Pioglitazone vs Placebo in Combination With Niacin Extended Release on Low HDL: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pioglitazone vs Placebo in Combination With Niacin Extended Release on Low HDL (NCT00300365), a randomized phase 2 study. The primary outcome is Mean Increase in High Density Lipoprotein Cholesterol (HDL-C) at Baseline and 12 Weeks.

2. Clinical Question

The comparison described by the trial arms is the addition of pioglitazone versus the addition of placebo while both groups receive open-label niacin ER and aspirin. The brief summary also states a hypothesis comparing combination treatment with niacin ER monotherapy; however, the arm descriptions presented in the registry record identify placebo plus niacin and aspirin as the comparator. These descriptions should not be treated as interchangeable.

The clinical question is therefore about the difference in mean HDL-C change between the randomized groups, not about a reported effect estimate. The registry record supplies no group means, between-group estimate, confidence interval, or p-value for the primary endpoint. It is not possible to determine the magnitude or precision of the observed difference from the record.

3. Trial Design

This is a randomized, phase 2, parallel-group trial with quadruple masking, according to the registry record. The listed enrollment is 38, and the status is completed. The sponsor is the University of Pennsylvania. The record identifies metabolic syndrome as the condition and describes the participants as non-diabetic people with low HDL-C and metabolic syndrome.

Both arms received open-label niacin ER and aspirin. In the active arm, pioglitazone was initially 30 mg and then titrated to 45 mg; the arm description also lists niacin ER 2.0 g/day and aspirin 325 mg/day. The comparator received pioglitazone placebo with niacin ER 2.0 g/day and aspirin 325 mg/day. The primary outcome description says niacin ER was titrated to 2 grams daily over 4 weeks and aspirin was taken 30 minutes before niacin ER. It describes adding blinded pioglitazone or placebo after baseline.

The endpoint is described as the mean increase in HDL-C from baseline, designated week -4, to 12 weeks after randomization. The intervention description also gives a pioglitazone schedule of 30 mg/day for 6 weeks followed by 45 mg/day for 6 weeks. The registry record does not explain how this schedule relates to the stated 12-week post-randomization endpoint and the niacin titration period.

The registry lists one outcome measure and one analysis. It does not report further details about the randomization procedure, allocation concealment, the masking of particular roles, or the statistical analysis plan. Those details are not reported in the registry record and should not be inferred from the design labels alone.

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: 4 weeks prior to randomization, subjects titrated to 2.0 g/day of niacin extended release. Subjects who could not tolerate niacin were excluded from the study.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All subjects for whom HDL-C measurements were recorded at baseline (week -4) and 12 weeks post randomization to placebo, niacin ER, and aspirin or pioglitazone, niacin ER, and aspirin

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 Mean Increase in High Density Lipoprotein Cholesterol (HDL-C) at Baseline and 12 Weeks, measured over Baseline, after 12 weeks of pioglitazone vs placebo.

The primary endpoint is the mean increase in high-density lipoprotein cholesterol, or HDL-C, measured in mg/dL. The record defines the change as the HDL-C value at 12 weeks after randomization minus the baseline value at week -4. It describes measurement at baseline and at the 12-week follow-up.

A continuous endpoint such as mean change summarizes the average change in a laboratory measurement across participants with recorded measurements. Comparing the mean changes between randomized groups addresses whether the average change differed between assignment to pioglitazone and assignment to placebo, against the shared niacin ER and aspirin background. This endpoint is the outcome named in the trial record; it does not by itself describe every individual participant's change.

HDL-C is used as the measured outcome because the trial's stated objective concerns increasing HDL-C in people with low HDL-C and metabolic syndrome. That explains the endpoint's role in the trial question, but a change in this measure alone does not establish effects on other health outcomes. The record does not report numerical group means, a between-group mean difference, or a confidence interval for the endpoint. It therefore does not establish the size or precision of any observed change.

The primary outcome population is described as all subjects for whom HDL-C measurements were recorded at both baseline and 12 weeks post-randomization. The record does not give the number with measurements at both time points or explain how missing measurements were handled.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: the mean increase in HDL-C, measured in mg/dL, from baseline to the 12-week assessment. The registry record identifies a 95% confidence interval as the dispersion measure, but gives no numerical endpoint estimate or interval. The descriptions below explain general statistical concepts; they do not establish which specific methods were used in this trial.

What does a mean change represent?

For each participant with measurements at both time points, a change score describes the difference between the follow-up value and the baseline value. The group mean change summarizes those individual changes. Comparing group mean changes focuses on the average difference between the randomized groups. The mean does not show the full distribution of individual responses: participants may have changes that differ from the group average.

The trial record names a mean increase as its primary outcome and describes the measurement times. It does not report the group-specific means or the between-group contrast. Without those results, it is not possible to assess the observed direction or size of the average difference.

What does a confidence interval convey?

In general, a confidence interval gives a range of estimates compatible with the observed data under a specified statistical procedure. It conveys uncertainty about an estimate. An interval that is narrow indicates greater precision than a wide interval, while neither width nor location alone provides all the information needed to assess a study. The registry identifies 95% confidence intervals as the endpoint's dispersion measure but does not report their numerical limits.

A confidence interval for each group's mean is not the same as a confidence interval for the contrast between groups. The latter directly addresses the randomized comparison. The registry record does not state which interval was intended for the primary comparison or provide an interval for it.

Why does the randomized comparison matter?

Random assignment is intended, in general, to make the groups comparable on average at the start of a trial. A comparison by randomized assignment can then estimate the effect of assignment to one study strategy rather than the other, subject to the trial's conduct and analysis. Here, both described arms receive niacin ER and aspirin, while the randomized difference is pioglitazone versus placebo. The result would therefore be interpreted against this shared background regimen, not as a comparison of pioglitazone alone with no treatment.

The record calls the trial randomized and parallel-group. It does not describe the random sequence generation, allocation concealment, or a specific analysis based on randomized assignment. These details are not reported in the registry record.

What is over-dispersion, and why can it matter?

In general, over-dispersion means that the observed variability is greater than a statistical model assumes. If a model understates variability, uncertainty estimates can be too small and statistical tests can appear more decisive than warranted. The relevance of over-dispersion depends on the outcome and model. For a continuous endpoint, analysts consider how the outcome varies and whether the chosen model represents that variation adequately. The registry record does not report a model or an assessment of over-dispersion, so no claim can be made about whether it arose in this trial.

What can be said about planning calculations?

The registry's description of sample-size assumptions refers to a planning illustration based on an observed effect, two-sided significance level 0.05, equal allocation, and a normal approximation. It says that the effect to detect is a stated fraction of the observed effect. These assumptions describe how planning illustrations are constructed; they do not establish the trial's original design assumptions. The registry record does not report the trial's own sample-size justification or numerical planning outputs.

In general, a sample-size calculation estimates the amount of information needed to detect a specified effect under chosen assumptions. A post hoc calculation using an observed effect is only an illustration of those assumptions. It is not observed power, and it should not be treated as evidence that the trial did or did not detect a meaningful effect. Because the primary numerical result is not reported, a quantitative interpretation of any effect-based planning illustration is also unavailable.

What is not specified?

The registry record does not describe the primary analysis model, covariate adjustment, handling of missing measurements, or whether the analysis used all randomized participants. It lists one analysis, but that count does not supply the details of the method. The statistical plan, including any interim analyses, is not reported in the registry record. These omissions limit the ability to reproduce or fully evaluate the primary analysis.

10. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes, time points, comparisons, or statistical tests are considered. In general, testing many hypotheses can increase the chance of at least one apparently notable finding by chance. Prespecified priorities and multiplicity-control methods can help readers interpret a family of tests. The registry record lists one outcome measure and one analysis, but its skip note indicates that the primary outcome has several categories or time points and was not reduced to one comparison. The record does not explain how these categories or time points relate to the listed analysis or whether any multiplicity adjustment was planned or used.

The primary endpoint description refers to baseline and 12 weeks post-randomization, and the primary outcome population requires HDL-C measurements at both times. The number of participants with complete measurements is not reported in the registry record. Nor does the record explain whether missing measurements were excluded, imputed, or handled by another approach. The analysis population and its relationship to the randomized groups are therefore unclear.

The record does not report a statistical analysis plan, prespecified rules for handling missing data, or interim analyses. These items should not be inferred from the phase, masking, or design labels. Without them, readers cannot determine from the registry alone whether all planned comparisons were reported or how incomplete follow-up affected the endpoint analysis.

11. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hospitalization for shortness of breath (2.6%), small intestinal bleed (2.6%), surgery for back (2.6%), hospitalization for depression (2.5%), rectal cancer (2.5%).

The registry record reports serious adverse events for two groups. In the placebo plus open-label niacin and aspirin group, 3 participants are listed as affected, with 38 at risk. In the active pioglitazone plus open-label niacin and aspirin group, 2 are listed as affected, with 40 at risk. These are the recorded counts and denominators; the registry record does not explain why the at-risk counts differ from one another or how they relate to the listed enrollment of 38.

For other adverse events, the corresponding affected counts are 16 of 38 in the placebo group and 15 of 40 in the active group. These categories should be kept distinct from the serious adverse event counts. The record lists serious-event terms including hospitalization for shortness of breath, small intestinal bleed, surgery for back, hospitalization for depression, and rectal cancer, with percentages of 2.6%, 2.6%, 2.6%, 2.5%, and 2.5%, respectively. It does not give further context for interpreting these term-level figures.

Deaths are not reported in the registry record. The safety counts describe events recorded during the trial, but they do not by themselves establish that an event was caused by a study intervention. Differences in counts should also be interpreted cautiously given the unexplained denominators and the absence of further safety-analysis details.

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

The registry record defines a randomized comparison of pioglitazone versus placebo added to open-label niacin ER and aspirin, with mean HDL-C increase as the primary endpoint. It also states a hypothesis about a 12% greater increase with combination treatment than with niacin ER monotherapy. These details describe the trial question and its intended comparison; they are not evidence of the observed treatment effect.

The numerical primary outcome results are not reported in the registry record. No mean change by group, between-group contrast, confidence interval, or p-value is available. Accordingly, the record does not establish how large the observed difference was, which group had the larger mean increase, or how precisely the difference was estimated. It cannot support a conclusion about statistical significance or the practical importance of a difference.

Even if an HDL-C difference were reported, the endpoint would describe a change in a laboratory measurement over the specified period. It would not on its own establish effects on other outcomes. The record's stated question is about HDL-C change in non-diabetic patients with metabolic syndrome and low HDL-C, under the described treatment regimens. Broader conclusions would require evidence addressing those outcomes and populations.

The safety record supplies event counts and at-risk denominators, but the denominators are not fully reconciled with each other or with the enrollment figure. The listed serious-event counts should therefore be read as registry entries, not as a complete account of the comparative safety profile. Deaths are not reported in the registry record.

Overall, the record allows readers to identify the design, regimens, endpoint, and some safety counts. It does not provide the numerical efficacy results or enough analysis detail to evaluate the primary comparison. The trial's stated hypothesis should remain separate from any claim about what the posted results show.

13. Why This Trial Matters Statistically

This trial illustrates an important distinction between a registered research question and an interpretable numerical result. The registry record identifies a primary endpoint and states a hypothesis, but does not report the mean HDL-C changes or the comparison between randomized groups. A clear endpoint definition is useful, yet it cannot substitute for an estimate and a measure of its uncertainty.

The arm descriptions also make the comparison's background treatment important. Both groups received open-label niacin ER and aspirin, while pioglitazone or placebo was added. Interpreting the intended contrast requires keeping that shared regimen in view. The brief summary's hypothesis refers to niacin ER monotherapy, whereas the described comparator arm includes aspirin as well as niacin ER. The record does not resolve that difference in wording.

The endpoint is continuous, and a mean-change comparison would ordinarily be accompanied by an estimate of uncertainty. A confidence interval helps readers assess precision rather than relying only on whether a test crosses a significance threshold. Here, the record names a 95% confidence interval as the dispersion measure but supplies no numerical interval for the primary outcome.

The record also demonstrates why analysis details matter. The number of participants with measurements at both endpoint times, the handling of missing data, and the statistical model are not reported. Safety denominators differ between groups and are not explained. These uncertainties limit what can be inferred from the design and counts alone.

Statistically, the main lesson is that a trial's hypothesis, planned endpoint, and numerical findings are distinct pieces of evidence. Without the estimate and its uncertainty, neither the direction nor precision of the primary comparison can be assessed from the registry record.

14. 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 most important limitation for interpreting the primary endpoint is that numerical results are not reported in the registry record. Although the outcome is defined as mean HDL-C increase from baseline to 12 weeks and the dispersion measure is identified as a 95% confidence interval, no group values, treatment contrast, interval bounds, or p-value are given.

The record's descriptions of the hypothesis and comparator are not fully aligned. The brief summary states a hypothesis comparing combination treatment with niacin ER monotherapy, while the arm descriptions identify placebo plus open-label niacin ER and aspirin as the comparator. The record does not clarify this difference. It also describes a pioglitazone titration schedule and a 12-week endpoint without explaining how those timing descriptions fit together.

The listed enrollment is 38, but the safety data give at-risk counts of 38 and 40 across the two groups. The record does not explain the difference or specify the number included in the primary analysis. It also does not report how missing HDL-C measurements were handled or how many participants had measurements at both baseline and follow-up.

Further details about randomization, allocation concealment, the statistical model, covariate adjustment, interim analyses, and the statistical analysis plan are not reported in the registry record. The record indicates one analysis, but this does not provide enough information to reproduce it. These limitations restrict evaluation of both the primary comparison and the safety summaries.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Pioglitazone vs Placebo in Combination With Niacin Extended Release on Low HDL — NCT00300365.
• Registry record last updated 2017-07-28; status last verified 2017-06.
• Results first posted 2017-07-28.
• 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.

18. Record Summary

The registry record describes a completed randomized, quadruple-masked, parallel-group phase 2 trial sponsored by the University of Pennsylvania, with listed enrollment of 38. It studied non-diabetic patients with metabolic syndrome and low HDL-C.

The active arm received pioglitazone with open-label niacin ER and aspirin; the comparator received placebo with open-label niacin ER and aspirin. The primary endpoint was mean HDL-C increase from baseline, week -4, to 12 weeks after randomization, measured in mg/dL. The stated hypothesis concerned a 12% greater increase with combination treatment than with niacin ER monotherapy.

Numerical primary outcome results and key analysis details are not reported in the registry record. Safety counts are listed, but the group denominators are not explained. The record therefore supports a description of the trial design and endpoint, but not a quantitative assessment of its primary comparison.