Inspiratory Flow Rates Achieved by the COPD Patients Through Breezhaler®, Ellipta® and Handihaler® Inhaler Devices: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Inspiratory Flow Rates Achieved by the COPD Patients Through Breezhaler®, Ellipta® and Handihaler® Inhaler Devices (NCT02596009), a randomized phase 4 study. The primary outcome is Peak Inspiratory Flows Rates Summary by Inhalation Devices - PPS.

2. Clinical Question

The study addresses a measurement question about flow achieved while using three devices, not a comparison of drug treatment. The endpoint was the mean peak inspiratory flow rate, reported in L/min with a standard deviation. Because the same patients were required to use all three devices, the design allows measurements from the devices to be considered within a crossover framework. The registry record reports paired t-tests for the sponsor's comparisons of Breezhaler® with each of the other devices.

The clinical question is limited to the recorded flow measurements in the per-protocol population and at the stated visit. The registry record does not report a clinical outcome linked to these flow rates, a threshold for judging a measurement, or a drug-treatment effect. The endpoint therefore describes device-specific inspiratory flow in this study rather than establishing a broader clinical consequence.

3. Trial Design

This was a Phase 4, multicenter, randomized, open-label crossover study of three dry powder inhaler devices: Breezhaler®, Ellipta® and Handihaler®. The registry record lists the status as completed, enrollment of 97, a start date of 2015-12-16 and primary completion on 2016-04-29. Masking is listed as none. No active drug or placebo was administered.

In a crossover design, participants receive or experience more than one study condition, so measurements can be compared within the same participant. In this study, each patient was required to generate inhalation flow profiles through all three devices in a randomized crossover sequence, with measurements taken in each device at the same visit. The listed device sequences were BEH, EHB, HBE, BHE, EBH and HEB, where B denotes Breezhaler®, E denotes Ellipta® and H denotes Handihaler®.

The primary outcome population was the per-protocol set (PPS), defined in the record as randomized patients who completed inhalation-flow maneuvers through all three devices. Posted device-specific means are based on 93 observations for Breezhaler®, 88 for Ellipta® and 93 for Handihaler®. These counts describe the reported endpoint groups; they should not be read as separate randomized arms in a parallel-group trial, because the study used crossover measurements.

The primary endpoint was measured at Visit 2 (Day 1). The record does not report the detailed sequence-generation procedure beyond the listed sequences, the planned handling of period or sequence effects, or the original statistical design assumptions. It also does not report an explanation for the difference between the enrollment figure and endpoint group counts. These details 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: No active drug or placebo was administered in this study. The patients were to continue on their respective current maintenance COPD treatment as deemed appropriate by the treating physician. Each patient was required to inhale through all three study devices at the same visit in a randomized cross-over sequence.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Per-protocol set (PPS): This PPS population consisted of all randomized patients, who completed inhalation flow maneuvers through all three inhaler devices.

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 Peak Inspiratory Flows Rates Summary by Inhalation Devices - PPS, measured over Visit 2 (Day 1).

The primary endpoint is the mean peak inspiratory flow rate measured while patients with COPD inhaled through each of the three dry powder inhaler devices. It is expressed in L/min, and the accompanying measure of spread is the standard deviation. Measurements were collected at Visit 2 (Day 1) from inhalation flow profiles generated through Breezhaler®, Ellipta® and Handihaler®, without drug or placebo administration.

In general, a peak flow measure captures the highest flow rate recorded during an inhalation maneuver. Comparing the means describes how the average recorded peak differed across devices in the study population. The standard deviation describes how much individual measurements varied around each device-specific mean; it is not the uncertainty interval for the mean itself.

This measure is useful for the study's stated purpose because the research question concerns the inspiratory flow rates achieved through the devices. It provides a continuous, quantitative comparison rather than a count of clinical events. However, the registry record does not report a clinically meaningful threshold, a target flow value, or a link between the measured flow and a clinical outcome. The endpoint therefore supports interpretation of the observed device-specific flow measurements, but the record alone does not establish what any particular value means for an individual patient.

7. Results Posted in the Registry

• Ellipta®: mean 80.73 versus 107.52 on control; difference -26.79 (95% CI -32.90 to -20.68), standardized effect d = -1.27; z-test p = <0.001.
• Handihaler®: mean 53.59 versus 107.52 on control; difference -53.93 (95% CI -60.26 to -47.60), standardized effect d = -2.45; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean describe?

The endpoint is continuous: it records a flow rate in L/min rather than a category or event count. A mean is the arithmetic average of the measurements included in a group. The posted means describe average peak inspiratory flow rates for each device in the per-protocol set. The standard deviation gives a measure of the spread of individual values around that mean. It is not a confidence interval and does not, by itself, express how precisely the mean has been estimated.

In general, a standard deviation can be useful for understanding variability among measurements, while a confidence interval addresses uncertainty in an estimated quantity. Those two summaries answer different questions. The posted standard deviations are 23.34 for Breezhaler®, 18.44 for Ellipta® and 20.63 for Handihaler®, all in L/min.

Why does the crossover design matter?

Each patient was required to perform inhalation maneuvers using all three devices, in a randomized crossover sequence. Thus, the device measurements are not simply from three unrelated groups of participants. In a paired comparison, the analysis first considers the difference between two measurements from the same participant, then summarizes those within-person differences. This approach can account for the fact that one participant's measurements may tend to be higher or lower across devices than another participant's measurements.

The sponsor's reported comparisons use paired t-tests. In general, a paired t-test evaluates whether the mean of the within-participant differences is consistent with zero, under the assumptions of that test. It uses information about the pairing, which cannot be recovered from device-specific means and standard deviations alone. The registry record lists paired comparisons of Breezhaler® with Ellipta® and of Breezhaler® with Handihaler®.

What are the recalculated comparisons?

The separately posted recalculations are simple unadjusted estimates derived from the device-level summary statistics. They compare the mean for each comparator with the Breezhaler® mean and use pooled standard deviations for the comparison. They do not reproduce the participant-level paired differences. For Ellipta®, the recalculated mean difference is -26.79 L/min, with a 95% confidence interval from -32.9 to -20.68. For Handihaler®, it is -53.93 L/min, with a 95% confidence interval from -60.26 to -47.6. The negative sign reflects the recalculation's comparator-minus-Breezhaler® direction.

These estimates are useful as a transparent, summary-statistic comparison, but the crossover design means they omit the within-person covariance used by the paired analysis. That omission can affect estimated uncertainty. In general, when measurements are paired, treating them as independent can misrepresent precision because the relationship between measurements from the same person is not included. The recalculations should therefore not be treated as equivalent to the sponsor's paired t-tests.

How should the p-values be read?

The sponsor reported p < 0.001 for each paired comparison. In general, a p-value is calculated under a specified statistical model and describes how unusual data at least as incompatible with the null hypothesis would be if that hypothesis and the model assumptions held. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of an effect.

The recalculated comparisons also list p-values of <0.001. These belong to the unadjusted summary-statistic calculations, not to the sponsor's paired analyses. Because the methods differ, their p-values should not be read as alternate versions of the same paired test. The registry record does not report a multiplicity adjustment for the two comparisons, so the record does not establish whether or how the sponsor addressed multiple testing.

What do confidence intervals add?

A confidence interval gives a range of estimates compatible with the data under a particular method and its assumptions. The sponsor's intervals are reported as 95% confidence intervals for paired mean differences. The recalculated intervals likewise describe the summary-statistic comparisons, but use a different calculation. In general, an interval's width reflects both the variability in the data and the amount of information available; a narrower interval indicates greater statistical precision for the quantity and method being summarized.

It is important to keep the comparison direction in mind. The sponsor reports Breezhaler® minus the comparator, giving positive mean differences of 26.97 L/min for Ellipta® and 53.93 L/min for Handihaler®. The recalculations report comparator minus Breezhaler®, yielding negative differences. Reversing the order of a subtraction reverses the sign and interval endpoints, but does not change the underlying contrast. In this record, the sponsor's own paired analysis is the directly reported analysis for the crossover comparisons.

What assumptions are not documented?

The registry record does not report the original sample-size justification, planned analysis of period or sequence effects, a detailed missing-data method, or an adjustment strategy for multiplicity. Those omissions limit assessment of the full analysis plan. General statistical interpretation should not fill those gaps by assuming a protocol method that is not recorded. The reported endpoint summaries and paired tests can be described, while unreported design and analysis details remain uncertain.

10. Confidence Intervals and Precision

The sponsor's paired comparison of Breezhaler® with Ellipta® estimated a mean difference of 26.97 L/min, with a 95% confidence interval from 23.70 to 30.24. The paired comparison of Breezhaler® with Handihaler® estimated a mean difference of 53.93 L/min, with a 95% confidence interval from 49.15 to 58.72. Both sponsor-reported intervals are presented in the direction Breezhaler® minus the comparator.

In general, a 95% confidence interval is a method-based range of values for the effect that is compatible with the observed data and the analysis assumptions. It reflects uncertainty in the estimated mean difference; it is not a range containing 95% of individual patients' flow measurements, and it does not imply that there is a 95% probability that a fixed effect lies in the particular interval. The standard deviations reported for each device describe variability among measurements, while the confidence intervals describe uncertainty in the estimated comparison.

The unadjusted recalculations give different intervals because they are based on device-level summary statistics rather than paired participant-level differences. Their direction is comparator minus Breezhaler®: -26.79 L/min, with an interval from -32.9 to -20.68 for Ellipta®, and -53.93 L/min, with an interval from -60.26 to -47.6 for Handihaler®. The sign difference reflects the order of subtraction. Because the study is a crossover trial, the paired estimates are the more directly relevant reported comparisons; the recalculated intervals do not incorporate the within-person covariance.

Precision is not the same as clinical importance. The record does not report a threshold for a meaningful flow-rate difference or a clinical outcome to interpret these intervals against. The intervals quantify statistical uncertainty for the reported comparisons, not the implications of a particular flow rate for an individual patient.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are expressed as participants per arm for 80% power and participants per arm for 90% power. They are illustrations based on two-arm comparisons, not counts from the crossover study. In general, 80% power means that a study using the stated assumptions would have an 80% chance of detecting the specified effect if that effect were the true one; 90% power describes a higher chance under the same type of assumptions. Power is a design property under a model, not a measure of whether the observed result is reliable.

The illustrations use a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect set to the stated fraction of the observed effect. For each device comparison, the planning rows consider 100%, 75% and 50% of the observed effect. A smaller effect is harder to detect and therefore requires more participants under the same assumptions. These figures are not observed power.

For the Ellipta® comparison, the posted planning values range from 10 participants per arm for 80% power and 14 participants per arm for 90% power when planning for 100% of the observed effect, to 39 participants per arm for 80% power and 53 participants per arm for 90% power when planning for 50% of the observed effect. For the Handihaler® comparison, the corresponding values range from 3 and 4 participants per arm at 100% of the observed effect to 11 and 15 participants per arm at 50% of the observed effect.

These calculations are planning illustrations based on the observed comparisons, not the trial's original design justification. The original sample-size assumptions are not reported in the registry record. The figures also use a two-arm, equal-allocation framing and should not be read as a reconstruction of the randomized crossover study's design or its enrollment.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -26.790; pooled standard deviation σ = 21.098.
2. n = 2 × (1.96 + 0.842)² × 21.098² / -26.790² = 9.7, rounded up to 10 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -26.79                      # observed difference in means
sigma <- 21.0981965789183                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(2596009)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 10 (80% power), 14 (90% power)
 75% of observed effect: n per arm = 18 (80% power), 24 (90% power)
 50% of observed effect: n per arm = 39 (80% power), 53 (90% power)
power.t.test: 10.8 
Simulated power at n = 10 per arm: 0.771 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Analyses Reported by the Sponsor

The registry record reports the sponsor's comparisons as paired t-tests. For Breezhaler® versus Ellipta®, the sponsor reports a mean difference of 26.97 L/min, a 95% confidence interval from 23.70 to 30.24, and p < 0.001. For Breezhaler® versus Handihaler®, the sponsor reports a mean difference of 53.93 L/min, a 95% confidence interval from 49.15 to 58.72, and p < 0.001.

The recalculated results are simple unadjusted comparisons of posted means and standard deviations. Their differences are expressed in the reverse direction, comparator minus Breezhaler®, so they are negative. For Ellipta®, the recalculated estimate is -26.79 L/min; for Handihaler®, it is -53.93 L/min. These calculations do not use participant-level paired differences and do not incorporate the within-person covariance of the crossover measurements.

The sponsor's reported analysis is the directly reported paired analysis and should be distinguished from the recalculations. The registry record does not flag a discrepancy. The recalculations can provide a summary-statistic comparison, but they do not replace the sponsor's paired estimates or establish the result of a different paired analysis.

14. Multiplicity, Missing Data and Analysis Populations

The registry record lists two sponsor-reported primary comparisons: Breezhaler® versus Ellipta® and Breezhaler® versus Handihaler®. Each is reported with a paired t-test, a 95% confidence interval and p < 0.001. The record does not report whether these comparisons were part of a prespecified testing hierarchy, whether the significance level was adjusted for multiple comparisons, or whether another multiplicity strategy was used. The multiplicity approach is therefore not reported in the registry record.

Multiplicity matters in general when several statistical hypotheses are tested: the chance of at least one false-positive finding can rise as more tests are considered if no adjustment or other control strategy is used. This general point does not establish that the reported results are false or that a particular adjustment was required. It identifies an analysis detail that is not available in the registry record. The number of outcome measures and analyses listed is not a substitute for a complete statistical analysis plan.

The primary endpoint population is the per-protocol set, defined as randomized patients who completed inhalation-flow maneuvers through all three devices. The endpoint counts are 93 for Breezhaler®, 88 for Ellipta® and 93 for Handihaler®. The registry record does not describe the reasons for the differing endpoint counts, the missing measurements, or a missing-data method. It does not report whether any sensitivity analysis assessed the effect of missing endpoint data. These details are not reported in the registry record.

In general, excluding participants who do not complete all required measurements can make a per-protocol summary describe a selected subset, particularly if completion is related to the outcome. The record does not provide information to evaluate whether that concern applies in this study. The posted endpoint summaries should therefore be understood as describing the PPS rather than assumed to represent all randomized patients.

The study's crossover design also makes sequence and period effects relevant statistical considerations in general. The listed sequence allocation and same-visit measurements establish the crossover structure, but the registry record does not report whether sequence or period effects were examined or adjusted for. It also does not report the original detailed statistical analysis plan. These absences limit reconstruction of the full analysis without justifying assumptions about methods that may or may not have been used.

15. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

The registry record lists one combined group, Breezhaler/Ellipta/Handihaler, with 97 at risk for serious adverse events and zero serious adverse events reported. It also lists 97 at risk for other adverse events and zero other adverse events reported. These figures should be read as the posted event counts for the combined device group.

No active drug or placebo was administered in the study, and the record describes the interventions as devices. The adverse-event figures therefore belong to this device-measurement study and should not be interpreted as a comparative drug-safety analysis. The registry record does not list serious adverse-event terms. A death count is not reported in the registry record.

Zero reported events does not establish that an event is impossible or that a device has no safety risks. It describes only what is recorded for this study and its stated at-risk group. The registry record does not provide further detail about event ascertainment or an explanation of the adverse-event reporting period.

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

The posted results describe average peak inspiratory flow rates recorded when patients with COPD performed inhalation maneuvers through three dry powder inhaler devices. In the per-protocol set, the mean was 107.5 L/min for Breezhaler®, 80.73 L/min for Ellipta® and 53.59 L/min for Handihaler®. The sponsor's paired analyses report mean differences between Breezhaler® and each comparator, with confidence intervals and p-values that indicate statistical separation under the reported tests.

These are measurements of flow through devices, not outcomes from administering a medication. The registry record explicitly states that no active drug or placebo was given. Accordingly, the findings do not establish a drug effect or compare treatment efficacy. Nor does the record report a clinical outcome, a target flow threshold, or a minimum difference considered meaningful. The statistical comparisons alone cannot answer whether the observed differences lead to a clinical benefit or harm.

The results also apply to the recorded per-protocol population: randomized patients who completed maneuvers through all three devices. They are not automatically generalizable to all patients with COPD or to people who did not complete all three maneuvers. The registry record does not report enough detail about missing measurements or the reasons for endpoint group counts to assess how completion may have influenced the summaries.

In general, statistical significance and clinical importance are separate concepts. A small p-value is evidence against a specified no-difference model under the analysis assumptions; it does not tell readers whether an effect is meaningful in practice. Likewise, confidence intervals quantify uncertainty in the estimated differences but cannot supply a clinical threshold that is absent from the record.

The results therefore support a narrow conclusion about the device-specific flow measurements collected at Visit 2 (Day 1) in this study and the sponsor's paired comparisons. They do not establish consequences for clinical outcomes, medication delivery, treatment response or individual use. Those questions are not answered by the endpoint and information recorded for this trial.

17. Why This Trial Matters Statistically

This trial illustrates why design structure matters when interpreting continuous outcomes. The same patients performed maneuvers through all three devices, so the measurements have a crossover, within-person structure. The sponsor's paired t-tests use that structure by analyzing within-participant differences. A calculation based only on device means and standard deviations cannot reproduce the paired analysis because it does not include the relationship between measurements from the same person.

It also shows why a difference in the direction of an estimate should not be mistaken for a disagreement. The sponsor reports Breezhaler® minus comparator, while the unadjusted recalculations report comparator minus Breezhaler®. The signs consequently point in opposite directions, even though both express the same ordering of device means. The estimates and intervals still differ because the calculations use different statistical information and methods.

Finally, the endpoint has a specific scope: peak inspiratory flow rate measured through devices without medication administration. Keeping the measurement separate from a clinical outcome avoids treating statistical evidence about flow as evidence about an outcome that was not measured. The record's unreported details about multiplicity, missing data, and sequence or period analyses also show why a posted result is best interpreted alongside its stated population and analysis method rather than by a p-value alone.

18. 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 endpoint results are reported for the per-protocol set, which includes randomized patients who completed maneuvers through all three devices. The posted group counts are 93 for Breezhaler®, 88 for Ellipta® and 93 for Handihaler®, while enrollment is listed as 97. The registry record does not explain the differences between the endpoint counts or describe missing measurements, reasons for noncompletion, or a missing-data analysis. This limits assessment of how the PPS summaries relate to all randomized patients.

The record describes the study as open-label with no masking. The endpoint is a measured flow rate, but the registry record does not provide details about measurement quality control or whether assessors were masked. It also does not report a statistical assessment of sequence or period effects, despite the randomized crossover design. These details are not reported in the registry record.

The sponsor reports paired t-tests, while the unadjusted recalculations use posted summary statistics and do not use paired participant-level differences. The latter should not be treated as a replication of the paired analysis. The registry record does not report a multiplicity strategy for the two comparisons or the original sample-size justification and statistical analysis plan.

Finally, the study measured inspiratory flow rather than a clinical outcome and administered no active drug or placebo. The registry record does not report a threshold for interpreting flow-rate differences as clinically meaningful. Consequently, the measurements and comparisons have a narrow interpretation and do not establish a clinical consequence or treatment effect.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Inspiratory Flow Rates Achieved by the COPD Patients Through Breezhaler®, Ellipta® and Handihaler® Inhaler Devices — NCT02596009.
• Registry record last updated 2018-06-26; status last verified 2018-06.
• Results first posted 2018-06-26.
• 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.

22. Record Summary

NCT02596009 is a completed Phase 4, multicenter, randomized, open-label crossover study of Breezhaler®, Ellipta® and Handihaler® in patients with COPD. Enrollment is listed as 97. No active drug or placebo was administered.

The primary endpoint was mean peak inspiratory flow rate in L/min at Visit 2 (Day 1), reported for the per-protocol set. The sponsor reported paired t-tests comparing Breezhaler® with Ellipta® and Handihaler®. The posted results are device-measurement comparisons, not drug efficacy results.

The registry record does not report the original sample-size justification, multiplicity approach, detailed missing-data handling, or analyses of sequence or period effects. It also does not report a clinical threshold for interpreting the flow-rate differences.