Study HZA106829: Efficacy/Safety Study of Fluticasone Furoate/Vilanterol (GW642444) in Adult and Adolescent Asthmatics: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study HZA106829: Efficacy/Safety Study of Fluticasone Furoate/Vilanterol (GW642444) in Adult and Adolescent Asthmatics (NCT01134042), a randomized phase 3 study. The primary outcome is Change From Baseline in Clinic Visit Trough (Pre-bronchodilator and Pre-dose) Forced Expiratory Volume in One Second (FEV1) at the End of the 24-week Treatment Period.

2. Clinical Question

The registry record describes a phase 3, randomized, double-blind, parallel-group study in people with persistent asthma. Its primary endpoint was change from baseline in clinic-visit trough FEV1 at the end of the treatment period. The trial also listed outcomes concerning rescue-free and symptom-free periods and asthma-related quality of life. The primary result should therefore be read as a comparison of lung-function measurements over the stated treatment period, not as a complete account of every outcome relevant to asthma.

3. Trial Design

This completed phase 3 study, NCT01134042, was randomized, double-blind, parallel-group, and multicentre. The registry lists enrollment as 587. The study started in 2010-06 and its primary completion was 2011-10. It enrolled adults and adolescents 12 years of age and older with persistent bronchial asthma and compared three active inhaled treatment regimens over 24 weeks.

The regimens were fluticasone furoate/vilanterol 200/25 µg once daily, fluticasone furoate 200 µg once daily, and fluticasone propionate 500 µg twice daily. Each regimen included a placebo inhalation powder on the schedule needed to support the stated dosing pattern. The registry does not describe a placebo-only treatment arm. The sponsor was GlaxoSmithKline.

For the primary endpoint, the registry lists arm-level analyzed sample sizes of 186 for fluticasone furoate, 187 for fluticasone furoate/vilanterol, and 190 for fluticasone propionate. These are the counts associated with the posted endpoint results; the registry’s description of the analysis population also requires receipt of at least one dose, non-missing covariates, and a post-baseline FEV1 measurement. The relationship between those endpoint counts and all randomized participants is not further explained in the registry record.

The registry’s description says the primary analysis used ANCOVA and included baseline trough FEV1, country, sex, age, and treatment group as covariates. It also says missing measurements were imputed using last observation carried forward. Details about the original sample-size justification, allocation ratio, concealment procedures, and any interim analyses 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 (par.) meeting eligibility criteria at the Screening visit entered a 4-week Run-in Period for completion of Baseline (BL) safety evaluations and to obtain BL measures of asthma status. Par. were then randomized to a 24-week Treatment Period. 1206 par. were screened, 587 were randomized, and 586 received \>=1 dose of study treatment.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent-to-Treat (ITT) Population: all participants randomized to treatment who received at least one dose of the study medication. Only those participants with non-missing covariates and a post-Baseline FEV1 measurement were analyzed.

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 in Clinic Visit Trough (Pre-bronchodilator and Pre-dose) Forced Expiratory Volume in One Second (FEV1) at the End of the 24-week Treatment Period, measured over Baseline and Week 24.

The primary endpoint was change from baseline in clinic-visit trough FEV1 at the end of the 24-week treatment period, measured in liters. FEV1 is the maximal amount of air that can be forcefully exhaled in one second. Trough FEV1 means a clinic measurement taken before bronchodilator use and before the study dose, while the participant remained on treatment. Measurements were made electronically by spirometry at baseline and scheduled clinic visits through Week 24. The highest of three technically acceptable measurements was recorded at a visit.

Change from baseline was defined as the Week 24 value minus the baseline value. A positive change therefore represents a higher measured FEV1 at the endpoint than at baseline; it does not by itself describe symptoms, daily functioning, or every aspect of asthma control. FEV1 is used as an objective measure of lung function, allowing the study to compare a physiologic outcome across randomized treatment groups.

The registry reports least-squares means and standard errors for the primary endpoint. Least-squares means are model-based group estimates that account for the covariates included in the ANCOVA, rather than simple unadjusted averages. The registry’s analysis description also says that the last non-missing on-treatment measurement after baseline was carried forward when a scheduled measurement was missing. The potential implications of that missing-data approach depend on how missingness arose and on whether carried-forward values represent later outcomes; those details are not reported in the registry record.

7. Results Posted in the Registry

• FF/VI 200/25 µg OD: mean 0.39 versus 0.20 on control; difference 0.19 (95% CI 0.11 to 0.28), standardized effect d = 0.47; z-test p = <0.001.
• FP 500 µg BID: mean 0.18 versus 0.20 on control; difference -0.02 (95% CI -0.10 to 0.07), standardized effect d = -0.04; z-test p = 0.673.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: it measures change in FEV1 in liters. The registry reports least-squares means with standard errors and describes an ANCOVA model. The simple recalculations are unadjusted comparisons reconstructed from posted summary statistics; they are not the sponsor’s covariate-adjusted analysis. When the registry’s own reported analysis and a recalculation differ, the sponsor’s analysis takes precedence.

What does ANCOVA do?

ANCOVA compares outcome values across groups while adjusting for specified covariates. In this trial, the registry says the model included baseline trough FEV1, country, sex, age, and treatment group. Adjustment can account for measured baseline or participant characteristics included in the model, which may improve precision or address chance differences in those characteristics. It does not adjust for unmeasured factors, and it does not remove the need to consider missing data or model assumptions. Further details about the model specification, including diagnostics and any transformations, are not reported in the registry record.

The registry identifies the outcome estimates as least-squares means. These are model-based estimates conditional on the fitted model and its covariates. They are not necessarily identical to raw arithmetic means. The registry’s analysis results use the label “Median Difference (Final Values),” while the primary endpoint description specifies least-squares means for change from baseline. Because those labels differ, the analysis details should be interpreted cautiously; additional clarification is not reported in the registry record.

What does the simple recalculation represent?

The recalculated comparisons use the posted arm-level estimates and their standard errors to form an unadjusted contrast. They do not reproduce the sponsor’s ANCOVA model, which uses participant-level data and the listed covariates. The recalculated standard deviation is described as approximate, because it is derived from the posted standard errors rather than from participant-level data. A recalculated p-value is therefore a secondary statistical reading of the posted summaries, not a replacement for the registered analysis. In particular, the recalculated p-value does not account for the covariate adjustment described by the sponsor.

A p-value describes how compatible the observed data are with a specified null model, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of a difference. The unadjusted p-value for fluticasone furoate/vilanterol versus fluticasone furoate is <0.001; for fluticasone propionate versus fluticasone furoate, it is 0.673. The sponsor’s reported p-values apply to its own analyses and should be read separately from these recalculations.

Why do standard errors and confidence intervals matter?

A standard error describes the uncertainty in an estimate under a statistical model. It is not the same as the spread of individual participant outcomes. A confidence interval gives a range of values compatible with the data and model at its stated confidence level. Wider intervals indicate less precision, while narrower intervals indicate greater precision, all else being equal. Neither a p-value nor a confidence interval alone establishes whether an effect is clinically important. The sponsor reports 95% confidence intervals for its stated analyses.

The registry reports standard errors for the arm-level least-squares mean estimates, but does not provide confidence limits for each arm in the posted primary results. The recalculation uses approximations from those summary statistics. The sponsor’s estimates and confidence intervals are therefore the more appropriate reference for the adjusted trial comparisons, while the recalculation can help readers understand the direction and approximate size of contrasts from the posted summaries.

What is over-dispersion, in general?

In general, over-dispersion means that the observed variability is greater than a statistical model expects. It is especially relevant in count or binary-data settings, where underestimated variability can make confidence intervals too narrow and p-values too small. This primary endpoint is continuous and the registry describes ANCOVA, so an over-dispersion adjustment is not identified in the record. The registry does not report diagnostic checks or an alternative variance model. General discussion of over-dispersion should not be mistaken for a reported feature of this trial.

How does last observation carried forward affect interpretation?

The registry says missing post-baseline measurements were imputed by carrying forward the last non-missing on-treatment measurement from a scheduled clinic visit. This method fills a missing later value with an earlier observed value. As a general statistical matter, this approach assumes that the carried-forward measurement is a reasonable substitute for the missing value. That assumption can be questionable if the outcome would have changed after the last observed measurement, and the method can understate uncertainty because the imputed value is treated as if it were observed. The registry record does not provide the amount or pattern of missingness, reasons for withdrawal, or sensitivity analyses using other missing-data assumptions.

The registry defines the intent-to-treat population as randomized participants who received at least one dose, then states that only participants with non-missing covariates and a post-baseline FEV1 measurement were analyzed. This is the population description attached to the reported endpoint, not an unqualified statement that every randomized participant contributed to the model. The record does not give further detail about exclusions from the endpoint analysis or how they were handled.

10. Confidence Intervals and Precision

Precision concerns how much uncertainty surrounds an estimated treatment contrast. The sponsor’s analysis reports a 95% confidence interval from 0.108 to 0.277 liters for fluticasone furoate/vilanterol compared with fluticasone furoate. The interval lies above zero and accompanies a reported p-value of <0.001. This indicates that, under the sponsor’s model, the estimated contrast is separated from zero at the stated confidence level. It does not establish the clinical importance of the difference, for which the registry does not report a threshold.

For fluticasone furoate compared with fluticasone propionate, the sponsor reports a 95% confidence interval from -0.066 to 0.102 liters. This interval includes zero and covers both negative and positive values, so the estimate is not precise enough to rule out contrasts in either direction under that analysis. The registry labels this comparison as non-inferiority or equivalence, but does not report the margin needed to interpret such a conclusion or a p-value for the comparison.

The sponsor’s comparison of fluticasone furoate/vilanterol with fluticasone propionate has a reported 95% confidence interval from 0.127 to 0.294 liters. It is centered on the reported estimate of 0.210 liters and lies above zero. The width of an interval reflects both the data and the model; it should be considered alongside the endpoint, analysis population, and adjustment strategy.

The posted arm-level standard errors are similar in scale, but they are not themselves confidence intervals and do not describe the full range of individual outcomes. The simple recalculation intervals are approximate and are not a substitute for sponsor-reported intervals from ANCOVA. The registry does not report a clinically important difference threshold, so statistical precision cannot by itself determine whether an estimate is important to patients or practice.

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 sample-size figures are planning illustrations based on the observed effect, not the trial’s original design calculations. For these continuous-outcome comparisons, the figures are participants per arm, not total events. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning row uses the stated fraction of the observed effect as the effect to detect. The trial’s own original design assumptions, including its sample-size justification, are not reported in the registry record.

The 80% power column gives the planned participants per arm for a hypothetical study with an 80% chance of detecting the selected effect under the stated assumptions, if that effect is real and the assumptions hold. The 90% power column gives the corresponding planning number for a higher target chance of detection. These are design calculations, not guarantees about a future study. The figures do not describe the probability that the observed result is correct.

For the fluticasone furoate/vilanterol versus fluticasone furoate comparison, the planning illustration uses the observed mean difference of 0.193 liters. Planning for 100% of the observed effect gives 72 participants per arm for 80% power and 97 for 90% power. Planning for 75% of the observed effect gives 128 participants per arm for 80% power and 172 for 90% power. Planning for 50% of the observed effect gives 288 participants per arm for 80% power and 386 for 90% power.

For fluticasone propionate versus fluticasone furoate, the observed recalculated difference is -0.018 liters. The associated illustrations require very large planned group sizes because the observed contrast is close to zero. Smaller effects need more participants to detect, all else equal. These calculations are not observed power: they are hypothetical planning illustrations based on the estimated effects and stated assumptions, not a retrospective measure of what the completed trial’s results could detect.

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 = 0.193; pooled standard deviation σ = 0.413.
2. n = 2 × (1.96 + 0.842)² × 0.413² / 0.193² = 71.9, rounded up to 72 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 <- 0.193                      # observed difference in means
sigma <- 0.413107519403364                     # 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(1134042)
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 = 72 (80% power), 97 (90% power)
 75% of observed effect: n per arm = 128 (80% power), 172 (90% power)
 50% of observed effect: n per arm = 288 (80% power), 386 (90% power)
power.t.test: 72.9 
Simulated power at n = 72 per arm: 0.793 (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 sponsor’s own analyses are reported as ANCOVA comparisons, while the recalculations are simple unadjusted comparisons based on posted summary statistics. They answer related but not identical statistical questions. The sponsor’s models include baseline trough FEV1, country, sex, age, and treatment group, according to the primary endpoint description. The recalculations do not reproduce those covariate adjustments or use participant-level records.

For fluticasone furoate/vilanterol compared with fluticasone furoate, the sponsor reports an estimate of 0.193 liters, a 95% confidence interval from 0.108 to 0.277, and p <0.001. The simple recalculation also gives an estimate of 0.193 liters and an unadjusted p-value of <0.001. Their close agreement in the point estimate does not make the methods interchangeable; the sponsor’s own analysis takes precedence.

For fluticasone propionate compared with fluticasone furoate, the simple recalculation estimates -0.018 liters with an unadjusted p-value of 0.673. The sponsor reports the reverse ordering, fluticasone furoate compared with fluticasone propionate, with an estimate of 0.018 liters and a 95% confidence interval from -0.066 to 0.102. The registry labels that analysis non-inferiority or equivalence, but its method and p-value are not reported. The registry also reports fluticasone furoate/vilanterol compared with fluticasone propionate as 0.210 liters, with a 95% confidence interval from 0.127 to 0.294 and p <0.001. The sponsor’s reported analyses should guide interpretation of these contrasts; missing details limit independent assessment of the non-inferiority or equivalence result.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists 11 outcome measures and reports 2 secondary outcomes: change from baseline in the percentage of rescue-free and symptom-free 24-hour periods at the end of treatment, and change from baseline in total Asthma Quality of Life Questionnaire (+12) score at Week 12 and Week 24 or early withdrawal. The posted primary result concerns trough FEV1. The registry also lists five analyses. It does not report a multiplicity adjustment strategy, a hierarchy for testing endpoints, or whether the reported comparisons were prespecified in a way that controls the chance of false-positive findings across multiple outcomes and contrasts. The nominal p-values should therefore be understood as applying to their stated comparisons, without assuming a family-wise adjustment that is not described.

Missing-data handling is described for the primary endpoint: the last non-missing post-baseline on-treatment measurement at a scheduled clinic visit was carried forward to impute a missing measurement. The registry record does not report how many measurements were missing, the timing or reasons for missingness, or whether missingness differed across arms. It does not describe sensitivity analyses under alternative missing-data assumptions. These omissions make it difficult to judge how robust the endpoint estimates are to the imputation approach.

The registry calls the analysis population intent-to-treat, defining it as randomized participants who received at least one dose, while also limiting analysis to participants with non-missing covariates and a post-baseline FEV1 measurement. The endpoint arm counts are 186, 187, and 190. The record does not explain in detail how participants who were randomized but did not meet the analysis criteria were handled, or whether every randomized participant contributed to each comparison.

The registry reports an ANCOVA model and its listed covariates, but does not report model diagnostics, a full statistical analysis plan, or details about any interim looks at the data. It also does not state the original sample-size justification. Those items are not reported in the registry record and should not be inferred from the planning illustrations, which use observed effects and general planning assumptions.

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 3% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: asthma (0.5%), lower limb fracture (0.5%), haemoptysis (0.5%), limb traumatic amputation (0.5%), atrial fibrillation (0.5%).

The registry reports serious adverse event counts for each treatment arm, with the number at risk stated alongside each count. There was 1 serious adverse event among 194 participants at risk in the fluticasone furoate arm, 6 among 197 at risk in the fluticasone furoate/vilanterol arm, and 2 among 195 at risk in the fluticasone propionate arm. These are reported counts; they should not be read as proof that treatment caused an event or as a complete comparison of safety.

The listed serious event terms, each reported as 0.5%, are asthma, lower limb fracture, haemoptysis, limb traumatic amputation, atrial fibrillation, and inguinal hernia. The record does not provide further event-level detail in the information, such as timing, relatedness, or outcomes. Death counts are not reported in the registry record.

The registry also reports counts for other adverse events: 66 among 194 at risk for fluticasone furoate, 62 among 197 at risk for fluticasone furoate/vilanterol, and 73 among 195 at risk for fluticasone propionate. The serious-event counts and other-adverse-event counts are separate reported summaries. They should not be added together, and they do not by themselves establish comparative risk. Differences in exposure, event definitions, follow-up, and chance can affect such counts. No safety conclusion should be drawn solely from these figures.

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

The primary endpoint results describe changes in a lung-function measure over the 24-week treatment period in adults and adolescents 12 years of age and older with persistent asthma. The reported change estimates were 0.394 liters for fluticasone furoate/vilanterol, 0.201 liters for fluticasone furoate, and 0.183 liters for fluticasone propionate. The sponsor’s adjusted comparison of fluticasone furoate/vilanterol with fluticasone furoate was 0.193 liters, with a 95% confidence interval from 0.108 to 0.277 and p <0.001.

Those results address measured trough FEV1, not every dimension of asthma or treatment experience. The registry lists secondary outcomes involving rescue-free and symptom-free periods and asthma-related quality of life, but the information used for these results does not include their estimates. Effects on those outcomes are therefore not reported in the registry record in the results. Nor do the FEV1 results establish effects on long-term outcomes beyond the stated treatment period.

A confidence interval and p-value describe statistical uncertainty under an analysis model. They do not establish whether a difference is large enough to matter clinically. The registry does not report a threshold for a clinically important FEV1 change. The results also depend on the analysis population, covariate adjustment, and the last-observation-carried-forward approach for missing measurements. Missingness patterns and sensitivity analyses are not reported, limiting assessment of how alternative assumptions might affect the estimates.

The sponsor’s analysis should take precedence over the simple recalculations because the latter are unadjusted and the registry flags a discrepancy. For the fluticasone furoate versus fluticasone propionate comparison, the registry provides a confidence interval that includes zero and labels the analysis non-inferiority or equivalence, but does not state the margin or method. That record does not support an independent conclusion about the non-inferiority or equivalence claim. In general, a failure to detect a difference is not itself proof that treatments are equivalent.

These findings apply to the trial population, treatments, endpoint, and follow-up described in the registry. They do not constitute medical advice and do not determine what treatment is appropriate for an individual.

17. Why This Trial Matters Statistically

This trial illustrates why a registry result needs to be read at several levels. The endpoint is a continuous physiologic measure, the sponsor analyzed it with covariate adjustment, and a separate recalculation can only approximate a contrast from summary statistics. The reported point estimates may look similar, but the statistical procedures are not identical. Recognizing that distinction prevents an unadjusted calculation from being mistaken for the trial’s primary analysis.

It also shows why estimates need uncertainty intervals, not just point estimates or p-values. For the sponsor’s comparison of fluticasone furoate/vilanterol with fluticasone furoate, the reported confidence interval is entirely above zero. For the reported fluticasone furoate versus fluticasone propionate comparison, the interval spans zero, and the registry’s non-inferiority or equivalence label cannot be fully assessed without the margin and method. These are different inferential situations and should not be collapsed into a simple statement that groups were or were not different.

The endpoint population and missing-data method also matter. The registry calls the analysis intent-to-treat but restricts it to people with specified data available, and it describes last observation carried forward. Without information about missingness and sensitivity analyses, the effect of those choices cannot be evaluated fully. These details are central to assessing whether an estimate is robust.

Finally, planning calculations based on an observed effect are not evidence of the trial’s original design rationale or of retrospective power. They show how sample-size requirements change with the effect targeted under stated assumptions. Keeping planning illustrations separate from observed trial results is important when interpreting a completed study.

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 registry record gives the primary endpoint definition and names the ANCOVA covariates, but does not provide the complete statistical analysis plan or model diagnostics. It does not report the original sample-size justification, allocation ratio, or interim-analysis plan. Those omissions prevent a full assessment of how the analysis and design were specified before results were observed.

The analysis population description combines an intent-to-treat label with eligibility conditions for inclusion in the endpoint analysis: receipt of at least one dose, non-missing covariates, and a post-baseline FEV1 measurement. The record does not explain the consequences of these restrictions or give a fuller account of participants excluded from the endpoint analysis. The endpoint arm counts should therefore be understood as the counts attached to the reported analysis, not assumed to describe every randomized participant.

Last observation carried forward was used for missing primary outcome measurements. The registry does not report missingness by arm, reasons for missing measurements, or sensitivity analyses. Since carrying an earlier measurement forward relies on assumptions about the unobserved later value, the absence of those details limits evaluation of robustness.

The registry’s sponsor analysis labels results as “Median Difference (Final Values),” whereas the primary endpoint description specifies least-squares means for change from baseline. The record does not explain this labeling difference. It also labels one comparison as non-inferiority or equivalence without reporting the analysis method, p-value, or margin. That limits independent evaluation of the stated inferential claim.

The simple recalculations are approximate, unadjusted comparisons based on summary statistics and do not reproduce the sponsor’s covariate-adjusted analysis. The registry reports multiple outcomes and analyses but does not describe multiplicity control. Finally, no clinically important threshold for FEV1 change is reported, so statistical results cannot by themselves establish clinical importance.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Study HZA106829: Efficacy/Safety Study of Fluticasone Furoate/Vilanterol (GW642444) in Adult and Adolescent Asthmatics — NCT01134042.
• Registry record last updated 2017-01-11; status last verified 2016-11.
• Results first posted 2013-08-08.
• 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

NCT01134042 is a completed phase 3 randomized, double-blind, parallel-group, multicentre study sponsored by GlaxoSmithKline. It enrolled people 12 years of age and older with persistent asthma and compared fluticasone furoate/vilanterol, fluticasone furoate, and fluticasone propionate over 24 weeks. Enrollment was 587; the study started in 2010-06 and primary completion was 2011-10.

The primary endpoint was change from baseline in clinic-visit trough FEV1 at the end of treatment. The sponsor reported an ANCOVA estimate of 0.193 liters for fluticasone furoate/vilanterol compared with fluticasone furoate, with a 95% confidence interval from 0.108 to 0.277 and p <0.001. The record also reports comparisons involving fluticasone propionate, including a non-inferiority or equivalence label for fluticasone furoate versus fluticasone propionate. Important details about that analysis, multiplicity, missingness, and the original sample-size justification are not reported in the registry record.