A Study Investigating the Effect of Montelukast in Patients With Seasonal Allergic Rhinitis (MK-0476-192)(COMPLETED): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study Investigating the Effect of Montelukast in Patients With Seasonal Allergic Rhinitis (MK-0476-192)(COMPLETED) (NCT00960141), a randomized phase 3 study. The primary outcome is Mean Change From Baseline in Daytime Nasal Symptoms Score.

2. Clinical Question

The registry record describes a study of montelukast in patients with seasonal allergic rhinitis, with loratadine included as an active control and placebo as the other comparator. The clinical question was whether the average daytime nasal symptom score changed from baseline during treatment, and how the reported changes differed descriptively among the three treatment groups.

The score gives a concise summary of several daytime nasal symptoms rather than a result for each symptom separately. A reduction in the score corresponds to movement toward the better end of the scale. The estimates are least-squares means with 95% confidence intervals, so they describe estimated average changes and their uncertainty. They should not be read as individual patient outcomes or as proof that one arm differed from another.

3. Trial Design

This was a completed Phase 3, randomized, double-blind, parallel-group study sponsored by Organon and Co. The official title describes a multicenter study in patients with seasonal allergic rhinitis, and the registry lists enrollment of 829. The study began in 2000-08 and had primary completion in 2000-10. These record details establish the broad design and timing, but do not provide the full protocol or the operational details of how the trial was conducted.

Participants were assigned to montelukast, loratadine, or placebo. The double-blind designation indicates that masking was part of the study design; the registry information in this record does not specify who was masked or how masking was maintained. The parallel-group model means participants were assigned to a group rather than receiving each study treatment in a crossover sequence. Allocation is listed as randomized, but the method used to generate or conceal assignments is not reported in the registry record.

The posted primary outcome is mean change from baseline in the daytime nasal symptoms score, with baseline and Week 2 as the stated timeframe. The arm-level estimates are based on 326 participants for montelukast, 163 for loratadine, and 331 for placebo. These endpoint counts are distinct from the overall registry enrollment figure; the record does not explain the difference between them.

The registry also lists secondary outcomes covering nighttime symptoms, daytime eye symptoms, patient and physician global evaluations, and rhinoconjunctivitis quality of life. It reports six outcome measures and five posted secondary outcomes. The analysis details for these outcomes are not present in the record summarized by the posted primary result. Likewise, specific eligibility criteria, treatment duration beyond the stated endpoint timeframe, and other protocol procedures are not reported in the registry record.

The study has three arms, but the posted continuous endpoint information gives arm estimates rather than the numerical comparisons between arms. The registry record marks the two-arm comparison as missing or not estimable for this continuous endpoint. Consequently, the descriptive arm results can be discussed, while a quantified estimate of the treatment contrast and its uncertainty cannot be reconstructed from the posted result.

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: Patients who required excluded medication and those who did not meet a minimum predefined level of combined daytime nasal symptoms score during the run-in period were excluded from randomization.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The primary efficacy analyses were based on the intention-to-treat (all-patients-treated) principle, i.e., all patients who had a baseline and at least one posttreatment measurement were included.

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 Daytime Nasal Symptoms Score, measured over Baseline and Week 2.

The primary endpoint is the mean change from baseline in the daytime nasal symptoms score at Week 2. The score is the average of four individual nasal symptom scores: congestion, rhinorrhea, itching, and sneezing. Each symptom contributes to a scale ranging from 0, the best score, to 3, the worst score. The endpoint therefore combines several relevant nasal symptoms into one continuous summary measure.

Change from baseline expresses how the score shifted over the stated follow-up interval, rather than only describing the score at Week 2. Because lower scores are better, a negative mean change indicates a decrease in symptom score from baseline. The reported arm estimates are negative in all groups. They describe estimated average within-arm changes, not the difference in change between an active-treatment group and placebo or between the active treatments.

A continuous summary can use information across the scale rather than reducing outcomes to a binary category such as improved or not improved. That can make average shifts visible, but an average does not reveal how changes varied among individual participants. Nor does the composite score show whether each symptom contributed equally to the overall result in practice. Separate symptom-level results are not reported as the primary endpoint in this record.

The registry reports least-squares means and 95% confidence intervals for the endpoint. The confidence intervals express uncertainty around each estimated group mean under the analysis used. The record does not post the full model specification or the numerical between-arm estimates, so the arm means should not be treated as substitutes for a direct randomized comparison.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary result uses a continuous outcome: the daytime nasal symptoms score. A continuous measure can take values across a scale, and a mean summarizes the average score or average change within a group. Here, the endpoint is change from baseline, so negative values indicate that the group average moved downward on a scale where lower scores are better. This general interpretation follows from the scale definition in the registry; it does not imply that every participant had a decrease.

The reported parameter is a least-squares mean. In general, a least-squares mean is an estimated group mean from a statistical model, rather than necessarily the simple arithmetic average of observed values. Such means can be useful when a model includes terms that account for features of the design or analysis. The registry record does not report the model specification, covariates, or other details needed to determine exactly how these estimates were adjusted. The least-squares label should therefore be retained, without assuming a particular adjustment.

Each arm estimate is accompanied by a 95% confidence interval. In general, a confidence interval expresses the uncertainty associated with an estimate under the statistical model and its assumptions. It is not a range that contains a fixed proportion of individual participant values, and it is not by itself a probability statement about the true value. For this endpoint, the intervals give uncertainty around the arm-level mean changes. To assess the comparative effect, the relevant quantity would be a contrast between group means, with its own uncertainty interval and, where applicable, a test. Those numerical comparison results are not reported in the registry record.

The randomized design is important because a properly conducted randomized comparison can support estimation of differences between assigned groups. The three posted arm means do not themselves complete that comparison. A difference between two separately estimated means could be calculated arithmetically, but doing so would not supply the model-based uncertainty or the sponsor's intended analysis. The registry record specifically indicates that values are missing or not estimable for a two-arm comparison for this continuous endpoint. No recalculated comparison is therefore presented.

The primary analysis population is described as intention-to-treat, with the registry clarifying that it included all patients who had a baseline and at least one posttreatment measurement. This definition is sometimes called an all-patients-treated principle in the record. In general, intention-to-treat analyses aim to preserve the comparison created by random assignment. The eligibility condition stated means that participants without the required measurements were not included in the reported primary efficacy analysis. The record does not state how missing measurements among included participants were handled, whether values were imputed, or what assumptions were used for the analysis.

The reported endpoint timeframe is baseline and Week 2. This identifies the measurement interval used for the primary outcome, but does not describe the full assessment schedule. The record does not report whether other time points were collected, how the Week 2 measurement window was defined, or whether an analysis used repeated measurements. Those details should not be inferred from the endpoint label.

What does a group estimate tell the reader?

An arm mean describes an average change for participants included in that arm's endpoint analysis. It does not show the full distribution of outcomes, the number whose score improved, or the degree of change for any individual. The arm estimate is also not the treatment effect by itself. In a randomized trial, a comparative treatment effect is typically expressed as a contrast between groups, with a corresponding measure of uncertainty. That contrast is absent from the posted continuous result.

How should the scale direction be read?

The score ranges from 0, best, to 3, worst. Thus, a negative change is in the direction of a lower, better score, while the sign alone does not establish a comparative advantage. The magnitude remains in the score's units. The registry does not provide a threshold for what would count as an important change, so the estimates cannot be classified using an unreported clinical-importance criterion.

What do the planning illustrations mean?

General sample-size calculations depend on assumptions about the effect to detect, significance level, allocation, and statistical model. The planning illustrations accompanying the results are described as based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They consider detecting a stated fraction of the observed effect. These are planning illustrations, not the trial's design calculations, and they are not observed power. The registry record does not report the trial's own sample-size justification or its original assumptions. The illustrations should not be used to infer that the trial was designed around those assumptions.

In general, a smaller target effect requires a larger sample to estimate or detect with comparable precision, while greater variability can also require more observations. This is a general statistical relationship, not a claim about the calculations behind this trial. The posted information does not give the underlying individual-level variability or enough detail to reproduce an independent calculation. Because the direct contrast is missing, any planning illustration based on a stated fraction of the observed effect also depends on what observed effect is used and should be read as a scenario rather than a report of trial performance.

The primary endpoint is continuous, and the registry record does not report a dispersion statistic alongside the arm estimates. It does report confidence intervals, which convey uncertainty in the means, but these are not a substitute for all the quantities needed to reconstruct the analysis. The posted summary should therefore be interpreted on its own terms: least-squares mean changes and confidence intervals for each arm, with no numerical comparison estimate or complete model description.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures and five posted secondary outcomes in addition to the primary endpoint description. The secondary outcomes cover nighttime symptoms, daytime eye symptoms, patient and physician global evaluations of allergic rhinitis, and rhinoconjunctivitis quality of life. Multiple outcomes create a general multiplicity issue: when many hypotheses are examined, the chance of at least one apparently notable result can increase if each is tested without adjustment. Whether and how that concern was addressed depends on the prespecified analysis plan.

The registry record does not report a multiplicity adjustment, a hierarchy of endpoints, or the statistical plan for secondary outcomes. It also reports zero analyses for the posted outcome data. This means the available record does not give a set of inferential results for the secondary outcomes that could be assessed alongside the primary result. The absence of posted analyses should not be interpreted as evidence that no analyses were conducted; it means the analyses are not reported in the registry record.

The primary analysis population included participants with a baseline measurement and at least one posttreatment measurement. The record does not specify how missing values were handled, how many participants lacked required measurements, whether missingness differed among arms, or whether sensitivity analyses were performed. Without those details, the effect of missing data on the estimates cannot be evaluated from the record.

In general, missing-data methods rely on assumptions about why measurements are unavailable and how unavailable outcomes relate to observed information. Different assumptions can lead to different estimates. This general point does not identify which assumptions were used in this trial; that information is not reported in the registry record. A complete interpretation would require the missing-data procedures and the analysis plan, neither of which is posted with these results.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: depression (0.3%).

The registry lists serious adverse event counts for the safety groups: 0 among 333 participants at risk in the placebo group, 1 among 326 at risk in the montelukast group, and 0 among 170 at risk in the loratadine group. These are counts, not estimates of comparative risk. The record does not provide a statistical comparison of serious adverse events between groups.

The registry also lists other adverse events for 19 of 333 participants in the placebo group, 13 of 326 in the montelukast group, and 8 of 170 in the loratadine group. These counts should be read with their corresponding group denominators. The registry's serious-event term list includes depression at 0.3%; the record does not identify an arm for that listed term in the information.

Deaths are not reported in the registry record. The posted counts alone do not describe event severity beyond the serious-event classification, timing, duration, relatedness, or the full range of events experienced by participants. They also do not establish whether observed differences between groups reflect chance. Safety comparisons require appropriate context, definitions, and analysis; the registry record does not provide those details for these counts.

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

The posted primary result describes average changes in a daytime nasal symptom score from baseline to Week 2. The score combines congestion, rhinorrhea, itching, and sneezing, with lower values indicating better symptom scores. The estimated changes are negative in each group, and the accompanying confidence intervals describe uncertainty around each group mean.

Those estimates do not by themselves answer the central comparative question. The treatment effect of interest in a randomized three-arm study is a contrast between assigned groups, not a within-group change considered alone. The registry record does not provide the numerical comparison estimates, their confidence intervals, or their statistical tests for this continuous endpoint. It therefore does not allow a precise account of the magnitude or uncertainty of the differences between montelukast, loratadine, and placebo based on this posted result.

The result also does not show how individual participants changed. A group mean can conceal a range of responses, and the record does not report the distribution of individual changes or the proportion meeting a defined response criterion. No threshold for a clinically important change is reported, so the group estimates cannot be classified against such a threshold from this record.

The listed secondary outcomes indicate that the trial considered other symptom, global evaluation, and quality-of-life measures. Their presence does not supply evidence about their results: the record posts five secondary outcomes but no analyses. The primary endpoint should not be treated as a complete summary of every measured outcome.

These limits are about what the registry result reports, not a judgment about the trial's conduct or the treatments. A fuller interpretation would require the comparative estimates and analysis methods, details of missing data and participant flow, and results for the other prespecified outcomes. None of those details can be reconstructed from the arm-level means alone. The posted data support a description of estimated average within-group changes, with stated uncertainty, and no stronger comparative conclusion.

13. Why This Trial Matters Statistically

This trial illustrates a central distinction in randomized-trial interpretation: a change within a treatment group is not the same as a treatment effect relative to a comparator. Baseline-to-follow-up changes can be useful descriptive outcomes, but the randomized design's comparative question depends on estimating differences between groups. Here, the registry posts arm-level least-squares mean changes while indicating that two-arm values are missing or not estimable for the continuous endpoint. This limits the comparative inference available from the posted result.

The endpoint also shows why the scale's direction and units matter. Negative changes correspond to lower average symptom scores, and the scale runs from best to worst in the opposite direction from a measure where higher is better. Reading the sign without the scale definition could lead to a mistaken interpretation. Reporting the component symptoms and the scale's anchors helps clarify what the average represents, though the composite still cannot reveal individual patterns.

The record is a useful example of how confidence intervals should be interpreted. The intervals describe uncertainty around each arm mean; they do not directly provide uncertainty around a treatment contrast. Readers need the contrast estimate and its interval to assess the size and precision of a comparative effect. Separate arm intervals cannot replace that information.

The posted analysis-population definition further demonstrates why endpoint counts and total enrollment should be distinguished. The efficacy analysis required baseline and posttreatment measurements, while the registry gives a separate overall enrollment figure. Without participant flow and missing-data handling, readers cannot fully trace who contributed to the primary result.

Finally, planning illustrations should not be confused with the trial's original design rationale or observed power. The registry record does not report the original sample-size justification. Clear separation of posted results, missing inferential details, and hypothetical planning assumptions is important for transparent statistical interpretation.

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 of the posted primary result is that it contains arm-specific estimates but not the numerical between-arm comparisons for the continuous endpoint. The record also does not report comparison confidence intervals or p-values. As a result, the magnitude and precision of treatment contrasts cannot be assessed from these figures.

The registry gives the endpoint definition and the intention-to-treat inclusion rule, but not a full statistical analysis plan. The model used to produce the least-squares means, any covariate adjustment, the handling of missing observations, and any sensitivity analyses are not reported in the registry record. The reason that endpoint arm counts differ from total enrollment is also not explained in the posted information.

The reported primary outcome is a composite score, so it does not show the separate results for congestion, rhinorrhea, itching, or sneezing. It reports group means rather than the spread or distribution of individual changes. The record does not state a threshold for clinically important change or provide a responder analysis for this endpoint.

Although secondary outcomes are listed, the registry reports zero analyses. The record therefore does not provide their results or explain how multiplicity across outcomes was addressed. Safety information is also limited to posted event counts and denominators, without comparative analysis, timing, relatedness, or a complete account of event definitions. These limitations constrain interpretation of the registry results; they do not establish what additional information may exist outside the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study Investigating the Effect of Montelukast in Patients With Seasonal Allergic Rhinitis (MK-0476-192)(COMPLETED) — NCT00960141.
• Registry record last updated 2022-02-03; status last verified 2022-02.
• Results first posted 2010-06-09.
• 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

The registry record describes a completed, randomized, double-blind, parallel-group Phase 3 study of montelukast in seasonal allergic rhinitis, with loratadine and placebo comparator groups. Enrollment was 829. The primary endpoint was mean change from baseline in daytime nasal symptoms score at Week 2, reported as least-squares means with 95% confidence intervals.

The posted arm estimates were -0.38 for montelukast, -0.47 for loratadine, and -0.32 for placebo. The record does not report numerical between-arm estimates for this continuous endpoint. The primary analysis included participants with baseline and at least one posttreatment measurement; further missing-data methods are not reported.

Five secondary outcomes are listed, with zero analyses reported. Serious adverse event counts were 1 for montelukast and 0 for each comparator group. The registry does not report deaths. The record does not include the full statistical plan, the original sample-size justification, or detailed participant flow.