This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Trial-specific numerical results on this page are restricted to the registry-reported REVEAL trial data.
1. Trial at a Glance
REVEAL was a randomized, parallel-group, quadruple-masked, phase 3 prevention trial evaluating allocation to anacetrapib versus placebo in participants with atherosclerotic cardiovascular disease. The registered primary endpoint was the number of participants with a major coronary event during the randomized treatment phase, with the primary assessment based on an intention-to-treat comparison and the first major coronary event.
| Feature | REVEAL |
|---|---|
| Trial name | REVEAL: Randomized EValuation of the Effects of Anacetrapib Through Lipid-modification |
| Phase | Phase 3 |
| Therapeutic area | Cardiovascular |
| Condition | Atherosclerotic Cardiovascular Disease |
| Design | Randomized, parallel-group, quadruple-masked |
| Primary purpose | Prevention |
| Enrollment | 30449 |
| Interventions | Anacetrapib; placebo anacetrapib |
| Lead sponsor | University of Oxford |
| Status | ACTIVE_NOT_RECRUITING |
| ClinicalTrials.gov | NCT01252953 |
2. Clinical Question
The central statistical question was whether allocation to anacetrapib, compared with placebo anacetrapib, was associated with a different time to the first major coronary event during the scheduled randomized treatment period.
Population
Participants in the phase 3 REVEAL trial with atherosclerotic cardiovascular disease.
Intervention
Anacetrapib.
Comparator
Placebo anacetrapib.
Primary question
Does allocation to anacetrapib versus placebo alter the occurrence of the first major coronary event during the randomized treatment phase?
3. Trial Design
Anacetrapib
- Study intervention: anacetrapib
- Randomized treatment assignment
Placebo
- Study comparator: placebo anacetrapib
- Randomized treatment assignment
The design is important for statistical interpretation because randomization establishes the comparison of treatment allocation before follow-up begins, while masking reduces the opportunity for knowledge of treatment assignment to influence trial conduct or outcome assessment.
4. Endpoints
The registered primary endpoint was a time-to-first-event assessment embedded within the registry's endpoint description. The registry defines the primary assessment as an intention-to-treat comparison of allocation to anacetrapib versus placebo for major coronary events during the scheduled treatment period.
| Endpoint | Registry definition | Time frame | Statistical role |
|---|---|---|---|
| Number of Participants With Major Coronary Event | Occurrence of coronary death, myocardial infarction or coronary revascularization procedure; data reported for the first major coronary event. | Randomized treatment phase during median follow-up period of 4.1years | Primary |
| Number of Participants With Major Atherosclerotic Event | Registry outcome measure analyzed as time to first event. | Randomized treatment phase during median follow-up period of 4.1years | Secondary |
| Number of Participants With Presumed Ischaemic Stroke | Registry outcome measure; statistical method is not reported in the ClinicalTrials.gov record. | Randomized treatment phase during median follow-up period of 4.1years | Secondary |
| Number of Participants With Major Vascular Event | Registry outcome measure analyzed as time to first event. | Randomized treatment phase during median follow-up period of 4.1years | Secondary |
5. Primary Endpoint Result
Major Coronary Event
The primary analysis compared anacetrapib with placebo for the time to the first major coronary event during the randomized treatment phase. The reported method was the log-rank test, and the reported effect measure was a rate ratio.
Primary rate ratio
95% CI: 0.85–0.97 · P = 0.004
Time to first event during the randomized treatment phase; median follow-up period of 4.1years.
| Primary analysis feature | Reported result |
|---|---|
| Outcome | Number of Participants With Major Coronary Event |
| Groups | Anacetrapib vs Placebo Anacetrapib |
| Endpoint type in statistical analysis | Time-to-event |
| Method | Log Rank |
| Effect measure | Rate Ratio |
| Estimate | 0.91 |
| 95% CI | 0.85–0.97 |
| P-value | 0.004 |
| Analysis note | Time to first event |
A reported rate ratio of 0.91 indicates that the estimated event rate over the analyzed time-to-event comparison was lower in the anacetrapib group than in the placebo group. Expressed descriptively, 0.91 corresponds to an estimated rate that is 9% lower relative to the comparator.
The estimate does not mean that exactly 9% of participants avoided an event, that every participant experienced a 9% reduction, or that the absolute probability of an event was reduced by 9 percentage points. A rate ratio is a relative measure and incorporates the time-to-event framework.
The 95% confidence interval of 0.85–0.97 describes statistical uncertainty around the estimated rate ratio under the analysis framework. It is an interval for the population-level effect estimate, not a range containing the effects experienced by individual participants.
The P-value of 0.004 quantifies evidence against the null hypothesis represented by the statistical test; it does not measure the size or clinical importance of the effect. The effect size is conveyed by the rate ratio and its confidence interval.
Because this is a time-to-event analysis, interpretation also depends on how censoring and follow-up were handled. The ClinicalTrials.gov record identifies the log-rank method and time-to-first-event framework but do not provide additional details about censoring rules or a proportional-hazards model.
6. Statistical Methodology
Log-rank test
The primary statistical method reported in the registry analysis is the log-rank test. The log-rank test compares time-to-event experience between randomized groups while taking the timing of events into account.
At each event time, the log-rank framework considers how many participants remain at risk in each group and how the observed events compare with the events expected if the groups had the same event experience.
This differs from a simple comparison of the proportion of participants who eventually experienced an event. In a time-to-event analysis, two participants who both experience an event do not necessarily contribute the same information if their events occur at different follow-up times. Similarly, participants who do not experience an event during observed follow-up can contribute information through their time at risk.
Rate ratio
The registry reports the treatment effect as a rate ratio. A value below 1 indicates a lower estimated event rate in the anacetrapib group relative to placebo under the reported analysis.
This is a relative interpretation. It should not be converted into an absolute risk reduction without the underlying event and person-time information required for that calculation.
Confidence interval
The primary 95% confidence interval extends from 0.85 to 0.97. The interval provides a measure of precision around the reported rate-ratio estimate. Its width reflects uncertainty in the estimate; it should not be interpreted as a prediction interval for individual participant outcomes.
Intention-to-treat comparison
The registered primary assessment specifies an intention-to-treat comparison among all randomized participants. This means the comparison is based on the treatment allocation assigned at randomization rather than redefining groups according to later treatment exposure.
The ITT principle is particularly important in randomized trials because the treatment groups are created by randomization. Maintaining participants according to randomized allocation preserves the interpretation of the comparison as an effect of assignment to the intervention versus assignment to the comparator, subject to the assumptions and follow-up structure of the trial.
Time-to-event structure
The primary analysis is explicitly identified as time-to-event and concerns the first major coronary event. This means the analysis distinguishes not only whether an event occurred but also when the first event occurred during follow-up.
7. Secondary Endpoint Results
The registry contains three posted secondary analyses. Two are explicitly reported as log-rank survival analyses. The presumed ischaemic stroke endpoint has a reported rate ratio but no statistical method is reported in the ClinicalTrials.gov record.
Major Atherosclerotic Event
Secondary rate ratio
95% CI: 0.86–1 · P = 0.052
Time to first event during the randomized treatment phase; median follow-up period of 4.1years.
The reported rate ratio of 0.93 corresponds to an estimated event rate approximately 7% lower in the anacetrapib group relative to placebo. The 95% confidence interval is 0.86–1.
The P-value of 0.052 should not be treated as a measure of the size of the effect. It also should not be used by itself to turn a continuous body of evidence into a binary claim about whether an effect exists. The estimate and confidence interval provide the more direct description of the observed relative effect and its precision.
Because this endpoint is a time-to-first-event analysis, the timing of events and follow-up contribute to the comparison. The ClinicalTrials.gov record does not provide additional model details beyond the reported log-rank method and rate ratio.
Presumed Ischaemic Stroke
Reported rate ratio
95% CI: 0.87–1.12
Randomized treatment phase during median follow-up period of 4.1years.
The reported rate ratio of 0.99 is close to 1, indicating little difference between the estimated event rates on the reported relative scale. The 95% confidence interval of 0.87–1.12 spans values below and above 1, indicating substantial uncertainty around the estimate.
The ClinicalTrials.gov record identifies the endpoint as a binary outcome but report the effect as a rate ratio and note that the analysis concerns time to first event. Importantly, the statistical method itself is listed as not reported. Therefore, no specific formal test should be attributed to this endpoint beyond what is explicitly contained in the registry data.
There is also no P-value reported for this analysis in the ClinicalTrials.gov record. It would therefore be inappropriate to infer one from the confidence interval or to assign an unreported analysis method.
Major Vascular Event
Secondary rate ratio
95% CI: 0.88–0.99 · P = 0.02
Time to first event during the randomized treatment phase; median follow-up period of 4.1years.
The reported rate ratio of 0.93 corresponds to an estimated event rate approximately 7% lower in the anacetrapib group relative to placebo. The 95% confidence interval of 0.88–0.99 quantifies uncertainty around that estimate.
The P-value of 0.02 provides the registry's reported evidence from the corresponding statistical test. It does not measure effect magnitude. The rate ratio and confidence interval remain necessary for understanding the size and precision of the estimated relative difference.
This result is a secondary endpoint, so it should be distinguished from the primary major-coronary-event analysis. The ClinicalTrials.gov record does not state a formal multiplicity-adjustment strategy or hypothesis hierarchy for these secondary analyses.
8. Results Summary
| Endpoint | Role | Method | Rate ratio | 95% CI | P-value |
|---|---|---|---|---|---|
| Major Coronary Event | Primary | Log-rank | 0.91 | 0.85–0.97 | 0.004 |
| Major Atherosclerotic Event | Secondary | Log-rank | 0.93 | 0.86–1 | 0.052 |
| Presumed Ischaemic Stroke | Secondary | Not reported | 0.99 | 0.87–1.12 | Not reported |
| Major Vascular Event | Secondary | Log-rank | 0.93 | 0.88–0.99 | 0.02 |
The four posted analyses illustrate why an effect estimate should be interpreted together with its confidence interval, statistical method, endpoint role, and time frame. The primary analysis reports a rate ratio of 0.91 with a 95% CI of 0.85–0.97, while the secondary analyses span a range from 0.93 to 0.99 and have different levels of statistical information reported by the registry.
9. Safety Results
The ClinicalTrials.gov record reports serious adverse events by randomized arm. These data provide the number affected and the number at risk in each group.
| Safety measure | Anacetrapib | Placebo Anacetrapib |
|---|---|---|
| Serious adverse events, affected / at risk | 8898 / 15225 | 8912 / 15224 |
The ClinicalTrials.gov record reports serious adverse events as affected participants over participants at risk rather than as a separately analyzed comparative effect measure. Consequently, this page does not calculate a risk ratio, rate ratio, confidence interval, or P-value for serious adverse events.
10. Statistical Methods Explained
Why was a log-rank test used for the primary endpoint?
The primary endpoint is analyzed as a time-to-first-event outcome. A log-rank test is designed to compare the event experience of two groups across follow-up while accounting for when events occur. This makes it different from a simple comparison of the final proportion of participants with an event.
What does a rate ratio of 0.91 mean?
A rate ratio of 0.91 indicates an estimated event rate 9% lower in the anacetrapib group relative to placebo on the reported relative scale. It does not mean a 9-percentage-point reduction in absolute event probability and does not imply that every participant's individual risk changed by exactly 9%.
What does the 95% confidence interval of 0.85–0.97 tell us?
It describes the statistical uncertainty around the estimated primary rate ratio. The interval is relatively narrow around 0.91 compared with the much wider range reported for the presumed ischaemic stroke analysis, illustrating how confidence intervals communicate precision in addition to the point estimate.
Why does the P-value not measure effect size?
A P-value describes the evidence reported by a statistical test against its null hypothesis. It is affected by both the observed data and the amount of information available. Effect magnitude is communicated by measures such as the rate ratio, while the confidence interval provides information about precision.
Why is time-to-first-event different from simply counting events?
Two treatment groups can have the same number of observed events while having different timing of those events. A time-to-event analysis uses the follow-up information and event timing. The REVEAL primary analysis specifically identifies the first major coronary event and reports a time-to-event log-rank analysis.
Does the reported analysis establish a proportional-hazards assumption?
The ClinicalTrials.gov record does not report a Cox proportional-hazards model or a proportional-hazards diagnostic. Therefore, a proportional-hazards assumption should not be attributed to the posted primary analysis. The reported method is the log-rank test and the reported effect measure is a rate ratio.
11. Confidence Intervals Across the Reported Analyses
One useful way to read the REVEAL results is to compare the effect estimates with the corresponding uncertainty intervals rather than focusing on P-values alone.
| Endpoint | Rate ratio | Lower 95% CI | Upper 95% CI | Interpretive point |
|---|---|---|---|---|
| Major Coronary Event | 0.91 | 0.85 | 0.97 | The interval is entirely below 1. |
| Major Atherosclerotic Event | 0.93 | 0.86 | 1 | The upper confidence limit reaches 1. |
| Presumed Ischaemic Stroke | 0.99 | 0.87 | 1.12 | The interval spans values below and above 1. |
| Major Vascular Event | 0.93 | 0.88 | 0.99 | The interval is below 1 at its upper limit. |
The confidence intervals should be read in conjunction with endpoint hierarchy and statistical method. A confidence interval that includes 1 does not prove that the treatment and comparator have identical effects; it indicates that the reported data and analysis do not exclude a null rate ratio at that confidence level.
12. Multiplicity and Secondary Endpoints
The ClinicalTrials.gov record identifies one registered primary endpoint and three secondary analyses. They do not state a multiplicity-adjustment procedure, alpha allocation, endpoint hierarchy beyond the primary-versus-secondary designation, or a formal hypothesis-testing strategy for the secondary outcomes.
| Analysis | Role | Reported statistical information |
|---|---|---|
| Major Coronary Event | Primary | Log-rank; rate ratio 0.91; 95% CI 0.85–0.97; P = 0.004 |
| Major Atherosclerotic Event | Secondary | Log-rank; rate ratio 0.93; 95% CI 0.86–1; P = 0.052 |
| Presumed Ischaemic Stroke | Secondary | Method not reported; rate ratio 0.99; 95% CI 0.87–1.12 |
| Major Vascular Event | Secondary | Log-rank; rate ratio 0.93; 95% CI 0.88–0.99; P = 0.02 |
13. Randomization and Blinding
REVEAL was randomized and parallel-group, with quadruple masking. These design features address different sources of bias. Randomization establishes the treatment allocation mechanism, while masking limits knowledge of treatment assignment among the relevant trial participants and personnel.
Randomization
Participants were randomly allocated to anacetrapib or placebo, creating the basis for an intention-to-treat comparison.
Quadruple masking
The registry identifies the trial as quadruple-masked, reducing the opportunity for treatment knowledge to influence trial conduct or assessment.
Neither randomization nor masking guarantees that every aspect of a trial is free from bias. They are design mechanisms that strengthen the credibility of the treatment comparison when implemented as specified.
14. Missing Data, Censoring, and Analysis Details
The ClinicalTrials.gov record identifies the primary analysis as a time-to-event comparison but do not provide detailed censoring rules, missing-data handling, imputation methods, sensitivity analyses, or analysis-population exclusions beyond the stated intention-to-treat comparison.
For a time-to-event endpoint, censoring is a fundamental part of the statistical framework. Participants who have not experienced the event by the end of their available observation can contribute information up to their censoring time. The validity of the resulting comparison depends on the assumptions underlying the censoring and follow-up process.
15. Interim Analysis, Non-Inferiority, and Bayesian Methods
The ClinicalTrials.gov record does not report an interim-analysis strategy, alpha-spending approach, non-inferiority margin, equivalence margin, Bayesian method, posterior probability, or Bayesian decision rule for the posted analyses.
| Design topic | What the registry-reported REVEAL data support |
|---|---|
| Non-inferiority margin | Not reported in the ClinicalTrials.gov record. |
| Equivalence margin | Not reported in the ClinicalTrials.gov record. |
| Interim analysis | Not reported in the ClinicalTrials.gov record. |
| Alpha spending | Not reported in the ClinicalTrials.gov record. |
| Bayesian analysis | Not reported in the ClinicalTrials.gov record. |
| Imputation | Not reported in the ClinicalTrials.gov record. |
| Stratification factors | Not reported in the ClinicalTrials.gov record. |
This distinction is important because the absence of a reported design feature is not evidence that the feature was absent from the full protocol or statistical analysis plan. It means only that the ClinicalTrials.gov record does not provide enough information to describe it here.
16. What the Primary Rate Ratio Does — and Does Not — Mean
The primary rate ratio of 0.91 means that the estimated event rate for the major coronary event endpoint was 0.91 times the corresponding rate in the placebo group under the reported analysis.
Equivalently, this corresponds to an estimated relative event-rate difference of approximately 9% in the direction of a lower rate for anacetrapib.
It does not mean that 9% of participants were prevented from having an event, that every participant experienced a 9% reduction, or that the absolute probability of an event fell by 9 percentage points.
The 95% CI of 0.85–0.97 shows the uncertainty around the estimated rate ratio. A point estimate without its confidence interval gives an incomplete picture because it does not communicate the precision of the estimate.
The P-value of 0.004 belongs to the statistical test associated with the primary analysis. It does not indicate that the treatment effect is "0.004 in size," nor does it distinguish a clinically large effect from a clinically small effect.
17. Primary vs Secondary Statistical Evidence
Primary endpoint
The major coronary event analysis has a reported log-rank method, rate ratio of 0.91, 95% CI of 0.85–0.97, and P-value of 0.004.
Secondary endpoints
The secondary analyses provide additional information about major atherosclerotic, presumed ischaemic stroke, and major vascular events, but their statistical reporting is not identical.
This distinction matters because an endpoint's position in the trial hierarchy is part of its statistical interpretation. The primary endpoint was prespecified as the principal assessment of the treatment comparison. Secondary endpoints can provide supportive information, but the ClinicalTrials.gov record does not establish a multiplicity-adjusted confirmatory framework for them.
18. Important Limitations and Interpretation Issues
- Registry-level statistical detail: the ClinicalTrials.gov record identifies the principal methods and effect measures but do not provide a complete statistical analysis plan.
- No reported Cox model: the primary analysis is described as a log-rank test with a rate ratio. A proportional-hazards assumption should not be attributed to the analysis without supporting documentation.
- Incomplete secondary-method reporting: the presumed ischaemic stroke analysis reports a rate ratio and confidence interval but no statistical method or P-value in the ClinicalTrials.gov record.
- Multiplicity: the ClinicalTrials.gov record does not specify a multiplicity-adjustment strategy for the secondary endpoints.
- Censoring details: the time-to-event framework is clear, but detailed censoring rules are not provided in the ClinicalTrials.gov record.
- Absolute effects: the ClinicalTrials.gov record provides relative rate ratios but do not provide enough event-count and follow-up information to calculate an absolute risk difference without introducing unsupported assumptions.
- Safety analysis: serious adverse events are reported descriptively as affected participants over participants at risk; no comparative inferential analysis is reported.
- Endpoint definition: the registry's primary endpoint is expressed as a number of participants with a major coronary event, while the statistical analysis identifies it as time-to-event and specifies time to first event. Both descriptions should be retained rather than collapsing the endpoint into a simple binary outcome.
19. Why This Trial Matters Statistically
REVEAL is a useful teaching example because it connects randomized trial design with the analysis of clinical events that occur at different times. The primary result cannot be understood fully from the rate ratio alone; its interpretation depends on randomization, time-to-event follow-up, the log-rank test, confidence intervals, and the distinction between primary and secondary endpoints.
| Concept | How it appears in REVEAL |
|---|---|
| Randomization | Participants were randomly allocated to anacetrapib or placebo. |
| Quadruple masking | The trial is registered as quadruple-masked. |
| Intention-to-treat | The primary assessment is an intention-to-treat comparison among all randomized participants. |
| Time-to-event endpoint | The primary analysis evaluates time to the first major coronary event. |
| Log-rank test | The reported primary method compares the time-to-event experience between randomized groups. |
| Rate ratio | The primary treatment effect is reported as a rate ratio of 0.91. |
| Confidence interval | The primary 95% CI is 0.85–0.97. |
| P-value | The primary analysis reports P = 0.004. |
| Secondary endpoints | Three additional event outcomes have posted analyses with varying levels of statistical detail. |
| Multiplicity | The ClinicalTrials.gov record does not report an adjustment strategy for the secondary endpoints. |
| Safety | Serious adverse events are reported by arm as affected participants over participants at risk. |
20. Related Tutorials
Learn more about the methods used in this trial:
21. Related Calculators
22. Sources
- ClinicalTrials.gov: REVEAL — NCT01252953. Official trial registry record.
- PubMed record: PMID 28847206.
- PubMed record: PMID 31331193.
- PubMed record: PMID 34910136.
- PubMed record: PMID 37750555.
- PubMed record: PMID 41843779.
Continue through the Clinical Biostats statistical pathway
Connect the trial's randomized design, time-to-event endpoint, log-rank analysis, rate ratio, and confidence interval to deeper statistical tutorials and practical calculators.
23. Record Summary
REVEAL provides a clear example of a large randomized cardiovascular prevention trial analyzed through a time-to-event framework. The primary endpoint was the first major coronary event, defined as coronary death, myocardial infarction, or coronary revascularization procedure, during the randomized treatment phase. The registry reports a log-rank analysis with a rate ratio of 0.91, a two-sided 95% confidence interval of 0.85–0.97, and P = 0.004.
The secondary analyses demonstrate why trial interpretation requires more than reading a single P-value. Major atherosclerotic events had a reported rate ratio of 0.93 with a 95% CI of 0.86–1 and P = 0.052; presumed ischaemic stroke had a rate ratio of 0.99 with a 95% CI of 0.87–1.12 and no reported statistical method or P-value in the ClinicalTrials.gov record; and major vascular events had a rate ratio of 0.93 with a 95% CI of 0.88–0.99 and P = 0.02.
From a statistical perspective, the most important lesson is the relationship among randomization, time-to-event follow-up, the log-rank test, the rate ratio, and the confidence interval. These elements describe different parts of the same evidence: randomization defines the comparison, time-to-event analysis uses the timing of events, the log-rank test supplies the formal comparison, the rate ratio communicates relative effect magnitude, and the confidence interval communicates uncertainty around that estimate.