A Randomized, Double-Blind, Placebo-Controlled Study to Evaluate the Efficacy and Safety of Fluticasone Furoate Nasal Spray for 2 Weeks in Chinese Adult and Adolescent Subjects With Allergic Rhinitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Randomized, Double-Blind, Placebo-Controlled Study to Evaluate the Efficacy and Safety of Fluticasone Furoate Nasal Spray for 2 Weeks in Chinese Adult and Adolescent Subjects With Allergic Rhinitis (NCT01231464), a randomized phase 3 study. The primary outcome is Mean Change From Baseline Over the Entire Treatment Period in the Daily Reflective Total Nasal Symptom Score (rTNSS).

2. Clinical Question

3. Trial Design

This was a Phase 3 randomized, parallel-group study of fluticasone furoate nasal spray and vehicle placebo in Chinese adult and adolescent subjects with allergic rhinitis. The registry reports enrollment of 365. The title describes the study as double-blind, while the masking field records triple masking. The specific masked roles are not reported in the registry record, so the masking description cannot be elaborated beyond those labels.

The brief summary describes three stages: a 2 to14 day run in, a 2 week treatment period, and a 3 to 5 day follow up. After run in, at least 360 subjects were to be randomized to FFNS 110ug once daily or placebo for 2 weeks. These are descriptions in the brief summary; the trial's own original sample-size assumptions and justification are not reported in the registry record. The registry lists the study as completed, with a start in 2009-09 and primary completion in 2010-05.

The primary outcome record gives analysis results for FFNS 110 mcg and placebo. Its arm-level analysis counts are 179 and 178, respectively. These are the participants included in the reported primary outcome analysis, which required baseline and post-baseline observations. They should not be treated as a complete account of every participant enrolled or randomized. The adverse-event record uses different at-risk counts, 181 for FFNS 110 mcg and 182 for placebo, reflecting the populations identified for those safety summaries.

The registry lists two secondary outcomes: change in nasal finding score by rhinoscopy and change in severity of overall interference in activities of daily living. Reflective total ocular symptom score was described as an exploratory endpoint for participants with severe ocular symptoms. Results for that exploratory measure are not included in the registry record for this analysis.

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 who were randomized and received at least one dose of study medication. Only participants for whom both baseline and post-baseline data were available were included in this analysis.

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 Over the Entire Treatment Period in the Daily Reflective Total Nasal Symptom Score (rTNSS), measured over Baseline through entire treatment period (Day 1 through Day 14).

The primary endpoint was mean change from baseline over the entire treatment period in daily reflective total nasal symptom score, or rTNSS. The assessment covered Day 1 through Day 14. TNSS combines participant ratings for rhinorrhea, nasal congestion, nasal itching, and sneezing. Each symptom is rated from 0=None to 3=Severe, and the total score can range from 0-12. The daily reflective score averages morning and evening assessments, each of which covers symptoms over the preceding 12 hours.

Change was defined as treatment-period rTNSS minus baseline rTNSS. Thus, a negative mean change indicates a decrease in the score relative to baseline. The posted group means are -4.226 for FFNS 110 mcg and -2.728 for placebo. The endpoint is useful for comparing changes in a common, participant-assessed symptom measure across randomized groups during the treatment period. The average offers a compact summary across several nasal symptoms and repeated daily assessments, rather than describing any single symptom or any one participant's experience.

The primary outcome is a continuous measure, so its mean difference describes the average contrast in score changes. It does not show the distribution of individual responses, the share of participants with a particular degree of improvement, or the clinical importance of a given score change. Those interpretations require additional information, including a threshold for meaningful change, which is not reported in the registry record.

7. Results Posted in the Registry

• FFNS 110 mcg: mean -4.23 versus -2.73 on control; difference -1.50 (95% CI -1.96 to -1.04), standardized effect d = -0.68; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: it is a score, and the analysis compares mean changes between randomized groups. A mean summarizes the central level of a set of observations, but does not describe the full spread or shape of individual values. The registry reports group least-squares means and a least-squares mean difference. Least-squares means are model-based estimates; they need not be identical to raw arithmetic averages. The registry identifies ANCOVA as the sponsor's method but does not report the covariates, model details, or diagnostics.

What does the change score represent?

For each participant, the change measure compares the treatment-period rTNSS with baseline rTNSS. The outcome description defines change as treatment-period score minus baseline score. Negative changes therefore indicate lower scores during treatment than at baseline. The group comparison then contrasts the average changes between FFNS 110 mcg and placebo. It is a difference in changes, not a comparison of only the scores at the end of treatment.

The daily rTNSS averages morning and evening symptom assessments. Each assessment asks about symptoms over the preceding 12 hours, and the daily score combines ratings for rhinorrhea, congestion, itching, and sneezing. Repeated measurement can represent symptom experience across the treatment period, while also creating a need to specify how measurements are aggregated and how incomplete records are handled. The registry describes the daily score and treatment-period change, but detailed rules for missing daily assessments are not reported in the registry record.

Why use ANCOVA?

In general, ANCOVA compares group outcomes while accounting for selected covariates in a regression model. For a change-from-baseline endpoint, models may include baseline outcome values, but the particular covariates used in this trial are not reported in the registry record. Adjusting for covariates can affect the estimated contrast and its standard error. The trial's reported least-squares mean difference and confidence interval should therefore be understood as model-based results, not as a simple subtraction of two unadjusted raw averages.

A recalculation from posted summary statistics is a different kind of estimate. It uses the available group means and standard errors to form a simple unadjusted comparison. In general, such a recalculation can help readers check the broad scale and direction of a reported contrast, but it does not reproduce an ANCOVA model when its covariates and details are unavailable. Here, the recalculated point difference is -1.498, with a 95% interval from -1.956 to -1.04 and p <0.001. The sponsor's reported ANCOVA difference is -1.498, with a 95% interval from -1.897 to -1.009 and p < 0.001.

What does the p-value say?

In general, a p-value measures how compatible the observed data are with a specified null hypothesis under the assumptions of the analysis. A small p-value is not the probability that the null hypothesis is true, and it does not quantify the chance that a result is due to bias or error. The posted sponsor analysis reports p < 0.001, while the unadjusted recalculation reports p <0.001. Both are reported as less than their respective thresholds; they are not exact p-values. The p-value should be interpreted alongside the estimated difference and its confidence interval, not as a measure of clinical importance.

Why does over-dispersion matter in general?

For count outcomes, over-dispersion means that observed variability exceeds what a simple count model assumes. It can make standard errors too small if the model does not allow for the extra variability. This primary endpoint is a continuous score rather than a count outcome, and the registry does not report an over-dispersion analysis. The general concept is not a claim about the model used in this trial.

The registry identifies the primary analysis population as randomized participants who received at least one dose, with both baseline and post-baseline data available for this analysis. This population definition indicates that not every randomized participant necessarily contributes to the primary outcome estimate. The record does not report the missing-data strategy, sensitivity analyses, or whether the analysis model included a particular baseline adjustment. These methodological details would be needed to assess the robustness of the estimate fully.

In general, randomization supports a comparison between assigned groups by reducing systematic differences at allocation, while masking can reduce some risks of expectation-related bias. The record identifies the study as randomized and parallel-group. It labels masking as triple, whereas the title calls it double-blind; the specific masked roles and procedures are not reported in the registry record. Neither design label by itself establishes that all potential sources of bias were eliminated.

10. Confidence Intervals and Precision

The sponsor's ANCOVA estimate of the mean difference is -1.498 points, with a 95% confidence interval from -1.897 to -1.009. In general, a confidence interval expresses a range of values compatible with the data and the assumptions of the statistical method. Its width reflects statistical precision: a narrower interval indicates less sampling uncertainty under those assumptions. It is not a range containing 95% of individual participant outcomes, nor does it account automatically for bias, missing data, or model misspecification.

The sponsor interval lies below zero throughout. For this comparison, zero would correspond to no difference in mean change between the groups. The interval therefore conveys both the direction of the estimated contrast and uncertainty around its magnitude. The interval still includes a range of possible differences, rather than establishing one exact treatment effect. Whether values in that range are important to participants cannot be judged from the confidence interval alone; a meaningful-change threshold is not reported in the registry record.

The separate unadjusted recalculation gives a 95% interval from -1.956 to -1.04. This interval is not identical to the sponsor's ANCOVA interval, even though both estimates of the difference are -1.498 at the reported precision. The difference reflects that the methods are not the same: one is a recalculation from summary statistics, and the other is the sponsor's model-based ANCOVA analysis. The recalculated interval should not replace the trial's reported analysis.

The group-level outcome standard errors are reported as 0.165 and 0.166, but those values are omitted because they exceed the permitted display precision. The registry outcome counts are 179 for FFNS 110 mcg and 178 for placebo. These counts and the confidence intervals help describe precision, but do not resolve the unreported details about missing observations, model specification, or the representativeness of the analysis population.

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 participants per arm, not total events. They illustrate the group sizes that a hypothetical future study would require under the stated assumptions, using fractions of the observed effect as planning targets. For 100% of the observed effect, the illustration gives 35 participants per arm for 80% power and 46 participants per arm for 90% power. For 75% of the observed effect, it gives 61 participants per arm for 80% power and 82 participants per arm for 90% power. For 50% of the observed effect, it gives 137 participants per arm for 80% power and 183 participants per arm for 90% power.

In general, power is the probability that a study will reject the null hypothesis under a specified alternative, if the assumptions used in planning are correct. The 80% and 90% columns describe different target probabilities of detecting the specified effect under those assumptions; the higher target generally requires more participants. The calculations assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect. Smaller effects need more participants to detect with the same target power, as illustrated by the increasing planning figures for smaller fractions.

These calculations are planning illustrations based on the observed effect, not the trial's original design justification. The trial's own sample-size assumptions and justification are not reported in the registry record. The figures are not observed power: they do not describe the probability that this completed trial detected its result. Actual study results depend on the observed data and the analysis, while planning illustrations depend on assumptions about the effect, variability, allocation, and statistical test.

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 = -1.498; pooled standard deviation σ = 2.206.
2. n = 2 × (1.96 + 0.842)² × 2.206² / -1.498² = 34.0, rounded up to 35 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 <- -1.498                      # observed difference in means
sigma <- 2.20578154785216                     # 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(1231464)
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 = 35 (80% power), 46 (90% power)
 75% of observed effect: n per arm = 61 (80% power), 82 (90% power)
 50% of observed effect: n per arm = 137 (80% power), 183 (90% power)
power.t.test: 35 
Simulated power at n = 35 per arm: 0.802 (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 posted primary analysis is an ANCOVA comparison, reported as a least-squares mean difference of -1.498 points. Its 95% confidence interval is -1.897 to -1.009, and the reported p-value is < 0.001. The group means in the primary outcome record are -4.226 for FFNS 110 mcg and -2.728 for placebo.

The simple recalculation from the posted summary statistics also gives a mean difference of -1.498, but its 95% confidence interval is -1.956 to -1.04 and its unadjusted p-value is <0.001. It is an unadjusted calculation, not a recreation of ANCOVA. The different intervals and p-value reporting reflect the use of different inferential methods and reporting precision. The registry entries do not indicate a discrepancy flag for this comparison.

The registry's reported analysis is the trial's own model-based result. The recalculation can provide a separate check using the posted group summaries, but it cannot recover covariates or other model details that are not reported in the registry record. Readers should distinguish the sponsor's inferential result from this unadjusted recalculation rather than treating the two confidence intervals as interchangeable.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome analysis and two secondary outcomes. It also describes reflective total ocular symptom score as an exploratory endpoint for a specific group of participants with severe ocular symptoms. The numerical results for the secondary outcomes and exploratory endpoint are not reported in the registry record. The record does not state whether the primary or secondary analyses were adjusted for multiple comparisons, or whether any hierarchy or other multiplicity procedure was prespecified.

In general, testing multiple outcomes or multiple comparisons can increase the chance of at least one apparently positive finding if the tests are treated as independent opportunities. A multiplicity strategy can control or otherwise address that concern. Without a reported strategy, the p-value for the primary outcome should be interpreted as the result of its stated analysis, not as evidence that all outcomes were tested under a particular family-wide error control plan. No adjustment should be assumed.

The primary analysis population was the Full Analysis Set, defined as participants who were randomized and received at least one dose. Only participants with both baseline and post-baseline data were included in the primary analysis. The record does not say how many participants were excluded for missing outcome data, what pattern of missingness occurred, or how missing observations were handled. It also does not report sensitivity analyses to assess the effect of missing data.

In general, missing-data methods rely on assumptions about why outcomes are unavailable and how missing outcomes relate to observed data. Different assumptions can lead to different estimates. The registry record does not identify such assumptions or methods, so the robustness of the primary result to missingness cannot be assessed from these registry entries alone. The record also does not describe an intention-to-treat analysis including all randomized participants regardless of post-randomization treatment exposure or outcome availability.

The reported primary outcome population and the at-risk populations used for adverse-event summaries are not identical in their stated counts. That difference is not itself evidence of an error: outcome availability and safety summaries may use different analysis populations. The registry does not explain the detailed relationship between these populations or give a participant flow account in the registry record for this analysis.

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 serious adverse event record reports 0 affected participants among 181 at risk in the FFNS 110 mcg group and 0 among 182 at risk in the placebo group. These are the posted serious adverse event counts for the groups. The registry does not list serious adverse event terms among the top serious terms. Death counts are not reported in the registry record.

The record also gives 13 participants with other adverse events among 181 at risk in the FFNS 110 mcg group, and 9 among 182 at risk in the placebo group. These are separate from the serious adverse event figures. The registry entries do not provide event types, severity breakdowns, duration, or an analysis comparing the groups, so those details should not be inferred from the counts alone.

In general, an observed count of zero does not establish that an event is impossible or that a treatment is free of risk. Safety summaries are also influenced by the number of participants observed, the duration of observation, and the definitions used to classify events. The brief summary states that safety measures included adverse event reports, ECGs, physical exams, vital signs, and nasal examination, but numerical results for those assessments are not reported in the registry record.

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

The primary result describes an average difference in change in daily reflective total nasal symptom score between the randomized FFNS 110 mcg and placebo groups. The sponsor's estimated difference is -1.498 points, with a 95% confidence interval from -1.897 to -1.009. The score combines four nasal symptoms, and negative changes indicate lower scores relative to baseline. The result therefore supports a difference in average score change in the posted analysis, with the interval indicating uncertainty around its estimated size.

This does not establish the experience of every participant. Group means can conceal substantial variation, and the registry does not report individual response patterns or the proportion reaching a particular improvement threshold. Nor does the score alone describe all aspects of allergic rhinitis, daily functioning, nasal examination findings, or ocular symptoms. The record lists secondary measures of rhinoscopy findings and interference with activities of daily living, but their numerical results are not reported in the registry record.

The p-value does not measure how important the difference is to participants, and a confidence interval does not establish a clinical threshold. A threshold for meaningful change in rTNSS is not reported in the registry record. Therefore, the statistical evidence for a difference and the clinical importance of its size are separate questions. The registry entries support describing the estimate and its precision, but not assigning a value judgment to the magnitude.

The findings are specific to the study population and design: Chinese adult and adolescent subjects with allergic rhinitis, a 2-week treatment period, and comparison with vehicle placebo. The registry entries do not establish effects beyond the recorded population, duration, outcome, or comparison. The study's safety counts provide a limited description of reported events during the recorded observation, not a complete characterization of all possible harms.

These results should be read as trial findings rather than medical advice. They describe what the registry reports about a group-level outcome and its uncertainty. They do not determine the best treatment for an individual, nor do they answer questions not measured or reported in the registry record.

17. Why This Trial Matters Statistically

This trial illustrates why a treatment comparison needs more than a p-value. The primary analysis reports a mean difference, a confidence interval, and a p-value. The estimate describes the scale and direction of the average contrast; the interval describes statistical precision under the analysis assumptions; and the p-value addresses compatibility with a null comparison. None of these alone establishes the clinical importance of the result.

It also shows why model-based trial results should be distinguished from simple recalculations. The sponsor reports ANCOVA, while the separate calculation uses posted summary statistics without adjustment. The point estimates agree at the reported precision, but the confidence intervals differ. Recognizing the estimand and method behind each result prevents an unadjusted check from being mistaken for a reproduction of the sponsor's analysis.

The endpoint itself involves a meaningful measurement choice: four participant-rated nasal symptoms are combined and assessed repeatedly through morning and evening reports. This creates a summary useful for group comparison, while leaving questions about symptom-specific effects and individual responses unanswered. The record's additional outcomes could add context, but numerical results are not reported in the registry record.

Finally, the record highlights the importance of transparent analysis populations and missing-data methods. The primary outcome analysis required baseline and post-baseline data, but the handling and extent of missingness are not reported. The interpretation of an estimate depends not only on its arithmetic but also on who contributes data and on the assumptions behind the model. Clear reporting of those details helps readers judge how securely the statistical result answers the trial question.

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 covariates or detailed specification of the sponsor's ANCOVA model. It also does not give diagnostics, sensitivity analyses, or information sufficient to reproduce the model-based confidence interval. The unadjusted recalculation is therefore informative only as a separate summary-statistic analysis, not as an independent verification of all aspects of the sponsor's model.

Missing-data handling is not described. The primary analysis required both baseline and post-baseline observations, but the number and pattern of missing outcomes and the method used to address them are not reported in the registry record. The analysis population may consequently differ from the full randomized population, and the available information does not show how sensitive the result is to that distinction.

The registry gives a 2-week treatment period. This result alone does not describe outcomes over longer periods. The record lists secondary outcomes and an exploratory ocular symptom measure, but their numerical results are not included in the registry record for this analysis. It also does not report a threshold for a meaningful change in rTNSS or results showing the distribution of individual changes. The average difference therefore cannot establish how many participants experienced a change of a particular size or whether the magnitude is meaningful to participants.

The title calls the trial double-blind, while the masking field records triple masking. The specific roles masked and details of the masking procedure are not reported in the registry record. The brief summary describes the intended regimen and study stages, but the trial's own sample-size justification and original design assumptions are not reported. Finally, the posted serious adverse event summary gives zero affected participants in each group, but deaths are not reported and the available safety details do not include a detailed account of event types or follow-up duration.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Randomized, Double-Blind, Placebo-Controlled Study to Evaluate the Efficacy and Safety of Fluticasone Furoate Nasal Spray for 2 Weeks in Chinese Adult and Adolescent Subjects With Allergic Rhinitis — NCT01231464.
• Registry record last updated 2017-07-06; status last verified 2017-06.
• Results first posted 2011-02-04.
• 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

NCT01231464 is a completed Phase 3 randomized, parallel-group study of fluticasone furoate nasal spray and vehicle placebo in Chinese adult and adolescent subjects with allergic rhinitis. The brief summary describes a 2-week treatment period. The primary outcome was mean change from baseline in daily reflective total nasal symptom score over Day 1 through Day 14.

The sponsor-reported ANCOVA comparison gives a least-squares mean difference of -1.498 points, with a 95% confidence interval from -1.897 to -1.009 and p < 0.001. The separate unadjusted recalculation gives the same point estimate at the reported precision, with a 95% interval from -1.956 to -1.04 and p <0.001. These are different analyses and should be read separately.

The registry lists two secondary outcomes, but their numerical results are not reported in the registry record. The primary analysis population required baseline and post-baseline data; missing-data methods and the trial's original sample-size assumptions are not reported in the registry record. Serious adverse event counts were 0 in both posted groups, while death counts are not reported.