Efficacy of Travoprost/Timolol for Uncontrolled Intraocular Pressure: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy of Travoprost/Timolol for Uncontrolled Intraocular Pressure (NCT02003391), a randomized phase 4 study. The primary outcome is Least Squares Mean Intraocular Pressure (IOP) at 8AM in the Study Eye.

2. Clinical Question

3. Trial Design

This was a phase 4, randomized, parallel-group, open-label study. The official title describes it as prospective and multi-center. The registry lists Alcon Research as sponsor, enrollment of 157, and a completed status. The study start is recorded as 2013-12 and primary completion as 2015-05. The listed conditions are open-angle glaucoma and ocular hypertension.

Participants were assigned to a DuoTrav arm or a beta-blocker arm. The DuoTrav regimen was travoprost 0.004% / timolol 0.5% fixed-combination solution, 1 drop in the study eye once daily in the evening for 8 weeks. The other arm used participants’ current beta-blocker monotherapy, 1 drop twice daily for 4 weeks, followed by the fixed combination once daily for 4 additional weeks. Because the primary endpoint was measured at Week 4, its contrast concerns the groups at that assessment, rather than a comparison of the two arms’ full 8-week regimens.

The primary outcome was least squares mean intraocular pressure at 8AM in the study eye at Week 4, measured by Goldmann applanation tonometry. The registry specifies an intent-to-treat population with a Week 4 study-eye measurement. The posted group counts for this endpoint are 78 for DuoTrav and 72 for beta-blocker; these are analysis counts, not the registry’s enrollment figure. Two secondary outcomes are also listed: mean change and percentage change from baseline in 8AM pressure in the study eye at Week 4.

The record identifies random allocation, parallel groups, and no masking. It does not report the allocation procedure, concealment details, masking rationale, recruitment setting, or a sample-size justification. Details of interim analyses and the full statistical plan are not reported in the registry record.

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: Of the 157 participants enrolled, 1 participant was exited as a screen failure. This reporting group includes all randomized participants (156).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent to treat with a measurement in the study eye at Week 4

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: Intent to treat population included participants who received study medication and had at least one on-therapy study visit.

6. Primary Endpoint

The registry lists the primary outcome as Least Squares Mean Intraocular Pressure (IOP) at 8AM in the Study Eye, measured over Week 4.

The primary endpoint is least squares mean intraocular pressure at 8AM in the study eye at Week 4, expressed in mmHg. The registry says pressure was assessed with Goldmann applanation tonometry and notes that higher pressure can be a risk factor for glaucoma development or progression, including optic-nerve damage. The description also specifies that one eye contributed to the analysis, so the endpoint is not an average across both eyes.

Pressure is relevant to the trial’s stated question because the study aimed to compare pressure control between the fixed-combination regimen and beta-blocker monotherapy in participants previously uncontrolled on a beta-blocker. Measuring at a specified morning time and using the study eye defines the outcome consistently for the reported comparison. The result is a measurement at the stated assessment, not a direct measure of every aspect of the condition or of longer-term clinical outcomes.

The sponsor reports least squares means, which are model-based adjusted group estimates, and an ANCOVA comparison of final values. The record does not provide the covariates used in that adjustment. Its primary population is described as intent to treat with a Week 4 measurement in the study eye. Thus, the endpoint is interpretable as a Week 4 pressure comparison for participants represented in that analysis, while the handling of missing measurements is not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary comparison is reported as an ANCOVA analysis of final values. ANCOVA is a general method for comparing group outcomes while adjusting for specified covariates. Adjustment can improve precision or account for measured baseline differences, depending on the model and its assumptions. The covariates and detailed model specification are not reported in the registry record, so the exact adjustment cannot be reconstructed from the posted summary.

What does a least squares mean represent?

In general, a least squares mean is a model-based group mean estimated after accounting for variables included in the model. It is not necessarily the same as a simple arithmetic mean of all observed measurements. The registry reports primary least squares means of 16.61 mmHg and 21.23 mmHg for the two arms, but does not give the model terms needed to reproduce those adjusted estimates independently.

How should the reported difference and p-value be read?

The sponsor’s reported mean difference in final values is -4.619, with p < 0.001. The listed group order is DuoTrav followed by beta-blocker, and the direction is consistent with a lower adjusted final pressure in the DuoTrav arm. In general, a p-value assesses how compatible observed data are with a specified null model under the test’s assumptions. It is not an effect-size measure and does not convey the full range of plausible effect values.

A confidence interval describes uncertainty in an estimate under a statistical procedure. The primary comparison is labeled as having a 95% confidence interval, but the lower endpoint is not reported; only the upper endpoint, -3.850, appears. Similarly, the arm summaries list upper limits but not lower limits. Since these intervals are incomplete in the registry record, a full interval-based assessment of primary-result precision is not possible from the reported values alone.

What do the secondary recalculations add?

For each secondary continuous outcome, the posted summaries include group means, standard deviations, and an unadjusted recalculated mean difference with a confidence interval and p-value. The mean-change comparison is 4.771 mmHg, with a 95% confidence interval from 3.806 to 5.736 and p < 0.001. The percentage-change comparison is 20.65 percentage points, with a 95% confidence interval from 16.54 to 24.76 and p < 0.001. The comparison is expressed as beta-blocker minus DuoTrav for these changes: because the DuoTrav group has the more negative mean change, the displayed difference is positive. These are simple unadjusted recalculations from the posted summary statistics, not sponsor-reported analyses.

In general, a standard deviation describes variation among observations within a group, whereas a confidence interval describes uncertainty around an estimated group contrast. The secondary recalculations also report standardized effects of 1.563 and 1.596. Standardization expresses a mean difference relative to a pooled measure of variability; it can aid comparison across scales, but does not replace interpretation in the original clinical units. The recalculated p-values and intervals rely on summary-statistic methods and do not account for any unreported covariate adjustment or analysis features.

Why can over-dispersion matter?

In general, over-dispersion means that observed variability is greater than a statistical model expects. When a model underestimates variability, standard errors can be too small and confidence intervals too narrow, which can make evidence appear more precise than it is. The registry’s primary analysis is described as ANCOVA, and the secondary outcomes are continuous measures with standard deviations. The record does not identify an over-dispersion issue or provide information that would establish one; this general point is not a claim about this trial.

How should the planning illustrations be used?

The sample-size illustrations are based on the observed secondary effects, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. They show participants per arm for 80% and 90% power when planning for 100%, 75%, or 50% of the observed effect. For the mean-change outcome, the displayed values are 7 and 9 for planning at 100% of the observed effect, 12 and 16 at 75%, and 26 and 35 at 50%. For percentage change, they are 7 and 9, 11 and 15, and 25 and 34 for those respective effect assumptions. In general, such calculations illustrate how sample-size requirements depend on the effect one plans to detect and the selected design assumptions. They are planning illustrations, not observed power, and they do not establish the trial’s original design rationale; that rationale is not reported in the registry record.

The registry does not report a multiplicity adjustment, missing-data method, interim-analysis plan, or fuller statistical analysis plan. Those absences matter when judging the strength and scope of evidence, particularly for secondary outcomes and for a primary analysis population defined by having a Week 4 measurement.

10. Analyses Reported by the Sponsor

The sponsor’s own primary analysis is the ANCOVA comparison of final values, reporting a mean difference of -4.619 mmHg and p < 0.001. Its direction agrees with the posted arm estimates: the DuoTrav least squares mean is lower than the beta-blocker least squares mean at Week 4. The registry’s comparison interval is labeled 95%, but the lower endpoint is not reported; the upper endpoint is -3.850. The lower limits for the individual arm estimates are also not reported.

The recalculated comparisons concern the two posted secondary outcomes and use unadjusted calculations from group summary statistics. They are not recalculations of the sponsor’s primary ANCOVA. The registry record identifies no discrepancy between the primary result and a recalculated primary result, and no recalculated primary comparison is presented. Accordingly, the sponsor’s own analysis remains the relevant reported primary analysis; the secondary recalculations can provide a separate, descriptive check on the pattern across outcomes, but do not supersede it.

The primary model’s covariates and full specification are not reported in the registry record, which limits independent reproduction. The incomplete confidence-interval reporting also limits assessment of the primary estimate’s precision. The small p-value is evidence against the stated null under the sponsor’s model assumptions, but does not remedy missing interval information or establish the effect’s persistence beyond the reported endpoint.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint and two secondary outcomes. The primary comparison is reported by the sponsor as an ANCOVA analysis. The secondary comparisons are recalculated from posted summary statistics and are unadjusted. The registry does not report whether the study used a testing hierarchy, family-wise error control, or another multiplicity procedure. Without that information, the secondary p-values should not be interpreted as if each had been the sole prespecified test. In general, multiple testing can increase the chance of a nominally unusual result when several hypotheses are examined.

The primary analysis population is described as intent to treat with a Week 4 measurement in the study eye. This wording indicates that the reported endpoint population is tied to the availability of that measurement, but the registry does not report how many participants lacked a measurement, why data were missing, or how missing values were handled. It also does not report sensitivity analyses for missing data. Those details are important for assessing whether the analysis could be affected by differences between participants with and without recorded outcomes.

The primary endpoint group counts are 78 for DuoTrav and 72 for beta-blocker. The registry enrollment is 157. These figures refer to different record elements: enrollment describes the trial, while the arm counts describe participants represented in the posted primary outcome analysis. The record does not provide enough detail to reconcile every participant’s disposition or explain any differences between enrollment and analysis counts. No further analysis population definition is reported for the secondary endpoints beyond the posted group summaries.

Randomization and an intent-to-treat label are relevant design features, but they do not by themselves establish how missing observations, protocol deviations, or other analysis decisions were handled. The full statistical analysis plan, sample-size justification, interim-analysis information, and missing-data method are not reported in the registry record. The interpretation should therefore stay with the estimates and analyses explicitly posted, while recognizing these limitations in the evidential record.

12. 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 1% reporting threshold.

Most frequent non-serious adverse event terms

The posted safety figures list serious adverse events affecting 0 of 81 participants at risk in the DuoTrav group and 0 of 75 at risk in the beta-blocker group. The registry lists no serious adverse-event terms among the top serious terms. These figures describe the reported serious-event counts and the corresponding at-risk group counts; they do not demonstrate that no serious events occurred outside the record’s reporting or observation framework.

Other adverse events are reported as affecting 18 of 81 participants at risk in the DuoTrav group and 9 of 75 in the beta-blocker group. These are participant counts, not event totals, and the record excerpt does not identify the specific other-event terms. The trial is open-label, and the safety data should be read as recorded rather than as a blinded comparison.

Deaths are marked as not reported in the registry record. The available figures do not establish comparative safety for rare events or events emerging beyond the recorded follow-up. They also should not be used to infer an individual participant’s risk. The safety counts are descriptive and do not alter the interpretation of the primary pressure endpoint.

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

The primary result is a Week 4 comparison of morning pressure in the study eye among participants with a reported measurement. The sponsor’s ANCOVA estimate and the arm-level least squares means point in the same direction: the adjusted pressure estimate is lower in the DuoTrav arm than in the beta-blocker arm. The reported p-value indicates incompatibility with the null comparison under the sponsor’s model assumptions. The incomplete confidence-interval reporting means the full interval-based precision of the primary contrast cannot be evaluated from the registry record.

The result does not establish how long the observed pressure difference persists, whether it translates into a difference in optic-nerve outcomes, or what happens beyond the recorded assessments. The primary endpoint is a pressure measure, not a direct measurement of glaucoma progression or visual function. The registry also does not report sufficient detail about missing-data handling or model specification to independently reconstruct every part of the primary analysis.

The secondary results have a similar direction, but they are unadjusted recalculations of related change-from-baseline outcomes. Their consistency is informative as a pattern, yet they do not remove uncertainty about multiplicity, missing data, or the primary model. The posted sample-size illustrations likewise describe hypothetical planning under stated assumptions, not the trial’s achieved power or its original rationale.

These findings are an account of the trial’s reported statistical results, not individualized evidence for a particular patient. They do not by themselves establish a treatment recommendation, an overall benefit-risk judgment, or a result for groups and outcomes not represented in the registry record. Such conclusions would require evidence beyond the limited endpoint and reporting details.

14. Why This Trial Matters Statistically

This trial offers a clear example of why the primary analysis method and the posted outcome summaries must be kept distinct. Its primary endpoint is a model-based least squares mean comparison analyzed with ANCOVA, while the secondary contrasts are unadjusted recalculations from group summary statistics. Treating these analyses as interchangeable would obscure what each estimate represents and how it was obtained.

The record also illustrates how direction and precision are separate questions. The primary estimate and arm summaries point toward lower pressure in the DuoTrav group at the stated assessment, and the sponsor reports a small p-value. Yet the primary confidence interval is incomplete in the registry record. A p-value does not fill in the missing interval endpoint or provide the same information about the plausible range of the estimate.

Outcome timing and definition matter as well. The primary outcome is an 8AM measure in the study eye at Week 4. The beta-blocker arm’s described regimen changes after its initial 4 weeks, while the primary assessment is at Week 4. This makes it especially important to interpret the endpoint at its stated time rather than as a general comparison of the entire treatment schedules.

Finally, the secondary outcomes are related pressure-change measures, and the registry does not report multiplicity handling. Their aligned direction adds context but should not be counted as independent confirmation without considering their relationship to the primary measure and the unadjusted nature of the recalculations. Statistical interpretation therefore depends on estimates, analysis methods, timing, and reporting completeness together.

15. 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.

Several limitations arise from details absent in the registry record. The primary ANCOVA model’s covariates and full specification are not reported, preventing complete independent reproduction of the sponsor’s adjusted analysis. The primary confidence interval has only an upper endpoint reported, and the arm-level intervals likewise lack reported lower endpoints. Consequently, the full precision of the primary estimates cannot be assessed from the posted interval information.

The analysis population requires a Week 4 measurement in the study eye, but the record does not describe missing observations, reasons for missingness, imputation, or sensitivity analyses. It does not provide participant disposition details sufficient to explain the relationship between enrollment and the posted primary analysis counts. The record also does not report the allocation procedure, concealment methods, or baseline characteristics needed to assess balance and generalizability in greater detail.

The study is listed as open-label, and the registry does not report masking. The primary endpoint is a measurement outcome, but lack of masking can still be relevant to study conduct and assessment. No claim about the direction or size of any such influence can be made from the record.

Multiplicity handling for the secondary outcomes is not reported. Those findings are unadjusted recalculations from summary data, not sponsor-reported analyses, and the outcomes are related measures of pressure at the same assessment point. The sample-size illustrations are based on observed effects and stated planning assumptions; the trial’s own sample-size justification is not reported. Interim analyses and a complete statistical analysis plan are also not reported. These limitations constrain interpretation without negating the posted estimates.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Efficacy of Travoprost/Timolol for Uncontrolled Intraocular Pressure — NCT02003391.
• Registry record last updated 2016-08-08; status last verified 2016-06.
• Results first posted 2016-08-08.
• 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.

19. Record Summary

The registry record describes a completed, randomized, parallel, open-label phase 4 study sponsored by Alcon Research. It enrolled 157 participants with open-angle glaucoma or ocular hypertension who had previously been uncontrolled on a beta-blocker. The primary outcome was least squares mean pressure at 8AM in the study eye at Week 4.

The sponsor reports an ANCOVA mean difference in final values of -4.619 mmHg, with p < 0.001; the lower endpoint of the labeled 95% confidence interval is not reported. Two secondary pressure-change outcomes are posted with unadjusted recalculations that follow the same general direction. Multiplicity handling, missing-data methods, the full primary model specification, and the trial’s sample-size rationale are not reported in the registry record. Safety figures list no serious adverse events in either group and provide counts for other adverse events; deaths are not reported.