A Cross-Over Study to Evaluate Lung Function Response After Treatment With Umeclidinium (UMEC) 62.5 Micrograms (mcg), Vilanterol (VI) 25 mcg, and Umeclidinium/Vilanterol (UMEC/VI) 62.5/25 mcg Once-Daily in Subjects With Chronic Obstructive Pulmonary Disease (COPD): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Cross-Over Study to Evaluate Lung Function Response After Treatment With Umeclidinium (UMEC) 62.5 Micrograms (mcg), Vilanterol (VI) 25 mcg, and Umeclidinium/Vilanterol (UMEC/VI) 62.5/25 mcg Once-Daily in Subjects With Chronic Obstructive Pulmonary Disease (COPD) (NCT02014480), a randomized phase 3 study. The primary outcome is Change From Baseline (BL) in Weighted Mean (WM) 0-6 Hour Forced Expiratory Volume in One Second (FEV1) Obtained Post-dose at Day 14 of Each Treatment Period (TP) by Response Type.

2. Clinical Question

FEV1 is forced expiratory volume in one second, a measure of the amount of air that can be forcefully exhaled in one second. The primary endpoint combines measurements over the six-hour post-dose interval into a weighted mean, then assesses change from the baseline for the relevant treatment period. The study also listed secondary lung-function outcomes, including responsiveness according to FEV1 at Day 1, the number of participants with a larger change in weighted mean FEV1 with the combination than with either component at Day 14, and change in pre-dose trough FEV1 at Day 15. Their results are not reported in the registry record. Thus the research question is clear, but the posted information does not establish the magnitude, precision, or statistical significance of any treatment contrast.

3. Trial Design

This was a Phase 3, randomized, quadruple-masked, 3-way crossover study in participants with COPD. The registry lists an enrollment of 207 and describes the study as multicenter. The three treatments were UMEC 62.5 mcg, VI 25 mcg, and UMEC/VI 62.5/25 mcg, administered once daily in the morning for 14 days per treatment period. Participants were assigned to a treatment sequence intended to ensure that each participant received all three treatments.

The study included a run-in phase of 5 to 7 days, three treatment periods separated by washout periods of 10 to 14 days, and a follow-up phase of 7 to 9 days after the final visit or early discontinuation. The registry describes the overall study duration as approximately 12 weeks. Spirometry was scheduled on Day 1 and Day 14 of each treatment period, with trough spirometry on Day 2 and Day 15. Vital signs were assessed on Day 1 and Day 14, and adverse events were recorded throughout the study.

A crossover design differs from a parallel-group design because participants receive more than one treatment over time. General statistical context: comparing treatments within the same participant can account for stable differences between participants, but interpretation also depends on the sequence of treatments, the time period, and whether effects carry over after a treatment ends. The washout periods are part of the recorded design; the registry record does not report whether carryover was assessed or how it was handled in the analysis.

The registry lists the study as completed, with a start date of 2013-02-01 and a primary completion date of 2013-06-01. Detailed sequence assignments, the statistical model, and the rationale for the enrollment target 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: Participants who met the eligibility criteria at Screening (Visit 1) completed a 5- to 7-day Run-in Period prior to being randomized to 1 of 6 treatment sequences. The treatment phase was comprised of three 14-day treatment periods, each separated by a 10- to 14-day washout period.

Treatment Period 1 (14 Days)

Reasons for not completing this period:

Washout Period 1 (10 to 14 Days)

Reasons for not completing this period:

Treatment Period 2 (14 Days)

Reasons for not completing this period:

Washout Period 2 (10 to 14 Days)

Analysis population for the primary outcome, as defined in the registry: Intent-to-Treat (ITT) Population: all participants (par.) randomized to treatment who received \>=1 dose of randomized study medication in a TP. Only par. available at the specified time points were analyzed. Different par. may have been analyzed for different parameters; the overall number of par, analyzed reflects everyone in the ITT Population.

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 (BL) in Weighted Mean (WM) 0-6 Hour Forced Expiratory Volume in One Second (FEV1) Obtained Post-dose at Day 14 of Each Treatment Period (TP) by Response Type, measured over Baseline and Day 14 of each treatment period (up to study day 85).

The primary endpoint is change from baseline in weighted mean FEV1 over the 0-6 hour period after dosing on Day 14 of each treatment period. FEV1 measures the maximal amount of air forcefully exhaled in one second. The registry reports this outcome in liters and describes the treatment estimates as least-squares means with standard errors.

The weighted mean is intended to represent lung function across a period after dosing rather than at just one measurement time. As described in the record, it is derived from the area under the FEV1-versus-time curve, calculated using the trapezoidal rule and divided by the time interval. Measurements include the pre-dose value and post-dose assessments at 15 minutes, 30 minutes, 1 hour, 3 hours, and 6 hours. The baseline for each period is based on pre-dose measurements on Day 1, and the change is the Day 14 value minus that period's baseline.

General statistical context: repeated measurements can capture the pattern of response over time, while a weighted mean provides a single summary for the specified interval. This endpoint may therefore describe the average lung-function response across the measured post-dose window rather than the response at one isolated time. The record identifies the endpoint as being reported by response type, but does not provide the response-type definitions or the treatment-specific estimates. Consequently, the endpoint's measurement approach is described, but its observed results are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The outcome is continuous: change in a lung-function measurement expressed in liters. The registry states that the primary endpoint is summarized using least-squares means and standard errors. Those terms indicate a model-based summary and its estimated sampling uncertainty, but the detailed model specification is not reported in the registry record. The available description should therefore be distinguished from a complete statistical analysis plan.

What does a weighted mean represent?

The endpoint combines FEV1 measurements taken at multiple points in the 0-6 hour post-dose interval. The registry says the weighted mean is calculated by finding the area under the FEV1/time curve using the trapezoidal rule and dividing by the time interval. In general, the trapezoidal rule approximates the area under a curve by joining adjacent observations with straight lines and summing the resulting areas. Dividing by the interval yields an average over that interval. The method gives a summary that reflects the scheduled measurements across the interval, rather than relying on a single post-dose reading.

The listed measurements include pre-dose, 15-minute, 30-minute, 1-hour, 3-hour, and 6-hour readings. The pre-dose value for Day 14 is described as the mean of the 23- and 24-hour assessments after Day 13 dosing. Baseline is the mean of the measurements taken 30 minutes and 5 minutes before dosing on Day 1 of each treatment period. The record defines change as the Day 14 value minus the baseline for that period. These definitions make the endpoint period-specific: each treatment period has its own baseline and post-dose summary.

What is a least-squares mean?

General explanation: a least-squares mean is an estimated group or treatment mean from a statistical model. It can account for terms included in that model, such as treatment or period, rather than simply averaging all observed values. Its interpretation depends on the model's structure and assumptions. The registry record does not report which model terms, covariance structure, or covariates were used, so it is not possible to explain the posted least-squares means beyond their stated role as the endpoint summary.

The standard error quantifies the estimated uncertainty of a mean estimate under the analysis model. It is not the same as the spread of individual participants' measurements. General statistical context: a confidence interval combines an estimate and its uncertainty to show a range of values compatible with the data under the stated method. Primary confidence intervals are not reported in the registry record, and no interval should be reconstructed without the relevant estimates and analysis details.

Why does the crossover structure matter?

Each participant was intended to receive all three treatments in a randomized sequence, with treatment periods of 14 days and washout periods of 10 to 14 days between periods. General explanation: within-participant comparisons can remove some variability associated with stable participant characteristics, because each person contributes observations under multiple treatments. At the same time, responses can vary by period, and a prior treatment may affect observations in a later period. Randomized sequence and washout are design features relevant to those concerns, but the registry record does not report tests or model adjustments for period, sequence, or carryover effects.

The registry identifies 207 participants enrolled, but the primary population description says that only participants available at the specified time points were analyzed. It also notes that the number of participants could differ across parameters. This means enrollment alone does not establish the number contributing to each primary estimate. The treatment-specific analysis counts and reasons for unavailable measurements are not reported in the registry record.

How should multiple analyses be considered?

The record lists 6 analyses and describes the primary outcome as having several categories or time points. General explanation: when several outcomes, time points, or treatment contrasts are tested, the chance of at least one apparently unusual result can increase if each is assessed separately. Statistical methods may address this multiplicity through prespecified testing sequences or adjustments, but the chosen approach must be known to interpret p-values as a family of tests. The registry record does not report a multiplicity procedure, a hierarchy among comparisons, or the detailed analysis plan.

Over-dispersion is relevant when observed variability exceeds what a particular model assumes. In general, accounting for it can affect standard errors and the resulting uncertainty. The endpoint is a continuous measure and the registry lists a standard error, but it does not report whether over-dispersion was evaluated or whether it was relevant to the selected model. There is not enough information to characterize that issue for this analysis.

Finally, the registry includes a sample-size planning description based on the observed effect, a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Such planning illustrations describe hypothetical participants per arm for a specified power and effect fraction; they are not observed power. The corresponding numerical planning outputs are not reported in the registry record. The registry also does not report the trial's own sample-size justification, so the planning description should not be treated as evidence of the original design rationale.

10. Multiplicity, Missing Data and Analysis Populations

The primary endpoint is described as being analyzed by response type, and the registry indicates that it has several categories or time points. It also lists 6 analyses. The response categories, specific contrasts, and testing sequence are not reported in the registry record. A multiplicity adjustment or other approach to controlling error across comparisons is not reported either. General statistical context: interpreting several tests together differs from interpreting a single prespecified test, because repeated testing can increase the chance of an apparently positive finding. Without the analysis plan and results, the role of multiplicity in this study cannot be assessed.

The stated analysis population is the Intent-to-Treat population, defined in the record as participants randomized to treatment who received at least one dose of study medication in a treatment period. The record further says that only participants available at the specified time points were analyzed and that different participants may have been analyzed for different parameters. It does not report treatment-specific denominators for the primary endpoint, the amount or pattern of missing data, reasons for missingness, or any imputation method. It is therefore not possible to determine how missing observations affected the estimates.

The registry record does not report whether the analysis followed a prespecified statistical plan, how the crossover structure was represented in the model, or how period, sequence, or carryover were handled. It also does not report an approach to sensitivity analyses for missing data. These omissions limit assessment of whether the reported analysis, if available elsewhere, would be robust to alternative assumptions. The registry's description of an Intent-to-Treat population should not be interpreted as meaning that every randomized participant contributed data to every endpoint; the record explicitly notes availability at the specified time points and potentially different participant counts by parameter.

11. 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 3% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: squamous cell carcinoma (0.5%), gastroenteritis (0.5%), orchitis (0.5%).

The registry lists serious adverse events separately for the three treatment groups. There was 1 participant with a serious adverse event among 202 at risk for UMEC 62.5 µg, 1 among 200 at risk for VI 25 µg, and 1 among 202 at risk for UMEC/VI 62.5/25 µg. The record lists 5, 10, and 10 participants with other adverse events in those respective groups, using the same at-risk counts. Death counts are not reported in the registry record.

These counts describe participants with events, not the total number of event episodes, and the registry does not provide event timing or participant-level detail in the information recorded. The serious-event terms listed are Squamous cell carcinoma, Gastroenteritis, and Orchitis, each reported as 0.5%. The record does not identify a treatment group for each listed term in this summary, so the terms should not be assigned to a particular treatment.

General statistical context: adverse-event counts are descriptive unless accompanied by a defined comparative analysis. In a crossover study, participants may receive several treatments, so group-level at-risk counts should be interpreted according to the registry's reporting structure rather than assumed to represent independent parallel groups. The record does not report a formal statistical comparison of serious or other adverse events. These figures alone do not establish whether event rates differ between treatments.

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

The registry record gives a clear outline of a randomized, double-blind crossover study comparing UMEC, VI, and their combination, along with a detailed definition of the primary lung-function endpoint. It also reports selected adverse-event counts. These registry entries help explain what the investigators intended to measure and how the primary measurement was constructed.

The record does not report the primary endpoint estimates or treatment comparisons. It therefore does not show the direction or magnitude of the differences in weighted mean FEV1, the precision of those differences, or whether any comparison met a statistical threshold. The listed secondary outcomes are also described without their results. No conclusion about comparative lung-function response can be drawn from the posted information alone.

General statistical context: a treatment estimate is most informative when considered alongside its uncertainty and the analysis choices that produced it. For a crossover study, interpretation also depends on how within-participant observations and treatment periods are modeled, whether missing measurements are handled appropriately, and how multiple comparisons are addressed. Those details are not reported in the registry record.

The adverse-event counts show that serious events were recorded in each treatment group, and that other adverse events were also recorded. The counts are descriptive, and no formal comparative safety analysis is reported. They do not by themselves establish that one treatment has more or fewer adverse events than another, nor do they support a comprehensive assessment of safety.

The distinction between absence of a reported result and evidence of no difference is important. Because the primary estimates and their uncertainty are not reported in the registry record, the absence of a numerical comparison cannot be interpreted as showing equal responses or no effect. The evidence available supports description of the trial's design and measurements, not a conclusion about which regimen had the greater response or about the clinical importance of any difference.

13. Why This Trial Matters Statistically

This study illustrates how a crossover design can be used to compare several treatments in the same participants. Since each participant was intended to receive UMEC, VI, and UMEC/VI, the design creates the possibility of within-participant comparisons. General statistical context: this can be useful when stable differences between participants contribute substantially to outcome variability, although interpretation still depends on period effects, carryover, and completeness of measurements.

The endpoint also illustrates a common measurement choice in clinical research: summarizing repeated observations over a defined time interval. The weighted mean FEV1 uses multiple scheduled measurements, rather than treating one time point as the whole response. The registry's description of the trapezoidal calculation makes the construction of that summary interpretable, even though the treatment estimates themselves are not reported.

The record also demonstrates why a detailed outcome definition is not a substitute for outcome results. It names the endpoint, measurement schedule, unit, analysis population, and summary statistic, but does not report the estimates, uncertainty intervals, or comparison results. Without these, readers cannot judge the magnitude or precision of the treatment contrasts. Similarly, a list of analyses without their contrasts and multiplicity approach does not show how the family of statistical tests was interpreted.

Finally, the different analysis availability across parameters matters in a crossover study with repeated assessments. Enrollment is not necessarily the denominator for every endpoint, and the registry notes that only participants available at specified time points were analyzed. Reporting the endpoint-specific counts and missing-data approach would help readers assess representativeness and uncertainty. Thus this trial is statistically informative as an example of endpoint construction and crossover structure, while the registry's missing outcome results limit interpretation of comparative effects.

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 principal limitation for interpreting efficacy is that the registry record does not report the primary endpoint estimates, treatment contrasts, confidence intervals, or p-values. The primary outcome is described by response type, but response definitions and category-specific results are not reported. The listed secondary outcomes likewise have no results in the record. As a result, the magnitude, direction, and precision of the lung-function comparisons cannot be assessed.

The analysis methods are incompletely described. Although the record identifies least-squares means and standard errors, it does not report the model specification, covariates, treatment-contrast definitions, or handling of treatment period, sequence, and possible carryover. The washout periods are described, but whether carryover was assessed is not reported in the registry record.

Missing-data information is also limited. The record says only participants available at the specified time points were analyzed and that participant counts could differ by parameter. Endpoint-specific denominators, missingness reasons, and imputation or sensitivity methods are not reported. The statistical approach to multiplicity across the 6 analyses and multiple response categories or time points is not reported either.

Safety information is limited to participant counts for serious and other adverse events, a short list of serious-event terms, and at-risk totals. Death counts and a formal comparative safety analysis are not reported in the registry record. These limitations mean that the available information describes important design and measurement features but does not allow a complete assessment of comparative efficacy or safety.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Cross-Over Study to Evaluate Lung Function Response After Treatment With Umeclidinium (UMEC) 62.5 Micrograms (mcg), Vilanterol (VI) 25 mcg, and Umeclidinium/Vilanterol (UMEC/VI) 62.5/25 mcg Once-Daily in Subjects With Chronic Obstructive Pulmonary Disease (COPD) — NCT02014480.
• Registry record last updated 2018-02-15; status last verified 2018-01.
• Results first posted 2014-02-11.
• 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

NCT02014480 is a completed Phase 3 randomized, quadruple-masked, 3-way crossover study of UMEC 62.5 mcg, VI 25 mcg, and UMEC/VI 62.5/25 mcg in participants with COPD. The registry reports enrollment of 207. Each participant was intended to receive all three once-daily treatments for 14 days per period, with washout periods of 10 to 14 days.

The primary outcome was change from baseline in weighted mean FEV1 over 0-6 hours after dosing on Day 14, expressed in liters and summarized as a least-squares mean with standard error. The record explains the endpoint calculation but does not report the primary treatment results, confidence intervals, or p-values. It lists selected adverse-event counts and serious-event terms, but no formal comparative safety analysis. Important details, including the full statistical model, multiplicity approach, and missing-data methods, are not reported in the registry record.