Assessment of Intraocular Pressure (IOP) Control in Subjects With Open-Angle Glaucoma or Ocular Hypertension Treated With Travoprost 0.004% (TRAVATAN® Z) or Bimatoprost 0.01% (LUMIGAN®): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Assessment of Intraocular Pressure (IOP) Control in Subjects With Open-Angle Glaucoma or Ocular Hypertension Treated With Travoprost 0.004% (TRAVATAN® Z) or Bimatoprost 0.01% (LUMIGAN®) (NCT01464424), a randomized phase 4 study. The primary outcome is Overall Mean Intraocular Pressure (IOP).

2. Clinical Question

The clinical question concerns IOP control during the after-office-hours period, rather than an average across the entire day. The scheduled measurement times corresponded to 20, 22, and 24 hours post dose. The registry’s brief summary describes efficacy and tolerability assessment after 6 weeks of treatment in each treatment period. Because this was a crossover trial, each sequence received both treatments in a different order; the comparison therefore sits within a design that includes treatment sequence and period as relevant features. Details of how those features were incorporated into the sponsor’s primary analysis are not reported in the registry record.

3. Trial Design

This was a randomized, single-masked, phase 4 crossover study sponsored by Alcon Research. The registry lists an enrollment of 84 and a completed status. The study began in 2011-10 and reached primary completion in 2012-06. Its stated focus was late-day IOP control among subjects with open-angle glaucoma or ocular hypertension.

Participants were assigned to one of two sequences. One sequence received travoprost 0.004% for 6 weeks, followed by bimatoprost 0.01% for 6 weeks. The other sequence received bimatoprost 0.01% first and travoprost 0.004% second, with the same stated dose and duration. Both were given as one drop to the study eye once daily at 8:00 pm. The total duration for each sequence was 12 weeks.

A crossover design compares treatments across periods within the assigned sequence, rather than assigning each participant to only one treatment for the entire study. This can be useful for continuous outcomes, but interpretation can depend on period effects, sequence effects, and whether effects from an earlier treatment carry into a later period. The registry record does not report details about washout, carryover assessment, or the statistical handling of period and sequence.

The primary endpoint was overall mean IOP at week 6, based on readings at 4 pm, 6 pm, and 8 pm. The stated analysis population was per protocol, and only the designated study eye was analyzed for efficacy. The record does not report how many participants completed each sequence or how many were included in the primary analysis beyond the posted arm counts. It also does not identify which parties were masked, or give further details about randomization procedures.

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: This reporting group includes all enrolled subjects. A washout-period based on prior medication preceded Period 1 dispense.

Period 1, First 6 Weeks

Reasons for not completing this period:

Period 2, Second 6 Weeks

Analysis population for the primary outcome, as defined in the registry: Per protocol: All subjects who received study medication, completed all study visits as per the protocol timelines and criteria, and satisfied inclusion/exclusion criteria.

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.

Population: This reporting group includes all enrolled subjects.

6. Primary Endpoint

The registry lists the primary outcome as Overall Mean Intraocular Pressure (IOP), measured over Week 6.

The primary endpoint was overall mean intraocular pressure at week 6, measured in millimeters of mercury (mmHg). It combined IOP measurements taken at 4 pm, 6 pm, and 8 pm into one overall mean for each treatment. Those evaluation times were described as 20, 22, and 24 hours post dose, so the endpoint focuses on late-day pressure after an evening dose rather than on a single measurement time.

IOP is a continuous measure, meaning it can take a range of values rather than representing a simple event count. Reporting a mean provides a summary of the group’s measured pressure, while the standard deviation describes how dispersed individual measurements were around that mean. The endpoint is suitable for comparing average values between treatments, although group means do not show the full distribution of individual responses.

The registry specifies that efficacy was assessed in one eye only: the designated study eye. It also identifies the per-protocol population as the prespecified population for the non-inferiority analysis. This population included subjects who received study medication, completed visits according to protocol timelines and criteria, and met inclusion and exclusion criteria. The registry does not report the non-inferiority margin, the clinical rationale for selecting this endpoint, or details of the sponsor’s analysis model.

7. Results Posted in the Registry

• LUMIGAN: mean 17.22 versus 17.45 on control; difference -0.23 (95% CI -1.04 to 0.58), standardized effect d = -0.09; z-test p = 0.579.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a continuous measure, so the registry reports an arithmetic mean and standard deviation for each treatment. A mean summarizes the central level of measured IOP. A standard deviation describes variability among the observations; it is not the same as a confidence interval and does not directly indicate how precisely the group mean has been estimated.

What does the treatment difference represent?

The posted contrast is the mean for LUMIGAN relative to the mean for TRAVATAN. Its estimate is -0.23 mmHg. A negative sign indicates that the posted LUMIGAN mean was lower than the posted TRAVATAN mean for this stated contrast. The estimate is an average difference, not a statement that every participant had a lower measurement on one treatment. Individual responses can vary around the group averages.

The registry also gives an unadjusted p-value of 0.579. An unadjusted p-value is calculated without an adjustment for multiple comparisons. It does not express the size or clinical importance of the difference, and it is not the probability that the null hypothesis is true. The record lists one primary outcome and one secondary outcome, but it does not report a multiplicity strategy or the sponsor’s complete testing plan. Therefore the unadjusted value should be understood as a limited summary, not as a substitute for the prespecified analysis.

Why does the crossover design matter?

In a crossover trial, treatment order is randomized and participants receive treatments in successive periods. This structure can allow comparisons that account for measurements from the same participant under different treatments. However, a valid analysis must also consider that observations within a participant are related, and that measurements can differ by period or sequence. Effects from an earlier treatment may also influence a later period if they persist. These are general statistical considerations for crossover studies; the registry record does not report whether or how the sponsor modeled within-person correlation, period, sequence, or carryover.

The registry reports two treatment sequence descriptions, each with a 6-week treatment period followed by the other treatment for 6 weeks. The primary endpoint timeframe is week 6. That timeframe identifies the endpoint timing but does not by itself explain which period observations entered the comparison or how the crossover data were analyzed. Those details are not reported in the registry record. The summary statistics can be described as posted, but they do not allow an independent reconstruction of a crossover model.

What does per protocol mean for this result?

The stated primary analysis population was per protocol. The registry defines it as subjects who received study medication, completed all study visits according to protocol timelines and criteria, and satisfied inclusion and exclusion criteria. A per-protocol analysis therefore pertains to participants meeting those stated conditions. It is not interchangeable with an analysis of everyone randomized, and it may be affected by who did not meet the criteria or lacked eligible measurements. The numbers and reasons for exclusions are not reported in the registry record.

Only the designated study eye was included in efficacy analysis. This avoids treating two eyes from one participant as if they were independent observations in the primary efficacy assessment. The registry does not report further details of eye selection or whether the study eye was selected before treatment assignment.

How should the non-inferiority label be understood?

Non-inferiority tests whether the data are sufficiently compatible with a prespecified limit for how much worse the experimental treatment could be while still meeting the study’s criterion. The margin and direction of the comparison are essential to that interpretation. Neither the margin nor a complete decision rule is reported in the registry record. Consequently, the presence of a non-inferiority label does not allow the criterion to be checked using only the posted contrast and unadjusted p-value.

The posted values permit a descriptive comparison of means and a look at the reported confidence interval. They do not reveal the full sponsor analysis, including whether adjustment or a specialized crossover model was used. The registry lists no analyses in its analysis count, and no further analysis specification is reported. Where an independent summary calculation differs from a sponsor analysis, the sponsor’s own analysis takes precedence; the record indicates no discrepancy for this trial.

What can the posted summaries establish?

They establish the reported group means, standard deviations, contrast, confidence interval, and unadjusted p-value for the stated endpoint and population. They do not establish how robust the result is to missing data, protocol deviations, alternative analysis populations, or assumptions about the crossover structure. General statistical interpretation therefore distinguishes the numerical estimate from the design-specific conclusion. The estimate gives the direction and size of the observed average contrast; the confidence interval expresses uncertainty; and the study’s non-inferiority conclusion would additionally require the prespecified margin and analysis details, which are not reported in the registry record.

10. Confidence Intervals and Precision

The posted mean difference for LUMIGAN relative to TRAVATAN is -0.23 mmHg. The reported confidence interval extends from -1.041 mmHg to a positive upper limit of 0.581 mmHg. The interval includes zero, indicating that the reported uncertainty is compatible with no average difference as well as with differences on either side of zero.

A confidence interval is a range of values that, under the statistical method and its assumptions, represents uncertainty about the estimated treatment difference. It is not a range containing a specified proportion of individual participants’ measurements, and it is not the probability that the true difference lies inside this particular interval. The standard deviations, 2.664 mmHg for TRAVATAN and 2.605 mmHg for LUMIGAN, describe variation in the posted outcome measurements rather than uncertainty in the treatment contrast.

The p-value of 0.579 is consistent with the interval’s inclusion of zero for an unadjusted comparison, but it does not answer the non-inferiority question by itself. That question requires a prespecified margin, which is not reported in the registry record. The precision of the estimate should therefore be considered alongside the missing margin and the unreported details of the crossover analysis.

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 sample-size figures are planning illustrations based on the observed effect, not the original trial’s design assumptions. They give participants per arm for 80% power and participants per arm for 90% power. They are not total event counts: the endpoint is continuous, and the planning figures are stated per arm.

The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to a stated fraction of the observed effect. The rows describe planning for 100%, 75%, and 50% of the observed effect. For the 100% scenario, the illustration lists 2,060 participants per arm for 80% power and 2,758 participants per arm for 90% power. For the 75% scenario, it lists 3,662 and 4,903 participants per arm, respectively. For the 50% scenario, it lists 8,240 and 11,031 participants per arm, respectively.

In plain terms, the 80% power column gives the planned number per arm under which the assumed calculation would have an 80% chance of detecting the specified effect if its assumptions hold. The 90% column uses a higher target chance of detection and therefore lists larger planning numbers. Smaller effects need more participants to detect under the same assumptions. These figures are not observed power. The trial’s original design assumptions, including its planned sample size rationale, 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, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -0.230; pooled standard deviation σ = 2.635.
2. n = 2 × (1.96 + 0.842)² × 2.635² / -0.230² = 2,059.8, rounded up to 2,060 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 <- -0.23                      # observed difference in means
sigma <- 2.6346651589908                     # 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(1464424)
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 = 2,060 (80% power), 2,758 (90% power)
 75% of observed effect: n per arm = 3,662 (80% power), 4,903 (90% power)
 50% of observed effect: n per arm = 8,240 (80% power), 11,031 (90% power)
power.t.test: 2060.8 
Simulated power at n = 2,060 per arm: 0.805 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome, overall mean IOP, and one secondary outcome, mean IOP at each after-office-hours evaluation timepoint. It reports two outcome measures in total. The registry record does not describe whether the analysis adjusted for testing more than one outcome, for multiple timepoints, or for other comparisons. No multiplicity procedure is reported, so the unadjusted p-value should not be interpreted as a multiplicity-adjusted result.

The primary endpoint combines readings at 4 pm, 6 pm, and 8 pm into an overall mean, while the secondary outcome concerns mean IOP at each of those evaluation timepoints. The record does not provide the posted secondary-outcome results in the registry record for this analysis. It is therefore not possible to describe their estimates or how they relate to the primary comparison.

Missing-data handling is also not reported in the registry record. The stated per-protocol population required completion of study visits according to protocol timelines and criteria, but the record does not give the number or reasons for missing measurements, exclusions, or withdrawals from the primary analysis. It does not specify whether imputation was used or how incomplete crossover periods were treated.

The analysis population was per protocol, and efficacy was assessed using one designated study eye. The registry does not report an intention-to-treat result or a sensitivity analysis comparing alternative populations. Details of the statistical model for the crossover design, including handling of period, sequence, within-participant correlation, and possible carryover, are not reported. The non-inferiority margin and full testing plan are likewise absent. These omissions limit independent assessment of how the posted result was derived and how robust it is to analytic choices.

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.

The registry reports serious adverse event figures for both treatment groups. For TRAVATAN, 84 were listed at risk and 0 were reported as affected by serious adverse events. For LUMIGAN, 84 were listed at risk and 0 were reported as affected by serious adverse events. The listed figures describe the registry’s serious-adverse-event reporting for the treatment groups; they do not establish that no adverse effects of any kind occurred.

The record also lists 0 affected for other adverse events in each group, with 84 at risk in each. However, the registry record does not provide event terms in the serious-adverse-event list, and it does not report death counts. It does not give further detail about event collection, severity grading, treatment relatedness, or the time period used for these safety summaries.

These counts should be read as posted safety summaries, not as a complete account of tolerability. A count of zero reported affected participants does not prove that an event is impossible or that treatment is risk-free, particularly without further information about ascertainment and reporting. The brief study summary states that tolerability was assessed, but additional tolerability results are not reported in the registry record.

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

The posted results show mean IOP values of 17.45 mmHg for TRAVATAN and 17.22 mmHg for LUMIGAN at the stated week 6 endpoint. The estimated mean difference for LUMIGAN relative to TRAVATAN is -0.23 mmHg. The reported confidence interval includes zero, and the unadjusted p-value is 0.579. Together, these figures describe a modest observed difference with uncertainty that includes no difference.

They do not show that the treatments are equivalent. They also do not, by themselves, establish non-inferiority. The registry says non-inferiority was prespecified, but the margin and decision rule are not reported in the registry record. A conclusion on that specific question cannot be recovered from the label alone.

The results apply to the per-protocol analysis population, using the designated study eye, and to the endpoint timing described in the registry. They should not be generalized automatically to other populations, other treatment schedules, measurements at other times, or analyses using different populations. The record does not provide details needed to assess the effect of missing data or the full crossover analysis.

The serious-adverse-event summaries list 0 affected in each treatment group, but these figures do not describe all dimensions of safety or tolerability. The registry’s brief summary includes tolerability as a study purpose, while further details are not reported in the registry record.

Statistical results describe data under specified design and analysis assumptions. They do not make treatment decisions for an individual, and the posted comparison is not a medical recommendation. The most defensible reading is limited to the reported group averages, their uncertainty, the stated analysis population, and the missing information needed to assess the intended non-inferiority claim.

16. Why This Trial Matters Statistically

This trial illustrates why a treatment comparison cannot be interpreted from point estimates alone. The mean difference is -0.23 mmHg, but the confidence interval spans values below and above zero. The estimate gives the observed direction and magnitude of the average contrast, while the interval conveys uncertainty about that contrast under the analysis method.

The non-inferiority framing makes the absent margin especially important. Non-inferiority is not established merely because a conventional difference test is not statistically significant. The acceptable boundary must be prespecified, and the estimated difference and its uncertainty must be assessed against it. Because the margin is not reported in the registry record, readers cannot verify the intended criterion from the posted summaries.

The crossover design also emphasizes the need to understand the analysis model. Each treatment was used in sequence, so treatment, period, and sequence can matter to interpretation. The registry record gives the randomized sequences and treatment durations, but not how the sponsor handled these features analytically. That leaves an important part of the inferential basis unavailable for independent review.

Finally, planning illustrations derived from an observed effect can help explain how assumptions about effect size influence planned sample size. They are not evidence of the trial’s design rationale and are not observed power. In this case, the examples make clear that aiming to detect a smaller effect requires more participants under the stated planning assumptions. Overall, the trial is a useful example of the distinction between descriptive findings, precision, and a design-specific claim that requires additional reported information.

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 record does not report the non-inferiority margin or the sponsor’s complete decision rule, so the principal non-inferiority objective cannot be independently assessed from the posted contrast. The record also does not provide a detailed statistical model for the crossover design. In particular, it does not describe handling of period or sequence effects, within-participant correlation, or possible carryover between treatments.

The primary analysis was per protocol and limited to the designated study eye. The registry does not report the number or reasons for exclusions from that population, missing-data methods, or results from an intention-to-treat or sensitivity analysis. These omissions limit assessment of the effect of protocol adherence and incomplete observations on the estimate.

The posted comparison is based on the overall mean across 4 pm, 6 pm, and 8 pm measurements. The registry record does not include the secondary-outcome values for each timepoint in the registry record for this analysis. Therefore, the overall average cannot show whether the treatment contrast varied across those measurement times.

Although the registry lists serious and other adverse-event counts, it provides no serious-event terms and no death counts. Further details about event ascertainment, timing, severity, and relatedness are not reported. The sample-size illustrations are based on the observed effect and stated planning assumptions, not the trial’s original design; the original sample-size justification is not reported in the registry record.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Assessment of Intraocular Pressure (IOP) Control in Subjects With Open-Angle Glaucoma or Ocular Hypertension Treated With Travoprost 0.004% (TRAVATAN® Z) or Bimatoprost 0.01% (LUMIGAN®) — NCT01464424.
• Registry record last updated 2013-09-13; status last verified 2013-09.
• Results first posted 2013-09-04.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

21. Record Summary

NCT01464424 is a completed, randomized, single-masked, phase 4 crossover study sponsored by Alcon Research. It enrolled 84 subjects with open-angle glaucoma or ocular hypertension and compared treatment sequences using travoprost 0.004% and bimatoprost 0.01%.

The primary endpoint was overall mean IOP at week 6, based on readings at 4 pm, 6 pm, and 8 pm. The per-protocol analysis used the designated study eye. Posted means were 17.45 mmHg for TRAVATAN and 17.22 mmHg for LUMIGAN; the mean difference was -0.23 mmHg, with an unadjusted p-value of 0.579.

The registry identifies the analysis as a prespecified non-inferiority analysis, but the non-inferiority margin and detailed crossover analysis methods are not reported in the registry record. Serious adverse-event summaries list 0 affected in each treatment group. These findings should be interpreted within the stated population and the limits of the reported analysis details.