Montelukast in Mild Asthmatic Children With Allergic Rhinitis (0476-367): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Montelukast in Mild Asthmatic Children With Allergic Rhinitis (0476-367) (NCT00442559), a randomized phase 4 study. The primary outcome is Change From Baseline for Daytime Asthma Symptom Score.

2. Clinical Question

The study asked how daytime asthma symptom scores changed during treatment in children with both conditions. It was described as assessing real-world effectiveness of montelukast. The posted primary results concern symptom scores in the participants with diary measurements, not every enrolled participant. They also do not establish a direct head-to-head estimate of treatment effect on the primary change endpoint: the sponsor’s tests compare baseline and final values within each group, whereas the unadjusted recalculations include comparisons against the montelukast baseline summary. These are different statistical questions, so their estimates and uncertainty should not be treated as interchangeable.

3. Trial Design

This was a completed phase 4 study with randomized allocation, no masking, and a parallel-group model. The registry lists 191 participants enrolled. The record identifies Organon and Co as the sponsor. The study start is recorded as 2005-01 and primary completion as 2007-10.

The study compared montelukast with inhaled corticosteroids over 12 months after randomization. Montelukast dosing depended on age: one 4 mg chewable tablet each evening for children aged 2 to 5 years and one 5 mg chewable tablet each evening for those aged 6 to 14 years. In the corticosteroid group, the physician selected the agent, dose, and regimen. The registry wording also refers to inhaled corticosteroid use if participants exacerbated from mild to moderate within 12 weeks, but the arm descriptions are truncated. The full details of that treatment rule and how it was applied are not reported in the registry record.

The primary measure was daytime asthma symptom score, with assessments at baseline and week 12. The score ranged from 0 to 5, and the stated change was week 12 minus baseline. The record says 138 participants without a daytime asthma symptom score from the participant diary were not included in the primary efficacy parameter. The posted primary arm summaries show 24 participants for montelukast and 29 for inhaled corticosteroids. Reasons for missing diary scores, the method used to handle missing measurements, and the full analysis-population definition are not reported in the registry record.

Randomization does not by itself ensure that the small subsets contributing to the posted efficacy summaries remain comparable. The registry record does not describe the allocation sequence, concealment, stratification, or any methods used to account for differences between groups. It also does not report the original sample-size justification or the statistical plan in enough detail to establish whether the posted analyses were prespecified.

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: Up to 1 week for wash-out - prior to baseline randomization.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The primary efficacy parameter was a mean change from baseline to treatment for daytime asthma symptom score. Therefore 138 participants who didn't have a daytime asthma symptom score from the participant diary were not 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 Change From Baseline for Daytime Asthma Symptom Score, measured over Baseline and Week 12.

The primary endpoint was change from baseline in daytime asthma symptom score, assessed from baseline to week 12. The scale ran from 0, meaning no symptoms, to 5, meaning most symptoms. Change was defined as the week 12 score minus the baseline score; accordingly, a negative change was designated favorable. The registry describes the primary efficacy parameter as a mean change.

Symptom scores provide a way to summarize the reported daytime symptom burden over the assessment period. The mean describes the group’s average score, while the standard deviation describes how spread out individual scores are around that average. These summaries can help characterize both the direction of average change and the variability among participants with available diary measurements.

There is an important distinction between the endpoint as described and the comparisons displayed in the posted analyses. The endpoint description emphasizes change from baseline to treatment, but the sponsor’s reported analyses are labeled as two-sided t-tests of mean differences in final values and are conducted within each group. The unadjusted recalculations compare posted group means using the montelukast baseline summary as the reference, including comparisons at baseline and week 12. Neither approach, as represented in the registry, should be assumed to be an adjusted between-group comparison of change scores.

The score is ordinal, so its ordered values indicate increasing symptom levels, but the registry does not explain why the numeric steps were treated as suitable for a mean-based analysis. The rationale for choosing this endpoint, its clinically meaningful change threshold, and the method for collecting or validating diary scores are not reported in the registry record.

7. Results Posted in the Registry

• Daytime Asthma Symptom Score at 12 Weeks - Montelukast: mean 0.16 versus 0.32 on control; difference -0.16 (95% CI -0.38 to 0.06), standardized effect d = -0.41; z-test p = 0.152.
• Daytime Asthma Symptom Score at Baseline - ICS: mean 0.29 versus 0.32 on control; difference -0.03 (95% CI -0.25 to 0.19), standardized effect d = -0.07; z-test p = 0.791.
• Daytime Asthma Symptom Score at 12 Weeks - ICS: mean 0.13 versus 0.32 on control; difference -0.19 (95% CI -0.38 to 0.00), standardized effect d = -0.55; z-test p = 0.056.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change From Baseline for Daily Allergic Rhinitis Symptom Score

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Daily Allergic Rhinitis Symptom Score at 12 Weeks- Montelukast: mean 0.23 versus 0.45 on control; difference -0.22 (95% CI -0.39 to -0.05), standardized effect d = -0.71; z-test p = 0.013.
  • Daily Allergic Rhinitis Symptom Score at Baseline - ICS: mean 0.31 versus 0.45 on control; difference -0.14 (95% CI -0.33 to 0.05), standardized effect d = -0.41; z-test p = 0.145.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Daily Allergic Rhinitis Symptom Score at 12 Weeks- Montelukast than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Daily Allergic Rhinitis Symptom Score at Baseline - ICS and the control. The interval gives the range of effects that remain compatible with the data.

The registry posts one secondary endpoint: change from baseline in daily allergic rhinitis symptom score, assessed at baseline and week 12. The overall pattern of posted summaries is that the average scores at week 12 are lower than the corresponding baseline averages in both groups. The sponsor also reports within-group tests for this endpoint, while the simple recalculations use contrasts against the montelukast baseline summary. As with the primary endpoint, these analyses do not all answer the same question.

The secondary results add information about symptom scores associated with allergic rhinitis, whereas the primary endpoint concerns daytime asthma symptoms. They may therefore complement the primary outcome descriptively, but they do not substitute for a direct between-group test of the primary endpoint or establish a comparative treatment effect. Similar directions across outcomes do not resolve the issues of missing diary measurements, small analysis groups, or differing estimands.

Unadjusted secondary findings generally carry less weight than a prespecified primary analysis because multiple outcomes and comparisons create more opportunities for apparently notable results. The registry lists two outcome measures and four analyses overall, but the multiplicity approach, if any, is not reported in the registry record. Without knowing whether and how the analysis plan accounted for multiple testing, a secondary p-value should not be treated as if it were an isolated confirmatory test. The registry also does not report a clinically meaningful threshold for the allergic rhinitis score.

The endpoint’s registry description identifies a mean change and standard deviation, but does not explain how the diary observations were collected, whether the analysis followed a prespecified model, or how missing secondary scores were handled. These limitations mean the secondary endpoint is best read as supportive descriptive evidence about the posted measurements rather than as independent confirmation of a treatment difference.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Change From Baseline for Daily Allergic Rhinitis Symptom Score

9. Statistical Interpretation

10. Statistical Methods Explained

Several distinct statistical ideas are needed to read these results carefully. The registry reports group means and standard deviations for a continuous summary, and it also includes sponsor-reported two-sided t-tests. Alongside those analyses are simple unadjusted recalculations from posted summary statistics. They are not all tests of the same estimand, meaning the specific quantity a statistical analysis is intended to estimate.

What does a mean change describe?

General statistical explanation: a mean is an average across the observations included in a calculation. A change score is formed by subtracting the baseline measurement from the later measurement for each participant and then summarizing those changes. In this study, the endpoint description defines the individual change as the week 12 score minus the baseline score; negative values are designated favorable. A group mean change can describe average within-group movement, but it does not automatically describe the difference in change between randomized groups.

To estimate a between-group difference in change, an analysis would need to compare the groups’ changes, or use another prespecified model that estimates the treatment contrast while accounting for baseline values. The registry record does not report enough detail to determine whether the sponsor conducted such an analysis for the posted primary result. The sponsor analysis is labeled as a mean difference in final values and is reported separately for each group.

What is the role of the t-test?

General statistical explanation: a t-test assesses whether a mean difference is compatible with a null value, relative to the estimated uncertainty in that difference and under the test’s assumptions. A two-sided test considers departures in either direction. A p-value is a measure of how unusual the observed result, or a more extreme one, would be under the null model and its assumptions. It is not the probability that the null hypothesis is true, nor does it measure clinical importance.

The sponsor reports two-sided t-tests for mean differences in final values. The reported p-values are 0.015 for montelukast and 0.027 for inhaled corticosteroids. Those tests pertain to each group’s reported baseline-to-final comparison, not a direct comparison of treatment groups. A small p-value for change within each arm cannot establish that the amount of change differs between arms. A treatment comparison requires a contrast that directly tests the groups against one another.

Why do the recalculations differ from the sponsor results?

The recalculations use unadjusted comparisons of posted summary means and standard deviations. For the primary measure, the montelukast baseline group is the reference in the recalculated contrasts. One contrast compares its baseline mean with its week 12 mean; another compares the inhaled corticosteroid baseline mean with that reference; and another compares the inhaled corticosteroid week 12 mean with that reference. The tests and intervals reflect those specific pairwise contrasts.

These are not the same as the sponsor’s within-group tests. A difference in the comparison being tested can lead to a different estimate, confidence interval, and p-value. The recalculations are also not adjusted for baseline characteristics or any other covariates. The registry record does not describe an adjustment model for the sponsor’s reported tests. Because a discrepancy is flagged, the sponsor’s own analysis takes precedence for its reported result; the recalculated results should be used only to understand the posted summaries under these alternative unadjusted comparisons.

What does the standard deviation contribute?

General statistical explanation: the standard deviation describes variation among individual measurements around their group mean. It is not the same as the standard error, which describes the estimated uncertainty of a mean or contrast. Standard errors and confidence intervals depend on both variability and the amount of information in the analysis. In small groups, estimates can be imprecise even when the group means differ numerically.

The outcome is a score on an ordinal scale. Treating such a score as a mean is one possible statistical summary, but it assumes that averaging the numeric score is informative for the question at hand. The registry identifies the measure as a mean and gives its scale, but does not report a rationale for this choice or a sensitivity analysis using another method.

How should confidence intervals and p-values be read together?

General statistical explanation: a confidence interval describes a range of values reasonably compatible with the data and the model used to calculate it. A wider interval indicates less precision than a narrower interval. When an interval for a difference includes zero, the data are compatible with no difference under that analysis, as well as with values on either side of zero. A p-value and interval are related summaries of uncertainty, but neither says whether a difference matters clinically.

For the sponsor’s primary tests, the reported 95% intervals exclude zero. For the recalculated comparisons of baseline and week 12 means against the montelukast baseline reference, some intervals cross zero. These findings are not a direct contradiction because the methods address different comparisons. The interpretation should therefore name the analysis and contrast rather than treating all values as interchangeable evidence about a single treatment effect.

What can be concluded about the analysis plan?

The record lists outcome measures and analyses but does not report the original statistical analysis plan, any interim analyses, a prespecified multiplicity strategy, or a sample-size justification. Those items cannot be inferred from the observed results. The registry also does not report whether the analyses were adjusted, whether assumptions were checked, or how diary measurements that were missing at some visits were handled. These omissions limit the ability to assess how closely the posted analyses follow the original plan.

11. Confidence Intervals and Precision

Confidence intervals communicate uncertainty around an estimate. General statistical explanation: an interval is not a guarantee that the true value lies within its endpoints; rather, under the method’s assumptions, it describes values compatible with the observed data. Narrower intervals indicate greater precision, while wider intervals leave more uncertainty about the size and direction of an effect.

The sponsor reports 95% confidence intervals for its within-group mean differences. For montelukast, the interval runs from -0.29 to -0.03; for inhaled corticosteroids, it runs from -0.3 to -0.02. Both intervals are on the negative side of zero, consistent with the reported p-values for those within-group tests. They do not estimate the uncertainty in a direct randomized comparison between groups.

The recalculated interval for the montelukast week 12 mean compared with the montelukast baseline reference extends on both sides of zero, as does the recalculated interval comparing the inhaled corticosteroid week 12 mean with that same reference. This indicates substantial uncertainty under those unadjusted contrasts. The baseline comparison for inhaled corticosteroids against the montelukast baseline reference also has an interval extending on both sides of zero. Those intervals should not be set alongside the sponsor’s intervals as though the estimates were identical: they use different contrasts and procedures.

Precision is limited by the numbers included in the primary summaries: 24 in the montelukast group and 29 in the inhaled corticosteroid group. The registry says 138 participants without diary scores were not included in the primary efficacy parameter. The resulting analyses therefore cover a restricted subset of the enrolled participants. The record does not report a missing-data analysis or enough information to determine how much missingness affects precision or could affect the direction of the estimates.

Statistical precision is also distinct from clinical importance. The registry does not give a threshold for a meaningful change in score, so the intervals cannot be classified against a stated clinical benchmark. They describe statistical uncertainty on the scale, not the practical importance of a change.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size illustrations describe hypothetical future planning based on the observed effect, not the number enrolled in this study. For these continuous outcomes, the figures are participants per arm, not total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Each planning scenario uses a stated fraction of the observed effect as the effect to detect. The original trial’s own design assumptions and sample-size justification are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm under the stated assumptions for a design with an 80% chance of detecting the chosen effect if that effect is real and the assumptions hold. The 90% power column gives the corresponding planning number for a design with a 90% chance. These are planning targets, not observed outcomes, and they do not establish the probability that the completed trial detected a true effect. This is not observed power.

Rows that plan for 75% or 50% of the observed effect use a smaller effect than the full observed-effect scenario. Smaller effects are harder to distinguish from random variation, so they require more participants under the same assumptions. The estimates are sensitive to the effect chosen for planning: an observed effect from a small analysis group may not represent the effect a future study should expect. The illustrations also do not account for missing measurements, unequal allocation, departures from the normal approximation, or other design features unless those are among the stated assumptions.

A planning curve, when read alongside the figures, represents how the required number of participants changes with the effect size under its assumptions. It is not a record of enrollment or a forecast of the number needed for every future study. The registry record does not report whether its original enrollment target was based on these assumptions or on a different calculation.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -0.160; pooled standard deviation σ = 0.387.
2. n = 2 × (1.96 + 0.842)² × 0.387² / -0.160² = 91.6, rounded up to 92 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -0.16                      # observed difference in means
sigma <- 0.386587635601554                     # 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(442559)
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 = 92 (80% power), 123 (90% power)
 75% of observed effect: n per arm = 163 (80% power), 219 (90% power)
 50% of observed effect: n per arm = 367 (80% power), 491 (90% power)
power.t.test: 92.6 
Simulated power at n = 92 per arm: 0.799 (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.

14. Analyses Reported by the Sponsor

The sponsor reports a two-sided t-test of mean differences in final values for each treatment group. Its montelukast result is -0.16, with a 95% confidence interval from -0.29 to -0.03 and p = 0.015. Its inhaled corticosteroid result is -0.16, with a 95% confidence interval from -0.3 to -0.02 and p = 0.027. These are the sponsor’s reported analyses and should take precedence over the simple recalculations when describing what the sponsor reported, because a discrepancy is flagged.

The recalculations use posted group means and standard deviations to form unadjusted contrasts. Their comparisons do not reproduce the sponsor’s within-group tests: they compare selected group summaries against the montelukast baseline summary. For example, the recalculated week 12 comparison for montelukast has p = 0.152, and the recalculated week 12 comparison for inhaled corticosteroids has p = 0.056. The estimates and their uncertainty therefore answer different questions.

The sponsor’s reported p-values indicate evidence against no baseline-to-final mean difference within each group under those tests. They are not evidence from a direct test that montelukast and inhaled corticosteroids differ from each other. The record does not report a sponsor analysis that directly compares treatment-group changes, and it does not provide enough detail to reconcile the differing comparisons further. The recalculations can help readers understand the posted summaries, but they should not replace the sponsor analysis or be interpreted as an adjusted treatment contrast.

15. Multiplicity, Missing Data and Analysis Populations

The registry record does not report a multiplicity strategy. This matters because the record includes a primary endpoint, a secondary endpoint, and multiple analyses, and testing several outcomes or contrasts increases the chance that at least one result may appear statistically notable by chance. General statistical explanation: a multiplicity plan may define which outcome is primary, which tests are confirmatory, and whether significance thresholds or interpretation are adjusted when more than one hypothesis is tested. Without such a plan, the reported p-values should be read as unadjusted evidence for their particular tests, not as a familywise guarantee across all analyses.

The primary endpoint description says 138 participants without a daytime asthma symptom score from the participant diary were not included in the primary efficacy parameter. The posted arm summaries contain 24 participants for montelukast and 29 for inhaled corticosteroids. The reasons for missing diary scores, their timing, and whether participants differed in measured characteristics from those with scores are not reported in the registry record. The record does not describe imputation, a missing-data model, sensitivity analyses, or any other method for assessing the effect of missing measurements.

Missingness can affect both precision and interpretation. General statistical explanation: if the chance of a score being missing is related to the score itself, treatment, or other participant characteristics, an analysis restricted to available measurements may not represent all randomized participants. The registry does not provide enough information to determine whether that concern applies in this study. No assumption about why scores were missing or how the missing values would have looked is justified by the posted record.

The record also does not report the full analysis-population definitions, the number randomized to each group, or whether participants were analyzed according to their original randomized assignment. The sponsor’s posted tests are labeled as within-group baseline and final comparisons, while the recalculations use available arm summaries. These descriptions do not establish how participants with incomplete follow-up were treated in every analysis. Interim analyses, protocol deviations, and the original statistical analysis plan are not reported in the registry record.

16. 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: asthma (3.0%), vomiting (1.1%), pyrexia (1.1%), chronic sinusitis (1.1%), pneumonia (1.1%).

The safety record lists serious adverse events for 100 participants at risk in the montelukast group, with 4 affected, and 91 at risk in the inhaled corticosteroid group, with 3 affected. These are the counts recorded for participants affected by at least one serious adverse event in the respective group summaries; they do not describe the number of all events or their duration. The registry also lists other adverse events for 83 of 100 participants at risk with montelukast and 70 of 91 at risk with inhaled corticosteroids.

The listed serious-event terms include asthma at 3.0%, and vomiting, pyrexia, chronic sinusitis, pneumonia, and headache at 1.1% each. The record does not provide enough context in these summaries to establish whether categories overlap for an individual participant or how each term was counted. No death counts are reported in the registry record.

These figures are descriptive counts, not a statistical comparison establishing a difference in safety between groups. The registry record does not report a formal comparative safety analysis, exposure-adjusted rates, or enough detail about event collection to assess differences in observation time. The counts should therefore be read as the reported adverse-event summaries, not as evidence that one treatment has greater or lesser risk.

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

The posted results describe changes in symptom-score summaries among participants with the relevant diary measurements. The mean daytime asthma symptom scores were lower at week 12 than at baseline in both groups. The sponsor’s within-group tests report negative mean differences with 95% confidence intervals that exclude zero. Those results concern change within each group, not a direct comparison of the randomized treatments.

The unadjusted recalculations illustrate a different set of comparisons. In particular, the recalculated intervals for the week 12 contrasts against the montelukast baseline reference cross zero. These results do not demonstrate a clear difference under those specific unadjusted contrasts. Because the sponsor analysis and recalculations address different comparisons, neither should be treated as a simple confirmation or refutation of the other. The discrepancy flag means the sponsor’s own analysis takes precedence for reporting the sponsor result.

These findings do not establish that one treatment is superior to the other. A within-group improvement cannot be attributed to treatment alone without an appropriate between-group contrast, because changes can also reflect factors such as the passage of time, measurement variation, or other influences. The registry record does not report enough detail to determine how these possibilities were handled analytically.

The endpoint uses a 0-to-5 ordinal symptom scale, and the record does not specify a threshold for a clinically meaningful change. Statistical significance in a within-group test therefore cannot be equated with a known degree of practical or clinical importance. The small numbers included in the posted efficacy summaries and the exclusion of participants without diary scores further limit how broadly the results can be applied to all enrolled participants.

The secondary allergic rhinitis endpoint has a similar descriptive pattern, but unadjusted secondary findings do not independently establish a treatment difference. The registry’s safety summaries give counts of participants affected by serious and other adverse events, but do not constitute a formal comparative safety analysis. Taken together, the posted record supports a cautious description of symptom-score changes and adverse-event counts in the analyzed groups; it does not justify a broader causal or clinical conclusion.

18. Why This Trial Matters Statistically

This trial is statistically instructive because it distinguishes a change within each treatment group from a difference between treatment groups. A result can show a statistically notable baseline-to-final change in each arm without testing whether the amount of change differs by randomized assignment. Readers need to identify the contrast behind each estimate before interpreting its p-value or interval.

The record also illustrates why analysis populations matter. Although 191 participants were enrolled, the primary efficacy description states that participants without daytime symptom scores from the diary were excluded, and the posted arm summaries contain 24 and 29 participants. A result from this subset may be less precise and less representative than one based on complete follow-up for all randomized participants. The record does not report why measurements were missing or how missingness was addressed.

Finally, this case shows why sponsor-reported analyses and simple recalculations may differ without one being a direct arithmetic correction of the other. The sponsor’s tests are labeled as within-group comparisons; the recalculations use alternative unadjusted contrasts. Naming the estimand and preserving that distinction is essential. Confidence intervals, p-values, planning illustrations, and safety counts each answer different questions, and none should be interpreted without its analysis context.

These distinctions are relevant beyond this study. Statistical significance is not the same as clinical importance, and a planning illustration is not evidence about the power actually achieved by a completed trial. Clear reporting of the original analysis plan, missing-data handling, multiplicity, and direct treatment contrasts would make the evidence easier to evaluate. Those details are not reported in the registry record.

19. 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 primary efficacy summaries are based on small groups: 24 participants for montelukast and 29 for inhaled corticosteroids. The registry states that 138 participants without a daytime asthma symptom score from the diary were not included in the primary efficacy parameter. Reasons for missing scores and their distribution across treatment groups are not reported in the registry record. This makes it difficult to assess the representativeness of the analyzed subset or the effect of missingness on the findings.

The posted sponsor analyses are within-group baseline-to-final tests, while the recalculations use unadjusted comparisons against the montelukast baseline summary. A direct, prespecified between-group comparison of change is not reported. The discrepancy between those approaches limits a unified interpretation of the primary result. The registry does not report enough of the statistical analysis plan to determine which analysis was intended to be primary for inference about treatment differences.

The study was open-label. The registry record does not report whether outcome assessors or those analyzing the data were masked, or how awareness of treatment assignment may have affected diary reporting. It also does not provide details of the allocation procedure, the original sample-size justification, interim analyses, multiplicity control, or missing-data sensitivity analyses.

The corticosteroid treatment was selected by the physician, so the comparator was not a single fixed agent, dose, or regimen. The arm descriptions also include an incomplete account of treatment changes following exacerbation. The full details of those rules and their implementation are not reported in the registry record. The endpoint is a mean on an ordinal scale, and no clinically meaningful score-change threshold is reported. Together, these limitations constrain interpretation of the size, precision, and generalizability of the posted findings.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Montelukast in Mild Asthmatic Children With Allergic Rhinitis (0476-367) — NCT00442559.
• Registry record last updated 2024-05-13; status last verified 2022-01.
• Results first posted 2008-11-18.
• 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.

23. Record Summary

This completed phase 4 randomized, open-label parallel study enrolled 191 children with asthma and allergic rhinitis. It compared 12 months of montelukast with inhaled corticosteroids selected by each participant’s physician. The primary endpoint was change in daytime asthma symptom score from baseline to week 12 on a 0-to-5 scale, with negative change designated favorable.

The sponsor reported within-group mean differences of -0.16 for both groups, with 95% confidence intervals excluding zero. Unadjusted recalculations use different contrasts and do not establish a direct treatment-group difference. Because a discrepancy is flagged, the sponsor’s own analysis takes precedence when describing its reported results. Interpretation is limited by the small primary analysis groups, diary-score exclusions, and missing details about the analysis plan and missing data. The registry also posts one secondary allergic rhinitis symptom-score outcome and descriptive adverse-event counts.