Study FFR116364, a Placebo-controlled Study of GW685698X in Paediatric Subjects With Perennial Allergic Rhinitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study FFR116364, a Placebo-controlled Study of GW685698X in Paediatric Subjects With Perennial Allergic Rhinitis (NCT01630135), a randomized phase 3 study. The primary outcome is Mean Change From Baseline in the 3 Total Nasal Symptom Score (3TNSS) Over the Entire Treatment Period.

2. Clinical Question

The study asked whether the mean change in nasal symptoms over the treatment period differed between children assigned to GW685698X and children assigned to placebo. The primary measure was the 3 Total Nasal Symptom Score (3TNSS), combining sneezing, rhinorrhea, and nasal congestion. Participants or their parent or guardian recorded symptoms in daily diaries.

The registry reports a negative mean change in both groups, with a larger decrease in the fluticasone furoate group. The reported ANCOVA comparison estimates the difference between groups while accounting for the model used in that analysis. The registry record does not report the covariates included in that model or a detailed description of who was masked.

3. Trial Design

This was a completed Phase 3, randomized, parallel-group, placebo-controlled study sponsored by GlaxoSmithKline. The registry lists an enrollment of 261 participants. The recorded study start is 2012-06, and the primary completion is 2012-08. The registered study title identifies the trial as a double-blind, placebo-controlled study; the masking field records quadruple masking.

Participants were assigned to fluticasone furoate, listed under the study name GW685698X, or placebo. The active arm is described as 55 µg per day, and the study summary specifies nasal spray given q.d. For 2 weeks. The registry record does not report the randomization procedure, allocation ratio, or the identities of the people included in the masking designation.

The primary endpoint was mean change from baseline in 3TNSS over the entire treatment period. The record also lists secondary assessments of nasal symptoms, including the three-symptom and four-symptom total scores, individual nasal symptoms, and ocular symptoms. There are 15 posted secondary outcomes and 16 outcome measures in total. Those additional outcomes provide context about the breadth of assessment, but the primary result is the 3TNSS comparison.

The registry record does not report the original sample-size justification, the planned statistical analysis document, or whether interim analyses were planned or conducted. Those details should not be inferred from the trial's completed status or from the number enrolled.

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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set (FAS): all participants meeting the primary criteria for enrollment, without any major good clinical practice (GCP) deviation, who received at least one dose of the assigned treatment and had diary assessment for 3TNSS after receiving a dose of study medication

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 Mean Change From Baseline in the 3 Total Nasal Symptom Score (3TNSS) Over the Entire Treatment Period, measured over Baseline through the entire treatment period (2 weeks).

The primary endpoint was mean change from baseline in the 3 Total Nasal Symptom Score (3TNSS) over the entire treatment period, identified as 2 weeks. The score adds ratings for sneezing, rhinorrhea, and nasal congestion. Each symptom is rated from 0 to 3, giving the total score a range from 0 to 9. Higher scores indicate more severe symptoms.

Baseline was defined as the average 3TNSS over the last 4 consecutive days before Visit 2, when the treatment period began. For the entire assessment period, each participant's treatment-period mean was calculated from available diary data, using the average of non-missing observations in that period. Change from baseline was calculated as the treatment-period mean minus the baseline score. A negative change therefore represents a lower average symptom score during treatment than at baseline.

This measure combines several nasal symptoms into one continuous outcome, which offers a single summary for comparing groups while retaining information about the degree of symptom severity. Daily diary scoring captures repeated assessments during the period rather than relying only on one assessment at its end. The registry record does not report a rationale for selecting this endpoint, or information about validation or interpretation thresholds for the score.

7. Results Posted in the Registry

• Fluticasone Furoate 55 µg Per Day: mean -1.98 versus -0.89 on control; difference -1.09 (95% CI -1.42 to -0.76), standardized effect d = -0.80; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a continuous measure: the score can take values across a numeric range, and the analysis compares averages. The registry identifies the arm summaries as least-squares means and the reported comparison as a least-squares mean difference from ANCOVA. ANCOVA is a regression-based method that estimates a group contrast while accounting for variables included in the model. The registry record does not identify those variables, so it is not possible to describe the adjustment in detail or independently reproduce the reported model.

What does a least-squares mean represent?

In general, a least-squares mean is a model-based estimate of a group's average outcome. It can differ from a raw arithmetic mean when the model adjusts for other variables or handles the data structure in a particular way. The registry lists least-squares mean change values of -1.98 for fluticasone furoate and -0.89 for placebo, and reports their modeled contrast as -1.089. The reported analysis is the relevant result for interpreting the registry's ANCOVA comparison.

How should the standard errors be read?

The posted standard error for each arm estimate is 0.12. In general, a standard error describes the uncertainty of an estimated quantity across repeated samples under the statistical model. A smaller standard error indicates a more precise estimate in that sense. It does not describe how much participants' scores varied from one another. The registry record does not report the arm-level standard deviations for this endpoint. The recalculation uses an approximate pooled standard deviation derived from the posted summaries, so that aspect of the recalculation is not a direct analysis of individual-level data.

What does the p-value address?

In general, a p-value measures how compatible the observed data are with a specified null model, assuming that model and the analysis assumptions hold. The reported p-value is <0.001 for the ANCOVA comparison. This is evidence against the null hypothesis of no group difference under that analysis; it is not the probability that the null hypothesis is true, nor is it a measure of the size or practical importance of the difference. The recalculated unadjusted p-value is also <0.001, but it comes from a simpler analysis of summary statistics and should not be treated as a reproduction of the ANCOVA.

Why can over-dispersion matter?

In general, over-dispersion occurs when observed variability is greater than a statistical model expects. If a model understates variability, standard errors can be too small and uncertainty intervals too narrow. This endpoint is a continuous score, and the registry record does not report an over-dispersion assessment or a model-based variance check. Over-dispersion should therefore be understood as a general issue in statistical modeling, not as a problem established for this trial.

What is the role of the recalculation?

The unadjusted recalculation is based on the posted arm estimates, standard errors, and group sizes. It is a simple comparison of the summaries, not an analysis using the original participant-level observations or the undisclosed ANCOVA model. It can serve as a numerical cross-check at a broad level, but it cannot recover adjustments, missing-data procedures, or other modeling choices absent from the registry record. The reported analysis is identified as ANCOVA and is the result to use when describing the trial's posted model-based comparison.

In general, the assumptions behind a continuous-outcome comparison include an appropriate model for the outcome and a suitable treatment of the observations and missing values. The record defines the primary analysis population and explains how available diary data contributed to period means, but it does not give enough detail to assess all assumptions or reproduce the complete analysis. The original statistical plan, including any prespecified diagnostics or sensitivity analyses, is not reported in the registry record.

10. Confidence Intervals and Precision

The reported ANCOVA estimate is a least-squares mean difference of -1.089, with a 95% confidence interval from -1.41 to -0.76. The interval lies entirely below zero, which is consistent with a greater mean reduction in 3TNSS in the fluticasone furoate arm than in the placebo arm under the reported analysis.

In general, a confidence interval gives a range of estimates compatible with the data and the model at the stated confidence level. It conveys precision as well as direction: a narrower interval indicates less statistical uncertainty about the estimated contrast, while a wider interval indicates more. The interval does not mean that there is a 95% probability that this particular fixed contrast lies within its limits. Its interpretation depends on the analysis and its assumptions.

The registry also reports a standard error of 0.12 for each arm estimate. These are standard errors for the separate estimates, not for the difference. The reported confidence interval for the ANCOVA contrast is the more direct measure of uncertainty for the primary comparison. The confidence interval from the unadjusted recalculation is based on approximate summary-statistic calculations and is distinct from the registry's model-based interval.

Although the confidence interval excludes zero, it does not establish the practical importance of the estimated change. The registry record does not report a threshold for a meaningful difference, nor does it report a precision target or a prespecified confidence-interval width. Thus the interval supports an interpretation about statistical uncertainty around the group contrast, not a definitive judgment about its importance to individual participants.

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, not as total events. They are illustrations calculated from the observed effect, not the trial's original design justification. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set to a stated fraction of the observed effect. The original design assumptions are not reported in the registry record.

For planning at 100% of the observed effect, the illustration gives 25 participants per arm for 80% power and 34 participants per arm for 90% power. For planning at 75% of the observed effect, it gives 45 participants per arm for 80% power and 60 participants per arm for 90% power. For planning at 50% of the observed effect, it gives 100 participants per arm for 80% power and 133 participants per arm for 90% power.

In general, the 80% power column represents the number of participants per arm estimated to give an 80% chance of detecting the specified effect under the planning assumptions if that effect is real. The 90% power column uses a higher target chance of detecting it, and therefore calls for more participants per arm in these illustrations. Smaller effects need more participants to distinguish them from random variation, which is why the figures rise as the planning effect becomes a smaller fraction of the observed effect.

The curve is a visual way to read the same planning relationship between the effect assumed and the sample size needed. These figures are planning illustrations, not observed power.

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 = -1.090; pooled standard deviation σ = 1.371.
2. n = 2 × (1.96 + 0.842)² × 1.371² / -1.090² = 24.8, rounded up to 25 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 <- -1.09                      # observed difference in means
sigma <- 1.37084929851089                     # 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(1630135)
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 = 25 (80% power), 34 (90% power)
 75% of observed effect: n per arm = 45 (80% power), 60 (90% power)
 50% of observed effect: n per arm = 100 (80% power), 133 (90% power)
power.t.test: 25.8 
Simulated power at n = 25 per arm: 0.787 (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's reported analysis is an ANCOVA, with a least-squares mean difference of -1.089, a 95% confidence interval from -1.41 to -0.76, and p < 0.001. This is the trial's posted model-based comparison. The registry record does not provide the covariates or enough detail to reproduce the ANCOVA from participant-level data.

The separate recalculation uses the posted arm summaries for a simple unadjusted comparison. It gives a mean difference of -1.09 and an unadjusted p-value of <0.001. Because it is unadjusted and based on summary statistics, it is not a substitute for the ANCOVA. It does not incorporate any model adjustments that may have been used in the reported analysis.

The registry does not flag a discrepancy between the recalculation and the reported result. Their estimates and p-values are broadly consistent, while their methods differ. For interpretation of the trial's stated analysis, the reported ANCOVA result remains the appropriate result to describe; the recalculation is only a simple check against the posted summaries.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists 16 outcome measures in total and 15 posted secondary outcomes. These include additional assessments of the three-symptom nasal score, the four-symptom nasal score, individual nasal symptoms, and total ocular symptom score. The primary endpoint is the mean change from baseline in 3TNSS over the entire treatment period.

When several outcomes or time points are tested, the chance of at least one nominally small p-value can increase. In general, multiplicity procedures address that issue by adjusting the testing strategy or interpretation. Whether the trial prespecified a multiplicity adjustment, which outcomes were included in any such strategy, and whether any adjusted p-values were calculated are not reported in the registry record. The record reports one analysis for the primary comparison; that fact alone does not establish the full testing plan for all outcomes.

The primary population is defined as the Full Analysis Set: participants meeting the primary criteria for enrollment, without a major good clinical practice deviation, receiving at least one dose of assigned treatment, and having a 3TNSS diary assessment after receiving a dose. The endpoint description says that the period mean used available diary data and averaged non-missing observations during the period.

The registry record does not report the amount or pattern of missing diary data, a method for imputing missing values, or sensitivity analyses for missingness. It also does not report whether the primary analysis population was changed from the population originally specified in a protocol or statistical plan. These omissions limit assessment of how missing data and analysis-population choices may have affected the estimate.

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 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse events affecting 0 of 131 participants in the fluticasone furoate group and 0 of 130 participants in the placebo group. These figures describe the number of participants recorded as affected by a serious adverse event, rather than the total number of all adverse events. No serious adverse-event terms are listed.

Other adverse events are reported separately: 23 of 131 participants in the fluticasone furoate group and 25 of 130 participants in the placebo group were affected. Those counts should not be combined with the serious adverse-event counts or treated as a description of event severity. The record does not list the individual terms for those other adverse events in the registry record for this analysis.

Death counts are not reported in the registry record. The available counts do not, on their own, establish whether the groups differ in safety, and they do not describe the nature, timing, duration, or severity of individual events. Safety interpretation is limited to the adverse-event figures recorded in the registry.

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

The primary result is a comparison of average change in a diary-based nasal symptom score between the randomized groups. The posted group estimates are negative in both arms, and the reported ANCOVA difference is also negative. With the comparison defined as fluticasone furoate relative to placebo, the estimate indicates a larger mean decrease in 3TNSS in the active-treatment group over the 2-week period. The 95% confidence interval and p-value describe statistical uncertainty and evidence for a group difference under the reported analysis.

This finding is about a group average on a particular score, not a prediction for any one participant. It does not show that every participant's symptoms changed in the same direction, and the group means do not reveal the distribution of individual responses. The registry record does not report a threshold for a meaningful 3TNSS difference, so statistical evidence for a difference should not be equated with a conclusion about practical importance.

The endpoint includes sneezing, rhinorrhea, and nasal congestion, but it does not capture every aspect of perennial allergic rhinitis or of participants' experience. Secondary outcomes cover other symptom measures, but their results are not part of the primary comparison. The result also pertains to the trial's stated study population, treatment, comparator, and assessment period; the registry record does not establish how the findings apply beyond those circumstances.

Randomization and masking are important design features for comparative interpretation, but the registry record does not report the detailed randomization procedure, the identities of those masked, or all analysis choices. The reported result therefore supports a statistical description of the posted comparison while leaving questions about model details, missing data, and broader interpretation unresolved. It should not be read as a complete account of every outcome or as individual medical advice.

17. Why This Trial Matters Statistically

This study illustrates how a repeated symptom diary can be converted into a continuous endpoint for group comparison. The 3TNSS combines three distinct nasal symptoms on a bounded scale, and the endpoint definition specifies how baseline and treatment-period averages were constructed. Clear definitions matter because the estimated change depends on which days contribute to baseline and how non-missing diary observations are summarized.

The result also illustrates the distinction between a model-based analysis and a simple recalculation. The registry reports an ANCOVA least-squares mean difference, while the unadjusted check uses the posted summaries. Their close numerical agreement is reassuring as a broad cross-check, but they are not identical procedures and the original model cannot be reproduced without its covariates and further analysis details.

Precision deserves attention alongside the point estimate and p-value. The reported confidence interval gives a range of values compatible with the ANCOVA analysis and excludes zero, but the registry does not state what difference would be important to participants. The statistical evidence therefore answers a question about the group contrast more directly than it answers a question about practical significance.

Finally, the record's missing information demonstrates why registry results require careful boundaries. The FAS and available-diary approach are described, but missing-data methods, multiplicity handling, and the original design justification are not reported. Those details affect how fully readers can evaluate the analysis and its scope, even when the posted primary result is clear.

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 does not report the ANCOVA covariates, so the model-based estimate cannot be fully reproduced or its adjustment assessed. The original statistical analysis plan, sample-size justification, and interim-analysis plans are also not reported. The planning sample-size illustrations are based on the observed effect and should not be mistaken for the trial's original design calculations.

Although the endpoint definition describes averaging available non-missing diary data, the amount and pattern of missing diary data are not reported. The record does not specify any imputation approach or sensitivity analysis. It also does not report a prespecified multiplicity strategy for the primary and secondary outcomes, despite listing 15 secondary outcomes and 16 outcome measures in total.

The posted endpoint summaries do not show the distribution of individual participant changes, and the record does not provide a threshold for interpreting the practical importance of a change in 3TNSS. The registry also does not provide detailed information about the randomization procedure, the people included in quadruple masking, or the individual terms for the non-serious adverse events. Death counts are not reported in the registry record.

These limitations restrict independent assessment of model choices, missing-data sensitivity, multiplicity, and the breadth of outcome interpretation. They do not change the reported ANCOVA estimate, but they identify areas where the registry entry does not support more detailed conclusions.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Study FFR116364, a Placebo-controlled Study of GW685698X in Paediatric Subjects With Perennial Allergic Rhinitis — NCT01630135.
• Registry record last updated 2017-01-11; status last verified 2016-11.
• Results first posted 2013-05-01.
• 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

NCT01630135 is a completed Phase 3 randomized, quadruple-masked, parallel-group, placebo-controlled study sponsored by GlaxoSmithKline. The registry lists 261 participants enrolled and records the study start as 2012-06 and primary completion as 2012-08.

The primary endpoint was mean change from baseline in 3TNSS over the 2-week treatment period among the Full Analysis Set. The posted arm changes were -1.98 for fluticasone furoate 55 µg per day and -0.89 for placebo. The reported ANCOVA difference was -1.089, with a 95% confidence interval from -1.41 to -0.76 and p < 0.001.

The registry also reports 0 participants with serious adverse events in either arm. Further detail on the original statistical plan, missing-data handling, multiplicity, and several design and safety features is not reported in the registry record.