Effects of Pioglitazone on High-density Lipoprotein (HDL) Function in Persons With Diabetes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effects of Pioglitazone on High-density Lipoprotein (HDL) Function in Persons With Diabetes (NCT01156597), a randomized phase 3 study. The primary outcome is Increased HDL-Cholesterol and Decreased Triglycerides.

2. Clinical Question

The trial asks whether treatment with pioglitazone is associated with changes in high-density lipoprotein cholesterol and triglycerides, and examines other aspects of HDL function in people with type 2 diabetes. Its stated rationale concerns reverse cholesterol transport: the movement of cholesterol through pathways involving HDL. The registry describes a hypothesis that pioglitazone might affect those pathways directly or through changes in HDL structure and composition.

The registry identifies increased HDL cholesterol and decreased triglycerides as the primary endpoint, expressed as percent change from baseline. It describes measurements after 12 or 24 weeks and gives 24 weeks as the endpoint timeframe. The primary endpoint has multiple categories or time points, and the registry record does not reduce it to one comparison. No primary endpoint result is reported in the registry record, so the magnitude, direction, and precision of any between-group difference cannot be assessed from that record.

3. Trial Design

This is a phase 3, randomized, parallel-group study sponsored by the University of Miami. The registry lists the study as completed. Its recorded start is 2008-04, and its primary completion is 2010-09. Enrollment was 30. The condition under study was type 2 diabetes mellitus.

The registry describes two groups: a pioglitazone group and a comparator group receiving either metformin or a sulfonylurea. The comparator description says the intent was to control blood sugar to a level comparable to the pioglitazone group. The pioglitazone arm description calls the study a baseline-versus-treatment study and says that participants were compared with a matched group. The registry also identifies the allocation as randomized and the model as parallel. It does not explain how matching was conducted or how that description relates to random allocation.

Masking is listed as none. Thus, the registry does not describe the study as blinded. It does not report further details about assignment procedures, concealment, treatment adherence, or the use of other therapies during follow-up. Those design details should not be inferred from the group descriptions.

The primary endpoint was specified as percent change in HDL cholesterol and triglycerides from baseline, with a timeframe of 24 weeks. The endpoint description refers to measurements after 12 or 24 weeks, so the primary outcome includes multiple categories or time points rather than one comparison. The registry lists HDL apolipoprotein levels at study end-point and cholesterol efflux capacity of HDL as secondary outcomes. It records three outcome measures and one analysis, but does not report the detailed analysis methods or results for those measures.

The final analysis population is described as all subjects who completed the study. The registry does not report how many participants completed, how many were assigned to each group, or how any missing outcome data were handled. The adverse-event record lists 16 participants at risk in the pioglitazone group and 14 in the comparator group; those counts should not be assumed to define the analysis population for the lipid endpoints.

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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: All subjects that completed the study were used for the final analysis

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 Increased HDL-Cholesterol and Decreased Triglycerides, measured over 24 weeks.

The primary endpoint is described as increased high-density lipoprotein cholesterol and decreased triglycerides, measured as percent change from baseline. The registry gives the calculation as 100% multiplied by the difference between the endpoint value and baseline value, divided by the baseline value. A percent-change measure expresses the change relative to a participant’s starting value rather than as an absolute change in the original measurement unit.

HDL cholesterol and triglycerides are lipid measures relevant to the trial’s stated interest in diabetic dyslipidemia and reverse cholesterol transport. The registry’s rationale is that changes in HDL levels, and potentially in HDL structure or function, might be related to cholesterol movement through this pathway. HDL apolipoprotein levels and cholesterol efflux capacity are listed as secondary outcomes intended to assess other aspects of HDL function.

The endpoint is specified as a mean with standard deviation, and its stated timeframe is 24 weeks. The description also refers to changes after 12 or 24 weeks, leaving more than one time point or category in the primary endpoint. The registry record does not reduce these to one comparison, and it does not report observed means, standard deviations, or percent changes. Therefore, the endpoint definition explains what was intended to be measured, but does not establish what changes were observed or how precisely they were estimated.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The endpoint is a continuous measure, and the registry identifies its parameter as a mean with standard deviation. In general, a mean describes the central value of a set of measurements. A standard deviation describes how spread out individual measurements are around that mean. These summaries answer different questions: the mean concerns the average level, while the standard deviation concerns variability among observations. Neither is reported for the primary endpoint in the registry record.

What does percent change from baseline represent?

The registry defines percent change using the endpoint value and baseline value. This expresses a change in relation to the starting measurement. Such a scale can make changes easier to compare across participants whose starting values differ, but it also means that the baseline value is part of the calculation. The registry does not report the individual measurements, group means, or variability needed to evaluate the resulting percent changes.

The primary endpoint description refers to both HDL cholesterol and triglycerides, as well as measurements after 12 or 24 weeks. The record therefore covers multiple outcome components or time points. In general, analyzing multiple outcomes or time points raises questions about which comparison is primary and how the possibility of chance findings across analyses is addressed. The registry does not specify a single comparison or report a multiplicity procedure. This absence does not establish that no such choices or procedures existed; they are not reported in the registry record.

How should an estimate and its uncertainty be read?

In general, a treatment-effect estimate describes the size and direction of a comparison between groups. A confidence interval conveys uncertainty around that estimate under the chosen statistical method. A narrow interval indicates greater precision than a wide interval, while an interval spanning values compatible with different interpretations may leave the result uncertain. A p-value, when reported, addresses compatibility with a specified statistical model and null hypothesis; it does not measure the size or clinical importance of an effect. The registry record reports no primary endpoint estimate, confidence interval, or p-value, so none of these interpretations can be applied to the trial’s primary results.

The registry records one analysis, but does not describe the statistical model, whether the analysis compared changes between groups, how baseline values were incorporated, or whether the reported means were adjusted. The endpoint description alone is insufficient to establish those choices. The record also does not provide standard errors, individual-level data, or other information that would support an independent calculation of uncertainty. No recalculated comparison is available to interpret.

What do the planning assumptions mean?

The registry includes assumptions for planning illustrations based on an observed effect: a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. It describes the effect to detect as a stated fraction of the observed effect. These are planning assumptions, not a report of the trial’s original design or a result of the trial. The registry explicitly distinguishes the illustration from the trial’s own design. No numerical planning sample sizes are stated in this record.

In general, a sample-size illustration estimates how many participants might be needed under stated assumptions to detect a chosen effect with a specified probability, often called power. Such an illustration depends on the assumed effect and analysis conditions. When the assumed effect is based on an observed result, it should not be treated as an independent confirmation of that result. A planning illustration is not observed power, and it cannot substitute for a reported estimate with its uncertainty. The registry does not report the original sample-size justification, so the planning assumptions should not be used to infer what informed the enrollment target.

Why can variability affect interpretation?

In general, continuous biological measures can vary considerably among participants. Greater variability makes an average group difference harder to estimate precisely for a given amount of information. Some outcomes may also show over-dispersion, meaning that variability is greater than a simple statistical model expects. In that situation, a model that does not account for the extra variability can give overly narrow uncertainty estimates. These are general statistical considerations; the registry does not report whether over-dispersion was assessed or whether it was relevant to these outcomes.

The use of a mean and standard deviation is consistent with describing a continuous measure, but it does not by itself establish that the data followed a particular distribution or that a particular test was used. The registry does not report distributional checks, transformations, robust methods, or model diagnostics. It is therefore not possible to evaluate those aspects of the analysis from the registry record.

Finally, missingness can affect the analysis of change from baseline. In general, participants without a follow-up measurement cannot contribute a complete baseline-to-endpoint change unless an analysis method handles incomplete data. Different approaches may rely on different assumptions. The registry states that subjects who completed the study were used for final analysis, but does not report the number of incomplete observations, the reasons for missingness, or any method for handling missing outcome data. The consequences for the estimated lipid changes cannot be determined from the record.

10. Multiplicity, Missing Data and Analysis Populations

The primary outcome combines HDL cholesterol and triglycerides and refers to measurements after 12 or 24 weeks. The registry notes that the primary outcome has several categories or time points and is not reduced to one comparison. This makes it important to know which result was designated as the principal test and whether the analysis accounted for multiple opportunities to find a result. The registry record does not report a multiplicity strategy, a hierarchy among outcomes, or a prespecified approach for choosing a time point.

The registry also lists HDL apolipoprotein levels at study end-point and cholesterol efflux capacity of HDL as secondary outcomes. In general, testing multiple outcomes can increase the chance of at least one apparently unusual result under a model in which there is no underlying effect. Statistical procedures can address this issue in different ways, but the registry does not report whether any were used. This does not demonstrate that multiplicity was ignored; the approach is not reported in the registry record.

The final analysis population is described as all subjects who completed the study. The registry does not state how many participants completed, how many had each measurement, or why any data may have been missing. It does not describe an intention-to-treat analysis, imputation, or another missing-data method. The number of participants at risk in the adverse-event record does not establish the number with complete primary endpoint data.

These omissions limit assessment of selection and missing-data effects. In general, when analysis is restricted to completers, results may differ from those that would have been obtained from all randomized participants if completion is related to outcomes or treatment. That is a general consideration, not evidence that such bias occurred in this study. The registry does not provide the information needed to assess the issue for this trial.

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 0% reporting threshold.

The adverse-event record lists 16 participants at risk in the pioglitazone group and 14 in the comparator group. For serious adverse events, it records 0 affected participants in each group. For other adverse events, it also records 0 affected participants in each group. These are the counts recorded for the groups, not an estimate of comparative safety.

The record does not list any serious adverse-event terms. Death counts are recorded as null, which means that a death result is not reported in the registry record; null should not be read as a reported count of zero. The adverse-event information does not explain the surveillance period, event-collection procedures, or how events were classified.

In general, a count of zero recorded events does not establish that a treatment has no risk. It describes what was recorded in the study under its reporting and observation procedures. The counts also do not provide information about event severity beyond the serious-versus-other categories, timing, or individual circumstances. With the registry’s limited event detail, the figures support only a narrow description of what was recorded, not a broad safety conclusion.

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

The record establishes that a randomized, parallel-group phase 3 study was completed to examine pioglitazone and HDL-related measures in people with type 2 diabetes. It describes a primary endpoint involving percent changes in HDL cholesterol and triglycerides, alongside secondary measures of HDL apolipoprotein levels and cholesterol efflux capacity. These details explain the trial’s focus on lipid levels and aspects of reverse cholesterol transport.

What the record does not establish is the result. It contains no reported primary endpoint values or between-group estimate, and it does not report numerical results for the secondary outcomes. The endpoint title and rationale describe intended questions; they do not show that HDL cholesterol increased, triglycerides decreased, or HDL function changed. Nor does the record allow a judgment about the size or precision of any such changes.

The listed comparator was standard treatment with metformin or a sulfonylurea, with an intent to achieve comparable blood sugar control. The registry does not report whether that control was achieved. It also does not explain the matching description or give enough detail about the analysis to establish how the comparison was made. Those limits matter when interpreting what a comparison between the groups would represent.

The adverse-event record lists no affected participants for serious or other adverse events in either group, while death counts are not reported in the registry record. These entries describe the recorded event counts but cannot establish absence of risk. The record also does not report enough about follow-up and event collection to interpret the counts beyond their stated categories.

In general, a trial’s design and rationale can be informative even when outcome results are unavailable. They clarify which questions were intended to be tested and which measurements were considered relevant. But statistical conclusions about effects require observed results and their uncertainty. The registry record therefore supports a description of the study’s purpose and recorded design, not a conclusion about whether pioglitazone changed the measured lipid or HDL-function outcomes.

13. Why This Trial Matters Statistically

This study is statistically notable for the distinction between a clearly stated scientific rationale and the information needed to evaluate an observed effect. The rationale connects lipid measures to reverse cholesterol transport and identifies HDL-related measures as relevant outcomes. Yet the registry’s outcome description spans more than one measure and time point, and it does not reduce the primary endpoint to one comparison. That makes clear endpoint specification important for interpreting any analysis.

The record also shows why a trial label or design description cannot replace outcome reporting. Randomized allocation and a parallel model are recorded, but the registry does not give the analysis model, an effect estimate, or a measure of uncertainty for the primary endpoint. Without those results, readers cannot assess whether an observed group difference was large or small, precise or uncertain, or consistent across outcome components.

The completer-based analysis description is another statistically important feature. In general, analyses that include only participants who complete follow-up can be affected if completion is related to treatment or outcomes. This is a possible source of uncertainty to consider, not a claim that bias occurred. The number of completers and the approach to missing data are not reported in the registry record, so the issue cannot be evaluated for this trial.

Finally, the planning assumptions described in the record are explicitly illustrations rather than the trial’s own design justification. Keeping planning calculations separate from observed evidence prevents a sample-size illustration from being mistaken for proof that a result is reliable or likely to replicate. Overall, the record is useful for understanding the intended question and several limitations on interpretation, while leaving the central outcome comparison unavailable.

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 registry’s primary endpoint description combines HDL cholesterol and triglycerides and refers to more than one follow-up time point. It does not identify one comparison for interpretation, and no primary endpoint results are reported. This prevents assessment of the direction, size, or precision of the lipid changes.

Although allocation is recorded as randomized, the arm descriptions include a matched group and do not explain how matching relates to randomization. The record does not report assignment procedures, concealment, adherence, or the statistical model. Masking is listed as none, but the registry does not discuss how outcome measurements were obtained or whether measurement procedures could have been influenced by knowledge of group assignment.

The registry says the final analysis included subjects who completed the study, but does not report the number of completers or how missing measurements were handled. The adverse-event counts should not be used as a substitute for the number analyzed for the primary outcome. The record also does not report whether the intended comparability of blood sugar control between the groups was achieved.

Secondary outcome names are listed without numerical findings. The registry does not report a multiplicity approach for the multiple endpoint components, time points, or secondary measures. The original sample-size justification and detailed statistical plan are also not reported in the registry record. Consequently, it is not possible to assess whether the completed study had adequate precision for its intended comparisons or how the analysis addressed multiple outcomes.

These limitations describe what the registry record leaves unresolved. They do not establish that procedures were absent or that the trial’s findings were invalid; the record does not provide the information needed to make those judgments.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Effects of Pioglitazone on High-density Lipoprotein (HDL) Function in Persons With Diabetes — NCT01156597.
• Registry record last updated 2014-11-21; status last verified 2014-11.
• Results first posted 2014-11-11.
• 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

NCT01156597 is a completed, randomized, unmasked, parallel-group phase 3 study sponsored by the University of Miami. It enrolled 30 people with type 2 diabetes and compared pioglitazone with standard treatment using either metformin or a sulfonylurea. The registry lists HDL cholesterol and triglyceride percent changes as the primary endpoint, with HDL apolipoprotein levels and cholesterol efflux capacity as secondary outcomes.

The primary endpoint description includes multiple measures or time points, and the registry record does not reduce it to one comparison. It reports no numerical primary or secondary outcome findings, no endpoint uncertainty estimates, and no detailed analysis method. The final analysis is described as including study completers, but the number of completers and the handling of missing data are not reported in the registry record.

The adverse-event record lists 0 serious and 0 other affected participants in each group; death counts are not reported. The available record supports a description of the study’s question and design, but not a conclusion about the treatment’s measured effects.