time in the intervention group from a pattern also seen in a control group. In a nonrandomized study, however, the groups may differ in characteristics that influence their starting values or subsequent changes. A group contrast therefore needs cautious interpretation, especially if the record does not describe adjustment for baseline differences.

AHI is used as a quantitative sleep-study measure of sleep-disordered breathing. Its interpretation depends on the sleep-study measurements and on how the outcome was summarized and analyzed. The record does not report the observed mean changes, group variability, a between-group estimate, or a confidence interval. It therefore does not establish the direction, size, or precision of the primary outcome result.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The endpoint is a mean change in AHI, a continuous measure expressed as apnea/hypopnea events per hour of sleep. The registry specifies standard deviation as the measure of dispersion. In general, a mean describes the average measurement in a group, while a standard deviation describes how spread out individual measurements are around that average. Neither statistic, on its own, describes the uncertainty in a between-group comparison. The registry record does not report the endpoint means, standard deviations, or comparison estimate.

What does a change score represent?

A change score compares a participant’s follow-up measurement with their baseline measurement. Group summaries of change can then be compared. In general, this approach focuses on within-participant movement over the study period, but it does not by itself ensure that groups are comparable. Baseline imbalances, differences in follow-up, or other factors can influence the contrast. Because allocation was nonrandomized, any between-group interpretation would need to account for the possibility that groups differed for reasons other than treatment. The record does not state whether the analysis adjusted for baseline AHI or other characteristics.

Why does nonrandom allocation matter?

Random allocation is generally used to balance, on average, both measured and unmeasured characteristics between groups. Nonrandom allocation does not provide that design-based protection. Statistical adjustment can sometimes address measured differences, but it cannot guarantee removal of bias from unmeasured factors. This is a general design principle, not a claim about the actual balance of this study’s groups. Baseline characteristics and an adjustment strategy are not reported in the registry record.

How should variability and precision be understood?

A standard deviation describes participant-to-participant variability in a group. A confidence interval for an estimated group contrast addresses uncertainty around that contrast. These answer different questions: a group can have substantial variation while its mean is estimated with some precision, or a mean difference can remain uncertain even when group summaries are available. The registry identifies standard deviation as the endpoint’s dispersion measure but does not provide the values or a confidence interval. As a result, precision cannot be judged from the endpoint entry alone.

In general, a p-value summarizes how compatible the observed data are with a specified null hypothesis under assumptions of the statistical model. It does not measure the probability that the null hypothesis is true, and it does not show whether an effect is clinically important. Here, the registry record does not report a p-value or a statistical test for the primary endpoint. It also does not report a prespecified hypothesis, test direction, or model details.

What does the planning illustration represent?

The stated planning assumptions describe an illustration based on the observed effect, using a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. The illustration considers the effect at a stated fraction of the observed effect. This is a planning exercise, not the trial’s own design or an observed-power calculation. Sample-size requirements based on an observed effect are sensitive to uncertainty in that estimate; a small or unstable estimate can make planning projections uncertain. The registry record does not report an original sample-size justification, and the endpoint results needed to inspect the observed effect are not reported. Planning figures should not be interpreted as evidence that the study had adequate power or that a result was likely to be detected.

For event-rate outcomes, over-dispersion can matter when the variability exceeds what a simple model assumes; in general, ignoring it can make uncertainty appear too small. That concept does not describe the listed primary endpoint, which is a mean change in AHI rather than an event rate per unit of exposure. The registry classifies the outcome measure in its record, but no event-rate primary result is listed. The outcome’s stated unit and summary type are the relevant details for interpreting this endpoint.

What cannot be reconstructed?

The registry record does not report individual participant data, group-level outcome summaries, a between-group estimate, an analysis model, or uncertainty measures. No recalculated effect can therefore be described from the registry record. The record also does not report whether covariates, repeated-measures methods, or any other adjustment were used. Statistical concepts can clarify what those methods would mean in general, but they cannot fill in an undocumented analysis or result.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure, the change in AHI from baseline at four weeks, and reports zero analyses. The presence of a listed primary outcome does not establish what hypothesis tests were conducted or whether additional analyses were planned. A statistical analysis plan, including any approach to multiple outcomes or multiple comparisons, is not reported in the registry record.

In general, multiplicity arises when several hypotheses or outcomes are tested; testing many hypotheses without an appropriate strategy can increase the chance of a false-positive result. The record does not provide enough information to determine whether multiplicity was an issue or how it would have been handled. It would be inappropriate to assume that a single listed outcome means there was no multiplicity, since other analyses or outcomes may not be described.

Missing-data details are also not reported. The record does not state how many participants had baseline and follow-up sleep-study measurements, why any measurements may have been unavailable, or whether missing values were imputed, excluded, or handled by another method. The overall enrollment and the group counts in the adverse-event record do not identify the endpoint analysis population.

The record does not specify whether the primary analysis included all enrolled participants, only those who received treatment, or participants with complete measurements. It also does not report exclusions, protocol deviations, or sensitivity analyses. These gaps matter because different analysis populations and missing-data assumptions can lead to different estimates and uncertainty. No conclusions about the robustness of a primary outcome analysis can be drawn without those details and the outcome results.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The adverse-event record lists zero serious adverse events and zero other adverse events for the liraglutide group, with 18 participants at risk, and zero in both categories for the control group, with 9 participants at risk. It also lists zero deaths in each group and no serious-event terms. These are the counts recorded in the registry, not proof that no events occurred outside the registry’s reporting definitions or observation period.

The registry summary describes a four-week treatment period, but the adverse-event entry does not specify the event-collection window or provide further definitions. It does not report a detailed accounting of event types, severity, relatedness, or how events were solicited and verified. The group counts at risk should not be assumed to be the same as the number included in the primary endpoint analysis.

In general, zero recorded events in a small study do not establish that a treatment has no safety risks. They indicate only that the registry lists no affected participants in the stated categories for the recorded groups. The nonrandomized design and limited reported safety detail also constrain comparisons. These figures should be read as descriptive registry counts, not as a definitive estimate of comparative safety.

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

This record documents a completed, open-label, nonrandomized pilot study of liraglutide for four weeks in adults without diabetes who had sleep apnea. The primary question concerned change in AHI, a measure of apnea and hypopnea events per hour of sleep. The record specifies the outcome and its timeframe, but does not report the outcome results. It therefore does not show whether mean AHI changed in either group, how the groups compared, or how uncertain an estimated difference might be.

The design can describe a treatment group and a control group, but nonrandom allocation means that the comparison is not equivalent to a randomized comparison. Differences between groups could reflect factors other than the intervention. Open-label conduct is another design feature to consider. The registry does not report baseline balance, adjusted analyses, or other details that would help assess these issues.

The adverse-event entry records zero serious events, zero other events, and zero deaths in each group, with different numbers of participants listed at risk. Those counts are descriptive and do not establish absence of risk. The registry does not provide enough detail about the observation window or event ascertainment to support a broader safety conclusion.

The enrollment and group descriptions also need careful reading: the record lists an enrollment of 34, describes groups of 20 and 10 in the summary, and lists 18 and 9 at risk in the adverse-event record. It does not explain how these figures relate to the primary-outcome analysis. Thus, no inference about completion or outcome availability should be made from these counts alone.

Overall, the registry record identifies a pilot study and its intended primary measure, but leaves the central efficacy result and key analysis details unreported. It cannot establish the size, direction, or precision of a group difference. It also cannot show whether any apparent change would be attributable to liraglutide rather than group differences, time, or other factors.

13. Why This Trial Matters Statistically

This study is statistically informative as an example of why the design and reporting context matter as much as the endpoint name. It has a clearly identified short-term measure—change in AHI from baseline at four weeks—and a control group. But nonrandom allocation and open-label conduct mean that a group contrast would require cautious interpretation, even if outcome estimates were available.

The registry record also illustrates the distinction between enrollment, the groups described in a study summary, and the participants listed in safety reporting. Those counts are not interchangeable, and the record does not explain how they relate to the primary endpoint analysis. Without the analysis population and outcome availability, readers cannot tell which participants contributed to an estimate.

Reporting the mean change alone would not be enough to establish precision. Group variability, a between-group estimate, and an uncertainty interval would help show how much the data constrain plausible effects. A p-value alone would not supply that information. In this record, those primary outcome results are not reported, so the key statistical questions about magnitude and precision remain unanswered.

The planning assumptions also emphasize a separate distinction: sample-size illustrations are not observed power and do not show what the study was designed to detect. The registry record does not report an original sample-size justification. Any planning exercise based on an observed effect should be considered an illustration, not a substitute for the trial’s actual analysis or a guarantee of future study performance.

For clinical researchers, the record underscores the value of transparent reporting of outcome summaries, analysis populations, missing-data handling, adjustment methods, and uncertainty. Without these elements, a completed pilot can be described by its design and intended question, but its primary statistical result cannot be independently interpreted from the registry record.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The main limitation for interpreting this trial record is that it does not report the primary outcome results. No group-specific AHI changes, between-group estimate, confidence interval, or p-value are available in the registry record. The central question therefore cannot be answered from the information recorded.

The nonrandomized allocation and lack of masking limit causal interpretation. The record does not report baseline group characteristics or whether an adjusted analysis was conducted, so it is not possible to assess measured group comparability or how potential confounding was addressed. The record also does not describe allocation procedures in enough detail to evaluate selection into groups.

Enrollment and group counts are reported in different contexts: overall enrollment is 34; the study summary describes groups of 20 and 10; and the adverse-event record lists 18 and 9 at risk. The record does not explain the relationship among these figures, report endpoint-specific denominators, or describe participant flow. Reasons for nonparticipation in outcome analyses, if any, are not reported.

Missing-data methods, analysis populations, protocol deviations, and sensitivity analyses are not reported. The registry also does not report a statistical analysis plan or an original sample-size justification. These omissions prevent assessment of whether the endpoint analysis followed a prespecified approach and how robust it might be to alternative assumptions.

The intervention period and primary outcome timeframe are short, and the population described is adults without diabetes who have sleep apnea. The registry record does not provide evidence about longer-term outcomes or populations beyond those described. Safety reporting is limited to event counts, without a detailed account of ascertainment or the adverse-event observation window.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Pilot Study of the Effects on Sleep Disordered Breathing (SDB) When Using the Drug Liraglutide for 4 Weeks — NCT01832532.
• Registry record last updated 2020-10-20; status last verified 2020-09.
• Results first posted 2020-10-20.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

NCT01832532 is a completed, open-label, nonrandomized parallel-group pilot study listed as Phase 1 and Phase 2. It compared four weeks of daily liraglutide, titrated from 0.6 mg/day to 1.8 mg/day or the highest tolerated dose, with a control group among adults without diabetes who had sleep apnea. Both groups were described as continuing standard care.

The primary endpoint was mean change in AHI from baseline at four weeks, measured in apnea/hypopnea events per hour of sleep. The registry record does not report the endpoint results, statistical comparison, or uncertainty estimates. It lists zero serious adverse events, zero other adverse events, and zero deaths in both groups. The record does not report the primary analysis population, missing-data methods, or statistical analysis plan.