A Study to Evaluate the Safety and Effectiveness of a Nasal Spray to Treat Seasonal Allergies: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate the Safety and Effectiveness of a Nasal Spray to Treat Seasonal Allergies (NCT00824473), a randomized phase 3 study. The primary outcome is Change From Baseline in 12-hour Reflective Total Nasal Symptom Score(rTNSS)for the Entire 14-day Study Period Compared to Placebo (AM and PM Combined)at 14 Days.

2. Clinical Question

The primary symptom score captures runny nose, itchy nose, nasal congestion, and sneezing. Each symptom is rated from 0 to 3, with higher ratings corresponding to greater symptom severity. The measure is intended to represent participants’ recalled nasal symptoms over a 12-hour period, rather than symptoms at a single instant. The listed primary endpoint concerns change from baseline across the study period, so interpretation depends on both the participants’ initial scores and their subsequent symptom reports.

This question is comparative: it concerns the difference in symptom-score change between the medication group and the placebo group, not simply whether either group’s score changed over time. The trial’s stated purpose was to determine whether the medication was more effective than placebo alone. The registry record identifies the outcome and the comparison but does not include the primary outcome values, an estimated between-group difference, or its uncertainty. Therefore, the question can be described, but the result of the primary comparison cannot be determined from the information reported.

3. Trial Design

This was a randomized, double-blind, placebo-controlled, parallel-group phase 3 trial sponsored by Meda Pharmaceuticals. The record lists seasonal allergic rhinitis as the condition and 506 as the enrollment. The study is marked completed. Its recorded start was 2008-12, and its primary completion was 2009-02.

Random allocation is intended to create groups that are comparable, on average, with respect to both measured and unmeasured characteristics. It does not guarantee that the groups will be identical. A placebo comparator allows the observed outcome in the active-treatment group to be considered alongside the outcome in a group receiving placebo, under the trial conditions. The parallel-group structure means participants were assigned to a study arm rather than described as switching between the two interventions.

The record describes masking as quadruple and the study as double-blind in its official title. It does not identify which specific roles were masked, so the exact people or groups whose knowledge of assignment was restricted are not reported in the registry record. The masking description is relevant because knowledge of assignment can influence participants’ reports, care, or outcome assessment, but the record does not detail procedures used to maintain masking or whether masking was assessed.

The listed arms were placebo and 0.15% azelastine hydrochloride. The safety counts list 255 participants at risk in the placebo group and 251 in the Astepro 0.15% group. Those counts describe the denominators used for the posted safety figures; the registry record does not state whether they are also the analysis populations for the primary endpoint.

The study was designed to evaluate safety and effectiveness, but the primary outcome description available in the record does not include the numerical comparison. The record also does not report the planned sample-size justification, allocation ratio, randomization method, concealment procedures, or detailed protocol. These omissions limit reconstruction of the design and analysis beyond the features explicitly recorded.

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: Subjects participated in a 7 day screening period to Identify appropriate subjects based on symptom scores.

Overall Study

Reasons for not completing this period:

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 12-hour Reflective Total Nasal Symptom Score(rTNSS)for the Entire 14-day Study Period Compared to Placebo (AM and PM Combined)at 14 Days, measured over baseline and 14 days.

The primary endpoint is change from baseline in the 12-hour reflective Total Nasal Symptom Score (rTNSS) for the entire 14-day study period, comparing 0.15% azelastine hydrochloride with placebo. The score covers runny nose, itchy nose, nasal congestion, and sneezing. Each symptom is rated on a scale from 0-3: 0 means none, 1 mild, 2 moderate, and 3 severe. The maximum is 12 points per assessment, and the total possible score is 24 per day.

The endpoint is based on reflective reports of symptoms over a 12-hour interval, with morning and evening assessments combined. It is used to represent symptom burden over repeated assessments rather than relying on a single observation. Because the score assigns higher values to more severe symptoms, a decrease in the score corresponds to a decrease in recorded symptom severity. The treatment comparison, however, requires comparing change between randomized groups; within-group change alone would not establish a difference between treatments.

The registry description says least-squares means were controlled for study day as a within-patient effect, treatment group and site as between-patient effects, the treatment-by-study-day interaction, and baseline as a covariate. This describes an adjusted approach to repeated symptom assessments. The numerical primary outcome result, including the estimated contrast and its precision, is not reported in the registry record, so the magnitude and uncertainty of any group difference cannot be evaluated.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a least-squares mean analysis for the primary symptom outcome. The following explanations are general statistical interpretation of the methods named in the record, not additional details about how the trial was conducted.

What does a least-squares mean represent?

A least-squares mean is an adjusted mean from a statistical model. Unlike a simple arithmetic mean, it reflects the model’s treatment of other included variables. Here, the record says the analysis controlled for baseline and included treatment group and site as between-patient effects. The intended role of adjustment is to estimate treatment-group means while accounting for those model terms. The record does not report the resulting adjusted means or the between-group contrast, so their values cannot be interpreted.

Adjusted means are model-based summaries, not direct counts or necessarily the same as the raw averages calculated from all recorded observations. Their interpretation depends on the model specification, the outcome data available to the model, and the way missing observations were handled. The registry description names some model terms but does not provide a complete statistical analysis plan, details of model fitting, or the handling of incomplete symptom records.

Why include study day and a treatment-by-day interaction?

General explanation: repeated symptom assessments from the same participant tend to be related. A participant’s ratings at different times are not independent in the way measurements from unrelated people might be. A model that accounts for study day can represent changes over time. A treatment-by-study-day interaction allows the pattern over time to differ by treatment group, rather than requiring the groups to follow the same time pattern. The registry says study day was treated as the within-patient effect and lists this interaction, but it does not provide the model’s covariance structure or other technical specifications.

Accounting for repeated observations matters because treating correlated measurements as independent can make uncertainty estimates misleading. The appropriate approach depends on the pattern of correlation and the model assumptions. Those assumptions, diagnostics, and sensitivity analyses are not reported in the registry record. The listed repeated-measures structure explains the general intent of the model, but it is not enough to reconstruct the analysis in full.

What is the role of baseline adjustment?

General explanation: baseline adjustment accounts for participants’ starting outcome values when estimating differences after baseline. In a randomized study, baseline adjustment can improve precision and address chance imbalances in starting scores. The registry specifically identifies baseline as a covariate. The outcome is described as change from baseline, but the model description also says baseline was included as a covariate. Without the complete model and result output, it is not possible to determine exactly how the change outcome and baseline adjustment were implemented together.

Adjustment should not be confused with proof that the groups were identical at baseline. Randomization supports a fair comparison in expectation, while observed baseline values can still differ by chance. The record does not report baseline outcome summaries or the extent of any observed imbalance.

Why might over-dispersion matter?

General explanation: over-dispersion arises when observed variability is greater than a statistical model expects. It is particularly discussed for count outcomes, where a simple model may understate variability if outcomes are more heterogeneous than assumed. The primary outcome in this trial is described as a continuous symptom score, and the registry names least-squares means rather than a count model. The record does not report an over-dispersion assessment. This concept is therefore a general statistical consideration, not a reported feature of this trial’s primary analysis.

More broadly, model assumptions matter because uncertainty estimates depend on how well the model represents the data. For a repeated symptom score, relevant considerations can include the distribution of scores, repeated-measure correlation, and the form of the time effect. The registry entry does not report diagnostics, alternative models, or sensitivity analyses, so readers cannot assess those aspects from the recorded description.

How should precision and significance be interpreted?

General explanation: an estimated treatment contrast summarizes the observed difference between groups on the chosen scale. A confidence interval expresses uncertainty around that estimate under the analysis assumptions. A p-value addresses how compatible the data are with a specified null hypothesis under a statistical model; it does not measure the size or practical importance of an effect. Interpreting an outcome requires considering the estimate, uncertainty, outcome scale, and study context together.

No numerical primary contrast, confidence interval, or p-value is reported in the registry record. Consequently, there is no basis in the recorded details for assessing precision, statistical significance, or the size of the primary difference. General explanations of these statistics cannot fill in the absent result.

What can be inferred from the named model terms?

The listed terms establish that the primary outcome analysis was intended to account for treatment assignment, site, baseline, study day, and treatment-related differences in the time pattern. They do not establish the sample-size rationale, the precise analysis population, imputation rules, multiplicity adjustment, interim monitoring, or the handling of protocol deviations. Those matters are not reported in the registry record. The model description should therefore be read as a partial account of the analysis, not as a complete protocol or a substitute for numerical results.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures and five analyses. The secondary outcome descriptions include morning instantaneous nasal symptom scores, combined morning and evening nasal symptom scores, ocular symptom scores, a quality-of-life measure for participants 18 years of age and older, and direct visual nasal examinations. This indicates that the study considered multiple outcomes and assessment approaches. The record does not explain how the listed measures correspond to the analysis count, which analyses were primary or secondary in the prespecified plan, or whether any outcomes were added or changed during the study.

General explanation: when several outcomes or comparisons are examined, the chance of finding at least one apparently notable result can increase if each is interpreted without accounting for the collection of tests. Multiplicity procedures can control a chosen error rate or prioritize an outcome hierarchy. The registry record does not report whether such procedures were planned or used. No conclusion about adjustment, or the absence of adjustment, can be drawn from the number of outcomes alone.

The record describes a least-squares mean model for the primary endpoint but does not report a full statistical analysis plan. It does not report whether interim analyses were conducted, whether stopping rules were used, or how the primary analysis population was defined. These details are not reported in the registry record.

Missing-data handling is also not reported. Symptom diaries and repeated assessments can have incomplete observations, but the registry entry does not say whether data were missing, how missing values were treated, or whether sensitivity analyses assessed the effect of missingness. It would be inappropriate to assume that all participants completed every assessment or that a particular imputation method was used.

The record does not identify an intention-to-treat, per-protocol, or other analysis population for the primary outcome. The safety denominators do not establish the efficacy analysis population. Without explicit population definitions, missing-data methods, and prespecified multiplicity procedures, the analysis cannot be fully reconstructed from the registry record.

11. 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 2% reporting threshold.

Most frequent non-serious adverse event terms

The posted safety figures list 255 participants at risk in the placebo group and 251 in the Astepro 0.15% group. Serious adverse events are reported for zero participants in each group. The entries for other adverse events report two affected participants in the placebo group and 25 in the active-treatment group.

These are counts, not rates or estimates of comparative risk. They should be read alongside the group denominators and the category labels. The record does not give descriptions of the events counted as other adverse events, the criteria used to classify seriousness, the timing of events, or details of how events were collected. It also does not report a statistical comparison of these safety counts. The figures therefore do not establish whether events were related to treatment or explain the difference in the reported counts.

Deaths are listed as null for both groups; death information is not reported in the registry record. A reported count of zero serious adverse events means none were recorded in the posted serious-event fields, not that the record supplies a detailed assessment of all possible harms. Safety interpretation is limited by the absence of event descriptions and follow-up details.

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

The registry record establishes that a randomized, double-blind, placebo-controlled phase 3 study compared 0.15% azelastine hydrochloride with placebo in participants with seasonal allergic rhinitis. It identifies the primary endpoint as change from baseline in the 12-hour reflective Total Nasal Symptom Score over the 14-day study period, with morning and evening assessments combined. It also describes the score’s component symptoms and the intended adjusted analysis.

Those details describe the question and design; they do not provide the primary outcome result. The numerical treatment contrast and its uncertainty are not reported in the registry record. Consequently, the record cannot support a statement about the size or direction of the primary comparison, its statistical significance, or how precisely any difference was estimated. The stated study purpose is not itself evidence of the result.

The endpoint’s scale can be understood from its definitions: it reflects reported severity of four nasal symptoms, with higher item ratings indicating more severe symptoms. That information helps explain what the outcome is intended to measure. It does not establish that the measure captures every aspect of seasonal allergic rhinitis, nor does it show how participants’ scores changed during the trial.

The safety entries report zero serious adverse events in both groups and different counts in the other-adverse-event category. These counts do not describe event types, establish causation, or replace a complete safety analysis. The registry record also does not report missing-data methods, the primary analysis population, multiplicity procedures, or a detailed analysis plan. Those gaps limit assessment of how robustly outcomes were analyzed.

General explanation: a trial result is most informative when the estimated comparison is interpreted with its uncertainty, the outcome’s scale, and the design and analysis assumptions. Statistical significance alone would not establish the practical importance of a difference, and an observed difference alone would not guarantee precision. Because the primary estimate and uncertainty are absent from the record, these interpretive steps cannot be completed for the central efficacy question.

The appropriate reading is therefore bounded. The record documents the study design, endpoint definition, and selected safety counts, while leaving the primary numerical result and several analysis details unreported. It does not justify a conclusion that the intervention did or did not produce a particular symptom-score effect.

13. Why This Trial Matters Statistically

This trial illustrates why an endpoint definition and a numerical result serve different purposes. The primary outcome entry carefully describes the symptom components, rating scale, assessment schedule, baseline comparison, and adjusted model terms. That information helps readers understand what the investigators intended to measure. Yet without the treatment contrast and its uncertainty, readers cannot determine what the study found on that endpoint.

The design also illustrates the value of specifying the comparison. In a randomized placebo-controlled trial, the central question is about the difference between assigned groups, not whether symptoms changed within either group in isolation. The record identifies that comparative structure and names the model terms intended to account for repeated assessments and baseline values. It does not report the estimated comparison, which is necessary to interpret the efficacy question.

Repeated symptom measurements raise statistical issues that are common in clinical research. Reports from the same participant are related, and treatment effects may vary over time. The registry’s description of study day and a treatment-by-study-day interaction makes those modeling considerations visible. At the same time, the absence of full model details, diagnostics, and missing-data methods shows why a brief model description is not equivalent to a reproducible analysis account.

The multiple outcome measures also emphasize the importance of transparent plans for interpreting several comparisons. Different symptom domains and assessment methods may provide complementary information, but examining multiple outcomes can complicate interpretation. The registry record does not report how multiplicity was managed or which analysis hierarchy was prespecified.

Finally, the posted safety counts show why category definitions and denominators matter. Counts without event descriptions or analysis details are informative as recorded observations but do not answer questions about event type, causality, or comparative risk. Across efficacy and safety, transparent numerical estimates, uncertainty, analysis populations, and methods are needed to move from a study description to a statistically interpretable result.

14. 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 main limitation for interpreting the primary question is that the numerical primary outcome comparison is not reported in the registry record. The outcome definition is available, but no group estimates, between-group contrast, confidence interval, or p-value is included in the recorded details. The magnitude, direction, and precision of the primary result therefore cannot be assessed.

The statistical description is partial. It names least-squares means and several model terms, including study day, treatment group, site, a treatment-by-study-day interaction, and baseline. It does not provide the complete model specification, assumptions, diagnostics, covariance structure, or sensitivity analyses. The registry record also does not report how missing observations were handled or which participants were included in the primary efficacy analysis.

Although multiple outcome measures and analyses are listed, the record does not describe the prespecified hierarchy, multiplicity strategy, or relationship between the listed outcomes and analyses. It does not report whether interim analyses were conducted. No assumptions about these features should be made from the trial’s phase or design label.

The record gives limited operational detail. It does not report the dosing schedule, eligibility criteria beyond the condition, randomization method, allocation concealment procedures, or which roles were masked. The official title describes the study as double-blind while the design field lists quadruple masking, but the exact masked parties and procedures are not reported.

Safety information is also limited to selected counts. Event descriptions, classification details, timing, attribution, and a detailed safety analysis are not reported. Death information is not reported in the registry record. These limitations restrict interpretation of both benefit and harm, without altering the fact that the study is recorded as completed and randomized.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study to Evaluate the Safety and Effectiveness of a Nasal Spray to Treat Seasonal Allergies — NCT00824473.
• Registry record last updated 2010-02-23; status last verified 2010-02.
• Results first posted 2010-01-28.
• 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.

18. Record Summary

NCT00824473 is a completed randomized, double-blind, placebo-controlled, parallel-group phase 3 trial sponsored by Meda Pharmaceuticals. It enrolled 506 participants with seasonal allergic rhinitis and compared placebo with 0.15% azelastine hydrochloride. The registry record lists the study start as 2008-12 and primary completion as 2009-02.

The primary endpoint was change from baseline in the 12-hour reflective Total Nasal Symptom Score over the 14-day study period, with morning and evening assessments combined. The score covers runny nose, itchy nose, nasal congestion, and sneezing. The record describes an adjusted least-squares mean analysis, but the numerical primary comparison and its uncertainty are not reported in the registry record.

Safety entries report zero serious adverse events in each group, and other adverse events affecting two participants in the placebo group and 25 in the Astepro 0.15% group. Event details and several analysis features, including missing-data handling, the primary analysis population, and multiplicity methods, are not reported in the registry record.