← Clinical Trials
Peripheral Artery Disease Phase 3 Randomized NCT02504216

VOYAGER PAD: Complete Statistical Analysis of Rivaroxaban in Peripheral Artery Disease

An independent statistical review of the randomized phase 3 VOYAGER PAD trial evaluating rivaroxaban 2.5 mg twice daily plus aspirin 100 mg once daily versus rivaroxaban placebo twice daily plus aspirin 100 mg once daily in subjects with symptomatic peripheral artery disease undergoing peripheral revascularization procedures of the lower extremities.

Completed trial  ·  Enrollment: 6564  ·  Sponsor: Bayer
Scope of this record

This page separates reported trial results from statistical interpretation. Trial facts and numerical results on this page are restricted to the ClinicalTrials.gov data posted on ClinicalTrials.gov for VOYAGER PAD. The registry provides the official trial record.

Registry context: This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record.

1. Trial at a Glance

VOYAGER PAD was a randomized, parallel-group, quadruple-masked phase 3 trial evaluating rivaroxaban plus aspirin against rivaroxaban placebo plus aspirin in subjects with symptomatic peripheral artery disease undergoing peripheral revascularization procedures of the lower extremities.

6564
Enrollment
Randomized trial
2
Arms
Parallel design
0.85
Primary efficacy HR
95% CI 0.76–0.96
1.43
Primary safety HR
95% CI 0.97–2.10
FeatureVOYAGER PAD
TrialVOYAGER PAD
PhasePhase 3
Therapeutic areaCardiovascular
ConditionPeripheral Artery Disease
DesignRandomized, parallel-group, quadruple-masked
AllocationRandomized
Primary purposePrevention
Enrollment6564
Primary endpointsPrimary efficacy outcome and primary safety outcome
Results postedYes
Statistical analyses posted11
ClinicalTrials.govNCT02504216
Lead sponsorBayer
Sponsor typeIndustry

2. Clinical Question

The central statistical question was whether rivaroxaban 2.5 mg twice daily added to aspirin 100 mg once daily was associated with a different time-to-event profile for the composite primary efficacy outcome compared with rivaroxaban placebo twice daily plus aspirin 100 mg once daily in subjects with symptomatic peripheral artery disease undergoing lower-extremity peripheral revascularization procedures.

Population

Subjects with symptomatic peripheral artery disease undergoing peripheral revascularization procedures of the lower extremities.

Intervention

Rivaroxaban 2.5 mg Bid + Aspirin 100 mg od.

Comparator

Rivaroxaban Placebo Bid + Aspirin 100 mg od.

Primary question

How do the randomized groups compare for the first occurrence of the composite efficacy outcome, while also evaluating TIMI Major Bleeding as the primary safety outcome?

3. Trial Design

01
Randomize6564 participants
02
Two armsRivaroxaban or placebo
03
Aspirin100 mg od in both groups
04
Follow eventTime-to-event outcomes
05
AnalyzeStratified log-rank framework
ARM A

Rivaroxaban combination

  • Rivaroxaban 2.5 mg Bid
  • Aspirin 100 mg od
  • Randomized treatment assignment
ARM B

Placebo combination

  • Rivaroxaban placebo Bid
  • Aspirin 100 mg od
  • Randomized treatment assignment

The registry describes the study as randomized, parallel, and quadruple-masked. Those design features are statistically important because randomization establishes the basis for a causal comparison, the parallel structure means participants remain assigned to their randomized treatment pathway rather than forming sequential treatment groups, and masking is intended to reduce knowledge of assignment during trial conduct.

4. Trial Timeline and Registry Status

2015-08-18

Trial start

The ClinicalTrials.gov record lists 2015-08-18 as the study start date.

2019-11-27

Primary completion

The registry lists 2019-11-27 as the primary completion date.

Completed

Registry status

The trial is listed as COMPLETED, with results posted and 11 statistical analyses posted.

5. Randomization, Masking, and Analysis Populations

Randomization is the foundation of the primary efficacy comparison. Because treatment assignment is randomized rather than selected after observing participant outcomes, the treatment groups can be compared according to their assigned interventions using the randomized population.

FeatureRegistry information
AllocationRANDOMIZED
Design modelPARALLEL
MaskingQUADRUPLE
Primary purposePREVENTION
Enrollment6564
Efficacy analysis populationIntention-to-treat analysis set: all randomized participants
Safety analysis populationSafety analysis set: all treated participants, defined as randomized participants who received at least one dose of study drug

The distinction between the ITT and safety populations matters. The primary efficacy analysis includes all randomized participants, preserving treatment assignment as the basis for comparison. The safety analysis instead uses participants who actually received at least one dose of study drug, because safety events are interpreted in the context of treatment exposure.

6. Primary Endpoints

EndpointRegistered definition / time framePosted analysis
Primary Efficacy Outcome Number of Participants With Composite of Myocardial Infarction (MI), Ischemic Stroke, Cardiovascular Death, Acute Limb Ischemia (ALI) and Major Amputation Due to a Vascular Etiology. For each participant, the first occurrence of the composite primary efficacy outcome after randomization up until the efficacy cut-off date (08-Sep-2019) was considered. The mean time in follow-up survival time until ECOD that date was 1109.76 days. Time-to-event; stratified log-rank test; hazard ratio
Primary Safety Outcome Number of Participants With TIMI (Thrombolysis in Myocardial Infarction) Major Bleeding. For each participant, the first occurrence of the primary safety outcome after randomization up until 2 days after permanent stop of study drug (rivaroxaban or rivaroxaban placebo). Time-to-event; stratified log-rank test; hazard ratio
Endpoint-definition detail: The registry defines the primary outcomes as participant-level binary endpoints, with only the first occurrence of the outcome event under analysis within the stated data scope considered. The posted statistical analyses classify both primary analyses as time-to-event analyses. That distinction is important: the outcome is ultimately whether an event occurs, but the analysis incorporates when the first event occurs and censoring.

7. Statistical Methodology

Intention-to-treat analysis

The primary efficacy outcome was analyzed in the intention-to-treat analysis set, defined by the registry as all randomized participants. ITT analysis maintains the original randomized comparison even when participants do not perfectly follow the assigned treatment over time.

For a randomized trial, this is more than a bookkeeping choice. Randomization creates the initial comparability between groups. If efficacy analyses subsequently exclude participants based on post-randomization behavior, the analysis can change the population being compared and potentially weaken the protection reported by randomization.

Safety analysis set

The primary safety outcome used the safety analysis set, consisting of all treated participants: randomized participants who received at least one dose of study drug, either rivaroxaban or rivaroxaban placebo.

This creates a deliberate distinction between the efficacy and safety estimands represented by the registry analyses. Efficacy follows randomized assignment, whereas the safety analysis requires actual exposure to study drug.

Stratified log-rank test

The registry reports the log-rank test as the method for both primary endpoints. For the primary efficacy outcome, the analysis notes specify that the P-value was based on a log-rank test stratified by type of procedure and clopidogrel use per IxRS assignment, with treatment as a factor.

For the primary safety outcome, the registry similarly reports a log-rank test stratified by type of procedure and clopidogrel use per IxRS assignment, with treatment as a factor. The safety analysis specifies that the reported P-value is two-sided.

Conceptual survival comparison
H0: the treatment groups have the same event-time distribution

The log-rank procedure compares the accumulated evidence about event timing between groups. Stratification allows the comparison to account for prespecified grouping factors in the analysis rather than treating every participant as though those factors were irrelevant.

Hazard ratio

The effect measure reported for every posted statistical analysis is the hazard ratio. In a time-to-event setting, an HR below 1 indicates a lower estimated instantaneous event rate in the rivaroxaban group relative to the placebo group under the fitted comparison; an HR above 1 indicates a higher estimated instantaneous event rate.

Conceptual interpretation
HR = hazard in rivaroxaban group ÷ hazard in placebo group

The hazard ratio is a relative time-to-event measure. It is not an absolute risk difference, a probability that an individual patient will experience an event, or a statement that every participant experiences the same proportional change.

Confidence intervals

All 11 posted statistical analyses use 95% two-sided confidence intervals. A confidence interval communicates statistical uncertainty around the estimated hazard ratio. Narrower intervals generally indicate greater precision, while wider intervals indicate that the estimate is less precisely determined.

For a hazard ratio, the value 1.00 is the usual no-relative-difference reference point. An interval entirely below 1.00 is consistent with a lower estimated hazard in the rivaroxaban group under the model; an interval that crosses 1.00 is compatible with both a lower and a higher hazard.

8. Primary Efficacy Result

The primary efficacy outcome was the first occurrence after randomization of the composite of myocardial infarction, ischemic stroke, cardiovascular death, acute limb ischemia, and major amputation due to a vascular etiology. The analysis used the ITT set and a stratified log-rank test.

Hazard ratio for the primary efficacy outcome

0.85

95% CI: 0.76–0.96   ·   P =0.0043

Superiority analysis; intention-to-treat population.

Primary efficacy featureReported result
ComparisonRivaroxaban 2.5 mg Bid + Aspirin 100 mg od vs Rivaroxaban Placebo Bid + Aspirin 100 mg od
Analysis populationIntention-to-treat analysis set
MethodLog Rank
Effect measureHazard Ratio (HR)
Estimate0.85
95% CI0.76–0.96
P-value=0.0043
Hypothesis typeSuperiority
Clinical Biostats interpretation

The HR of 0.85 means that the estimated hazard of experiencing the first event in the composite primary efficacy outcome was approximately 15% lower in the rivaroxaban group than in the placebo group under the reported time-to-event analysis.

That does not mean that 15% of participants avoided an event, that each individual participant had a 15% reduction in risk, or that the absolute probability of an event was reduced by 15 percentage points. The HR is a relative measure of event rates over time.

The 95% CI of 0.76–0.96 describes uncertainty around the estimated HR. Because the interval lies below 1.00, the reported interval is consistent with a lower hazard in the rivaroxaban group throughout the interval under the stated statistical framework.

The P-value of =0.0043 addresses evidence against the null hypothesis used for the superiority comparison; it does not measure the size or clinical importance of the treatment effect. Effect size is described by the HR, while precision is described by the confidence interval.

The analysis was stratified by type of procedure and clopidogrel use per IxRS assignment. Because this is a time-to-event analysis, censoring and the assumptions underlying hazard-based interpretation also matter. A single HR should not be interpreted as though it were an absolute risk ratio at every point in time.

9. Primary Safety Result

The primary safety outcome was the first occurrence of TIMI Major Bleeding after randomization within the registry-defined time frame. The safety analysis set consisted of randomized participants who received at least one dose of study drug.

Hazard ratio for TIMI Major Bleeding

1.43

95% CI: 0.97–2.10   ·   P =0.0695

Two-sided log-rank analysis; safety analysis population.

Primary safety featureReported result
ComparisonRivaroxaban 2.5 mg Bid + Aspirin 100 mg od vs Rivaroxaban Placebo Bid + Aspirin 100 mg od
Analysis populationSafety analysis set
MethodLog Rank
Effect measureHazard Ratio (HR)
Estimate1.43
95% CI0.97–2.10
P-value=0.0695
Hypothesis typeOther / not stated
Clinical Biostats interpretation

The HR of 1.43 means that the estimated hazard of the first TIMI Major Bleeding event was higher in the rivaroxaban group relative to the placebo group under the reported time-to-event model. It does not mean that the probability of bleeding was 43% higher for every patient, nor does it provide an absolute increase in bleeding probability.

The 95% CI of 0.97–2.10 is relatively broad and includes 1.00. Thus, the interval encompasses a range of effects that includes little relative difference as well as a higher hazard in the rivaroxaban group.

The P-value of =0.0695 is a two-sided measure of evidence under the reported analysis. It is not a measure of effect size and should not be read as a 6.95% probability that the null hypothesis is true.

This result also illustrates why efficacy and safety should be examined separately. A treatment comparison can have an HR below 1 for one clinical outcome and an HR above 1 for a bleeding outcome because the endpoints represent different event processes.

10. Serious Adverse Events by Arm

The ClinicalTrials.gov record reports serious adverse events by treatment arm as follows:

ArmAffected / at risk
Rivaroxaban 2.5 mg Bid + Aspirin 100 mg948/3256
Rivaroxaban Placebo Bid + Aspirin 100 mg927/3248
Serious adverse events: affected participants
Rivaroxaban
948
Placebo
927

These counts describe serious adverse events using the affected/at-risk format reported by the registry data. They should not be substituted for the primary TIMI Major Bleeding analysis. The primary safety endpoint is a specific time-to-event bleeding outcome, whereas serious adverse events are a broader safety category.

11. Secondary Efficacy Results

In addition to the primary composite, the registry reports several secondary time-to-event analyses. Each uses the ITT analysis set and the same basic treatment comparison. The results illustrate how different endpoint definitions can yield different estimated treatment effects even within the same randomized population.

Secondary endpointHR95% CIP-value
Composite of MI, ischemic stroke, CHD death, ALI, and major amputation of a vascular etiology 0.800.71–0.91=0.0004
Unplanned index limb revascularization for recurrent limb ischemia 0.880.79–0.99=0.0140
Hospitalization for a coronary or peripheral cause of a thrombotic nature 0.720.62–0.85<0.0001
Composite of MI, ischemic stroke, all-cause mortality, ALI, and major amputation of a vascular etiology 0.890.79–0.99=0.0145
Composite of MI, all-cause stroke, CV death, ALI, and major amputation of a vascular etiology 0.860.76–0.96=0.0051
All-cause mortality 1.080.92–1.27=0.8322
Venous thromboembolic events 0.610.37–1.00=0.0235

How to read this table

The estimates should be read endpoint by endpoint rather than collapsed into a single overall treatment effect. For example, an HR of 0.72 for hospitalization for a coronary or peripheral cause of a thrombotic nature represents a different estimand from the HR of 1.08 for all-cause mortality.

The confidence intervals also communicate important differences in precision. The VTE estimate of 0.61 has a 95% CI of 0.37–1.00, which is substantially wider than several of the other intervals. That width matters when interpreting the stability of the estimate.

A statistical reading of the secondary results

The secondary outcomes show a range of estimated hazard ratios, from below 1 to above 1. The appropriate interpretation is not simply that "all outcomes improved" or "nothing changed." Each endpoint has its own clinical definition, event process, analysis population, uncertainty interval, and P-value.

In particular, the all-cause mortality estimate of 1.08 with a 95% CI of 0.92–1.27 differs materially from the primary efficacy estimate of 0.85. This demonstrates why a composite endpoint cannot automatically be interpreted as though each component has the same treatment effect.

12. Secondary Safety Results

The registry reports two additional bleeding outcomes using the safety analysis set. Both were analyzed as time-to-event outcomes with a log-rank test and hazard ratio.

Secondary safety endpointHR95% CIP-value
ISTH Major Bleeding 1.421.10–1.84=0.0068
BARC Type 3b and Above Bleeding Events 1.290.95–1.76=0.0979

The different estimates demonstrate that bleeding classifications are not interchangeable. TIMI Major Bleeding, ISTH Major Bleeding, and BARC Type 3b and Above Bleeding Events are separate registered outcomes. A statistical result for one should not be transferred to another endpoint simply because all are bleeding measures.

ISTH Major Bleeding

HR 1.42 with a 95% CI of 1.10–1.84 and P =0.0068. The interval is entirely above 1.00 under the reported analysis.

BARC Type 3b and Above

HR 1.29 with a 95% CI of 0.95–1.76 and P =0.0979. The confidence interval includes 1.00.

13. Statistical Methods Explained

Why was an intention-to-treat analysis used for the primary efficacy outcome?

The registry defines the ITT Set as all randomized participants. An ITT analysis keeps participants associated with the treatment assignment generated by randomization. This preserves the principal design advantage of the randomized trial and avoids making post-randomization adherence or treatment exposure the criterion for inclusion in the primary efficacy comparison.

Why was a log-rank test used?

The posted primary and secondary analyses are time-to-event analyses. A log-rank test is designed to compare event-time distributions between groups while using the ordering of event times and accounting for censored observations. It therefore fits the structure of endpoints defined by the first occurrence of an event after randomization.

What does an HR of 0.85 mean?

An HR of 0.85 means that the estimated instantaneous event rate in the rivaroxaban group was approximately 85% of that in the placebo group under the reported time-to-event comparison. Equivalently, the estimate corresponds to an approximately 15% lower estimated hazard. It does not mean that 15% fewer participants necessarily experienced the event in absolute terms.

Why does the primary safety HR have the opposite direction?

The primary efficacy outcome and TIMI Major Bleeding are different endpoints. An HR below 1 for one outcome and an HR above 1 for another is statistically possible and clinically interpretable because the two analyses measure different event processes. Treatment effects should therefore be described separately for efficacy and safety outcomes.

What does a confidence interval crossing 1.00 tell us?

For a hazard ratio, 1.00 represents no relative difference in hazard. A 95% CI that includes 1.00 means the reported interval is compatible with no relative difference as well as with effects on either side of 1.00, within the assumptions of the statistical procedure.

Why doesn't the P-value measure effect size?

A P-value quantifies the compatibility of the observed result with a specified null hypothesis under the statistical model. It depends on both the magnitude of the observed effect and the amount of information available. The HR communicates relative effect size, while the confidence interval communicates the precision and uncertainty surrounding that estimate.

Why is stratification important in this trial?

The registry's primary efficacy analysis was stratified by type of procedure and clopidogrel use per IxRS assignment, with treatment as a factor. Stratification allows the time-to-event comparison to incorporate those analysis strata rather than ignoring them. It is particularly relevant when the randomization or analysis framework recognizes clinically or operationally meaningful groups.

14. Understanding Composite Time-to-Event Endpoints

The primary efficacy endpoint is a composite containing five event types: myocardial infarction, ischemic stroke, cardiovascular death, acute limb ischemia, and major amputation due to a vascular etiology. The registry specifies that only the first occurrence of the outcome event under analysis is considered for each participant.

Composite endpoint structure
First qualifying event = MI OR ischemic stroke OR CV death OR ALI OR major vascular amputation

The composite analysis treats the first qualifying event as the participant's event for the primary composite outcome. Consequently, the HR describes the composite endpoint as defined, not necessarily each component separately.

This distinction is central to interpreting composite outcomes. If a treatment changes the frequency or timing of different components by different amounts, the overall composite HR represents their combined contribution to the first-event endpoint.

The secondary endpoints reinforce this point because the registry contains several related but nonidentical composites. One replaces cardiovascular death with CHD death; another includes all-cause mortality; another specifies all-cause stroke. These definitions are not interchangeable.

15. Time-to-Event Analysis: What the Hazard Ratio Does — and Does Not — Mean

Relative effect

The primary efficacy HR of 0.85 is a relative time-to-event estimate. It indicates a lower estimated hazard for the rivaroxaban group under the reported analysis. It is not an absolute percentage-point reduction in event probability.

Precision

The 95% CI of 0.76–0.96 indicates uncertainty around the primary efficacy estimate. The interval does not represent the range of individual treatment effects experienced by participants.

Safety direction

The primary safety HR of 1.43 points in the opposite direction from the efficacy HR: the estimated hazard of TIMI Major Bleeding was higher in the rivaroxaban group. Its 95% CI of 0.97–2.10 includes 1.00.

Censoring

Time-to-event analyses can include participants who have not experienced the event during their available follow-up. Such observations contribute information up to their censoring time rather than being treated as though the event had occurred.

A hazard ratio also should not automatically be interpreted as a constant relative risk at every point in time. The statistical meaning of the HR depends on the time-to-event model and its assumptions. The registry reports HRs and log-rank analyses but does not provide, in the ClinicalTrials.gov record, additional diagnostics for proportional hazards.

16. Primary Analysis vs Secondary Analyses

VOYAGER PAD has two registered primary endpoints and nine additional statistical analyses classified as secondary in the ClinicalTrials.gov record. The distinction matters because the primary endpoints define the principal confirmatory questions of the trial, while secondary outcomes provide additional information about related efficacy and safety domains.

Analysis groupPopulationMethodEffect measure
Primary efficacyITTStratified log-rankHR 0.85
Primary safetySafety setStratified log-rankHR 1.43
Secondary efficacy outcomesITTLog-rankHRs 0.61–1.08
Secondary safety outcomesSafety setLog-rankHRs 1.29–1.42

The ClinicalTrials.gov record does not report an explicit multiplicity-adjustment strategy, interim-analysis procedure, non-inferiority margin, Bayesian method, crossover analysis, or missing-data/imputation strategy. Those topics are therefore not used to infer additional properties of the trial.

Multiplicity caution: several secondary P-values are reported, including P <0.0001, P =0.0004, P =0.0043, P =0.0051, P =0.0068, P =0.0140, P =0.0145, P =0.0235, P =0.0695, P =0.0979, and P =0.8322. The ClinicalTrials.gov record does not specify an adjustment procedure for treating this collection of tests as a single family. Therefore, these P-values should be read as reported results for their individual analyses rather than automatically interpreted as a multiplicity-controlled hierarchy.

17. Stratified Analysis in VOYAGER PAD

The primary efficacy analysis contains an important design detail: the log-rank test was stratified by type of procedure and clopidogrel use per IxRS assignment, with treatment as a factor.

Type of procedure

This was one of the stratification variables specified in the reported primary efficacy analysis. The analysis therefore compares treatment groups within the stratified framework rather than ignoring procedure type.

Clopidogrel use

Clopidogrel use per IxRS assignment was another stratification variable in the reported primary efficacy analysis.

IxRS

The analysis notes identify the Interactive web/voice response system as the source of the procedure and clopidogrel assignment used for stratification.

Treatment factor

Treatment was included as the comparison factor in the stratified log-rank analysis.

Stratification does not create separate randomized trials. It provides a structured way to compare the treatment groups while accounting for the specified strata in the analysis. The overall HR remains the reported treatment effect for the randomized comparison under that analytical framework.

18. Secondary Endpoint Detail

Secondary outcomeAnalysis populationMethodEstimate95% CIP-value
Composite of MI, ischemic stroke, CHD death, ALI, and major amputation of a vascular etiology ITTLog RankHR 0.800.71–0.91=0.0004
Unplanned index limb revascularization for recurrent limb ischemia ITTLog RankHR 0.880.79–0.99=0.0140
Hospitalization for a coronary or peripheral cause of a thrombotic nature ITTLog RankHR 0.720.62–0.85<0.0001
Composite of MI, ischemic stroke, ACM, ALI, and major amputation of a vascular etiology ITTLog RankHR 0.890.79–0.99=0.0145
Composite of MI, all-cause stroke, CV death, ALI, and major amputation of a vascular etiology ITTLog RankHR 0.860.76–0.96=0.0051
All-cause mortality ITTLog RankHR 1.080.92–1.27=0.8322
Venous thromboembolic events ITTLog RankHR 0.610.37–1.00=0.0235

These secondary results are best interpreted as a collection of endpoint-specific analyses. For instance, the estimate of 0.72 for hospitalization for a coronary or peripheral thrombotic cause has a confidence interval of 0.62–0.85, while all-cause mortality has an estimate of 1.08 with a confidence interval of 0.92–1.27. The difference is informative because it shows that the trial's statistical evidence is not reducible to a single universal hazard ratio.

19. Safety Interpretation Beyond the Primary Endpoint

The primary safety endpoint and secondary bleeding endpoints provide several related measures of bleeding. Their estimates are directionally similar, but the confidence intervals differ:

Safety endpointHR95% CIStatistical reference point
TIMI Major Bleeding1.430.97–2.10CI includes 1.00
ISTH Major Bleeding1.421.10–1.84CI above 1.00
BARC Type 3b and Above Bleeding Events1.290.95–1.76CI includes 1.00

This table illustrates why endpoint definition and uncertainty should be reported together. The three estimates are all above 1.00, but the corresponding confidence intervals do not have the same relationship with 1.00.

It would be inappropriate to describe these three analyses as though they were three measurements of exactly the same endpoint. Each uses its own bleeding classification. The differences in estimates and precision are therefore part of the statistical story rather than inconsistencies that should be averaged away.

20. What the P-values Mean in Context

The primary efficacy analysis reports P =0.0043 under a superiority hypothesis, while the primary safety analysis reports a two-sided P =0.0695 with hypothesis type listed as other / not stated.

P =0.0043

This is the reported P-value for the primary efficacy comparison. It quantifies evidence against the relevant null hypothesis under the reported stratified log-rank analysis; it is not the probability that the treatment effect is real.

P =0.0695

This is the reported two-sided P-value for the primary TIMI Major Bleeding analysis. It should not be converted into a probability that bleeding is or is not increased.

The distinction between a P-value and an effect estimate is especially important here. The primary efficacy HR is 0.85, while its P-value is =0.0043. The HR describes the estimated relative effect; the P-value describes the statistical evidence under the specified test. Neither quantity alone provides a complete clinical interpretation.

21. Limitations

22. Why This Trial Matters Statistically

VOYAGER PAD is a useful statistical teaching case because it connects randomized trial design with multiple layers of time-to-event analysis. The primary efficacy outcome is a composite first-event endpoint analyzed in the ITT population, while the primary safety endpoint is analyzed in a treated safety population. Both use log-rank methods and hazard ratios, but they answer different clinical questions.

ConceptHow it appears in VOYAGER PAD
RandomizationRandomized phase 3 parallel-group design with 6564 enrolled participants
BlindingQuadruple masking
Intention-to-treatPrimary efficacy analysis includes all randomized participants
Safety analysis setIncludes randomized participants who received at least one dose of study drug
Time-to-event endpointsPrimary and secondary analyses evaluate first occurrences after randomization
Log-rank testReported as the statistical method for all posted analyses
Stratified analysisPrimary efficacy analysis stratified by type of procedure and clopidogrel use per IxRS assignment
Hazard ratioReported effect measure for all 11 statistical analyses
Confidence intervalsAll reported estimates use 95% two-sided CIs
Composite endpointsMultiple analyses combine different cardiovascular and limb events
Safety endpointsTIMI, ISTH, and BARC bleeding classifications are analyzed separately
MultiplicityMultiple primary and secondary analyses require careful interpretation of P-values

The trial also illustrates why statistical interpretation should proceed in layers. First identify the randomized comparison. Then identify the analysis population. Next identify the endpoint definition and whether it is time-to-event. Then identify the statistical test and effect measure. Finally interpret the estimate, confidence interval, and P-value together.

23. Statistical Concepts in This Trial

Learn more about the methods used in this trial:

24. Related Statistical Calculators

25. Sources

Continue through the Clinical Biostats statistical pathway

Use the trial's endpoints and methods as a starting point for deeper study of survival analysis, hazard ratios, confidence intervals, randomization, and time-to-event methods.

26. Record Summary

VOYAGER PAD provides a detailed example of how a randomized cardiovascular trial can use time-to-event methodology to evaluate both efficacy and safety. The primary efficacy analysis used the ITT population and a stratified log-rank test, producing an HR of 0.85 with a 95% CI of 0.76–0.96 and P =0.0043. The primary safety analysis used the treated safety population and produced an HR of 1.43 with a 95% CI of 0.97–2.10 and P =0.0695.

The secondary analyses show why endpoint definition matters. Related cardiovascular and limb composites produced HR estimates ranging from 0.72 to 0.89, while all-cause mortality had an HR of 1.08. The bleeding analyses also differed by classification: ISTH Major Bleeding had an HR of 1.42, whereas BARC Type 3b and Above Bleeding Events had an HR of 1.29.

The most useful statistical reading therefore combines randomization, analysis population, endpoint definition, time-to-event methodology, hazard ratio, confidence interval, and P-value. No single number captures the entire trial. The statistical evidence is best understood by examining efficacy and safety endpoints separately and by keeping each estimate tied to its precise registered outcome.

Clinical Biostats methodology: A trial-results page should not merely repeat reported numbers. The goal is to reconstruct the statistical story of the trial while clearly separating registry-reported evidence from educational interpretation and avoiding unsupported reconstruction of missing trial information.