Comparative Study of Error Rates Between ELLIPTA® Dry Powder Inhaler (DPI) and Other DPIs: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparative Study of Error Rates Between ELLIPTA® Dry Powder Inhaler (DPI) and Other DPIs (NCT03114969), a registered phase not stated study. The primary outcome is Percentage of Participants Making at Least One Critical Error at Visit 1-Primary Device Comparisons.

2. Clinical Question

The study’s stated rationale was that ELLIPTA requires fewer steps to use and might therefore be associated with fewer use errors. The clinical question concerns observed inhaler technique, not medication efficacy, COPD symptoms, or longer-term clinical outcomes. The listed primary endpoint captures whether a participant made a critical error during the demonstration; it does not measure how often errors occur during routine use over time.

3. Trial Design

NCT03114969 is described as an open-label, low-interventional clinical study in adults with COPD. The registry lists enrollment of 450 and gives the status as completed. The recorded study dates are 2017-06-08 for the start and 2018-03-09 for primary completion. The brief summary describes two clinical visits and a total study duration of approximately 6 weeks.

Participants were assessed while demonstrating use of their maintenance DPI or DPIs. At Visit 1, a health care practitioner recorded errors using a checklist based on the correct-use steps in the relevant patient instruction leaflets. The primary endpoint concerns critical errors at Visit 1, before any retraining. The brief summary says participants were instructed on correct use, or informed of their correct use, after the assessment. The primary endpoint’s timeframe is Day 1.

The device regimens included ELLIPTA products and other devices, used either alone or in combinations. These included ICS/LABA combinations and LAMA monotherapies or combination regimens. The registry describes comparisons among these device groups, but allocation is not stated. It also cautions that the arms are not a like-for-like control comparison. The record therefore does not establish randomized assignment or a common control group.

The primary analysis population is defined as the ITT Population, comprising enrolled participants who demonstrated use of their primary DPI. The registry lists 24 outcome measures and 96 analyses, spanning critical and overall errors and several device comparisons. Detailed allocation procedures, a statistical analysis plan, and a sample-size justification are not reported in the registry record. The number enrolled is not, by itself, the number analyzed for every outcome.

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: A total of 461 participants were screened of which 11 failed screening and 450 participants were enrolled. The study was conducted in two countries-Netherlands and the United Kingdom.

Only the first 6 of 9 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: 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.

Only the first 6 of 10 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants Making at Least One Critical Error at Visit 1-Primary Device Comparisons, measured over Day 1.

The primary endpoint is the percentage of participants who made at least one critical error when demonstrating use of their primary DPI at Visit 1. Errors were recorded by a health care practitioner against a checklist based on the correct-use instructions in the patient instruction leaflet for each DPI. The registry defines a critical error as one that resulted in the participant receiving a lesser or no dose.

This endpoint is a binary participant-level measure: a participant either met the criterion of making at least one critical error during the demonstration or did not. The percentage expresses how common that outcome was within the analysis population. It does not distinguish participants by the number of errors they made, nor does it describe the severity or frequency of errors beyond the critical-error definition.

As a general statistical point, a percentage is useful for comparing the frequency of an event across groups, while still depending on how the event and analysis population are defined. Here, the measure focuses on errors with a potential consequence for dose delivery, rather than every departure from the correct-use checklist. That focus makes the endpoint relevant to the study’s question about inhaler technique, but it does not establish an effect on medication delivery in routine practice or on clinical outcomes.

The registry record identifies the endpoint and its population but does not report the observed percentages, group denominators for the primary outcome, or a comparative effect estimate. The endpoint’s numerical result and its precision are therefore not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is expressed as a percentage of participants who made at least one critical error during a demonstration of their primary DPI. The record defines the event as an error resulting in a lesser or no dose. This is a binary outcome at the participant level, even though the assessment itself used a checklist of inhaler-use steps. The primary endpoint is not a count of all errors, and the registry does not state that the number or type of critical errors per participant was the primary measure.

What does a participant-level percentage describe?

As a general statistical explanation, a participant-level percentage describes the fraction of an analysis group meeting a defined event criterion. Its interpretation depends on the event definition, the denominator, and who was included in the analysis. For this study, the ITT Population is defined as enrolled participants who demonstrated use of their primary DPI. The outcome-specific denominators and observed percentages are not reported in the registry record, so the endpoint cannot be reconstructed from the enrollment entry alone.

How should group comparisons be understood?

The registry lists comparisons involving ELLIPTA products and other DPIs, including single-device regimens and combinations. It does not state that participants were randomly allocated to these device groups. The record specifically notes that allocation is not stated and that the arms are not a like-for-like control comparison. As a general matter, nonrandomized group comparisons may reflect differences in participant characteristics, prescribed regimens, or other factors in addition to the device being assessed. The record does not report an adjustment method that would address such differences.

The observed percentage in each group would describe the participants assessed in that group. A between-group contrast would describe the difference or relative relationship between those observed frequencies, depending on the chosen effect measure. No primary-outcome group percentages or comparison estimates are reported, and the statistical method for the comparison is not reported in the registry record. It is therefore not possible to characterize the size, direction, or uncertainty of a group contrast from this record.

What do confidence intervals and p-values add?

In general, a confidence interval communicates the uncertainty around an estimate under a specified statistical method. A narrower interval indicates greater precision, while a wider interval indicates less precision. It does not by itself identify which values are clinically important. A p-value describes how compatible the observed data are with a null hypothesis under the model used; it does not measure the probability that the null hypothesis is true, or the magnitude of an effect. The registry record does not report a confidence interval or p-value for the primary endpoint.

Why do the many listed analyses matter?

The registry lists 24 outcome measures and 96 analyses. As a general statistical point, when many outcomes or comparisons are examined, the chance of seeing an apparently notable result somewhere among them can increase. Methods that account for multiple testing can address that concern, but the chosen approach depends on the analysis plan and research question. Whether multiplicity was controlled, and how any primary and secondary comparisons were prioritized, is not reported in the registry record.

What does over-dispersion mean for this endpoint?

Over-dispersion is a general concern for count data when variability exceeds what a statistical model expects. The primary endpoint is a binary participant-level percentage, not a count outcome, and the registry does not identify an over-dispersion parameter or model. No model details are reported, so there is no basis for describing an over-dispersion adjustment for the primary analysis.

How should planning calculations be read?

Any sample-size calculations associated with these results are planning illustrations, not observed power. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect to detect defined as a stated fraction of the observed effect. The record does not report the numerical observed effect or a sample-size justification from the trial’s own design. These calculations should not be interpreted as evidence that the study had a particular probability of detecting an effect.

The registry does not report the statistical analysis plan, model specifications, handling of missing outcomes, or any interim analyses. Those omissions limit the ability to assess how the endpoint was analyzed beyond the stated definition and analysis population.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Relvar Ellipta", "Symbicort Turbuhaler", "Seretide Diskus", "Spiriva Handihaler", "Incruse/Anoro Ellipta", "Ultibro/Seebri Breezhaler")
n      <- c(50, 50, 50, 50, 51, 49)
events <- c(5, 20, 13, 17, 5, 16)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Relvar Ellipta  50  5  10.0  3.3  21.8  4.3  21.4  4.4  21.4
ARM  Symbicort Turbuhaler  50  20  40.0  26.4  54.8  27.6  53.8  27.6  53.9
ARM  Seretide Diskus  50  13  26.0  14.6  40.3  15.9  39.6  15.9  39.6
ARM  Spiriva Handihaler  50  17  34.0  21.2  48.8  22.4  47.8  22.4  47.9
ARM  Incruse/Anoro Ellipta  51  5  9.8  3.3  21.4  4.3  21.0  4.4  21.0
ARM  Ultibro/Seebri Breezhaler  49  16  32.7  19.9  47.5  21.2  46.6  21.2  46.7

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 24 outcome measures and 96 analyses, with outcomes addressing critical and overall errors, several device comparisons, and assessments at Visit 1 and Visit 2. Examining multiple outcomes and comparisons raises a multiplicity question: whether the analysis strategy accounts for the increased chance of finding an apparently notable result across many tests. Whether any adjustment was used, which comparisons were designated as primary for testing, and how secondary outcomes were prioritized are not reported in the registry record.

The primary population is described as the ITT Population, defined as all enrolled participants who demonstrated use of their primary DPI. The record does not report how missing demonstrations, incomplete checklists, or other missing endpoint data were handled. It does not state whether participants with unavailable outcome assessments were excluded, imputed, or addressed through another method. Outcome-specific denominators are also not reported, so the extent of missingness cannot be assessed.

The registry gives no detailed analysis plan, model specification, interim-analysis information, or sample-size justification. The planning assumptions described for sample-size illustrations do not establish the trial’s original design rationale or analysis plan. In the absence of these details and the numerical outcome results, the record does not support an assessment of the robustness of the primary analysis or the consequences of missing data and multiple testing.

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 3% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: chest pain (2.0%), chronic obstructive pulmonary disease (2.0%), appendicitis (2.0%).

The registry reports serious-event and other-event figures separately for the listed device groups. In the serious-event category, the affected-participant counts are 0 for Relvar Ellipta, Symbicort Turbuhaler, Seretide Diskus, Spiriva Handihaler, and Relvar Ellipta+LAMA; they are 1 for Incruse/Anoro Ellipta, Ultibro/Seebri Breezhaler, and Symbicort Turbuhaler+LAMA; and 0 for Seretide Diskus+LAMA. The corresponding numbers at risk are reported for each group in the registry. No deaths are reported for any listed group.

For the other-event category, the registry reports affected-participant counts of 4 for Relvar Ellipta, 1 for Symbicort Turbuhaler, 0 for Seretide Diskus, 1 for Spiriva Handihaler, 2 for Incruse/Anoro Ellipta, 2 for Ultibro/Seebri Breezhaler, 3 for Relvar Ellipta+LAMA, 3 for Symbicort Turbuhaler+LAMA, and 5 for Seretide Diskus+LAMA. These are descriptive counts, not comparative estimates. The registry also lists Chest pain, Chronic obstructive pulmonary disease, and Appendicitis among its top serious terms, each at 2.0%.

As a general statistical caution, small event counts provide limited information about comparative safety, particularly when groups are not randomized or like-for-like controls. These figures do not establish that a device caused an event or that groups differ in risk. The registry does not provide additional detail about event timing, attribution, or comparative statistical analyses.

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

The study addresses a practical measurement question: how often adults with COPD made a critical error while demonstrating use of their prescribed DPI before retraining. Its endpoint is specifically tied to a checklist-based assessment at Visit 1. The registry also describes secondary outcomes involving overall errors and later assessments. This provides a framework for studying inhaler technique across several devices and regimens.

The record does not report the primary outcome percentages or comparisons. Therefore, it does not establish whether critical errors were more common with any particular DPI, whether the stated hypothesis was supported, or how precisely any difference was estimated. Completion status and enrollment do not substitute for outcome results. The record also does not provide outcome-specific denominators, confidence intervals, or p-values.

Even if comparative outcome estimates were available, this design would need careful interpretation. Allocation is not stated, and the registry says the arms are not a like-for-like control comparison. A difference in observed percentages could reflect differences in the participants or prescribed regimens as well as differences in inhaler technique. The posted information does not describe an adjustment strategy that would separate these explanations.

The endpoint concerns a demonstration, not repeated use in everyday settings. It does not directly measure dose delivery over time, symptom control, exacerbations, or other clinical outcomes. A critical error is defined by its potential consequence for dose receipt, but the record does not report a direct measurement of medication received or a clinical consequence attributable to an error.

Thus, the registry record identifies a relevant and clearly described technique outcome but does not contain the numerical results needed to evaluate the comparative findings. It supports understanding what the investigators intended to measure and how the assessment was framed; it does not support a conclusion about which device or regimen had fewer errors, or about clinical effectiveness.

14. Why This Trial Matters Statistically

Inhaler technique is a measurable aspect of medication use, and errors that may result in a lesser or no dose are a meaningful subject for clinical research. This study’s primary endpoint focuses on that defined form of error rather than treating every checklist deviation as equivalent. The registry also distinguishes critical errors from overall errors, allowing the research question to address different levels of technique performance.

The design illustrates why the comparison structure matters as much as the endpoint. Participants were assessed across multiple DPI regimens, including monotherapies and combinations, but the registry does not state allocation and explicitly warns that the groups are not a like-for-like control comparison. Statistical comparisons of such groups require caution because observed differences need not arise from device design alone.

The record also illustrates the importance of complete reporting. A defined outcome and completed study status are not enough to assess a comparative result. Group percentages, denominators, effect estimates, and uncertainty measures are needed to judge both the direction and precision of findings. Those primary outcome results are not reported in the registry record.

Finally, the registry lists many outcomes and analyses. That breadth can support examination of several questions, but it also makes prespecification and multiplicity handling important for interpreting statistical evidence. The relevant analysis plan and any multiplicity strategy are not reported. The study therefore raises useful statistical questions about outcome definition, group comparability, and reporting, while the posted information remains insufficient to answer the principal comparative question.

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 principal limitation for interpreting the study is that the registry record does not report the primary endpoint results. It gives no observed critical-error percentages, primary comparison estimates, confidence intervals, or p-values. Without those results, neither the direction nor the precision of the comparison can be evaluated.

The record does not state how participants came to use particular devices or regimens. It explicitly notes that allocation is not stated and that the arms are not a like-for-like control comparison. Differences in the groups could therefore reflect differences beyond the inhaler being demonstrated. The record does not report adjustment for potential group differences.

The primary endpoint is based on a demonstration at Visit 1 and a checklist derived from the relevant patient instruction leaflets. The record does not describe assessor training, consistency between assessors, or whether assessments were independently reviewed. It also does not report how missing or incomplete demonstrations were handled. Those details are not reported in the registry record, so their impact on measurement and analysis cannot be assessed.

The registry lists 24 outcome measures and 96 analyses, but the analysis plan and multiplicity approach are not reported. It also does not report outcome-specific denominators, the statistical model, or a sample-size justification from the trial’s own design. Enrollment of 450 does not establish how many participants contributed to each outcome comparison.

Finally, the primary outcome concerns observed technique at a study visit, not long-term use or clinical outcomes. The record does not establish that observed errors translated into changes in medication delivery or health outcomes. These limitations constrain both causal interpretation and the scope of any clinical conclusions.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Comparative Study of Error Rates Between ELLIPTA® Dry Powder Inhaler (DPI) and Other DPIs — NCT03114969.
• Registry record last updated 2020-10-28; status last verified 2020-10.
• Results first posted 2019-06-10.
• 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

NCT03114969, titled “Comparative Study of Error Rates Between ELLIPTA® Dry Powder Inhaler (DPI) and Other DPIs,” is a completed, open-label, low-interventional study in adults with COPD. The registry lists enrollment of 450, a start date of 2017-06-08, and primary completion on 2018-03-09.

The primary endpoint was the percentage of participants making at least one critical error while demonstrating use of their primary DPI at Visit 1. A critical error was defined as an error resulting in a lesser or no dose. The primary analysis population was the ITT Population, consisting of enrolled participants who demonstrated use of their primary DPI.

The registry describes comparisons among ELLIPTA and other DPI regimens, alone or in combination. Allocation is not stated, and the record cautions that the arms are not a like-for-like control comparison. Although it lists 24 outcome measures and 96 analyses, numerical primary endpoint results and their uncertainty are not reported in the registry record.