← Clinical Trials
Cardiovascular Diseases Phase 3 Time-to-Event NCT02993406

CLEAR Outcomes: Complete Statistical Analysis of Bempedoic Acid in Cardiovascular Disease

An independent statistical analysis of the randomized phase 3 CLEAR Outcomes trial evaluating bempedoic acid 180 mg versus placebo in participants with, or at high risk for, cardiovascular disease who are statin intolerant.

Trial period: 2016-12-22 to 2022-11-07  ·  Enrollment: 13,970  ·  Primary analysis: four-component MACE
Scope of this record

This page separates reported trial results from statistical interpretation. Trial-specific numerical results are restricted to the information provided in the ClinicalTrials.gov-derived ClinicalTrials.gov record. The ClinicalTrials.gov record provides the official registry record.

1. Trial at a Glance

CLEAR Outcomes was a completed, randomized, parallel-group, triple-masked phase 3 trial evaluating bempedoic acid 180 mg versus matching placebo in participants with, or at high risk for, cardiovascular disease who are statin intolerant.

13,970
Enrollment
Randomized phase 3 trial
2
Arms
Bempedoic acid vs placebo
0.87
Primary MACE HR
95% CI 0.79–0.96
0.004
Primary P-value
Two-sided superiority test
FeatureCLEAR Outcomes
Trial nameCLEAR Outcomes
ClinicalTrials.gov identifierNCT02993406
PhasePhase 3
StatusCompleted
Therapeutic areaCardiology
ConditionsCardiovascular Diseases; Statin Adverse Reaction
AllocationRandomized
Design modelParallel
MaskingTriple
Primary purposeTreatment
Enrollment13,970
Primary endpoint typeTime-to-event
Hypothesis typeSuperiority
Primary statistical methodLog-rank test
Primary effect measureHazard ratio
Lead sponsorEsperion Therapeutics, Inc.
Sponsor typeIndustry

The registry-derived record contains seven posted outcome measures and seven posted statistical analyses. One of those analyses is identified as the primary endpoint analysis; the other six are secondary endpoint analyses.

2. Clinical Question

The central statistical question was whether treatment with bempedoic acid 180 mg differed from matching placebo with respect to the time to first occurrence of the trial's prespecified four-component major adverse cardiovascular event (MACE) composite in the enrolled population.

Population

Participants with, or at high risk for, cardiovascular disease who are statin intolerant.

Intervention

Bempedoic acid 180 mg tablet.

Comparator

Matching placebo tablet.

Primary question

Does bempedoic acid produce a different time to first four-component MACE event than placebo?

3. Trial Design

01
Randomize 13,970 enrolled
02
Parallel arms Two randomized groups
03
Masked treatment Triple masking
04
Follow events Time-to-event outcomes
05
Compare Log-rank and HR
Allocation
Randomized
Participants were assigned to the two parallel treatment groups through randomized allocation.
Masking
Triple masked
The registry classifies the study as triple masked.
Primary purpose
Treatment
The registered primary purpose is treatment.
Hypothesis
Superiority
The posted statistical analyses are identified as superiority analyses.
ARM 1

Bempedoic Acid

  • Bempedoic acid 180 mg tablet
  • Drug intervention
  • Included in the primary and secondary efficacy comparisons
ARM 2

Placebo Comparator

  • Matching placebo tablet
  • Drug comparator
  • Reference group for the randomized comparisons

Trial timing

2016-12-22

Trial start

The registered trial start date was December 22, 2016.

2022-11-07

Primary completion

The registered primary completion date was November 7, 2022.

4. Endpoints

EndpointRegistry definitionTime frameType
Four-component MACE Number of Participants With First Occurrence of Four Component Major Adverse Cardiovascular Events (MACE) Up to 68 months Time-to-event
Three-component MACE Number of Participants With First Occurrence of Three Component MACE Up to 68 months Time-to-event
Myocardial infarction Number of Participants With First Occurrence of Myocardial Infarction Up to 68 months Time-to-event
Coronary revascularization Number of Participants With Time to First Occurrence of Coronary Revascularization Up to 68 months Time-to-event
Stroke Number of Participants With Time to First Occurrence of Stroke Up to 68 months Time-to-event
Cardiovascular death Number of Participants With Time to Cardiovascular Death Up to 68 months Time-to-event
All-cause mortality Number of Participants With Time to All-cause Mortality Up to 68 months Time-to-event

Primary endpoint definition

The registered primary efficacy endpoint was a four-component composite of adjudicated MACE, defined as death from cardiovascular causes, nonfatal myocardial infarction, nonfatal stroke, or coronary revascularization, assessed in a time-to-first-event analysis.

This distinction matters statistically. The endpoint is not simply a count of every cardiovascular event experienced during follow-up. The registered definition concerns the first occurrence of any component of the composite. Once a participant experiences a qualifying first event, that participant's primary endpoint event has occurred.

5. Statistical Methodology

Intention-to-treat analysis

The posted analyses specify the Intention-to-Treat Population as the analysis population for the primary and secondary time-to-event comparisons. Under the intention-to-treat principle, randomized participants are analyzed according to their assigned treatment group rather than being reassigned according to later treatment exposure.

This preserves the treatment comparison created by randomization. It also means that the resulting estimate is a comparison of randomized treatment strategies as analyzed under the trial's specified framework, rather than a comparison restricted to participants who perfectly adhered to treatment.

Log-rank test

The statistical method reported for all seven posted analyses is the log-rank test. This is a standard method for comparing time-to-event distributions between randomized groups. Rather than reducing follow-up to a single proportion, the log-rank approach uses the ordering of event times and the number of participants at risk when those events occur.

Conceptual comparison
Observed events vs expected events under the null hypothesis

The log-rank test evaluates whether the event experience over follow-up differs between the treatment groups. Its purpose is hypothesis testing; it is not itself a measure of the magnitude of the treatment effect.

Hazard ratio

The posted effect measure is the hazard ratio (HR). A hazard ratio compares the estimated instantaneous event rate between groups over the analyzed time period.

Interpretation of the hazard ratio
HR < 1  →  lower estimated instantaneous event rate in the bempedoic-acid group

An HR of 1 represents equal hazards under the model. An HR below 1 indicates a lower estimated hazard in the intervention group relative to placebo; an HR above 1 indicates a higher estimated hazard.

Confidence intervals

Each posted hazard ratio is accompanied by a two-sided 95% confidence interval. The confidence interval quantifies statistical uncertainty around the estimated hazard ratio under the analysis framework. It is not a range containing 95% of individual treatment effects or a prediction interval for individual participants.

Superiority testing

All seven posted analyses are classified as superiority analyses. The relevant inferential question is therefore whether the data provide evidence of a difference between the randomized groups in the prespecified direction of the statistical hypothesis, rather than whether bempedoic acid satisfies a non-inferiority margin.

6. Statistical Methods Explained

Why use a time-to-event endpoint?

The primary endpoint is defined as time to the first occurrence of a composite cardiovascular event. Time-to-event analysis preserves information about when an event occurs and can appropriately account for participants who have not experienced the event during their available follow-up.

This is different from simply asking what proportion of participants experienced an event by a fixed date. Two participants may both eventually experience an event, but the timing of those events contains additional information that a simple binary endpoint would discard.

Why was a log-rank test used?

The log-rank test is designed for comparing survival or event-time distributions between groups. Because CLEAR Outcomes used time-to-first-event endpoints, the method aligns with the structure of the registered outcomes.

The test evaluates the evidence for a difference in event-time experience. It does not tell us by itself whether the difference is clinically large, how precise the estimated effect is, or how many participants would benefit.

What does a hazard ratio of 0.87 mean?

For the primary MACE endpoint, an HR of 0.87 means that the estimated hazard of the first four-component MACE event in the bempedoic-acid group was approximately 87% of the estimated hazard in the placebo group under the analysis framework. Equivalently, 1 − 0.87 = 0.13, so the point estimate corresponds to an approximately 13% lower estimated hazard.

That interpretation is about a relative event rate over time. It does not mean that exactly 13% of participants avoided MACE, nor does it mean that each individual participant had a 13% reduction in personal probability of an event.

Why does the confidence interval matter?

The primary 95% CI was 0.79–0.96. The interval describes the uncertainty surrounding the estimated HR of 0.87. It is substantially more informative than the point estimate alone because it shows that the analysis is compatible with a range of relative hazard effects rather than a single known value.

Because the interval is a confidence interval for a relative hazard measure, it should not be interpreted as a range of absolute risk reductions. Absolute risks would require corresponding event probabilities or survival estimates.

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

The primary p-value was 0.004. A p-value measures the compatibility of the observed data with the null hypothesis under the specified statistical test. It does not quantify how large the treatment effect is.

For example, the distinction is visible directly in this trial: the effect size is communicated by the HR of 0.87, while the precision is communicated by its 95% CI of 0.79–0.96 and the hypothesis-testing evidence by the p-value of 0.004. These are different pieces of statistical information.

Why is the intention-to-treat population important?

The posted analyses use the intention-to-treat population. This means the primary statistical comparison remains anchored to randomized assignment. That is important because randomization is the design mechanism that creates comparable treatment groups in expectation.

An ITT analysis therefore answers a question about the groups as randomized. It should not automatically be interpreted as an estimate of the biological effect among only those who adhered perfectly to the assigned treatment.

What does “first occurrence” mean for a composite endpoint?

The primary endpoint combines four event types: cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, and coronary revascularization. The analysis concerns the first qualifying event.

This means the composite can increase statistical efficiency by allowing several clinically relevant event types to contribute to the endpoint, but it also requires careful interpretation. A composite result is not necessarily evidence that every component is affected to the same degree. Component-specific analyses can provide additional context, which is one reason the registry's secondary analyses of myocardial infarction, stroke, coronary revascularization, and mortality are informative.

7. Primary Result: Four-Component MACE

The primary endpoint was the first occurrence of the four-component MACE composite over up to 68 months. The analysis population was the intention-to-treat population, and the groups compared were bempedoic acid 180 mg versus placebo comparator.

Hazard ratio for first four-component MACE

0.87

95% CI: 0.79–0.96   ·   P = 0.004

Analysis: two-sided log-rank test   ·   Hypothesis: superiority

Primary endpointBempedoic acid 180 mg vs placebo
EndpointFirst occurrence of four-component MACE
Time frameUp to 68 months
Analysis populationIntention-to-Treat Population
MethodLog Rank
Effect measureHazard Ratio
Estimate0.87
95% CI0.79–0.96
P-value0.004
HypothesisSuperiority
Clinical Biostats interpretation

The HR of 0.87 indicates a lower estimated hazard of first four-component MACE for bempedoic acid relative to placebo under the posted time-to-event analysis. In relative terms, the point estimate corresponds to approximately a 13% lower estimated hazard.

The HR does not mean that 13% fewer participants experienced MACE, nor does it imply that every participant experienced the same reduction in risk. A hazard ratio is a relative time-to-event measure, not an absolute risk difference.

The 95% CI of 0.79–0.96 describes uncertainty around the estimated HR. It does not describe the range of effects across individual patients. The interval is also not an absolute-risk interval.

The P = 0.004 result provides evidence against the null hypothesis under the specified two-sided log-rank framework. It does not measure the magnitude of the treatment effect. The magnitude is communicated by the HR, while the confidence interval communicates its precision.

Because the endpoint is time-to-event, interpretation also depends on the structure of follow-up and censoring. A single HR summarizes the relative event-rate experience over the analyzed period and should not be treated as though it were a fixed cumulative risk ratio at one particular time point.

Endpoint definition matters: the primary result concerns the first occurrence of a four-component composite consisting of cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, or coronary revascularization. The HR therefore describes the composite endpoint as defined, not each component separately.

8. Secondary Endpoint Results

The registry-derived data contain six secondary statistical analyses. Each uses the intention-to-treat population, compares bempedoic acid 180 mg with placebo, uses a log-rank test, reports a hazard ratio, and is classified as a two-sided superiority analysis.

Secondary endpointHR95% CIP-value
First occurrence of three-component MACE 0.85 0.76–0.96 0.0058
First occurrence of myocardial infarction 0.77 0.66–0.91 0.0016
Time to first coronary revascularization 0.81 0.72–0.92 0.0013
Time to first stroke 0.85 0.67–1.07 0.1593
Time to cardiovascular death 1.04 0.88–1.24 0.6227
Time to all-cause mortality 1.03 0.90–1.18 0.6608

Three-component MACE

Hazard ratio

0.85

95% CI: 0.76–0.96   ·   P = 0.0058

The point estimate corresponds to an approximately 15% lower estimated hazard of the first three-component MACE event in the bempedoic-acid group relative to placebo. As with the primary result, this is a relative hazard interpretation rather than an absolute reduction in the number of participants experiencing events.

Myocardial infarction

Hazard ratio

0.77

95% CI: 0.66–0.91   ·   P = 0.0016

The HR of 0.77 corresponds to an approximately 23% lower estimated hazard of first myocardial infarction under the posted analysis. The confidence interval provides the uncertainty around that estimate; it does not establish the absolute probability that an individual participant will experience myocardial infarction.

Coronary revascularization

Hazard ratio

0.81

95% CI: 0.72–0.92   ·   P = 0.0013

The HR of 0.81 corresponds to an approximately 19% lower estimated hazard of first coronary revascularization. This endpoint concerns time to first revascularization, rather than a count of all revascularization procedures occurring during follow-up.

Stroke

Hazard ratio

0.85

95% CI: 0.67–1.07   ·   P = 0.1593

The point estimate for stroke is below 1, but the 95% CI extends from 0.67 to 1.07. The posted p-value is 0.1593. Thus, the point estimate alone should not be treated as proof of a treatment difference for this endpoint. The confidence interval communicates substantial uncertainty about the magnitude and direction of the relative hazard compatible with the analysis.

Cardiovascular death

Hazard ratio

1.04

95% CI: 0.88–1.24   ·   P = 0.6227

The cardiovascular-death estimate is close to 1, with a 95% CI of 0.88–1.24 and a p-value of 0.6227. The appropriate statistical interpretation is that the posted analysis does not provide evidence of a detectable difference under this test; it does not establish that the two treatments are identical.

All-cause mortality

Hazard ratio

1.03

95% CI: 0.90–1.18   ·   P = 0.6608

The all-cause mortality estimate is 1.03, with a 95% CI of 0.90–1.18 and a p-value of 0.6608. Again, the appropriate interpretation is uncertainty around the relative hazard rather than proof of equivalence or absence of any possible treatment effect.

Secondary-endpoint caution: The six secondary analyses should be interpreted in the context of their role as secondary endpoints. The ClinicalTrials.gov record identifies all seven analyses as superiority analyses and provide individual p-values, but they do not provide a multiplicity-adjustment scheme in the registry-reported statistical-analysis fields. Individual p-values should therefore not automatically be treated as though they were independent confirmatory tests under an explicitly documented familywise-error strategy.

9. Reading the Complete Pattern of Hazard Ratios

Looking across the seven posted analyses illustrates why a clinical trial should not be reduced to a single p-value. The estimates span HR values from 0.77 for myocardial infarction to 1.04 for cardiovascular death. The corresponding confidence intervals also differ materially in width.

EndpointHRInterpretive signal
Four-component MACE0.87Point estimate below 1; 95% CI 0.79–0.96
Three-component MACE0.85Point estimate below 1; 95% CI 0.76–0.96
Myocardial infarction0.77Point estimate below 1; 95% CI 0.66–0.91
Coronary revascularization0.81Point estimate below 1; 95% CI 0.72–0.92
Stroke0.85Point estimate below 1; 95% CI 0.67–1.07
Cardiovascular death1.04Point estimate above 1; 95% CI 0.88–1.24
All-cause mortality1.03Point estimate above 1; 95% CI 0.90–1.18

The table demonstrates an important statistical principle: direction, magnitude, and precision are separate concepts. A point estimate below 1 does not guarantee statistical evidence of a difference, while a point estimate close to 1 does not by itself prove equality.

The primary MACE result is a composite endpoint. Component-specific results help describe which types of events contributed to the overall pattern, but the component estimates should not simply be substituted for the primary endpoint. Each endpoint answers a different statistical question.

10. How to Interpret the Primary Confidence Interval

Point estimate

The primary HR of 0.87 is the central estimate from the posted analysis. Relative to an HR of 1, it indicates a lower estimated hazard for bempedoic acid.

Interval estimate

The 95% CI of 0.79–0.96 shows the statistical uncertainty surrounding the HR estimate. It is narrower than an interval that would result from a substantially less precise estimate, but it still represents uncertainty rather than a known treatment effect.

Hypothesis test

The P = 0.004 result addresses compatibility with the null hypothesis under the specified two-sided log-rank test. It is not a probability that the null hypothesis is true and is not the probability that the observed treatment effect occurred by chance.

Absolute effects

The registry analysis does not provide absolute event probabilities, cumulative incidence values, or Kaplan-Meier estimates for the primary endpoint. Consequently, the HR and confidence interval should not be converted into an absolute number-needed-to-treat or absolute risk reduction from the ClinicalTrials.gov record.

11. Safety Results

The ClinicalTrials.gov record reports serious adverse events by randomized treatment arm. The figures are reported as affected participants divided by participants at risk.

Safety measureBempedoic Acid 180 mgPlacebo Comparator
Serious adverse events1767 / 70011733 / 6964

The affected/at-risk figures should be read as reported rather than reconstructed into percentages. The ClinicalTrials.gov record does not provide a formal statistical comparison of serious adverse events, so no safety p-value, hazard ratio, or confidence interval is reported here.

Safety and efficacy are different estimands: the efficacy analyses evaluate prespecified cardiovascular time-to-event outcomes in the intention-to-treat population. Serious adverse events describe safety experience by treatment arm. These should be evaluated separately rather than collapsed into a single numerical measure of overall treatment value.

12. What the Hazard Ratio Does — and Does Not — Mean

What it means

An HR compares the estimated instantaneous event rates between the two groups over time. For the primary endpoint, 0.87 indicates a lower estimated hazard in the bempedoic-acid group relative to placebo.

What it does not mean

It is not a percentage of patients who benefit, not an absolute risk reduction, and not a statement that every individual experiences the same relative change.

What the CI adds

The confidence interval communicates uncertainty around the estimated HR and should be considered alongside the point estimate rather than after it.

What the p-value adds

The p-value evaluates evidence against the null hypothesis under the specified test. It does not measure clinical magnitude or importance.

13. Kaplan-Meier Estimation and Time-to-Event Interpretation

The trial's endpoints are explicitly classified as time-to-event outcomes. For this type of endpoint, Kaplan-Meier estimation is commonly used to describe the event-free distribution over follow-up, while the log-rank test provides a group comparison and a hazard ratio provides a relative effect estimate.

Kaplan-Meier concept
S(t) = ∏ti ≤ t (1 − di/ni)

Here, di represents events at an event time and ni represents participants at risk immediately before that time. The resulting survival function describes the estimated probability of remaining event-free through time.

A useful conceptual distinction is that the Kaplan-Meier curve describes the distribution over time, whereas the hazard ratio summarizes a relative hazard between treatment groups. A p-value from the log-rank test answers a third question: whether the observed event-time distributions provide evidence against the null hypothesis.

Educational note: the ClinicalTrials.gov record contains hazard ratios, confidence intervals, and p-values but do not contain the participant-level event and censoring information needed to reconstruct a valid Kaplan-Meier curve. No reconstructed curve is presented here.

14. Censoring and the Meaning of “Up to 68 Months”

The registered time frame for every posted endpoint is up to 68 months. In a time-to-event analysis, not every participant necessarily contributes an observed event. Participants without a qualifying event during their available follow-up can contribute information up to the point at which their follow-up ends or they are otherwise censored according to the analysis rules.

This is one reason time-to-event analysis differs from simply counting events. A participant followed for a substantial period without an event contributes information about the event-free experience during that period even if no event is ultimately observed for that participant.

The registry-reported statistical-analysis fields do not provide detailed censoring rules, competing-risk methodology, or missing-data procedures. Those details therefore are not added to this page.

15. Primary Analysis vs Secondary Analyses

RoleEndpointMethodEffect measure
Primary Four-component MACE Log-rank HR 0.87 (95% CI 0.79–0.96)
Secondary Three-component MACE Log-rank HR 0.85 (95% CI 0.76–0.96)
Secondary Myocardial infarction Log-rank HR 0.77 (95% CI 0.66–0.91)
Secondary Coronary revascularization Log-rank HR 0.81 (95% CI 0.72–0.92)
Secondary Stroke Log-rank HR 0.85 (95% CI 0.67–1.07)
Secondary Cardiovascular death Log-rank HR 1.04 (95% CI 0.88–1.24)
Secondary All-cause mortality Log-rank HR 1.03 (95% CI 0.90–1.18)

The hierarchy is statistically important. The primary endpoint is the prespecified main efficacy question in the ClinicalTrials.gov record. Secondary endpoints add information about specific cardiovascular outcomes and mortality, but they do not retroactively redefine the primary endpoint.

16. Interpreting Results Without Overreading Them

A statistically significant result does not automatically answer every clinical question. For the primary MACE endpoint, the HR of 0.87 and P = 0.004 provide evidence of a difference under the posted analysis. The magnitude of that difference is described by the HR, and its uncertainty is described by the confidence interval.

At the same time, the ClinicalTrials.gov record does not provide absolute event rates or time-specific survival probabilities. Therefore, the statistical evidence supports interpretation of the relative hazard, but the ClinicalTrials.gov record does not support a calculation of an absolute risk reduction.

The secondary results further illustrate why endpoints must be interpreted individually. Myocardial infarction has an HR of 0.77, while cardiovascular death has an HR of 1.04. These are not contradictory statistical statements: they concern different endpoints with different event processes and different estimates of treatment effect.

Similarly, the stroke HR of 0.85 has a confidence interval of 0.67–1.07. The fact that its point estimate is below 1 does not justify describing it in the same inferential terms as an estimate whose confidence interval excludes 1. Precision and hypothesis testing matter alongside direction.

17. Important Statistical Limitations

18. Why This Trial Matters Statistically

CLEAR Outcomes is a useful teaching case because it demonstrates how a large randomized clinical trial can use a relatively compact statistical framework to answer several related time-to-event questions.

ConceptHow it appears in CLEAR Outcomes
RandomizationThe trial used randomized allocation to two parallel groups.
Triple maskingThe registry classifies the study as triple masked.
Intention-to-treat analysisThe posted efficacy analyses use the intention-to-treat population.
Time-to-event endpointsThe primary endpoint and all six secondary endpoints are time-to-event analyses.
Log-rank testThe reported statistical method for all seven analyses is Log Rank.
Hazard ratioThe treatment effect is reported using HRs for every posted analysis.
Confidence intervalsEach posted HR includes a two-sided 95% confidence interval.
Superiority testingAll seven posted analyses are classified as superiority analyses.
Composite endpointThe primary endpoint combines four cardiovascular event components.
Component analysisSecondary analyses separately evaluate myocardial infarction, stroke, coronary revascularization, and mortality.
Safety analysisSerious adverse events are reported separately by treatment arm.

The statistical lesson is broader than the individual HR of 0.87. A complete trial interpretation requires keeping the estimand, endpoint definition, analysis population, effect measure, uncertainty interval, and hypothesis test aligned. Changing any one of those elements can change what the reported number actually means.

19. A Worked Statistical Reading of the Primary Result

Step 1 · Identify the estimand

The analysis compares the time to first four-component MACE between randomized bempedoic-acid and placebo groups in the intention-to-treat population over up to 68 months.

Endpoint = first cardiovascular death, nonfatal MI, nonfatal stroke, or coronary revascularization
Step 2 · Identify the effect measure

The effect measure is the hazard ratio.

HR = 0.87
Step 3 · Quantify uncertainty

The two-sided 95% confidence interval is:

95% CI = 0.79 to 0.96
Step 4 · Evaluate the hypothesis test

The posted log-rank p-value is:

P = 0.004
Step 5 · Translate carefully

The point estimate indicates approximately a 13% lower estimated hazard of first four-component MACE for bempedoic acid relative to placebo. This is a relative time-to-event interpretation, not an absolute risk reduction or a statement about individual patients.

20. Clinical Interpretation vs Statistical Interpretation

Statistical interpretation

The primary randomized comparison produced an HR of 0.87 with a two-sided 95% CI of 0.79–0.96 and P = 0.004 using a log-rank test in the intention-to-treat population.

Clinical interpretation

The statistical result describes a difference in the time-to-first-event experience for the defined four-component MACE composite. The ClinicalTrials.gov record does not provide absolute event probabilities needed to quantify absolute clinical benefit.

This distinction is important. Statistical significance concerns the evidence against a null hypothesis under the specified analysis. Clinical interpretation asks what the magnitude and nature of the observed effect mean in the context of the endpoint and population. The two concepts overlap, but neither is a substitute for the other.

21. What the Secondary Results Add

The secondary analyses allow the broad MACE result to be examined through more specific endpoints. Three of the secondary HRs are below 1 with confidence intervals of 0.76–0.96, 0.66–0.91, and 0.72–0.92 for three-component MACE, myocardial infarction, and coronary revascularization, respectively.

The stroke analysis provides a useful contrast: its HR is also below 1 at 0.85, but its 95% CI is 0.67–1.07 and its p-value is 0.1593. This demonstrates why statistical interpretation cannot rely only on whether the point estimate is below 1.

The mortality analyses provide another contrast. Cardiovascular death has an HR of 1.04 with a 95% CI of 0.88–1.24, while all-cause mortality has an HR of 1.03 with a 95% CI of 0.90–1.18. Their p-values are 0.6227 and 0.6608, respectively. These estimates do not provide evidence of a detectable difference under the posted tests, but they should not be converted into claims of equivalence.

22. Sources

Source boundary: Trial-specific numerical values and statistical-analysis details on this page are restricted to the registry-reported CLEAR Outcomes trial data. The PubMed links are provided as source-navigation links and are not used here to introduce additional trial numbers or findings.

23. Related Tutorials

Learn more about the methods used in this trial:

24. Related Calculators

Continue through the Clinical Biostats statistical library

Connect this trial's design and time-to-event methods to deeper tutorials and statistical calculation tools.

25. Record Summary

CLEAR Outcomes provides a clear example of randomized time-to-event analysis in a phase 3 cardiovascular trial. The primary endpoint was the first occurrence of a four-component MACE composite over up to 68 months, analyzed in the intention-to-treat population using a log-rank test and reported with a hazard ratio. The primary estimate was HR 0.87 (95% CI 0.79–0.96; P = 0.004).

The secondary analyses extend the statistical picture to three-component MACE, myocardial infarction, coronary revascularization, stroke, cardiovascular death, and all-cause mortality. Their hazard ratios range from 0.77 to 1.04, illustrating the importance of examining each endpoint's estimate, confidence interval, and p-value rather than treating the trial as a single numerical result.

From a statistical perspective, the central lesson is the relationship between randomization, intention-to-treat analysis, time-to-event endpoints, log-rank testing, hazard ratios, and confidence intervals. Together, these elements define what the reported results can and cannot establish.

Clinical Biostats methodology: A trial-results page should not merely repeat a reported statistic. The goal is to explain the estimand, endpoint, analysis population, statistical method, effect measure, uncertainty, and inferential limitations so that readers can understand exactly what each number represents.