Pioglitazone to Treat Fatty Liver in People With HIV Infection: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pioglitazone to Treat Fatty Liver in People With HIV Infection (NCT00441272), a randomized phase 2 study. The primary outcome is Hepatic Steatosis.

2. Clinical Question

The clinical question is whether assignment to pioglitazone, compared with placebo, was associated with a difference in the liver measure at the stated endpoint. The registry record describes random assignment and a placebo comparator, which are features intended to make a treatment comparison interpretable. But it does not report results needed to assess the direction, size, or precision of any difference. It also describes all 11 enrolled subjects as ineligible, leaving the relationship between the planned trial population and the reported outcome population unclear. The record lists an enrollment of 100, while the outcome-population description says 11 subjects enrolled; the reason for this discrepancy is not reported in the registry record.

3. Trial Design

The registry identifies this as a phase 2, randomized, quadruple-masked, parallel-group study. In a parallel-group design, assigned participants remain in one of the study groups rather than crossing between treatments. Random assignment is intended to reduce systematic differences between groups at the point of assignment. Masking is intended to limit the influence of knowing the assigned treatment on participants, study staff, or outcome assessment. The record does not specify which four roles were masked.

The planned comparison was pioglitazone 45 mg once daily versus placebo. The brief summary describes 48 weeks of study medication, after which all participants were to stop medication and be followed for an additional 48 weeks. It gives the primary endpoint timeframe as 96 weeks. Visits and procedures described in the summary included laboratory tests and insulin clamp testing at day 0 and weeks 24 and 48. The registry record does not report whether each planned procedure was completed or how adherence to treatment was assessed.

The listed enrollment is 100. However, the primary outcome population description says 11 subjects enrolled and states that 10 were deemed ineligible after a computerized tomography scan found a liver-to-spleen ratio > 1, while one was deemed ineligible because of concomitant medication use. These entries do not explain how enrollment, eligibility, and randomization relate to one another. The number actually randomized is not reported in the registry record.

The study is marked completed, with a start date of 2007-02 and a primary completion date of 2008-10. Completion status does not establish that the endpoint was measured for every participant or that results were analyzed. The registry record does not report a protocol, detailed statistical plan, allocation concealment procedures, or a sample-size justification. It also does not report whether any interim analyses occurred.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: A total of 11 subjects enrolled into the study. 10 were determined to be ineligible during the screening as the Computerized tomography scan revealed a liver-to-spleen ratio \> 1 and one was determined ineligible due to concomitant medication use.

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 Hepatic Steatosis, measured over 96 weeks.

The primary endpoint is hepatic steatosis, with the reported measure defined as a mean in Hounsfeld units at 96 weeks. The outcome description states that the evaluation concerns the safety and potential benefits of pioglitazone therapy on hepatic steatosis in HIV-infected men and women. The record does not give further detail about how the measure was derived, what direction of change would represent less steatosis, or how measurements were assessed.

As a general statistical point, a continuous endpoint retains information about the measured value rather than reducing each participant to a category such as improved or not improved. A group mean describes the central value of the measurements, while the standard deviation describes how spread out individual measurements are around that mean. Comparing group means can therefore address whether the average endpoint differed between assigned groups, while the variability helps characterize how consistently participants’ values were distributed.

The registry record identifies a 96-week timeframe, after the planned 48-week treatment period and the further 48-week follow-up without study medication. A measurement at that time would concern the outcome after the treatment and follow-up schedule described in the summary; it would not, by itself, describe the trajectory at earlier visits. The registry record does not report the endpoint values, standard deviations, participant counts contributing to the endpoint, or an estimated between-group contrast. Consequently, the endpoint’s observed distribution and the precision of any comparison cannot be assessed from the record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical methods available for interpretation are limited. The registry classifies the primary outcome as continuous and describes it using a mean and standard deviation, but it does not report the analysis method used to compare groups. The statistical plan, including model specification, covariate adjustment, confidence level, missing-data approach, and analysis population, is not reported in the registry record. The record also reports no analyses.

What does a mean and standard deviation describe?

General explanation: the mean is the arithmetic average of the observed values in a group. It provides a compact description of the group’s central tendency, but it does not show the full distribution. The standard deviation describes how dispersed observations are around the mean. A small standard deviation indicates that values are relatively concentrated; a larger one indicates greater variation among participants. Neither statistic alone gives the precision of a difference between groups. Precision also depends on how many outcome measurements are available and how variable those measurements are.

For this study, the endpoint is described as a mean in Hounsfeld units at 96 weeks. The registry record does not report the group means, standard deviations, or counts used to calculate them. Thus, even the basic descriptive comparison cannot be reconstructed from the posted results. It is also not reported whether the primary comparison used only the 96-week values, changes from baseline, or another analysis of repeated measurements.

How should a treatment contrast be read?

General explanation: a treatment contrast compares outcomes between assigned groups. A point estimate describes the estimated size and direction of the difference under the analysis used. A confidence interval expresses a range of values compatible with the data and analysis, subject to the method’s assumptions; a narrower interval indicates greater precision than a wider one. A p-value addresses how compatible the observed data are with a specified null hypothesis under the statistical model. It does not measure clinical importance, and it is not the probability that the null hypothesis is true.

No point estimate, confidence interval, or p-value is reported for the primary outcome in this registry record. The treatment contrast therefore cannot be described as favoring either group, and its precision cannot be assessed. No result can be reconstructed merely from the randomized design, stated endpoint, or study status.

What does randomization contribute?

General explanation: random assignment aims to make treatment groups comparable, on average, with respect to measured and unmeasured factors at assignment. It supports interpretation of a between-group comparison when the assigned groups, follow-up, and analysis are handled appropriately. It does not guarantee balance in a small realized sample, prevent missing outcome data, or eliminate bias introduced after assignment. The registry record does not report the number randomized, baseline characteristics, or the analysis population, so the success and consequences of randomization cannot be evaluated from these entries.

Quadruple masking is recorded, but the identities of the masked roles and the success of masking are not reported. General explanation: masking can reduce changes in behavior or assessment that arise from knowledge of assignment. It cannot correct incomplete follow-up or a poorly specified analysis, and its presence in a registry description is not evidence that masking remained intact.

What can be inferred from the planning assumptions?

The record describes a planning illustration based on an observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. It says the effect to detect is a stated fraction of the observed effect. These are planning assumptions, not trial results or evidence of observed power. No observed treatment effect, planning sample-size values, or corresponding outcome data are reported in the registry record. Without those values, the illustrations cannot be interpreted as a justification for the listed enrollment or as evidence that the study had a particular probability of detecting an effect.

General explanation: a sample-size calculation estimates how many participants may be needed under specified assumptions about variability, target effect, significance level, and desired power. If assumptions differ from the eventual data, the calculation may not describe the study’s actual precision. A post hoc calculation based on an observed effect is particularly dependent on that effect estimate and should not be treated as observed power. The record does not report the original design justification or a prospective power calculation.

The outcome description notes a screening-based ineligibility finding, but the record does not specify whether this affected the intended analysis, how many endpoint observations were available, or how missing outcomes were handled. Those details are needed to understand the estimand—the precise treatment effect the analysis is intended to estimate—and to judge whether the reported analysis, if any, answers the stated clinical question.

10. Multiplicity, Missing Data and Analysis Populations

The registry record does not report a statistical analysis plan or describe how multiple outcomes, time points, or comparisons were handled. Hepatic steatosis is listed as the primary outcome, and insulin resistance is listed as a secondary outcome. The summary describes several study visits, while the primary endpoint timeframe is 96 weeks. It is not reported whether the analysis tested only a single prespecified comparison or considered multiple time points or outcome measures, nor whether any multiplicity adjustment was planned or applied.

General explanation: when many statistical comparisons are made, the chance of at least one apparently notable result can increase. A prespecified analysis plan and an explicit approach to multiplicity help readers distinguish a primary confirmatory question from additional exploratory comparisons. The registry record does not give enough information to assess this issue for the study.

Missing-data methods are not reported in the registry record. It does not give the number of participants with a 96-week measurement, reasons for missing measurements, or whether missing outcomes were imputed, modeled, or excluded. The number randomized and the analysis population are also not reported. These details matter because the meaning of a treatment comparison depends on who contributes data and how incomplete follow-up is addressed.

The outcome-population description reports 11 enrolled subjects and identifies all 11 as ineligible for the reasons stated, while the registry lists enrollment as 100. The record does not explain how these figures relate to randomization or analysis populations. No participant flow or primary-outcome analysis is reported, so the extent of missingness and the handling of exclusions cannot be assessed.

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

The registry record establishes the study question, planned comparison, endpoint timeframe, and several design features. It does not report the results needed to estimate how hepatic steatosis differed between the pioglitazone and placebo groups. No conclusion about the size, direction, or precision of a treatment effect follows from the record’s design description alone.

The record also does not allow an assessment of whether any group difference persisted after medication stopped. Although the summary describes follow-up for 48 weeks after treatment and the primary endpoint is at 96 weeks, outcome values at that time are not reported. The record similarly does not provide estimates for insulin resistance, the listed secondary outcome.

The adverse-event entries require particular caution: both groups have zero participants listed at risk and zero affected. With no participants listed at risk, these entries do not establish an event rate or demonstrate an absence of events. Death counts are not reported in the registry record. Safety cannot be characterized by treating missing or unusable denominators as evidence of no harm.

General explanation: a registered study design describes what investigators intended to do; a results record is needed to assess what was observed. A completed status indicates a registry status, not that results are available or that the planned analysis is documented. The discrepancy between the listed enrollment of 100 and the outcome-population description of 11 enrolled subjects, all described as ineligible, further limits interpretation of the study’s realized sample and analysis population.

These limitations do not show that pioglitazone had or lacked an effect. They mean the registry record does not contain enough outcome information to determine the result. Clinical conclusions, including claims of benefit or lack of benefit, would go beyond the posted data.

12. Why This Trial Matters Statistically

This trial is statistically notable as an example of the gap between a registered design and an interpretable result. The record describes randomization, masking, a parallel comparison, and a continuous primary measure at a specified follow-up point. Those details define the intended structure of a treatment comparison, but they cannot replace group-level outcomes and an analysis estimate.

The enrollment information also illustrates why participant flow matters. The registry lists enrollment as 100, while its primary-outcome population description reports 11 enrolled subjects and says each was found ineligible for a stated reason. Without a clear account of how many people were screened, eligible, randomized, treated, and assessed, readers cannot determine which population the study results would represent. The reason for the discrepancy is not reported in the registry record.

In general, reporting group means alongside variability, participant counts, and a between-group estimate with a confidence interval would help readers judge both the magnitude and precision of a continuous outcome comparison. Information about missing observations, analysis populations, and the prespecified plan would help establish how that estimate was obtained. Those results and methods are not reported in the registry record.

The case also illustrates that sample-size planning requires transparent assumptions and should be distinguished from observed results. The record describes a planning illustration based on an observed effect, but no effect estimate or numerical planning values are reported. Such planning information cannot establish whether the eventual study was adequately sized or what its results showed. The statistical importance of this record is therefore mainly interpretive: its design can be described, but its primary treatment comparison cannot be evaluated from the reported data.

13. 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.

A central limitation is the absence of primary-outcome results. The registry record does not report 96-week group means, standard deviations, participant counts, an estimated treatment contrast, confidence intervals, or p-values. Consequently, neither the direction nor precision of a comparison can be assessed. The record also does not report results for the listed secondary outcome of insulin resistance.

The enrollment information is internally difficult to reconcile. The record lists enrollment as 100, but the primary-outcome population description says 11 subjects enrolled, with 10 ineligible after a computerized tomography scan found a liver-to-spleen ratio > 1 and one ineligible because of concomitant medication use. The reason for the difference is not reported in the registry record. The number randomized, treated, and assessed at 96 weeks is not stated, so the achieved sample and participant flow are unclear.

The endpoint description gives the measure, unit, and timeframe, but does not explain measurement procedures, the interpretation of higher or lower values, or whether the analysis used the endpoint value or change from baseline. The registry record does not report how missing data, protocol deviations, exclusions, or repeated measurements were addressed. It also does not provide the statistical analysis plan, multiplicity strategy, or the identities of the roles included in quadruple masking.

The adverse-event entries list zero participants at risk and zero affected in each group, which does not support estimation of event frequency. Death counts are not reported in the registry record. Finally, the study’s completed status and design features do not resolve these missing results and methods. The registry record does not explain the relationship between the stated planning assumptions and the listed enrollment, or provide an original sample-size justification.

14. Related Tutorials

15. Related Statistical Calculators

16. Sources

• ClinicalTrials.gov: Pioglitazone to Treat Fatty Liver in People With HIV Infection — NCT00441272.
• Registry record last updated 2010-02-19; status last verified 2008-10.
• Results first posted 2010-02-19.
• 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.

17. Record Summary

NCT00441272 is a completed phase 2 randomized, quadruple-masked, parallel-group study titled “Pioglitazone to Treat Fatty Liver in People With HIV.” It compared planned daily pioglitazone 45 mg with placebo for 48 weeks, followed by 48 weeks without study medication. The primary outcome was hepatic steatosis, described as a mean in Hounsfeld units at 96 weeks.

The registry lists enrollment as 100 and gives study dates of 2007-02 to 2008-10. The outcome-population description instead reports 11 enrolled subjects, all described as ineligible for screening or concomitant-medication reasons. The number randomized and the reason for this enrollment discrepancy are not reported in the registry record.

Group outcome values, a treatment comparison, statistical precision, and a detailed analysis plan are not reported. Insulin resistance is listed as a secondary outcome, but no result is reported. The adverse-event entries have zero participants listed at risk in each group, and death counts are not reported in the registry record.