Study to Compare the Efficacy of Pitavastatin With That of Atorvastatin in Lowering Cholesterol Levels: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study to Compare the Efficacy of Pitavastatin With That of Atorvastatin in Lowering Cholesterol Levels (NCT00249249), a randomized phase 3 study. The primary outcome is Percent Change From Baseline Low Density Lipoprotein-cholesterol (LDL-C) at Week 12.

2. Clinical Question

The clinical question is about a measured change in LDL-C over the stated follow-up, not about long-term clinical events. The primary endpoint is a continuous laboratory measure summarized as a mean. Its reported population is defined by having a Week 12 measurement, rather than by requiring that participants remained on drug or had no protocol violations.

The study also included comparisons involving pitavastatin 4 mg QD and atorvastatin 20 mg QD. In the posted pairwise results, each of those groups is contrasted with pitavastatin 2 mg QD. The registry record does not report a direct pairwise comparison between pitavastatin 4 mg QD and atorvastatin 20 mg QD, or between pitavastatin 2 mg QD and atorvastatin 10 mg QD as a formal sponsor analysis. The comparisons discussed in the statistical interpretation are unadjusted re-calculations from the posted summary statistics.

The registry record describes the purpose as comparing the efficacy of pitavastatin with that of atorvastatin. The results can address the observed Week 12 LDL-C changes in the reported groups, but the record does not report enough detail to infer how dose selection, titration, or other treatment decisions were handled during the study.

3. Trial Design

This was a randomized, double-masked, parallel-group phase 3 trial sponsored by Kowa Research Europe. The registry record lists an enrollment of 830 and marks the study as completed. It identifies primary hypercholesterolemia and dyslipidemia as the conditions under study.

The four posted treatment groups were pitavastatin 2 mg QD, atorvastatin 10 mg QD, pitavastatin 4 mg QD, and atorvastatin 20 mg QD. The arm-level primary-outcome summaries report 301.0 participants for pitavastatin 2 mg QD, 98.0 for atorvastatin 10 mg QD, 288.0 for pitavastatin 4 mg QD, and 100.0 for atorvastatin 20 mg QD. These are the numbers attached to the posted outcome summaries; they should not be treated as a description of all enrolled participants or of every analysis population.

Randomized allocation means treatment assignment was by chance, while double masking indicates that the record classifies the study as double-masked. The registry record does not report the randomization procedure or allocation-concealment method. It also does not specify who was masked, how masking was maintained, or whether masking was assessed.

The official title describes the comparisons as following titration, but the registry record does not report the titration schedule or other detailed treatment procedures. The primary endpoint was measured from baseline to 12 weeks. The registry record does not report the original sample-size justification, the planned statistical analysis, the planned number of interim analyses, or additional detail on the study calendar. These missing protocol details limit what can be inferred about the design beyond the characteristics explicitly recorded.

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 with major protocol violations were identified programmatically prior to unblinding to determine who should be excluded from the per protocol population.

Overall Study

Analysis population for the primary outcome, as defined in the registry: All patients, irrespective of protocol violations, who had Week 12 measurements, whether on drug or not.

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 Low Density Lipoprotein-cholesterol (LDL-C) at Week 12, measured over Baseline to 12 weeks.

The primary endpoint is percent change from baseline in low-density lipoprotein cholesterol (LDL-C) at Week 12. It summarizes the change in LDL-C relative to each participant’s baseline measurement, and the posted group result is the mean percent change. A negative mean indicates that the measured value at follow-up was lower than baseline on average; the size of that mean describes the average change, not how every individual participant changed.

LDL-C is used as a laboratory measure relevant to the trial’s cholesterol-lowering question. Expressing change as a percentage places the change in relation to baseline rather than reporting only an absolute concentration change. The endpoint is continuous, so the record reports a mean and standard deviation for each group. The standard deviation describes variation among the measurements within a group; it is not the uncertainty interval for the difference between groups.

The stated timeframe is baseline to 12 weeks. The outcome population includes all patients, irrespective of protocol violations, who had Week 12 measurements, whether on drug or not. This definition makes the endpoint broader than an analysis restricted to participants who remained on treatment, but it also means the primary summaries depend on having a Week 12 measurement. The registry record does not report how missing Week 12 measurements were handled or whether any imputation was used.

The endpoint captures a short-term laboratory change. It is not itself a measure of symptoms, cardiovascular events, or long-term outcomes. The registry record does not establish that a particular size of LDL-C change corresponds to a specific clinical benefit for an individual.

7. Results Posted in the Registry

• Atorvastatin 10 mg QD: mean -38.76 versus -37.89 on control; difference -0.87 (95% CI -4.15 to 2.41), standardized effect d = -0.06; z-test p = 0.604.
• Pitavastatin 4 mg QD: mean -45.50 versus -37.89 on control; difference -7.61 (95% CI -9.86 to -5.36), standardized effect d = -0.55; z-test p = <0.001.
• Atorvastatin 20 mg QD: mean -43.90 versus -37.89 on control; difference -6.01 (95% CI -9.49 to -2.53), standardized effect d = -0.42; 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 Total Cholesterol (TC)

• 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:
  • Atorvastatin 10 mg QD: mean -28.92 versus -27.57 on control; difference -1.35 (95% CI -3.90 to 1.20), standardized effect d = -0.13; z-test p = 0.299.
  • Pitavastatin 4 mg QD: mean -33.09 versus -27.57 on control; difference -5.52 (95% CI -7.22 to -3.82), standardized effect d = -0.52; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Atorvastatin 10 mg QD and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Pitavastatin 4 mg QD 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 deviations.
• Result:
  • Atorvastatin 10 mg QD: mean 3.42 versus 4.28 on control; difference -0.86 (95% CI -4.68 to 2.96), standardized effect d = -0.05; z-test p = 0.659.
  • Pitavastatin 4 mg QD: mean 4.93 versus 4.28 on control; difference 0.65 (95% CI -2.03 to 3.33), standardized effect d = 0.04; z-test p = 0.634.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Atorvastatin 10 mg QD 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 Pitavastatin 4 mg QD and the control. The interval gives the range of effects that remain compatible with the data.

TC:HDL-C Ratio

• 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:
  • Atorvastatin 10 mg QD: mean 3.76 versus 4.02 on control; difference -0.25 (95% CI -0.47 to -0.04), standardized effect d = -0.24; z-test p = 0.019.
  • Pitavastatin 4 mg QD: mean 3.58 versus 4.02 on control; difference -0.43 (95% CI -0.61 to -0.26), standardized effect d = -0.39; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Atorvastatin 10 mg QD 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 lower for Pitavastatin 4 mg QD than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary summaries broadly extend the primary laboratory question to other lipid measures. Total cholesterol changes generally follow the same broad pattern as the primary LDL-C results: the higher-dose pitavastatin group has a more negative mean change than the reference, while the atorvastatin 10 mg comparison with the reference is less distinct. The total cholesterol comparison with atorvastatin 20 mg is not among the posted pairwise re-calculations. The HDL-C summaries do not show a clear separation in the posted unadjusted comparisons. The TC:HDL-C ratio summaries show lower means in the comparator groups than in the reference group, with unadjusted comparisons reported for both listed contrasts.

These measures are related to the primary endpoint but are not interchangeable with it. LDL-C percent change is the primary outcome; other cholesterol components and ratios provide additional descriptions of lipid measurements. Similar patterns across related measures can provide context, but they do not independently establish the primary finding or clinical importance. The registry record also names triglycerides, non-HDL:HDL ratio, apolipoprotein B, apolipoprotein-A1, and their ratio among secondary outcomes. Detailed posted results for those outcomes are not reported in the registry record.

Secondary results generally carry less weight than a prespecified primary result because there may be more outcomes and comparisons to examine. Without a reported multiplicity plan, a p-value from one secondary comparison does not account for the full set of questions that may have been considered. The record does not state which secondary outcomes were prespecified for formal testing, whether testing was hierarchical, or whether any adjustment was used. Therefore, secondary patterns are best treated as supportive description rather than as a set of independent confirmatory conclusions.

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

• Percent Change From Baseline in Total Cholesterol (TC)
• Percent Change From Baseline in High Density Lipoprotein Cholesterol (HDL-C)
• TC:HDL-C Ratio
• Triglycerides (TG)
• Non-HDL:HDL Ratio
• Apolipoprotein B (Apo B)
• Apolipoprotein-A1 (Apo-A1)
• Apo-B:Apo-A1 Ratio

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: each participant has a percent change measurement, and the group summaries report the mean and standard deviation. The mean gives a central summary of the measurements. The standard deviation describes how dispersed those measurements are around the mean. A larger standard deviation means more variability within the group; it does not by itself indicate more uncertainty about the difference between group means.

What does a mean difference represent?

A mean difference contrasts the average percent changes in two groups. The displayed contrasts use pitavastatin 2 mg QD as the reference. For example, the atorvastatin 10 mg QD estimate of -0.87 is the difference between the two reported group means, with the comparator’s mean minus the reference mean. A negative estimate indicates a more negative average change in the comparator group. This is a difference in percentage-change units, not a relative percent difference between treatments.

The comparisons are labelled as unadjusted re-calculations from posted summary statistics. They do not account for covariates or other design features beyond the information reflected in those summaries. The registry record does not report the sponsor’s statistical model, covariate adjustments, or a formal primary hypothesis test. The record lists zero reported analyses, so these re-calculations should not be mistaken for a sponsor-reported analysis.

How should the p-values be read?

A p-value describes how incompatible the observed comparison is with a specified null model, under the assumptions of the statistical calculation. It is not the probability that the null hypothesis is true, and it does not describe the size or clinical importance of a difference. The unadjusted p-values are 0.604 for atorvastatin 10 mg QD, and <0.001 for both pitavastatin 4 mg QD and atorvastatin 20 mg QD, each compared with pitavastatin 2 mg QD.

These are unadjusted results: the record does not report a correction for making multiple comparisons or for evaluating multiple outcomes. When many comparisons are considered, the chance of seeing at least one apparently unusual result can increase. A multiplicity plan determines how results are interpreted across that family of tests; no such plan is reported in the registry record.

What does a confidence interval add?

A confidence interval gives a range of estimates compatible with the observed data under the calculation’s assumptions. It makes the uncertainty around the mean difference more visible than a single estimate does. The interval for atorvastatin 10 mg QD includes zero, while the intervals for pitavastatin 4 mg QD and atorvastatin 20 mg QD lie on the negative side. An interval that includes zero does not prove there is no difference; it indicates that the estimates compatible with the data include no difference as well as differences in either direction.

Intervals and p-values describe statistical uncertainty, not whether a difference matters to patients. Their interpretation also depends on the analysis being appropriate and on the endpoint data being adequately represented. The registry record does not provide details about distributional checks, handling of outliers, missing observations, or the procedure used for the sponsor’s original analysis.

Why does over-dispersion matter in general?

In general, over-dispersion means the observed variation is greater than a model expects. If a model assumes less variation than the data contain, its standard errors can be too small and its intervals too narrow, giving an overly confident impression. For a continuous outcome such as this one, the reported standard deviations give information about within-group spread, but the registry record does not report a specific over-dispersion assessment or a model that would allow one to evaluate it.

The displayed sample-size planning illustrations are separate from the observed-outcome comparisons. They use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect set to a stated fraction of the observed effect. Those assumptions describe the planning illustrations only; the registry record does not report the original trial’s own design assumptions or sample-size justification.

11. Confidence Intervals and Precision

Precision concerns how narrowly the data locate a group difference, not whether the result is important. The posted mean differences and their intervals convey both the estimated direction and the range of uncertainty. For atorvastatin 10 mg QD versus pitavastatin 2 mg QD, the estimate is -0.87 and the interval runs from -4.153 to 2.413. Because this range crosses zero, the posted summary statistics are compatible with a small difference in either direction as well as no difference.

For pitavastatin 4 mg QD, the estimated difference is -7.61, with an interval from -9.86 to -5.36. For atorvastatin 20 mg QD, the estimate is -6.01, with an interval from -9.488 to -2.532. Both intervals are entirely negative, indicating that the unadjusted estimates consistently favor a more negative mean change in the comparator group relative to pitavastatin 2 mg QD. The interval widths still matter: they show that the precise size of the difference is uncertain.

The intervals are based on unadjusted re-calculations from the posted summaries. The registry record does not identify a confidence level, so one should not attach a stated coverage percentage to these intervals based on this record alone. Nor does it report intervals for all outcomes or a multiplicity-adjusted interval procedure.

Statistical precision is affected by both the amount of information in the comparison and variation in the outcome. The group standard deviations indicate within-group variability, while the comparison intervals represent uncertainty about mean differences. These are related but not interchangeable quantities. The registry record’s reported endpoint population requires a Week 12 measurement; without details on missing measurements, the effect of missingness on precision and representativeness cannot be assessed.

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 for these continuous outcomes are expressed as participants per arm, not as a total across arms and not as total events. They are illustrations based on the observed effect, not the trial’s original design calculations. 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 under those assumptions for a planning scenario designed to have an 80% chance of detecting the selected effect if that effect is the true one. The 90% power column gives the corresponding planning figure for a 90% chance. Power is conditional on the assumptions and effect used; it is not a probability that a particular trial result is correct.

Rows planning for 100% of the observed effect use the full observed estimate as the target for the illustration. Rows planning for 75% or 50% of the observed effect use smaller target effects. Smaller effects are harder to distinguish from random variation, so they require more participants per arm under the same assumptions. The contrast is especially relevant when an observed estimate is small: a planning exercise based on that estimate can yield very large participant requirements, and those figures should not be interpreted as a recommendation or as evidence about the actual design of this trial.

This is not observed power. It is a forward-looking planning illustration calculated from the observed effect and stated assumptions. The registry record does not report the trial’s own sample-size justification, original target power, or original design assumptions.

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 = -0.870; pooled standard deviation σ = 14.026.
2. n = 2 × (1.96 + 0.842)² × 14.026² / -0.870² = 4,079.9, rounded up to 4,080 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 <- -0.869999999999997                      # observed difference in means
sigma <- 14.025664900074                     # 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(249249)
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 = 4,080 (80% power), 5,462 (90% power)
 75% of observed effect: n per arm = 7,254 (80% power), 9,710 (90% power)
 50% of observed effect: n per arm = 16,320 (80% power), 21,848 (90% power)
power.t.test: 4080.8 
Simulated power at n = 4,080 per arm: 0.802 (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 record lists no reported analyses, and it does not describe a statistical analysis plan. In particular, it does not report whether the primary comparisons were prespecified as separate hypotheses, whether secondary outcomes were tested formally, or whether a multiplicity adjustment was planned. The posted p-values for the pairwise re-calculations are unadjusted. When more than one comparison or outcome is examined, unadjusted p-values do not account for the wider family of tests; this can increase the chance of at least one apparently strong result arising by chance.

The primary-outcome population is described as all patients with Week 12 measurements, regardless of protocol violations and whether on drug or not. This specifies an outcome population, but it does not fully describe the analysis population or the handling of missing data. The registry record does not report how many participants lacked Week 12 measurements, why measurements were unavailable, whether any values were imputed, or whether a sensitivity analysis assessed missing-data assumptions. It also does not report whether all randomized participants were included in the analysis or how protocol violations were evaluated.

Missing data can affect both the estimated group means and their uncertainty, especially if the likelihood of having a measurement is related to treatment or outcome. The posted group summaries cannot establish whether that concern applies. The record also does not report how adherence, treatment discontinuation, or changes in treatment were incorporated into analysis beyond the statement that the endpoint population includes patients whether on drug or not.

The record’s silence on these matters is not evidence that no multiplicity or missing-data methods were used in the full study documentation. It means those methods are not reported in the registry record, so the posted summaries cannot be judged against a prespecified analysis strategy.

15. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: angina unstable (1.0%), angina pectoris (1.0%), deafness neurosensory (1.0%), encephalopathy (1.0%), myocardial infarction (0.3%).

The registry record lists serious adverse events and other adverse events by treatment group. For serious adverse events, the listed at-risk and affected counts are 316 and 3 for pitavastatin 2 mg QD, 102 and 1 for atorvastatin 10 mg QD, 300 and 1 for pitavastatin 4 mg QD, and 103 and 2 for atorvastatin 20 mg QD. These are registry counts; they do not by themselves establish that an event was caused by treatment.

For other adverse events, the listed at-risk and affected counts are 316 and 22 for pitavastatin 2 mg QD, 102 and 6 for atorvastatin 10 mg QD, 300 and 17 for pitavastatin 4 mg QD, and 103 and 4 for atorvastatin 20 mg QD. The registry record lists serious adverse event terms including unstable angina, angina pectoris, neurosensory deafness, encephalopathy, myocardial infarction, and generalized pruritus, with the recorded percentages ranging from 0.3% to 1.0% across those listed terms. The term list does not provide enough detail to attribute each term to a particular treatment group.

Deaths are not reported in the registry record. The counts should be read alongside the stated at-risk groups, not as proof of comparative safety. The registry record does not provide event timing, duration of follow-up for safety, adjudication methods, or a causal assessment.

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

The posted primary results describe mean percent change in LDL-C from baseline to Week 12 in four treatment groups. The means are negative in every group, and the estimates for pitavastatin 4 mg QD and atorvastatin 20 mg QD are more negative than the estimate for pitavastatin 2 mg QD. Their unadjusted re-calculated comparisons have intervals entirely below zero. The atorvastatin 10 mg QD comparison has an interval that includes zero.

Those observations do not establish that one treatment is clinically better for every participant. Group means do not show the full range of individual responses, and the registry record does not report a threshold for a clinically important difference. An interval including zero is not proof of equivalence, while an interval that excludes zero does not by itself establish a meaningful clinical advantage. The record reports no equivalence or non-inferiority margin.

The endpoint is a short-term laboratory measure, not a clinical event outcome. The findings do not establish effects on cardiovascular events, symptoms, survival, or long-term health. Nor do these summaries answer what would happen outside the trial population or under treatment approaches not described in the record.

The design is described as randomized, double-masked, and parallel-group, but the registry record does not report detailed methods for randomization, masking, missing-data handling, or the sponsor’s statistical analysis. The comparison estimates are unadjusted re-calculations from arm summaries. They are useful for understanding the direction and approximate precision of the posted group contrasts, but they do not replace a full prespecified analysis.

The results therefore support a limited statistical reading: the groups differed in their observed mean Week 12 changes to varying degrees, with uncertainty reflected by the reported intervals. They do not establish causation beyond the strength of the recorded design and analysis details, and they should not be translated into individualized medical conclusions.

17. Why This Trial Matters Statistically

This trial illustrates why a comparison needs more than a single group mean. The mean difference communicates the estimated contrast, while the confidence interval communicates how precisely that contrast is located. Here, the atorvastatin 10 mg comparison has an interval spanning zero, whereas the comparisons involving pitavastatin 4 mg and atorvastatin 20 mg have intervals on the negative side. The distinction is important: the point estimate alone would not show the uncertainty around the result.

It also illustrates that a statistically distinct result and a clinically important result are different questions. The registry record supplies estimates and unadjusted p-values but does not state a clinically important difference threshold. Statistical evidence should therefore be interpreted as evidence about the measured group contrast, not as a direct judgment about patient benefit.

The outcome population is another useful design detail. Week 12 measurements were included regardless of protocol violations and whether the patient was on drug or not, but the record does not say how absent measurements were addressed. Knowing the population definition helps readers understand what the means represent; knowing missing-data methods would be needed to assess how fully they represent the randomized groups.

Finally, the secondary outcome pattern gives context without replacing the primary outcome. Related lipid measures may point in similar or different directions, but multiple unadjusted tests require caution. The absence of a reported analysis plan and multiplicity approach limits confirmatory interpretation. These are practical lessons for reading registry results: distinguish observed summaries from prespecified analyses, estimates from uncertainty, and statistical contrasts from clinical conclusions.

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.

Several limitations are specific to the information recorded for this trial. First, the primary endpoint population requires a Week 12 measurement, but the registry record does not state how many measurements were missing or how missingness was handled. The effect of that restriction on the reported group means cannot be evaluated from the record.

Second, the registry record does not report the sponsor’s analyses or a statistical analysis plan. The pairwise contrasts discussed are unadjusted re-calculations from posted means, standard deviations, and group sizes. They do not reveal whether the original analysis used covariate adjustment, a different estimand, or a specified multiplicity strategy. The record’s discrepancy indicator is false, but no sponsor analysis is listed for direct comparison.

Third, the dose-group summaries do not provide direct pairwise results for every possible treatment contrast. In particular, the posted comparisons use pitavastatin 2 mg QD as the reference; they do not report a direct comparison between the two higher-dose groups. Results for several named secondary measures are also not reported in the registry record.

Fourth, the posted outcome window is baseline to 12 weeks, so these summaries do not address longer-term outcomes. The registry record does not report a clinical-importance threshold for the LDL-C difference, details of adherence or treatment changes, or event follow-up sufficient to interpret long-term safety. Deaths are not reported in the registry record.

Finally, descriptions of randomization and masking are limited to the broad design labels. The randomization procedure, allocation concealment, masking implementation, and detailed titration methods are not reported in the registry record. These omissions constrain assessment of study conduct and of how broadly the observed summaries can be interpreted.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Study to Compare the Efficacy of Pitavastatin With That of Atorvastatin in Lowering Cholesterol Levels — NCT00249249.
• Registry record last updated 2010-01-12; status last verified 2010-01.
• Results first posted 2009-12-16.
• 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

The registry record describes a completed, randomized, double-masked, parallel-group phase 3 study of pitavastatin and atorvastatin in people with primary hypercholesterolemia or combined dyslipidemia. Enrollment was 830. The primary endpoint was mean percent change in LDL-C from baseline to Week 12 among patients with Week 12 measurements, regardless of protocol violations or whether they were on drug.

The posted means were -37.89 for pitavastatin 2 mg QD, -38.76 for atorvastatin 10 mg QD, -45.5 for pitavastatin 4 mg QD, and -43.9 for atorvastatin 20 mg QD. Unadjusted re-calculated comparisons with pitavastatin 2 mg QD show different degrees of precision. The registry record lists secondary lipid outcomes and adverse event counts, but no sponsor analyses or statistical plan.

Missing-data methods, multiplicity handling, original sample-size justification, and detailed randomization and masking methods are not reported in the registry record.