Assessment of the Effect of PAP on Energy and Vitality in Mild OSA Patients: The Merge Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Assessment of the Effect of PAP on Energy and Vitality in Mild OSA Patients: The Merge Study (NCT02699463), a randomized na study. The primary outcome is Change in Quality of Life Using the Energy and Vitality Subscale of the Short Form 36 (SF-36) Questionnaire.

2. Clinical Question

The registry states that the investigators wanted to assess responses to CPAP among people with mild OSA, including participants identified using the AASM 2012 definition. Its background notes that treatment reimbursement for mild OSA varies across healthcare systems and says the study was intended to add evidence about treatment in mild OSA subgroups. Those statements describe the study rationale, not a result of the trial.

The primary endpoint focused on energy and fatigue as participants perceived them, rather than on a direct measure of sleep-disordered breathing. The outcome was the change in score at 3 months, calculated as the score at 3 months minus the baseline score, and compared between randomized groups. The registry reports group-level mean changes and a sponsor-reported ANCOVA test. It does not report the adjusted between-group estimate or its confidence interval.

The registry record describes a randomized, parallel-group study with no masking. It records 301 participants enrolled. The relationship between the enrollment figure and the group counts accompanying the primary outcome is not explained in the registry record.

3. Trial Design

The MERGE Study (NCT02699463) was a randomized, parallel-group trial sponsored by ResMed. The registry lists the study as completed, with a start date of 2016-12 and primary completion in 2019-05. The recorded enrollment was 301. The registry lists no applicable phase. Masking is recorded as none.

The intervention group was assigned continuous positive airway pressure, described as CPAP therapy used nightly for the 3 month duration of the trial. The control group received standard care, described as sleep hygiene counseling during the study. The registry does not report additional details about how participants were randomized, how assignments were concealed, or how the counseling was delivered.

The primary outcome was the change in the SF-36 Energy and Vitality subscale score from baseline to 3 months. The registry says participants completed the questionnaire at baseline and 3 months, and defines change as the 3-month value minus the baseline value. The outcome is continuous and is reported as a mean with a 95% confidence interval for each group. Group counts accompanying these summaries are 115 for CPAP and 118 for control. The registry record does not explain how these outcome group counts relate to total enrollment or describe the analysis population.

The posted sponsor analysis is an ANCOVA comparison, with a reported p-value of <0.01 and a superiority analysis classification. The adjusted effect estimate, its confidence interval, covariates, and details of the ANCOVA specification are not reported in the registry record. The record also does not report the randomization ratio, a protocol-specific sample-size justification, interim analyses, or a detailed statistical analysis plan.

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: 301 participants were randomised. 233 patients formed the primary analysis group. Of those, 115 were allocated to the CPAP group and 118 were allocated to standard care. The remaining 68 patients were not analysed as part of the primary analysis but were analysed for secondary analysis.

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 Quality of Life Using the Energy and Vitality Subscale of the Short Form 36 (SF-36) Questionnaire, measured over 3 months.

The primary endpoint was change in the Energy and Vitality subscale of the Short Form 36 (SF-36) questionnaire from baseline to 3 months. The questionnaire measures quality of life through questions about physical and mental functioning. This subscale captures participants’ perceptions of energy and fatigue. The registry states that its possible score ranges from 0 to 100 and that an increase indicates improvement.

Using change from baseline focuses the comparison on how each group’s score shifted over the follow-up period, rather than on the 3-month score alone. The registry defines change as the 3-month score minus the baseline score. A positive mean change therefore reflects an increase in the group’s score, while a negative mean change reflects a decrease. These interpretations concern the direction of this scale, not the clinical importance of a particular change.

Patient-reported energy and vitality are relevant to the study’s stated interest in how people with mild OSA experience their daily functioning. They complement measures focused on sleep or breathing by capturing participants’ reported experience. The registry does not report a threshold for a clinically important change, nor does it explain how a change of a given size should be interpreted for an individual participant.

The posted group summaries report mean changes with 95% confidence intervals. The sponsor’s comparison is reported as ANCOVA, but the adjusted group contrast and its interval are not reported in the registry record. The endpoint description therefore identifies what was measured and how the score is oriented, while the record does not give enough detail to reconstruct the sponsor’s adjusted estimate.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: it is a numerical scale score, and the registry reports a mean change for each group. A mean summarizes the average of the observed changes within a group. It does not reveal the full distribution of responses, such as whether most participants had similar changes or whether individual changes varied widely. The registry does not report a standard deviation or other measure of that individual-level spread for the primary outcome.

What does a confidence interval express?

In general, a confidence interval describes the uncertainty around an estimated quantity under the assumptions of the statistical method. A narrower interval indicates greater precision for that estimate than a wider interval, but it does not establish that the estimate is free of bias. The posted 95% confidence intervals are for the mean change in each group: 6.0 to 9.0 for CPAP and -1.5 to 1.5 for control. They are not reported as intervals for the adjusted between-group comparison.

The confidence intervals should be interpreted in relation to the endpoint’s direction. The scale ranges from 0 to 100, and the registry says an increase indicates improvement. Thus, the CPAP group’s interval is on the positive side of zero, while the control group’s interval includes zero. These group-specific intervals do not substitute for a direct test of the difference between groups. The registry’s direct reported comparison is the ANCOVA test, for which the p-value is <0.01.

What does ANCOVA do?

In general, analysis of covariance (ANCOVA) compares groups on an outcome while accounting for one or more additional variables in a statistical model. For a change-from-baseline outcome, an ANCOVA specification may include baseline information, but the particular covariates and model used in this trial are not reported in the registry record. It would therefore be inappropriate to assume which variables were included or to reconstruct the adjusted result from the group summaries.

The registry reports the ANCOVA p-value but not the adjusted group contrast or its confidence interval. A p-value addresses evidence against a null hypothesis within the model; it does not give the magnitude of the adjusted effect. Without the adjusted estimate and model details, the exact size and precision of the sponsor’s adjusted comparison cannot be assessed from the registry entry.

Why does over-dispersion matter?

In general, over-dispersion occurs when outcome variability is greater than a statistical model assumes. It is most often discussed for count outcomes, where an overly restrictive variance assumption can make uncertainty estimates too small. The primary outcome in this trial is continuous, and the registry does not report a count model or evidence about over-dispersion. The concept is therefore not a specific reported issue for this endpoint; it illustrates why a model’s assumptions need to fit the outcome being analyzed.

Other general assumptions can also affect inference, including the suitability of the model, the accuracy of measurement, and the way missing observations are handled. The registry does not report diagnostic checks, missing-data methods, or enough detail about the ANCOVA specification to evaluate these matters. These omissions do not establish that a particular assumption was violated; they limit what can be assessed from the record.

How should the reported p-value be read?

In general, a p-value is calculated under a specified null hypothesis and model. It describes how unusual results at least as incompatible with that null would be if the null and model assumptions held. The sponsor-reported p-value is <0.01. This indicates a result that the sponsor’s analysis classified as statistically significant under its approach, but the registry does not give the adjusted effect estimate needed to describe the size of the contrast or its precision.

Statistical significance and clinical importance are distinct questions. The registry provides no threshold for a clinically important change on the Energy and Vitality scale, so the p-value cannot establish that the score change crossed such a threshold. Nor does a group mean establish that every participant experienced a change in the same direction.

What can planning illustrations tell readers?

Sample-size calculations are planning illustrations, not observed results. In general, they depend on specified assumptions about the effect to detect, significance level, allocation, and statistical model. The registry record does not report the trial’s own sample-size justification. Any planning illustration based on the observed effect is separate from the trial design and should not be interpreted as observed power or as evidence about the trial’s conduct.

The stated assumptions for planning illustrations use a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation, with the effect defined as a fraction of the observed effect. These are illustration assumptions, not evidence that the trial was designed using them. They do not replace the registry’s missing protocol-specific sample-size rationale, and they do not provide a basis for inferring the design choices actually made by the investigators.

10. Analyses Reported by the Sponsor

The registry’s sponsor-reported analysis compares the groups using ANCOVA and gives p < 0.01. It does not report the adjusted between-group estimate, its confidence interval, or the covariates in the model. The posted group summaries are mean changes of 7.5 for CPAP and 0.0 for control, with their respective 95% confidence intervals, but those summaries alone do not recover the sponsor’s adjusted comparison.

A separate unadjusted between-group recalculation is not estimable from the posted information represented in the record. Accordingly, there is no recalculated contrast to set alongside the sponsor’s ANCOVA estimate. No discrepancy between a recalculation and the sponsor result is flagged in the registry record; the principal limitation is that the adjusted estimate and the inputs needed for a separate two-group calculation are not reported.

In general, an adjusted ANCOVA result and a simple unadjusted comparison can differ because they use different statistical models and may account for different information. The registry does not provide enough detail to determine how any such differences would apply in this trial. The sponsor-reported p-value is therefore the available direct inferential result, while the reported arm means and intervals describe changes within each group. Neither should be mistaken for the other.

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists one primary outcome measure and one reported analysis. It records no posted secondary outcomes. The record does not report whether additional outcomes or analyses were planned outside the posted results, whether the primary analysis was prespecified in a protocol, or whether any adjustment for multiple testing was used. The absence of reported secondary results does not show that no other analyses were conducted; it means they are not reported in the registry record.

In general, multiplicity arises when several hypotheses are tested or multiple analyses are used to support conclusions. If many tests are considered without an appropriate plan, the chance of finding at least one apparently unusual result can increase. How that concern applies depends on the full set of planned and conducted analyses. That full set and any multiplicity strategy are not reported in the registry record, so the extent of multiplicity cannot be assessed from the posted entry.

The registry gives outcome group counts of 115 for CPAP and 118 for control, compared with total enrollment of 301. It does not report participant flow, the number assigned to each group, reasons for absent outcome data, or whether the analysis followed an intention-to-treat population. It also does not describe imputation, exclusion rules, or other missing-data procedures. These omissions prevent assessment of how missing observations may have affected the endpoint summaries and ANCOVA result.

In general, missing-data methods rely on assumptions about why measurements are unavailable. Different assumptions can lead to different estimates, so a description of the analysis population and missing-data handling is important. The registry does not report enough information to judge those assumptions. It would be unwarranted to infer that the difference between enrollment and outcome group counts reflects a particular type of loss, exclusion, or protocol deviation.

The primary endpoint is identified as a change from baseline to 3 months, and the record describes administration of the questionnaire at both time points. It does not report the amount or pattern of missing baseline or follow-up information, or whether the posted mean summaries use all available observations. Interpretation should therefore remain focused on the results as recorded, without assuming a specific missing-data mechanism or analysis population.

12. 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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: stroke (0.9%), hospitalization for dizziness, existing cardiac condition (0.8%), symptoms of transient ischemic attack (tia) (0.8%).

The registry reports serious adverse events for each group and records zero deaths in both groups. In the CPAP group, 1 participant was reported with a serious adverse event among 115 at risk; in the control group, 2 were reported among 118 at risk. The record does not provide further detail about event timing, severity beyond the serious classification, or whether events were judged related to the assigned intervention.

The registry also reports other adverse events: 9 in the CPAP group among 115 at risk and 13 in the control group among 118 at risk. These are event counts in the registry record, and the record does not clarify whether each count represents participants or events in every case. They should not be interpreted as a causal comparison or as evidence that an intervention caused or prevented an event.

The listed serious adverse event terms include Stroke at 0.9%, hospitalization for dizziness, existing cardiac condition at 0.8%, and symptoms of transient ischemic attack (TIA) at 0.8%. The registry does not identify these terms by treatment group in the listed summary, so they cannot be assigned to either group from this information. These percentages are quoted as recorded and should not be combined or treated as group-specific rates.

Safety summaries are descriptive. They do not establish whether an event was caused by treatment, and the registry does not report an event-adjudication method or comparative safety analysis. The counts also need to be read in light of the reported group sizes and the absence of further participant-flow information.

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

The posted results show positive mean change in the SF-36 Energy and Vitality score for the CPAP group and a mean change of 0.0 for the control group at 3 months. Since higher scores indicate improvement on this subscale, the direction of the group summaries favors an increase in the CPAP group’s mean score relative to baseline. The sponsor’s ANCOVA analysis reports p < 0.01 for the group comparison.

These findings do not state how much an individual participant’s score changed, how responses varied across participants, or what proportion of participants experienced improvement. Group means can conceal varied individual experiences. The registry does not provide a threshold for a clinically important score change, so the practical importance of the reported mean change cannot be determined from the record alone.

The inferential result is also incomplete in the registry entry: the adjusted ANCOVA estimate and its confidence interval are not reported. The group-specific intervals quantify uncertainty around each group’s mean change, not around the adjusted contrast. The reported p-value indicates the result of the sponsor’s test under its model, but does not communicate the size or precision of the adjusted difference.

The study addresses the stated question for people with mild OSA and follows the outcome to 3 months. It does not establish what happens after that follow-up, whether results differ between the mild OSA definitions mentioned in the study rationale, or how findings apply to populations not described in the record. Subgroup-specific results are not reported.

The record also does not explain the relationship between total enrollment and the group counts accompanying the endpoint, or report missing-data handling and the analysis population. These are relevant uncertainties when judging how broadly the posted summaries represent trial participants. The results describe the registry-reported endpoint and analysis; they do not by themselves settle broader questions about treatment, long-term outcomes, or individual response.

14. Why This Trial Matters Statistically

This trial is statistically informative because its primary outcome is a patient-reported continuous measure and its stated comparison is randomized. The outcome concerns perceived energy and fatigue, dimensions that are not interchangeable with physiological measures. The registry’s definition of change and its explanation that higher Energy and Vitality scores indicate improvement make the direction of the group means interpretable.

The record also illustrates why arm summaries and a direct treatment comparison answer different questions. Mean changes and their confidence intervals describe each group’s average change and the uncertainty around that average. The ANCOVA p-value reports a test of the comparison between groups. Since the adjusted contrast and its interval are absent, the record offers evidence about the test result but limited information about the magnitude and precision of the adjusted comparison.

Interpretation of a participant-reported outcome warrants attention to masking. The trial is recorded as having no masking, and the outcome reflects participants’ perceptions. This does not establish that the result is biased, but it makes the absence of information about outcome-assessment procedures relevant. The registry also does not report how missing outcome data were handled or explain how the endpoint group counts relate to total enrollment.

The stated aim to include both AASM 2007 and AASM 2012 definitions of mild OSA makes subgroup reporting potentially important for understanding the study’s rationale. However, subgroup-specific outcomes are not reported in the registry record. The overall result cannot answer whether the outcome differed by definition or subgroup.

Finally, the case shows the value of distinguishing a sponsor’s model-based test from what can be reconstructed from posted summaries. The record contains a reported ANCOVA p-value but not the adjusted estimate or model details. A careful reading preserves that distinction rather than treating the arm means as a substitute for the adjusted comparison.

15. 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 reports the sponsor’s ANCOVA p-value but not the adjusted between-group estimate, its confidence interval, or the covariates included in the model. That limits assessment of the size and precision of the primary comparison. The arm-specific means and confidence intervals are useful descriptions, but they are not a substitute for the adjusted contrast.

The study is recorded as unmasked, and the primary endpoint is a participant-reported measure of energy and fatigue. The registry does not describe steps taken to limit possible expectation effects or provide further detail on outcome administration. This is a potential interpretive limitation, not evidence that a particular bias occurred.

The record lists 301 participants enrolled, while the primary outcome summaries give group counts of 115 and 118. It does not explain the relationship between these figures, report participant flow, or state the analysis population. Reasons for missing endpoint information and methods for handling it are not reported in the registry record. No conclusion about attrition or exclusions can be drawn from the counts alone.

The registry describes the study rationale as including mild OSA definitions from AASM 2007 and AASM 2012, but it does not report subgroup-specific primary results. It also does not report a threshold for a clinically important change in the Energy and Vitality score. These omissions restrict interpretation of subgroup patterns and the practical importance of the reported change.

Other details important for statistical appraisal are not reported in the registry record, including a protocol-specific sample-size justification, a detailed statistical analysis plan, randomization implementation, and any interim analyses. The record reports one primary outcome and one analysis but does not describe a broader testing plan or multiplicity handling. These limits concern what can be assessed from the registry entry and should not be read as proof that the procedures were absent.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Assessment of the Effect of PAP on Energy and Vitality in Mild OSA Patients: The Merge Study — NCT02699463.
• Registry record last updated 2021-01-26; status last verified 2021-01.
• Results first posted 2021-01-26.
• 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.

19. Record Summary

NCT02699463, the MERGE Study, was a completed randomized, parallel-group trial sponsored by ResMed. The registry records 301 participants enrolled, no masking, a start date of 2016-12, and primary completion in 2019-05. The study compared nightly CPAP for 3 months with standard care described as sleep hygiene counseling.

The primary endpoint was change from baseline to 3 months in the SF-36 Energy and Vitality score. Mean change was 7.5 (95% confidence interval 6.0 to 9.0) for CPAP and 0.0 (95% confidence interval -1.5 to 1.5) for control. The sponsor-reported ANCOVA comparison had p < 0.01. The adjusted contrast and its confidence interval are not reported in the registry record.

The record reports serious adverse events in both groups and zero deaths in either group. Participant flow, missing-data methods, analysis-population details, and subgroup-specific results are not reported. The results therefore describe the posted endpoint summaries and sponsor test, with important details about precision, analysis, and representativeness unavailable in the registry entry.