To Evaluate Ezetimibe/Simvastatin and Niacin (Extended Release Tablet) in Patients With Type IIa or Type IIb Hyperlipidemia (0653A-091)(COMPLETED): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for To Evaluate Ezetimibe/Simvastatin and Niacin (Extended Release Tablet) in Patients With Type IIa or Type IIb Hyperlipidemia (0653A-091)(COMPLETED) (NCT00271817), a randomized phase 3 study. The primary outcome is Percent Change From Baseline in Low-Density Lipoprotein-Cholesterol (LDL-C).

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, double-blind, parallel-group study. Its official title describes it as multicenter. The registry record lists enrollment of 1220 and status as completed. The recorded start was 2005-12, and primary completion was 2008-02. Organon and Co is listed as the sponsor.

The study evaluated ezetimibe/simvastatin and extended-release niacin in people with high cholesterol. The arm descriptions identify groups in Part 1 and Part 2, including combination therapy, ezetimibe/simvastatin, niacin, and a combination with placebo to niacin. The interventions list also includes placebo descriptions. The registry record does not report further details of the allocation sequence, concealment, treatment duration for every arm, or the specific masking procedures.

The primary result concerns a narrower comparison than the complete set of arm descriptions: combination therapy versus niacin alone. It is based on the completers population, not a population defined simply by randomization. For this analysis, participants needed a baseline measurement, at least 24 weeks of active study therapy, and an on-treatment measurement at the maximum titrated dose per protocol. The registry record does not report the number randomized to each primary-comparison group or the number excluded from this analysis.

The main endpoint was percent change from baseline in LDL-C at 24 weeks. Secondary outcomes listed in the record include changes in other lipid measures and later LDL-C follow-up. The registry record does not report the original design assumptions used to choose the trial enrollment target, nor does it give enough detail to reconstruct all trial procedures.

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.

Part 1

Reasons for not completing this period:

Part 2

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The participant population for this analysis is the Completers Population. This includes all patients with a baseline value, who receive at least 24 weeks of active study therapy, and who have an on-treatment measurement at the maximum titrated dose per the protocol.

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), measured over Baseline and 24 Weeks.

The primary endpoint is the mean percent change from baseline in low-density lipoprotein cholesterol (LDL-C), measured at baseline and 24 weeks. The endpoint description defines the change as the 24-week measure minus baseline. The registry reports a continuous outcome and presents the group means as percent changes.

Change from baseline expresses how an outcome shifted relative to each participant’s starting measurement. Comparing mean change between groups can help describe whether the average course differed between the assigned treatment strategies. In this registry analysis, the primary contrast is the mean difference between the combination group and the niacin-alone group.

LDL-C is one of the lipid measures the trial set out to assess. The endpoint is useful for describing the measured biological response over the stated period, but it is not by itself a measure of clinical events, symptoms, or quality of life. The posted result therefore answers a question about average LDL-C change in the specified completers population, rather than every possible effect of treatment or the experience of every randomized participant.

The record does not provide a clinical threshold for interpreting the size of this endpoint, or a patient-level distribution from which to judge how individual responses varied. Interpretation should remain tied to the endpoint, analysis population, and comparison actually reported.

7. Results Posted in the Registry

• Ezetimibe/Simvastatin + Niacin: mean -58.50 versus -20.10 on control; difference -38.40 (95% CI -41.39 to -35.41), standardized effect d = -2.38; z-test p = <0.001.

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 From Baseline in Non-High-Density Lipoprotein-Cholesterol (Non-HDL-C)

• 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 errors.
• Result:
  • Ezetimibe/Simvastatin + Niacin: mean -55.60 versus -22.00 on control; difference -33.60 (95% CI -36.59 to -30.61), standardized effect d = -2.09; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Ezetimibe/Simvastatin + Niacin than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percent Change From Baseline in High-Density Lipoprotein-Cholesterol (HDL-C)

• 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 errors.
• Result:
  • Ezetimibe/Simvastatin + Niacin: mean 30.20 versus 8.10 on control; difference 22.10 (95% CI 18.89 to 25.31), standardized effect d = 1.13; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Ezetimibe/Simvastatin + Niacin than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percent Change From Baseline in Low-Density Lipoprotein-Cholesterol (LDL-C)

• 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 errors.
• Result:
  • Ezetimibe/Simvastatin + Niacin: mean -54.00 versus -49.30 on control; difference -4.70 (95% CI -7.91 to -1.49), standardized effect d = -0.25; z-test p = 0.004.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Ezetimibe/Simvastatin + Niacin than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary lipid outcomes follow a generally coherent pattern with the primary endpoint: the combination group’s mean change is more negative for non-HDL-C and LDL-C, while its mean change is more positive for HDL-C than the relevant comparator. The longer-term LDL-C comparison is also in the same direction as the 24-week primary comparison, though it is a separate outcome assessment at a different follow-up. Taken together, these results are directionally consistent across several lipid measures, but they are not all identical comparisons or time points.

Secondary results complement the primary finding by showing whether the pattern extends beyond the primary LDL-C measure. They do not replace the primary endpoint, and their estimates should be interpreted on their own definitions and analysis populations. The registry lists additional secondary outcomes, but not all posted outcome entries have a corresponding detailed comparison in the registry record.

The reported secondary p-values are unadjusted. When many outcomes or comparisons are examined, the chance of at least one apparently strong result arising by chance can increase. Adjustment methods or a prespecified hierarchy can address that concern, but a multiplicity procedure is not reported in the registry record. Consequently, the secondary findings carry less confirmatory weight than a single prespecified primary comparison unless the full analysis plan establishes otherwise.

Statistical significance in a secondary comparison should not be treated as proof of a separate clinical benefit. These outcomes measure lipid changes, not clinical events. The record also does not provide enough information to establish how secondary endpoints were prioritized, whether analyses were prespecified in their posted form, or how missing values affected each comparison.

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

• Percent Change From Baseline in Non-High-Density Lipoprotein-Cholesterol (Non-HDL-C)
• Percent Change From Baseline in High-Density Lipoprotein-Cholesterol (HDL-C)
• Percent Change From Baseline in Triglycerides (TG)
• Percent Change From Baseline in High-Density Lipoprotein-Cholesterol (HDL-C)
• Percent Change From Baseline in Triglycerides (TG)
• Percent Change From Baseline in Non-High-Density Lipoprotein-Cholesterol (Non-HDL-C)
• Percent Change From Baseline in Low-Density Lipoprotein-Cholesterol (LDL-C)
• Percent Change From Baseline in Non-High-Density Lipoprotein-Cholesterol (Non-HDL-C)

9. Statistical Interpretation

10. Statistical Methods Explained

The posted primary result includes two kinds of estimates. The arm summaries describe each group’s mean change and standard error. The sponsor’s reported comparison uses ANCOVA and gives a mean difference, confidence interval, and p-value. A separate unadjusted re-calculation uses the posted group summary statistics. These methods can be related but are not interchangeable: an adjusted model may account for baseline values or other model terms, while a simple comparison of summary statistics does not reproduce unreported model details.

What does a mean difference represent?

A mean difference subtracts one group’s average outcome from the other group’s average outcome according to a stated comparison direction. For the primary result, the difference is combination therapy relative to niacin alone. Because the outcome is percent change from baseline, a negative difference indicates a more negative average change in the combination group. The estimate describes a group-average contrast; it does not imply that every participant had the same change or that individual outcomes can be inferred from the group means.

What does ANCOVA contribute?

ANCOVA is a model-based method for comparing groups while accounting for one or more covariates. In a baseline-adjusted analysis, the baseline measurement can be included as a covariate to improve the comparison of follow-up outcomes. The registry labels the sponsor’s method as ANCOVA and reports the parameter as a mean difference in final values. It does not provide the model specification, covariates, assumptions checks, or detailed handling of missing observations. Those elements are therefore not reported in the registry record and should not be inferred from the method label alone.

Why compare the sponsor analysis with a re-calculation?

A re-calculation from posted means and standard errors offers a transparent check on the scale and direction of the contrast. It is unadjusted and depends on the available summary statistics. The sponsor’s reported ANCOVA result is the trial’s own analysis. In this case the point estimates agree at the displayed precision, and the confidence intervals and p-values are also close. There is no recorded discrepancy. Agreement between the two approaches supports consistency of the summary, but it does not reveal model details that the registry does not report.

What does a p-value say?

A p-value describes how incompatible the observed data are with a specified null hypothesis under the statistical model and assumptions. A small value such as the reported p <0.001 is evidence against a zero mean difference under that analysis. It is not the probability that the null hypothesis is true, the chance that the finding is due to bias, or a measure of clinical importance. The p-value also does not describe the range of effect sizes compatible with the data as directly as a confidence interval does.

Why can dispersion and sample size matter?

For a continuous outcome, variability affects the precision of a group mean and the comparison between groups. Standard errors describe uncertainty in estimated means; they are not the spread of individual participant values. Greater variability generally makes a mean difference less precise when other features are held constant. The record reports standard errors for the primary group means and a pooled standard deviation in the re-calculation, but not the participant-level distribution.

Unequal group sizes and the analysis population also shape precision. The primary summaries are for the completers population, so their group sizes need not equal the originally randomized group sizes. The registry record does not report the full flow of participants through the study or enough information to determine how missing outcome data affected the analysis. These limitations matter when deciding how far to generalize the model-based estimate.

The secondary endpoints are continuous lipid measures, with some comparisons reported at 24 weeks and another LDL-C comparison at 64 weeks. Each estimate has its own outcome definition, comparator, and analysis population as listed in the record. Statistical interpretation should preserve those distinctions rather than treating all estimates as one common analysis.

11. Confidence Intervals and Precision

The sponsor’s primary estimate is a mean difference of -38.4, with a 95% confidence interval from -41.4 to -35.4. The unadjusted re-calculation gives -38.4, with a 95% confidence interval from -41.39 to -35.41. A confidence interval describes a range of values reasonably compatible with the data and model used; it is not a range containing a stated proportion of individual participant responses. Under repeated use of the same procedure and assumptions, intervals constructed this way would capture the true parameter at the stated confidence level.

The interval is narrow relative to the distance of its limits from zero, indicating that the primary mean difference is estimated with fairly high statistical precision within this analysis. Its limits remain negative, consistent with the direction of the point estimate. That precision is conditional on the completers population, the endpoint definition, the statistical model, and the assumptions used. It does not remove possible selection effects from analyzing completers or uncertainty about unreported model details.

The sponsor and unadjusted intervals are not identical in their displayed limits. The sponsor’s ANCOVA estimate is based on the trial’s reported adjusted analysis; the re-calculation is based on posted summary statistics and is not an exact reproduction of the model. The differences in interval limits should be read in that context, not as conflicting conclusions. The registry record does not provide enough detail to explain every modeling difference.

Precision is distinct from clinical importance. A confidence interval can be narrow while the relevance of its estimated values still requires clinical context that the registry record does not supply. It also does not describe the spread of individual outcomes or the chance of harms.

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 planning figures are participants per arm for the continuous-outcome comparisons, not total events. They are illustrations based on the observed effect rather than the trial’s own design. The listed 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 the stated fraction of the observed effect.

The 80% power column describes the illustrative number of participants per arm for a design with an 80% probability of detecting the assumed effect under those planning assumptions, if that effect and the model assumptions were correct. The 90% power column uses a higher target probability of detection. These are planning concepts, not guarantees about what a future study would find. Smaller effects need more participants for the same target power, which is why the planning requirements increase when the assumed effect is a smaller fraction of the observed effect.

The original trial’s own sample-size justification, target power, and design assumptions are not reported in the registry record. The planning calculations therefore do not explain why the study enrolled the number recorded or establish what power the completed trial had for its actual analysis. They are not observed power: they do not quantify the probability that the study detected its effect after the results were known.

Interpret these estimates as conditional planning illustrations rather than as a recommended sample size. They use the observed effect as an anchor, which may differ from the effect a future study should be designed to detect. They also assume equal allocation even though the posted primary analysis groups are not equal in size.

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 = -38.400; pooled standard deviation σ = 16.101.
2. n = 2 × (1.96 + 0.842)² × 16.101² / -38.400² = 2.8, rounded up to 3 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 <- -38.4                      # observed difference in means
sigma <- 16.1011834381237                     # 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(271817)
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 = 3 (80% power), 4 (90% power)
 75% of observed effect: n per arm = 5 (80% power), 7 (90% power)
 50% of observed effect: n per arm = 12 (80% power), 15 (90% power)
power.t.test: 4 
Simulated power at n = 3 per arm: 0.597 (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. Analyses Reported by the Sponsor

The sponsor reported an ANCOVA mean difference in final values of -38.4 for the primary comparison, with a 95% confidence interval from -41.4 to -35.4 and p <0.001. The unadjusted re-calculation from the posted summaries estimates the same mean difference at the displayed precision, with a confidence interval from -41.39 to -35.41 and p <0.001.

The close agreement is reassuring about the basic direction and approximate size of the contrast, but the methods answer slightly different computational questions. The recalculation is an unadjusted comparison built from group summaries. ANCOVA is the sponsor’s model-based analysis; details of its covariates and specification are not reported in the registry record. It should not be assumed that the recalculation reproduces the sponsor’s model or its exact standard error.

The registry does not identify a discrepancy between the two estimates. The sponsor’s result remains the trial’s reported analysis, while the recalculation serves as a check using the posted summary statistics. Neither comparison resolves limitations arising from the completers analysis population or missing information about the full analysis plan.

15. Multiplicity, Missing Data and Analysis Populations

The registry reports a primary ANCOVA comparison and several secondary lipid outcomes, but a complete statistical analysis plan is not reported in the registry record. In particular, the record does not describe whether secondary tests were adjusted for multiplicity, whether a testing hierarchy was specified, or how the analysis team controlled the overall chance of false-positive findings across multiple comparisons. The reported secondary p-values should therefore be treated as unadjusted evidence, not as though they were protected by an unreported correction.

The primary analysis uses a completers population. Its definition requires a baseline value, at least 24 weeks of active therapy, and an on-treatment measure at the maximum titrated dose per protocol. This is a specific subset definition, not a general statement about all randomized participants. The registry record does not report an intention-to-treat analysis for the primary endpoint, the number of participants who did not meet the completer definition, or the reasons for missing measurements.

Missing-data methods can affect results if participants with missing outcomes differ from those with observed outcomes. The registry record does not report whether missing observations were imputed, whether any sensitivity analyses were conducted, or what assumptions were used about unobserved outcomes. It also does not report whether the analysis population was determined before unblinding.

These omissions do not invalidate the posted estimate, but they limit assessment of robustness. The point estimate and its confidence interval describe the reported analysis, conditional on its population and methods. Without further information, it is not possible to determine how alternative missing-data assumptions, different analysis populations, or multiplicity adjustment would affect the interpretation.

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

The primary result estimates a difference in mean LDL-C percent change at 24 weeks between ezetimibe/simvastatin plus extended-release niacin and niacin alone among participants meeting the registry’s completers definition. The combination group had a more negative mean change, and the sponsor’s ANCOVA estimate was -38.4 with a 95% confidence interval from -41.4 to -35.4. The reported p-value was p <0.001. Those values describe the measured endpoint in that analysis.

The result does not establish that every participant experienced a reduction of that size. Group means conceal variation in individual responses, and the reported standard errors are uncertainty measures for means rather than summaries of individual spread. Nor does this lipid endpoint establish an effect on cardiovascular events, symptoms, quality of life, or long-term health. Those outcomes are not the primary measure reported in this comparison.

The result also should not be read as applying automatically to all people with high cholesterol or to all participants randomized in the trial. The analysis population was restricted to completers as defined by treatment duration and measurement requirements. The registry record does not provide enough detail about exclusions, missing data, or the full analysis plan to assess the effect of those choices fully.

Statistical evidence against a zero mean difference is not equivalent to proof of clinical importance or an overall judgment about benefits and harms. The registry’s adverse-event information is presented in separate group summaries, and denominators for those summaries are not reported. The posted findings should therefore be understood as evidence about a defined average lipid change, with limits on broader clinical interpretation.

17. Why This Trial Matters Statistically

This trial is statistically instructive because its primary comparison links a continuous endpoint, an analysis population, and a model-based estimate. The mean percent change is easy to express in the clinical scale of the outcome, while the difference and its confidence interval describe the contrast and its precision. The example also illustrates why an endpoint’s analysis population matters: a completers analysis answers a more restricted question than an analysis of all randomized participants might.

The comparison between the sponsor’s ANCOVA result and an unadjusted re-calculation shows how posted summary statistics can provide a useful consistency check without replacing the trial’s own model. Similar estimates do not make the analyses identical. The adjusted model, the unadjusted calculation, and their assumptions should remain distinct in interpretation.

The secondary lipid measures add context, but they also raise the familiar issue of multiple testing. A set of directionally consistent results may be informative, yet unadjusted secondary p-values do not carry the same confirmatory force as a prespecified primary test when the multiplicity plan is unknown. This distinction is central to reading a registry record carefully.

Finally, the planning illustrations show how sample-size needs depend on the effect a future study aims to detect. They are not a retrospective measure of power and do not reveal the original design rationale. Together, these features make the record a useful example of interpreting estimates while keeping design, analysis, and evidential limits separate.

18. 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 primary result is based on completers rather than a population explicitly described as all randomized participants. Participants had to meet the stated therapy-duration and on-treatment measurement criteria, so the estimate may not represent those who did not complete that course. The registry record does not report the numbers excluded from the analysis or reasons for exclusion.

The record gives the sponsor’s ANCOVA method but not its covariates, model specification, diagnostics, or missing-data approach. It does not report sensitivity analyses that might show whether the result changes under alternative assumptions. The registry also does not provide the original sample-size justification or the full prespecified statistical analysis plan.

Although secondary lipid results are posted, a multiplicity strategy is not reported. Their unadjusted p-values should therefore be interpreted cautiously. Several intervention descriptions and arms appear across Part 1 and Part 2, while the primary analysis compares only combination therapy with niacin; the registry summary does not provide enough procedural detail to reconstruct every comparison and treatment sequence.

The adverse-event summaries list affected counts by group, but the numbers at risk are not reported in the record, and deaths are not given as counts. Those summaries therefore cannot be converted into rates or used to make a complete comparative safety assessment from this record alone. Finally, lipid changes are surrogate measurements rather than direct reports of clinical events, and the record does not supply a clinical threshold for judging the importance of the estimated difference.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: To Evaluate Ezetimibe/Simvastatin and Niacin (Extended Release Tablet) in Patients With Type IIa or Type IIb Hyperlipidemia (0653A-091)(COMPLETED) — NCT00271817.
• Registry record last updated 2024-05-16; status last verified 2022-02.
• Results first posted 2009-08-03.
• 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.

22. Record Summary

NCT00271817 is a completed, randomized, double-blind, parallel-group Phase 3 study sponsored by Organon and Co. The registry lists enrollment of 1220, a start date of 2005-12, and primary completion in 2008-02.

The primary analysis compared ezetimibe/simvastatin plus extended-release niacin with niacin alone for mean percent change from baseline in LDL-C at 24 weeks. In the completers population, the reported ANCOVA mean difference was -38.4, with a 95% confidence interval from -41.4 to -35.4 and p <0.001. Secondary lipid outcomes show a broadly consistent direction, though their comparisons and follow-up differ.

The registry record does not report the full analysis plan, multiplicity handling, or missing-data methods. The primary result is specific to the stated completers analysis and should not be generalized beyond its endpoint and population without further evidence.