12-week Open-label, Phase IIIb Comparing Efficacy and Safety of Rosuvastatin (CRESTOR™) in Combination With Ezetimibe: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for 12-week Open-label, Phase IIIb Comparing Efficacy and Safety of Rosuvastatin (CRESTOR™) in Combination With Ezetimibe (NCT00525824), a randomized phase 3 study. The primary outcome is Percent Change From Baseline in Low-density Lipoprotein Cholesterol (LDL-C) After 6 Weeks Combination Treatment.

2. Clinical Question

The study record describes a 12-week comparison of rosuvastatin and simvastatin, given as monotherapy or in combination with ezetimibe, in patients with elevated cholesterol and cardiovascular risk characteristics. The primary outcome focuses on the change in LDL-C after combination treatment, rather than a clinical event or a long-term outcome. The posted arm results report average percent change for four regimens, while the reported comparisons use the rosuvastatin 10 mg combination regimen as the reference.

This framing matters when reading the estimates: they describe differences in an average laboratory measure over the specified treatment period. They do not by themselves establish differences in symptoms, cardiovascular events, or long-term health outcomes. The record does not report a minimum clinically important difference for LDL-C or explain how the chosen outcome difference would be interpreted clinically. The question addressed by the posted results is therefore a short-term, comparative biomarker question in the trial population, not a general claim about all patients or about outcomes beyond the study period.

3. Trial Design

The registry describes a Phase IIIb, randomized, open-label, parallel-group, multicentre study. Its official title identifies a 12-week comparison of rosuvastatin in combination with ezetimibe and simvastatin in patients with hypercholesterolaemia and coronary heart disease. The record lists 1,743 enrolled participants and a completed status. The start is recorded as 2007-08 and primary completion as 2008-09.

Random allocation means treatment assignment was planned to occur by chance, which can help balance participant characteristics across groups on average. Open-label means masking was not used; the record lists masking as none. Knowledge of treatment assignment can influence care or reporting, though the primary endpoint is a laboratory measurement. The record calls the model parallel-group, so participants were assigned to a group rather than crossing through all regimens in sequence. The specific allocation procedure and any measures to limit bias are not reported in the registry record.

The posted primary endpoint includes four regimens: rosuvastatin 10 mg followed by rosuvastatin 10 mg plus ezetimibe 10 mg; rosuvastatin 20 mg followed by rosuvastatin 20 mg plus ezetimibe 10 mg; simvastatin 40 mg followed by simvastatin 40 mg plus ezetimibe 10 mg; and simvastatin 80 mg followed by simvastatin 80 mg plus ezetimibe 10 mg. The arm descriptions indicate a monotherapy phase followed by combination therapy, but the precise schedule and duration of each phase are not reported in the registry record. The primary endpoint timeframe is the mean of Weeks 4 and 6 on combination therapy, using last observation carried forward.

The primary outcome results give group sizes of 210, 204, 199, and 201, respectively. The registry does not explain the relationship between these outcome group sizes and the total enrollment, nor does it report the analysis population definition. The original sample-size justification, prespecified statistical plan, and any interim analyses are not reported in the registry record. As a result, the posted numerical comparisons can be interpreted, but their relationship to a prespecified analysis plan cannot be fully assessed from this 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: Patients underwent screening procedures (Week -6; Visit 1), and entered a 6-week dietary lead-in period. Those who fulfilled all eligibility criteria (Week -6; Visit 1) and had qualifying lipid values at Visit 2 were randomly allocated (1:1:1:1) to 1 of 4 treatments for a period of 12 weeks.

Overall Study

Reasons for not completing this period:

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.

6. Primary Endpoint

The registry lists the primary outcome as Percent Change From Baseline in Low-density Lipoprotein Cholesterol (LDL-C) After 6 Weeks Combination Treatment, measured over Mean of Weeks 4 and 6 on combination therapy (Last observation carried forward).

The primary endpoint is percent change from baseline in low-density lipoprotein cholesterol (LDL-C) after six weeks of combination treatment. The registry defines the calculation as the combination-treatment value minus the baseline value, divided by the baseline value, then multiplied by 100. The reported endpoint is the mean of Weeks 4 and 6 on combination therapy, with last observation carried forward used for the stated timeframe.

LDL-C is a continuous laboratory measure, so the endpoint summarizes change numerically and can be compared across treatment groups using means. Expressing change as a percentage relates the observed change to each participant’s baseline value, rather than describing only an absolute level. This makes the endpoint a measure of relative change over the specified treatment period. The registry reports group means and standard deviations, which describe the average change and the spread of individual changes within each arm.

As a general statistical point, a continuous biomarker endpoint can detect differences in average measurements without requiring a clinical event to occur during follow-up. That can make it useful for a relatively short study, but the endpoint remains a surrogate laboratory measure. The posted result alone does not quantify how any difference translates into clinical outcomes, and the registry record does not report a threshold for a clinically meaningful change. The endpoint also captures the period specified in the record; it is not evidence about persistence after follow-up ends.

7. Results Posted in the Registry

• R20 to R20 + E10: mean -63.48 versus -59.72 on control; difference -3.76 (95% CI -6.75 to -0.77), standardized effect d = -0.24; z-test p = 0.014.
• S40 to S40 + E10: mean -55.22 versus -59.72 on control; difference 4.50 (95% CI 1.59 to 7.41), standardized effect d = 0.30; z-test p = 0.002.
• S80 to S80 + E10: mean -57.42 versus -59.72 on control; difference 2.30 (95% CI -1.12 to 5.72), standardized effect d = 0.13; z-test p = 0.187.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Percent Change in High-density Lipoprotein Cholesterol (HDL-C) After 6 Weeks Combination Treatment

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • R20 to R20 + E10: mean 7.46 versus 6.41 on control; difference 1.05 (95% CI -1.88 to 3.98), standardized effect d = 0.07; z-test p = 0.482.
  • S40 to S40 + E10: mean 3.92 versus 6.41 on control; difference -2.49 (95% CI -5.07 to 0.09), standardized effect d = -0.19; z-test p = 0.059.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between R20 to R20 + E10 and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between S40 to S40 + E10 and the control. The interval gives the range of effects that remain compatible with the data.

Percent Change in Total Cholesterol (TC) After 6 Weeks Combination Treatment

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • R20 to R20 + E10: mean -46.63 versus -43.00 on control; difference -3.63 (95% CI -5.95 to -1.31), standardized effect d = -0.30; z-test p = 0.002.
  • S40 to S40 + E10: mean -39.56 versus -43.00 on control; difference 3.44 (95% CI 1.12 to 5.76), standardized effect d = 0.29; z-test p = 0.004.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for R20 to R20 + E10 than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for S40 to S40 + E10 than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percent Change in Triglycerides (TG) After 6 Weeks Combination Treatment

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • R20 to R20 + E10: mean -35.00 versus -28.85 on control; difference -6.15 (95% CI -10.74 to -1.56), standardized effect d = -0.26; z-test p = 0.009.
  • S40 to S40 + E10: mean -22.95 versus -28.85 on control; difference 5.90 (95% CI 0.84 to 10.96), standardized effect d = 0.23; z-test p = 0.022.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for R20 to R20 + E10 than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for S40 to S40 + E10 than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary outcomes extend the lipid assessment beyond LDL-C to HDL-C, total cholesterol, triglycerides, non-HDL-C, apolipoproteins, and cholesterol ratios. Across the reported summaries, the measures do not all move in the same numerical direction or show the same degree of separation between regimens. This is unsurprising statistically: different measures have different distributions, variability, and relationships to the intervention. The record posts results for only some of the listed secondary outcomes, with 12 secondary outcomes posted overall.

These outcomes provide descriptive context for the primary LDL-C result, but they do not independently confirm or overturn it. They measure related but distinct aspects of the lipid profile, and the study’s primary question is specifically framed around LDL-C percent change. Any consistency across measures may be of interest, but the record does not report a prespecified method for combining these outcomes or treating them as a family of confirmatory tests.

The secondary results carry less inferential weight because the posted comparisons are unadjusted and the record does not describe multiplicity control, a testing hierarchy, or a complete statistical analysis plan. When many outcomes and comparisons are assessed, individual p-values can appear notable even without a broad underlying pattern. A result from a secondary measure should therefore be interpreted as exploratory unless the protocol states otherwise, and that protocol detail is not reported in the registry record.

Also, these outcomes remain laboratory measurements over the specified combination-treatment timeframe. They do not supply evidence about later clinical events or the persistence of changes after study treatment. Their most defensible role is to show the breadth of measures the trial recorded and to help readers see whether the primary endpoint sits within a broader descriptive pattern, while avoiding claims that the record does not support.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Percent Change in High-density Lipoprotein Cholesterol (HDL-C) After 6 Weeks Combination Treatment
• Percent Change in Total Cholesterol (TC) After 6 Weeks Combination Treatment
• Percent Change in Triglycerides (TG) After 6 Weeks Combination Treatment
• Percent Change in Non-high-density Lipoprotein Cholesterol (nonHDL-C) After 6 Weeks Combination Treatment
• Percent Change in Apolipoprotein B (ApoB) After 6 Weeks Combination Treatment
• Percent Change in Apolipoprotein A1 (ApoA-1) After 6 Weeks Combination Treatment
• Percent Change in TC/HDL-C After 6 Weeks Combination Treatment
• Percent Change in LDL-C/HDL-C After 6 Weeks Combination Treatment

9. Statistical Interpretation

10. Statistical Methods Explained

The outcome is continuous: each participant has a percent change in LDL-C, and each arm is described by a mean and standard deviation. The mean is the arithmetic average of the observed endpoint values. The standard deviation describes how dispersed those values are around the arm mean; it is not the uncertainty in the mean itself. A large standard deviation can coexist with a precisely estimated mean when the group is sufficiently large, while a small group can leave the mean less precise even when individual values are relatively clustered.

How to read the comparisons

The posted comparison estimates are mean differences against the rosuvastatin 10 mg followed by rosuvastatin 10 mg plus ezetimibe 10 mg reference regimen. The sign depends on the order of subtraction: the estimate is the comparison-group mean minus the reference-group mean. A negative estimate therefore corresponds to a lower, more negative mean percent change in the comparison group; a positive estimate corresponds to a higher, less negative mean percent change. This is a description of the measure’s direction, not a statement about clinical importance.

Each comparison also has an interval estimate. In general, a confidence interval describes a range of values compatible with the observed data under the analysis method and its assumptions. Its width reflects precision: a wider interval means the data are consistent with a broader set of possible differences. The confidence level and exact method used for the posted intervals are not reported in the registry record, so they should not be inferred from the numerical limits alone. Where an interval includes zero, the displayed results do not rule out no mean difference under the interval calculation. Where it does not include zero, that does not establish clinical importance or resolve multiplicity concerns.

What the p-values say

The listed p-values are unadjusted. In general, a p-value measures how unusual data at least as incompatible with a null hypothesis would be if that hypothesis and the statistical model were correct. It is not the probability that the null hypothesis is true, nor is it the probability that the observed difference arose by chance. A smaller p-value is not a measure of effect size or clinical importance. Because the results are unadjusted and the registry lists numerous outcomes and comparisons, the p-values do not account for the full set of opportunities to observe a nominally notable result.

The registry record does not report the prespecified statistical test, whether covariate adjustment was planned, or the statistical analysis population. The posted comparisons can be read as unadjusted group-mean comparisons, but the complete modeling and analysis procedures are not reported in the registry record. No reported analyses are listed for the primary outcome measure in the analysis listing, and the secondary measures likewise have no reported analyses listed. This limits the ability to confirm whether the posted analyses match a complete prespecified plan.

Repeated measurements and missing observations

The endpoint timeframe uses the mean of Weeks 4 and 6 on combination therapy and specifies last observation carried forward. In general, last observation carried forward fills a missing later measurement with the participant’s most recent available measurement. This approach can preserve a nominal analysis set, but it relies on an assumption that the last observed value is a reasonable stand-in for the missing value. If values would have changed after the last observation, the imputed value may not represent the unobserved outcome. The record does not report the amount or pattern of missing data, the reasons for missingness, or sensitivity analyses using alternative approaches.

Randomization is a design feature intended to support fair comparisons by reducing systematic differences between assigned groups. It does not guarantee that all groups are balanced in the realized sample, and it does not address missing outcomes or problems with the analysis after randomization. The record identifies the study as open-label and randomized, but the allocation details, concealment procedures, and analysis population are not reported. Those omissions matter when considering how well the design protections were implemented.

Multiplicity and interpretation

Multiplicity arises when several statistical comparisons or outcomes are assessed. Even if each test is considered separately, conducting many tests increases the chance that at least one result appears unusual under its null hypothesis. General methods to address this include adjusting significance thresholds or specifying a hierarchy of outcomes in advance, but the registry record does not report whether either was used. Therefore, the posted unadjusted p-values should be read as individual comparisons rather than as multiplicity-controlled evidence across the trial.

The same caution applies to secondary outcomes. Their means can describe patterns across measures, but a selected nominal p-value is not independent confirmation of the primary finding when many measures are being examined. The original statistical plan, sample-size justification, planned interim analyses, confidence-interval method, and any adjustment strategy are not reported in the registry record. General statistical principles help explain the posted estimates, but they cannot fill in those missing protocol details.

11. Confidence Intervals and Precision

The posted primary comparisons include intervals around the estimated mean differences. For rosuvastatin 20 mg followed by rosuvastatin 20 mg plus ezetimibe 10 mg versus the reference regimen, the estimate is -3.76 percentage points and the interval runs from -6.747 to -0.773. For simvastatin 40 mg followed by simvastatin 40 mg plus ezetimibe 10 mg, the estimate is 4.5 and the interval runs from 1.591 to 7.409. For simvastatin 80 mg followed by simvastatin 80 mg plus ezetimibe 10 mg, the estimate is 2.3 and the interval runs from -1.117 to 5.717.

An interval is useful because an estimate alone can make a result look more exact than the data warrant. In general, the interval represents a range of values compatible with the data under the method used, while its width conveys precision. The simvastatin 80 mg comparison interval includes zero and spans differences in either direction. The other two intervals do not include zero, but interval exclusion of zero does not establish a meaningful clinical difference, nor does it account for the multiple unadjusted comparisons. The confidence level and interval-construction method are not reported in the registry record.

The group standard deviations also show that individual participant changes varied within each regimen. Those spreads are not confidence intervals for the mean and should not be read as uncertainty in the between-group estimate. The group sizes reported for the primary endpoint range across the four posted arms, but the relationship of those outcome counts to total enrollment of 1,743 is not explained in the registry record. The extent and handling of missing endpoint data are also not reported beyond the use of last observation carried forward.

Precision is therefore best judged from the interval widths and the stated endpoint group sizes, with the limits of the analysis in mind. The intervals quantify uncertainty for the displayed comparisons only; they do not capture uncertainty about unreported design decisions, missing-data assumptions, multiplicity, or the applicability of these results outside the study population.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size estimates are planning illustrations based on the observed effect, not the trial’s original design calculation. For these continuous outcomes, each figure is the number of participants per arm, rather than the total across arms or a count of events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect.

The 80% power column gives the estimated participants per arm for a design that would have an 80% probability of detecting the specified effect under those assumptions, if that effect and the assumptions were correct. The 90% power column uses a higher target probability of detecting that effect, so its required per-arm count is larger. These are prospective planning quantities, not a measure of the chance that the completed trial found a true effect.

Rows planning for 100% of the observed effect use the full observed estimate as the target; rows planning for 75% or 50% use smaller target effects. Smaller effects are harder to distinguish from random variation, so they require more participants per arm to reach the same target power. For example, the primary comparison estimates vary in size, and the planning counts consequently vary as well. The figures are not observed power and do not describe the trial’s actual probability of detecting an effect after its results were known.

The original sample-size justification and design assumptions are not reported in the registry record. These planning illustrations should therefore be used to understand how the assumed effect size and desired power relate to participant requirements, not to infer what the investigators planned or whether the trial was adequately sized for its original objectives.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -3.760; pooled standard deviation σ = 15.465.
2. n = 2 × (1.96 + 0.842)² × 15.465² / -3.760² = 265.6, rounded up to 266 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -3.76                      # observed difference in means
sigma <- 15.4649244872269                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(525824)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 266 (80% power), 356 (90% power)
 75% of observed effect: n per arm = 473 (80% power), 633 (90% power)
 50% of observed effect: n per arm = 1,063 (80% power), 1,423 (90% power)
power.t.test: 266.5 
Simulated power at n = 266 per arm: 0.796 (20000 replicates)

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

14. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary LDL-C endpoint and multiple secondary lipid outcomes, with comparisons against a reference regimen. Multiple statistical testing raises a general interpretive concern: as the number of tests grows, the chance of at least one apparently unusual result also grows, even when no true differences exist. Adjustment procedures, a prespecified hierarchy, or a clear separation of confirmatory and exploratory analyses can help manage this issue. Whether such procedures were planned or used is not reported in the registry record. The posted p-values are unadjusted, so they should not be interpreted as controlling error across the set of outcomes and comparisons.

The registry record does not report an interim-analysis plan, stopping rules, or an analysis plan defining the primary comparison and analysis population. It also does not report a sample-size justification or the assumptions underlying the original trial design. These omissions make it difficult to determine whether the posted analyses correspond to the intended confirmatory tests, or whether other analyses were conducted but not included in the record.

The endpoint timeframe specifies last observation carried forward. The number of participants with missing measurements, reasons for missingness, and the timing of any missing outcomes are not reported in the registry record. The record also does not describe sensitivity analyses or alternative missing-data approaches. As a general matter, carrying an earlier value forward assumes that it adequately represents the missing later value; the reliability of that assumption depends on how outcomes evolve and why measurements are missing.

The registry does not state whether the primary analysis followed intention-to-treat principles, included only participants with available measurements, or used another population definition. The group sizes listed with the outcome do not by themselves establish which participants were included or how post-randomization events were handled. These gaps do not invalidate the posted numerical summaries, but they limit assessment of bias, robustness, and consistency with an original protocol. Further details about the analysis plan and missing-data handling are not reported in the registry record.

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

The results describe average percent changes in LDL-C for four regimens during the endpoint period defined as the mean of Weeks 4 and 6 on combination treatment. They also provide estimated mean differences between three regimens and the rosuvastatin 10 mg combination reference. The estimates and intervals show the magnitude and precision of those short-term comparisons in the participants represented by the posted outcome group sizes.

These findings do not establish that one regimen is clinically superior, or that any regimen does or does not work. The endpoint is a laboratory measure, and the registry record does not report a threshold for a clinically meaningful difference or connect these short-term changes to clinical outcomes. The results also do not establish how long the observed changes would persist beyond the specified treatment period. A biomarker comparison should not be treated as equivalent to evidence about symptoms, cardiovascular events, or long-term benefit and harm.

The comparisons are relative to one posted reference regimen, not direct comparisons among every possible pair of regimens. The study is described as open-label, and no masking was used. Although LDL-C is a laboratory endpoint, the record does not give enough detail to assess all measurement and analysis procedures. The posted p-values are unadjusted, and the original statistical plan and multiplicity handling are not reported in the registry record. Those limitations matter when judging the strength of any inferential claim.

The adverse-event record lists counts by regimen, but the numbers at risk are not reported and deaths are not reported in the registry record. Counts without denominators do not establish comparative event rates. The record therefore supports a bounded reading: it documents short-term mean changes in a lipid measure and provides selected safety counts, while leaving important questions about clinical relevance, long-term outcomes, missing data, and comparative safety unanswered.

16. Why This Trial Matters Statistically

This trial is statistically interesting because it compares several active regimens while reporting a continuous laboratory endpoint as both group means and mean differences. That structure makes clear that an observed group average, a comparison estimate, and uncertainty around that estimate answer related but distinct questions. The means describe each arm; the differences frame the comparisons against a reference; the intervals indicate how precisely those comparisons are estimated under the method used.

It also illustrates why direction and magnitude should be read together. Because percent changes are negative in the posted primary results, a negative between-group difference represents a more negative mean change relative to the reference, while a positive difference represents a less negative change. Reading a sign without checking the subtraction order can reverse the interpretation. The interval adds another layer: some estimates are more precisely bounded than others, and a point estimate should not be treated as exact.

The record is also a useful example of limits in registry-result interpretation. It identifies randomization and an open-label parallel design, but does not report the original analysis plan, sample-size justification, multiplicity approach, or full missing-data details. Those omissions prevent readers from assuming that an unadjusted comparison was the sole prespecified test or that the listed outcome population represents every randomized participant.

Finally, the distinction between a planning calculation and observed trial evidence is important. Sample-size illustrations based on observed effects can help show how power targets and effect assumptions relate to participant requirements, but they cannot reconstruct the study’s original design rationale. The statistical value of the record lies in reading the numerical evidence carefully while keeping its design, endpoint, and reporting limits in view.

17. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The study duration is stated as 12 weeks, and the primary endpoint is a short-term measurement on combination therapy. The record does not report follow-up for outcomes after that period, so durability and longer-term outcomes cannot be assessed from these results. The endpoint is a surrogate laboratory measure, and the record does not provide a clinical-importance threshold for the observed differences.

The registry describes the study as open-label with no masking. The primary endpoint is a laboratory measure, but open-label conduct can still influence participant behavior, treatment decisions, follow-up, and reporting. The record does not detail safeguards for these possibilities. It also does not report the allocation method or concealment procedure beyond identifying the study as randomized.

There is an information gap between the total enrollment of 1,743 and the smaller group sizes attached to the primary outcome. The registry record does not explain this relationship or define the analysis population. It does not report the number or pattern of missing endpoint measurements, although the endpoint timeframe specifies last observation carried forward. No alternative missing-data analyses are reported. The original sample-size justification and statistical analysis plan are also not reported in the registry record.

The posted comparison p-values are unadjusted, and the record does not state whether multiplicity was addressed. Several secondary outcomes are listed, but the posted secondary results cover only a subset. Adverse-event counts are reported without numbers at risk, and death counts are not reported in the registry record. These gaps limit interpretation of comparative safety. Finally, the registry’s participant description is broad and the record does not provide further detail to assess how well the results apply to populations beyond the stated study criteria.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: 12-week Open-label, Phase IIIb Comparing Efficacy and Safety of Rosuvastatin (CRESTOR™) in Combination With Ezetimibe — NCT00525824.
• Registry record last updated 2011-05-13; status last verified 2011-05.
• Results first posted 2009-10-07.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

21. Record Summary

NCT00525824 is a completed Phase IIIb, randomized, open-label, parallel-group, multicentre study sponsored by AstraZeneca. The record lists 1,743 participants enrolled, with a recorded start of 2007-08 and primary completion of 2008-09. It concerns patients with hypercholesterolaemia and coronary heart disease or related risk characteristics.

The primary endpoint is percent change from baseline in LDL-C, summarized as the mean of Weeks 4 and 6 on combination therapy using last observation carried forward. Four combination regimens are reported, with comparisons against rosuvastatin 10 mg plus ezetimibe 10 mg after rosuvastatin 10 mg monotherapy. Secondary lipid outcomes and adverse-event counts are also posted.

The record does not report the original sample-size justification, full statistical analysis plan, multiplicity approach, or detailed missing-data information. The posted comparisons and planning illustrations can be read as statistical descriptions of the recorded results, but they do not establish clinical importance or long-term outcomes.