The Leukotriene Modifier Or Corticosteroid or Corticosteroid-Salmeterol Trial: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Leukotriene Modifier Or Corticosteroid or Corticosteroid-Salmeterol Trial (NCT00156819), a randomized phase 4 study. The primary outcome is Treatment Failure.

2. Clinical Question

The registry describes a comparison of treatment strategies among participants whose asthma was controlled on fluticasone before the treatment assignments. The question is not limited to a single symptom or lung-function measurement: treatment failure could be triggered by any of several clinical or treatment-related events defined in the registry record.

The posted results allow comparisons of montelukast with fluticasone and fluticasone plus salmeterol with fluticasone. They do not report a direct statistical comparison between montelukast and the combination group. The estimates therefore address the two reported comparisons against continued fluticasone, rather than establishing a complete ranking of all three regimens. The registry describes the study as asking whether the treatments were equally effective at reducing the number and severity of asthma attacks; the posted primary outcome is the number of participants with treatment failure.

3. Trial Design

This was a Phase 4 randomized, parallel-group trial with quadruple masking. The registry lists an enrollment of 500 and a completed status. The recorded study start is 2003-06, and primary completion is recorded as 2005-08. The sponsor was JHSPH Center for Clinical Trials.

Participants were assigned to one of three medication groups. The fluticasone group continued fluticasone at 100 microgram twice daily. The montelukast group changed to 5 or 10 mg each night. The combination group received fluticasone at 100 microgram plus salmeterol at 50 microgram each night. The registry lists these as drug interventions.

“Randomized” indicates that assignment to groups was made using random allocation, but the method used to generate or conceal the assignment is not reported in the registry record. “Quadruple” is the registry’s masking classification; which specific people were masked is not reported in the registry record. The parallel model means the groups received their assigned regimens during the same trial period rather than following a reported sequence of treatments.

The primary outcome was treatment failure, counted as participants, over 16 weeks. The posted group counts were 168 for fluticasone, 165 for montelukast, and 162 for fluticasone plus salmeterol. The registry record does not report the original sample-size justification or the trial’s planned statistical analysis details. Its posted planning illustrations are separate calculations based on the observed effects, not a description of the original design.

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:

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 Treatment Failure, measured over 16 weeks.

The primary endpoint was treatment failure, recorded as the number of participants with an event over 16 weeks. A participant met the definition if any one of several events occurred. These included hospitalization or an urgent medical visit for asthma; use of systemic corticosteroids or a need for open-label inhaled corticosteroids; specified declines in prebronchodilator FEV1 or morning peak expiratory flow; high rescue beta-agonist use over consecutive days; refusal to continue because of dissatisfaction; or a physician decision to stop treatment for safety reasons.

This composite endpoint brings several kinds of clinically relevant deterioration or treatment discontinuation into one outcome. It can capture events that differ in severity and in how they arise, so the total count does not indicate which component was responsible for each participant’s failure. Component-specific counts are not reported in the registry record.

Counting participants with an event gives an event count rather than a measure of the timing of failure or the number of repeated failures. The posted results do not provide time-to-event estimates. The endpoint’s broad definition makes it useful for summarizing whether a participant experienced any listed failure event during the stated period, while also limiting interpretation of the overall count as a measure of any single type of asthma worsening.

7. Results Posted in the Registry

• Montelukast: 30.3% (50/165) versus 20.2% (34/168) on control; risk difference 10.1 percentage points (95% CI 0.8 to 19.3), risk ratio 1.50 (1.02–2.19); unadjusted two-proportion z-test p = 0.034.
• Fluticasone Plus Salmeterol: 20.4% (33/162) versus 20.2% (34/168) on control; risk difference 0.1 percentage points (95% CI -8.5 to 8.8), risk ratio 1.01 (0.66–1.54); unadjusted two-proportion z-test p = 0.976.

8. Statistical Interpretation

9. Statistical Methods Explained

The posted primary outcome is binary at the participant level: a person either met the treatment-failure definition during the 16 weeks or did not. For a binary outcome, the event count and the number of participants in each group describe the observed frequency of that outcome. The risk difference compares the group risks by subtraction; the posted estimates express that comparison on the risk-difference scale. The registry also posts risk-ratio and odds-ratio estimates, but the principal interpretation follows the risk differences and their confidence intervals.

What does a risk difference describe?

A risk difference is the difference between the proportions of participants with an event in two groups. A value above zero in the posted comparisons indicates a higher observed event risk in the named treatment group than in the fluticasone comparator. A value below zero would point in the opposite direction. The risk difference is an absolute contrast on the risk scale; it is not a relative measure such as a ratio. The estimates are based on the observed group counts and do not by themselves establish why the groups differed.

How should the confidence intervals be read?

A confidence interval describes statistical uncertainty around an estimated effect. It is not a range containing a fixed proportion of individual participants’ outcomes, and it is not a guarantee that the underlying effect lies within its endpoints. A narrower interval indicates greater precision than a wider interval on the same scale. For a risk difference, zero represents no difference in risk. An interval that includes zero is compatible with no difference as well as with effects on either side; an interval wholly above or below zero excludes zero within the interval’s stated confidence procedure. The confidence level is not reported in the registry record.

What does the p-value add?

A p-value summarizes how unusual data at least as incompatible with a null hypothesis of no difference would be under the assumptions of the test. It is not the probability that the null hypothesis is true, nor does it describe the size or clinical importance of an effect. The posted p-values are identified as unadjusted. The registry record does not report the test specification, the planned decision threshold, or whether any multiplicity procedure was used in the trial analysis. The confidence intervals and p-values should therefore be treated as posted statistical summaries, with that missing analysis-plan context in mind.

Why distinguish the trial analysis from a recalculation?

Recalculated comparisons are simple, unadjusted calculations from posted summary statistics. They do not reproduce an analysis that might use participant-level data, covariate adjustment, or another prespecified method. In this record the discrepancy indicator is false, so there is no reported conflict requiring the sponsor’s own analysis to take precedence. The registry nonetheless reports zero posted analyses, and the methods underlying the original trial analysis are not reported in the registry record.

Why can a composite endpoint be difficult to interpret?

Treatment failure combines multiple events into one endpoint. This can provide a single summary of whether any listed failure event occurred, but the overall count cannot show whether the difference is driven by urgent visits, medication escalation, lung-function thresholds, dissatisfaction, or safety decisions. The registry does not post component-level results. The binary summary also does not represent event timing or repeated events. These distinctions matter because the same overall participant count could arise from different mixtures of endpoint components.

What can randomization and masking establish?

Random allocation is intended to make treatment groups comparable in expectation, while masking can reduce the influence of treatment knowledge on care or assessment. The registry identifies the trial as randomized and quadruple-masked but does not describe the allocation procedure, the masked roles, or whether masking was maintained. The design labels alone do not establish that every source of bias was prevented. Interpretation also depends on the quality and completeness of outcome follow-up, details not reported in the registry record.

The posted results are therefore best read as group-level event counts and comparisons accompanied by measures of statistical uncertainty. They do not supply the protocol-level detail needed to evaluate all assumptions behind the original analysis.

10. Confidence Intervals and Precision

For montelukast versus fluticasone, the risk-difference estimate is 0.101 and its confidence interval runs from 0.008 to 0.193. The interval lies above zero, but its span indicates that the estimate is not pinned to one precise value. The unadjusted p-value is 0.034. The interval’s lower endpoint is relatively near zero compared with its upper endpoint, so the observed contrast should not be treated as an exact or certain effect size.

For fluticasone plus salmeterol versus fluticasone, the risk-difference estimate is 0.001 and its confidence interval runs from -0.086 to 0.088. This interval includes zero and covers differences in either direction. Its width means that the posted result does not resolve small or moderate differences with high precision. The unadjusted p-value is 0.976, but that value does not establish equivalence or show that the treatments have identical effects.

A confidence interval communicates uncertainty in an estimate under the statistical method used to construct it. The confidence level and method are not reported in the registry record. The interval should not be read as a range of all effects that are possible, nor as a measure of variability between individual participants. The posted interval is also not adjusted for multiplicity, as the registry reports the p-values as unadjusted and does not describe any adjustment procedure.

Precision is distinct from whether an estimate is near zero. A near-zero estimate with a broad interval can leave substantial uncertainty about the size and direction of the underlying contrast. Likewise, an interval above zero does not make the point estimate exact. These comparisons are limited to the posted endpoint, groups, and follow-up period.

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 for both comparisons; they are not total event counts. The calculations use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. They use the observed effect or a stated fraction of that effect as the effect to detect. The original trial’s sample-size assumptions and justification are not reported in the registry record.

The 80% power column gives the illustrated number of participants per arm for a design with an 80% chance of detecting the specified effect under the stated assumptions, if that effect is the true one. The 90% power column makes the same kind of illustration for a 90% chance. These are planning calculations, not a measure of how much power the completed trial actually had. This is not observed power.

For the montelukast comparison, planning for 75% of the observed effect gives 502 participants per arm for 80% power and 671 participants per arm for 90% power. For the combination comparison, the observed risk difference is close to zero, and the planning illustrations consequently require very large participant counts. The posted numbers should be understood as consequences of the observed estimates and stated assumptions, not as recommended enrollment targets or evidence that a trial of that size was conducted.

Smaller effects require more participants to detect under otherwise unchanged planning assumptions. That relationship explains why planning for a fraction of an observed effect yields larger counts than planning for the full observed effect. Because the effect estimates themselves are uncertain, these illustrations can change substantially if the true effect differs from the observed one. They do not account for any original design choices that are not reported in the registry record.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.202; treatment proportion p₁ = 0.303; average p̄ = 0.253.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6146 = 1.2045.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6104 = 0.5138.
4. n = (1.2045 + 0.5138)² / (0.1006)² = 291.5, rounded up to 292 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
p0 <- 0.202380952380952                          # control proportion
p1 <- 0.303030303030303                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(156819)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 292 (80% power), 390 (90% power)
 75% of observed effect: n per arm = 502 (80% power), 671 (90% power)
 50% of observed effect: n per arm = 1,089 (80% power), 1,457 (90% power)
power.prop.test: 291.5 
Simulated power at n = 292 per arm: 0.801 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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. Multiplicity, Missing Data and Analysis Populations

The registry posts one primary outcome and no secondary outcomes. It lists zero posted analyses. The two reported comparisons each use fluticasone as the comparator, and the p-values are labeled unadjusted. Whether the original protocol identified these comparisons in advance, specified an order of testing, or planned an adjustment for multiple comparisons is not reported in the registry record.

Multiplicity matters because testing several hypotheses can increase the chance of at least one apparently unusual result under a broader null scenario. An adjustment or testing hierarchy can address that issue, but the registry does not describe whether either was part of the trial’s original analysis plan. The posted unadjusted p-values should therefore not be treated as evidence that multiplicity was handled in a particular way.

The record does not report the analysis population, such as whether results were based on all randomized participants, participants who received treatment, or participants with complete outcome data. It also does not report how missing outcomes, withdrawals, adherence, or treatment changes were handled. The arm counts and event counts describe the posted summaries, but they do not reveal whether every randomized participant contributed outcome data under the same analysis rules.

Information about imputation, sensitivity analyses, or exclusions is not reported in the registry record. Without those details, the effect of missing data or analysis-population choices cannot be assessed from the posted results alone. The absence of these details does not establish that no such procedures were used; it means they are not documented in the registry record.

14. 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: abdominal pain (0.6%), depression (0.6%), allergic reaction to bactim (0.6%), elbow sprain (0.6%), drop in peak flow (0.6%).

The registry lists serious adverse events affecting 6 participants in the fluticasone group, 4 in the montelukast group, and 4 in the fluticasone plus salmeterol group. The corresponding groups at risk were 168, 165, and 162 participants. These are counts, not estimates of comparative safety. The registry also lists other adverse events affecting 159, 156, and 157 participants in those respective groups.

The serious-event terms listed include Abdominal pain, Depression, Allergic reaction to Bactim, Elbow sprain, Drop in peak flow, and Asthma exacerbation, each recorded as 0.6%. The record does not assign these listed terms to a particular group in the summary available. Serious-event deaths are not reported in the registry record.

Counts alone do not describe event severity beyond the registry’s serious-event classification, timing, duration, recurrence, or relationship to treatment. The number of participants at risk also matters when reading counts across groups. No adjusted safety comparison or statistical uncertainty for these counts is reported in the registry record, so the figures are descriptive and should not be used alone to infer comparative risk.

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

The posted primary results describe treatment-failure counts over 16 weeks in three randomized groups. In the comparison with continued fluticasone, the montelukast group had 50 participants with treatment failure among 165, compared with 34 among 168 in the fluticasone group. The risk-difference interval is above zero. In the comparison with fluticasone, the combination group had 33 participants with treatment failure among 162; its risk-difference interval includes zero and allows differences in either direction.

These results do not show that the combination and fluticasone have identical effects. A result whose interval includes zero is not the same as a demonstration of equivalence or non-inferiority; the registry does not report an equivalence margin or a corresponding analysis. Similarly, the montelukast comparison’s interval does not specify one exact effect size. Both comparisons concern the registry’s composite treatment-failure endpoint, not each of its separate components.

The posted information does not establish what happened beyond the recorded follow-up, whether the same patterns apply to other asthma populations, or how results would differ under other dosing schedules. The registry summary describes participants whose asthma was well controlled with fluticasone and mild to moderately severe asthma, but detailed eligibility criteria are not reported in the registry record.

The data are statistical summaries from this trial, not a complete account of its conduct or analysis. Important context—including analysis populations, missing-data handling, multiplicity procedures, and the original statistical plan—is not reported in the registry record. The results should be interpreted alongside those limits, without extending them to claims about treatment outside the studied comparison and period.

16. Why This Trial Matters Statistically

This trial illustrates how a common primary endpoint can combine a range of clinical events. A treatment-failure count is easy to describe at the participant level, but its meaning depends on the endpoint definition. Here, the definition includes urgent care, medication escalation, lung-function thresholds, rescue medication use, dissatisfaction, and a physician’s safety decision. A single overall estimate cannot reveal which components account for the observed group contrast.

It also demonstrates why an estimate and its precision need to be read together. The montelukast comparison has a positive risk-difference estimate with an interval above zero, while the combination comparison has an estimate close to zero and an interval that spans both directions. Neither the point estimate alone nor the p-value alone captures the full uncertainty. In particular, a result near zero with a wide interval does not demonstrate that two regimens are equivalent.

The registry’s planning illustrations show how sample-size requirements depend on the effect a future study is designed to detect. Smaller target effects require more participants under the stated assumptions, and a planning calculation based on an observed effect is not the same as the original trial’s design justification. That distinction is important when interpreting large planning figures.

Finally, the record highlights the boundary between posted results and a full statistical report. The outcome counts and comparisons are available, while analysis details such as missing-data methods, multiplicity handling, and analysis populations are not reported in the registry record. Recognizing both what the summaries quantify and what they leave unspecified is central to careful biostatistical interpretation.

17. 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 summary gives a broad description of the study population but does not provide detailed inclusion and exclusion criteria. The primary outcome definition combines events with different clinical meanings, and component-specific counts are not reported in the registry record. As a result, the overall treatment-failure comparison cannot identify which type of event contributed to the group results.

The posted outcome is a participant count over 16 weeks, without event timing or recurrent-event information. Longer-term outcomes are not reported in the registry record. The record also does not report detailed information about treatment adherence, changes in assigned medication, withdrawals, or missing outcome data, so their influence on the posted comparisons cannot be evaluated.

The original statistical analysis plan, sample-size justification, confidence-interval method, analysis population, and any adjustment for multiplicity are not reported in the registry record. The posted p-values are unadjusted, and the registry lists zero posted analyses. These gaps limit assessment of whether the displayed comparisons follow the trial’s prespecified analysis approach.

The reported contrasts compare each active strategy with fluticasone; a direct montelukast-versus-combination comparison is not reported. The safety summaries are also counts without reported adjusted comparisons or uncertainty estimates. These limitations restrict the scope of conclusions that can be drawn from the registry summaries, but they do not establish that unreported procedures or analyses were not performed.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: The Leukotriene Modifier Or Corticosteroid or Corticosteroid-Salmeterol Trial — NCT00156819.
• Registry record last updated 2015-12-09; status last verified 2015-11.
• Results first posted 2015-12-09.
• 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.

21. Record Summary

NCT00156819 is a completed Phase 4 randomized, parallel-group, quadruple-masked trial sponsored by JHSPH Center for Clinical Trials. The registry records an enrollment of 500, a start date of 2003-06, and primary completion in 2005-08. It compared continued fluticasone with montelukast and fluticasone plus salmeterol among people described as having asthma controlled with fluticasone.

The primary outcome was treatment failure over 16 weeks, defined as the occurrence of any one of several clinical, treatment, or safety events. The registry posts two risk-difference comparisons against fluticasone, with unadjusted p-values and confidence intervals, as well as descriptive safety counts. It does not report secondary outcomes or analyses. Details of the original statistical plan, sample-size justification, analysis population, and missing-data handling are not reported in the registry record.