CPAP to Treat Cognitive Dysfunction in MS: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for CPAP to Treat Cognitive Dysfunction in MS (NCT02544373), a randomized na study. The primary outcome is Change From Baseline in Performance on the Minimal Assessment of Cognitive Function in MS Battery (MACFIMS).

2. Clinical Question

The trial concerns people with multiple sclerosis (MS) and obstructive sleep apnea (OSA). Its stated objectives are to examine the relationship between OSA and cognitive function in people with MS, and to evaluate whether positive airway pressure (PAP) treatment could improve cognitive dysfunction in participants with both conditions. The randomized comparison focuses on the treatment question: cognitive test-score changes after immediate PAP therapy versus standard care in which PAP therapy is delayed.

The comparison addresses cognitive scores over the registry’s stated three-month interval; it is not a direct report of whether OSA causes cognitive dysfunction. The registry record does not report numerical MACFIMS results, an estimated between-group contrast, or a confidence interval. Consequently, it does not establish the direction or size of any treatment-group difference in cognitive scores.

3. Trial Design

The registry describes a randomized, parallel-group trial with no masking. The listed enrollment is 135, and the study status is completed. The sponsor is the University of Michigan. The registry lists the start date as 2015-11-12 and the primary completion date as 2021-06-25.

Participants were assigned to immediate PAP therapy or standard care PAP therapy. In the immediate group, the arm description says PAP treatment would begin as soon as possible after baseline polysomnography (PSG), followed by repeat baseline cognitive testing three months after PAP initiation. In the standard-care group, PAP treatment would be delayed for three months after the baseline sleep study, and cognitive testing would be repeated before PAP treatment began. The registry describes PAP as standard clinical care for OSA.

The parallel design means that the two groups were followed under their respective treatment schedules rather than being described as crossing over between treatment conditions during the primary comparison. The absence of masking means the registry does not describe participants or study personnel as blinded to group assignment. The record does not report additional details about allocation procedures, concealment, adherence, or the conduct of the sleep studies.

The primary outcome is described as mean change from baseline to month 3 on individual MACFIMS tests, with change calculated as the month 3 score minus baseline score. The primary-outcome population includes participants with MACFIMS testing at both timepoints. The registry lists 135 participants for enrollment, but participant flow and the number contributing to each endpoint are not reported in the registry record. Nor are numerical primary-outcome estimates or analysis results 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: Of 135 enrolled participants,131 received both baseline cognitive testing and polysomnography for Aim 1 (pre-intervention) analyses (outside of the interventional phase of the study. Among these 131 participants, 111 were randomized for this trial.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: These analyses were conducted to examine associations between apnea severity and cognitive performance prior to randomization and PAP intervention. Four out of 135 consented had missing data that did not allow analysis of associations.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Baseline data is provided for all randomized participants.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Performance on the Minimal Assessment of Cognitive Function in MS Battery (MACFIMS), measured over baseline, 3 months.

The primary outcome concerns cognitive performance measured with the Minimal Assessment of Cognitive Function in MS Battery (MACFIMS). The registry describes mean change from baseline to month 3 for individual tests in the battery. Scores are oriented so that higher values indicate better performance; with change defined as month 3 minus baseline, positive changes indicate improvement on that test.

The listed tests cover several cognitive tasks: verbal fluency, visuospatial judgment and memory, verbal learning, auditory processing and addition, and symbol-digit performance. Their score scales differ, so the registry describes results as changes on individual tests rather than as a single common-unit score. The description lists the Controlled Oral Word Association Test, Judgement of Line Orientation, Brief Visuospatial Memory Test Revised total and delayed scores, California Verbal Learning Test-II, two Paced Auditory Serial Addition Test versions, and Symbol Digit Modalities test.

Using a battery is relevant to a study of cognitive function because it assesses performance across more than one task. The test-level results would need to be interpreted in light of each test’s scale and direction. The primary outcome is limited to participants who had testing at both baseline and month 3, according to the registry description. Numerical test-specific means, standard deviations, and comparisons are not reported in the registry record, so the magnitude and precision of cognitive change cannot be assessed from the posted results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a continuous outcome: mean change in scores on individual cognitive tests. A continuous measure can take a range of values rather than being limited to categories such as event or no event. For this outcome, each participant’s change is defined as the month 3 score minus the baseline score, and the registry specifies the mean and standard deviation as the summary measures.

What does a mean change describe?

A mean change summarizes the average direction and size of score changes among participants included in the analysis. Because the tests use different scales, a change on one test should be interpreted using that test’s own scoring range and context, rather than treated as directly interchangeable with a change on another test. The registry states that higher scores mean better performance, so a positive change denotes a higher month 3 score than baseline for the test in question. This definition describes score direction; it does not imply that a group-level change is statistically reliable or that the groups differ.

What does the standard deviation add?

The standard deviation describes how dispersed individual scores or changes are around the group mean. A mean gives a central summary, while the standard deviation gives information about variability among participants. The standard deviation is not the same as the uncertainty in the mean or in a between-group contrast. Estimating precision also depends on the amount of information available and the analysis used. The record specifies standard deviation as the dispersion measure but does not report numerical values for the primary outcome.

Why compare randomized groups?

Random assignment is intended to create groups that can be compared without choosing treatment based on participant characteristics. In this study, the randomized contrast is immediate PAP therapy versus delayed PAP therapy under standard care. The groups’ score changes would be compared over the registry’s baseline-to-month-3 assessment interval. A direct between-group estimate is more informative for the treatment question than separately describing whether scores rose or fell in either group. The registry record does not report such an estimate.

How should uncertainty be read?

In general, a confidence interval conveys uncertainty around an estimated effect: narrower intervals indicate greater precision, while wider intervals indicate less precision. A confidence interval also helps readers evaluate which effect sizes remain compatible with the data and analysis assumptions. A p-value, in general, measures how surprising data at least as incompatible with a specified null hypothesis would be if that hypothesis and the analysis assumptions held. Neither measure describes clinical importance on its own. No primary-outcome confidence interval or p-value is reported in the registry record, so these concepts cannot be applied to a numerical trial estimate.

What is the role of planning illustrations?

The accompanying planning illustrations are described as using an observed effect, a two-sided significance level of 0.05, equal allocation between groups, and a normal approximation, with the target effect defined as a stated fraction of the observed effect. Such calculations are planning illustrations, not observed power and not evidence that the trial was designed around those assumptions. Sample sizes derived this way describe the participants per arm for the specified power under the calculation assumptions. The registry record does not report the trial’s own sample-size justification, and the registry entries do not give numerical illustration results to interpret.

Why can variability affect interpretation?

For continuous outcomes, greater variation among participants can make it harder to estimate a group mean or contrast precisely. In other settings, count outcomes can be over-dispersed: their variation exceeds what a simple model assumes, which can make standard errors and confidence intervals misleading if the extra variation is ignored. That general issue is not a reported feature of this cognitive endpoint; the registry describes means and standard deviations for score changes. No modeling details for the primary outcome are reported in the registry record.

These principles describe how the stated summaries and comparisons are usually interpreted. They do not fill in the missing estimates, analysis procedures, or uncertainty measures for this trial. Those details are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The primary outcome is described as change on individual tests in the MACFIMS battery, rather than a single score with one shared scale. The listed battery includes multiple cognitive tests and subtests. The registry also lists six outcome measures overall. It does not report a multiplicity adjustment, a hierarchy among tests, a designated single test for formal inference, or an explanation of how conclusions across the measures were to be controlled. These details are not reported in the registry record.

Multiplicity is a general statistical concern when several outcomes or comparisons are examined: the more questions considered, the greater the chance that at least one result may appear unusual by chance alone. Analysts may address this with prespecified outcome hierarchies or statistical adjustments, among other approaches. The record does not describe whether any such approach was planned or used. Without that information and without the numerical results, the strength of evidence across individual cognitive tests cannot be assessed.

The primary-outcome population is described as participants with cognitive testing at baseline and month 3. This is a complete-case requirement for those timepoints: a participant needs both measurements to contribute to the described change score. The record does not report how many participants had both assessments, how many lacked either assessment, why assessments were missing, or whether any missing-data method was used. It also does not report whether additional analysis populations were defined.

Those omissions limit interpretation because follow-up availability can affect which participants contribute to the endpoint. In general, an analysis restricted to participants with complete measurements may differ from an analysis that accounts for missing observations, depending on why data are missing and how they are handled. The registry record does not provide enough information to determine whether missingness altered the analysis or its interpretation.

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 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports adverse events separately for the immediate PAP therapy group and the standard-care PAP therapy group. In the immediate group, 71 participants are listed as at risk for serious adverse events and 0 as affected; 71 are listed as at risk for other adverse events and 11 as affected. In the standard-care group, 40 are listed as at risk for serious adverse events and 0 as affected; 40 are listed as at risk for other adverse events and 0 as affected. No deaths are listed in either group.

These figures are counts of affected participants relative to the listed at-risk group, not estimates of a treatment effect. The registry provides no serious adverse event terms in its summary. It does not report the nature, timing, severity, or attribution of the other adverse events, nor does it provide a statistical comparison of event counts. The record also does not explain whether the groups had different observation time or how adverse events were collected. As a result, the counts should be read descriptively and do not by themselves establish why the events occurred or whether the groups differed in safety.

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

The record describes a randomized trial of immediate versus delayed PAP therapy in people with MS and OSA, with cognitive testing planned at baseline and month 3. Its primary outcome is mean change on individual MACFIMS tests, with positive change defined as a higher score at month 3 than at baseline. The record is marked completed and lists an enrollment of 135.

What the record does not establish is the size or direction of the between-group cognitive difference. Numerical primary-outcome summaries and analysis estimates are not reported in the registry record. Without group changes, a direct comparison, and a measure of uncertainty, it is not possible to determine whether the observed cognitive data favored one group, how large any difference was, or how precise an estimate would be.

The distinction between a study question and a result is important. The trial’s stated objective is to evaluate whether OSA treatment with PAP could improve cognitive dysfunction in participants with MS and OSA. That objective is not itself a finding. Similarly, describing positive score changes as improvement defines the scoring direction, but it does not show that participants improved or that changes differed by randomized group.

The record also does not provide enough information to judge the completeness of follow-up, the handling of missing cognitive assessments, the statistical approach to multiple tests, or the analysis population beyond the stated requirement for testing at both timepoints. Those are relevant to understanding how any estimate was obtained and how broadly it applies.

The adverse-event summary reports no serious adverse events or deaths in either group, along with other adverse-event counts. These figures are descriptive and lack details about event types, timing, and attribution. They do not, by themselves, support a comprehensive comparative safety assessment.

Accordingly, the registry record supports a description of the randomized design, endpoint, and limited safety counts. It does not support a conclusion about cognitive efficacy or a definitive comparison of safety. Any such conclusion would require results and analysis details that are not reported in the registry record.

13. Why This Trial Matters Statistically

This trial is statistically notable because its question links a sleep-related condition with cognitive function in people with MS and tests a treatment schedule through random assignment. The design distinguishes immediate PAP therapy from a standard-care group whose PAP treatment is delayed for three months. That creates a defined comparison for the cognitive assessment over the stated interval.

The outcome also illustrates a common measurement challenge in clinical research. A cognitive battery includes different tasks, and the registry describes changes on individual measures rather than one shared score. Separate test results can preserve useful detail about distinct aspects of performance, but they also require careful interpretation across different scales. If many outcomes or comparisons are evaluated, multiplicity and prespecification become important; the record does not report how these issues were addressed.

The stated analysis population is based on having cognitive assessments at both baseline and month 3. This makes follow-up completeness a key interpretive issue. The registry does not report the number of participants with paired measurements, the amount or pattern of missing data, or the method used to handle missing assessments. A clear account of these points would help readers understand who contributed to the outcome estimates.

The record’s limitations are especially important because it provides the study design and endpoint description but no numerical primary-outcome results or analysis entries. Statistical interpretation depends on estimates and their uncertainty, not on the research question alone. The listed safety counts add descriptive information, but lack event-level detail and a comparative analysis.

As a result, the trial record offers a useful example of why registered design features and posted results should be distinguished. Randomization and a clearly defined score-change direction explain what comparison was intended. They cannot substitute for group estimates, uncertainty measures, missing-data information, and transparent analysis methods when judging what the trial found.

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 most important limitation for interpretation is that numerical results for the primary cognitive outcome are not reported in the registry record. There are no test-specific group summaries, between-group estimates, confidence intervals, or p-values. The trial status is completed, but completion status does not supply an outcome estimate.

The endpoint spans individual tests with different scales. The record describes the tests and the meaning of positive change, but does not report how results across the battery were prioritized or whether multiplicity was addressed. It also does not report a single composite score or identify one test as the sole focus of inference.

The primary-outcome population requires MACFIMS assessments at both baseline and month 3. The number of participants meeting that requirement is not reported in the registry record. Participant flow, reasons for missing assessments, missing-data methods, and the analysis population used for any group comparison are also not reported. The enrollment figure alone cannot establish how many participants contributed to the primary endpoint.

The record describes the trial as randomized, parallel, and unmasked, but does not report allocation-concealment procedures, adherence to PAP therapy, or details of how the assigned treatment schedules were implemented. These details are not reported in the registry record and should not be inferred from the brief arm descriptions.

The safety summary gives affected-participant counts but not event terms or further event-level details. The record does not report comparative statistical analyses of adverse events or enough information to assess differences in observation time. Finally, the registry does not report the trial’s own sample-size justification or the statistical plan for the primary comparison. These gaps constrain assessment of precision, robustness, and generalizability.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: CPAP to Treat Cognitive Dysfunction in MS — NCT02544373.
• Registry record last updated 2022-09-14; status last verified 2022-08.
• Results first posted 2022-09-14.
• 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

NCT02544373 is a completed, randomized, parallel-group trial with no masking, sponsored by the University of Michigan. The registry lists enrollment of 135, a start date of 2015-11-12, and a primary completion date of 2021-06-25. The study concerns PAP therapy for OSA in people with MS.

The comparison is immediate PAP therapy versus standard care PAP therapy delayed for three months. The primary outcome is mean change from baseline to month 3 on individual MACFIMS tests, among participants tested at both timepoints. The registry record does not report numerical primary-outcome results or analysis estimates. Its safety summary lists no serious adverse events or deaths in either group, and reports other adverse events for 11 of 71 participants in the immediate group and 0 of 40 in the standard-care group.