University of Texas H.S.C. San Antonio Pioglitazone in Non-Alcoholic Steatohepatitis Trial (UTHSCSA NASH Trial): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for University of Texas H.S.C. San Antonio Pioglitazone in Non-Alcoholic Steatohepatitis Trial (UTHSCSA NASH Trial) (NCT00994682), a randomized phase 4 study. The primary outcome is Liver Histology (Using Kleiner et al Criteria, Hepatology 2005).

2. Clinical Question

3. Trial Design

The registry identifies this as a phase 4, randomized, quadruple-masked, parallel-group trial. It lists an enrollment of 176 and a completed status. The registry dates the study start to 2008-12 and primary completion to 2014-12. The sponsor is the University of Florida, while the trial title names the University of Texas H.S.C. San Antonio. The registry record does not explain this difference in institutional names.

The principal comparison posted for the primary outcome is placebo versus pioglitazone at 18 months. The primary outcome denominators are 51 participants in the placebo group and 50 in the pioglitazone group. Those denominators are smaller than the stated enrollment, but the registry record does not explain how participants flowed from enrollment to the outcome analysis, or how missing outcome data were handled. The difference alone does not establish why the denominators differ.

The intervention descriptions state that participants with NASH and prediabetes or type 2 diabetes mellitus were to receive dietary counseling before starting pioglitazone or placebo. The pioglitazone description starts with a dose of 30 mg/day, but the remainder of the dosing schedule is not reported in the registry record. The listed interventions also include pioglitazone open label. Adverse-event categories distinguish the first 18 months, placebo participants during a pioglitazone open-label period, and pioglitazone participants during months 18-36. These entries show that the record includes later treatment periods, but it does not fully describe the sequence or eligibility for those periods.

The study summary describes a multiethnic cohort as an aim and names several locally common ethnic groups, but actual demographic distributions are not reported in the registry record. Allocation concealment, the randomization method, masking procedures, and the original trial sample-size justification are not reported in the registry record. The planning illustrations accompanying the posted outcomes are separate from the trial's original design assumptions.

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: Once eligibility was met by histologically confirmed NASH, patients were randomized. The number of patients consented was 176, screen failed was 38, did not complete run-in phase was 37 (12 consent withdrawn, 9 lost to follow-up, 4 discontinued by PI decision, and 12 for other reasons). The remaining 101 patients were randomized.

Double Bilnd

Reasons for not completing this period:

Pioglitazone Open Label

Reasons for not completing this period:

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 Liver Histology (Using Kleiner et al Criteria, Hepatology 2005), measured over At 18 months.

The primary endpoint is a participant count at 18 months. A participant meets the stated response definition if the nonalcoholic fatty liver disease activity score (NAS) falls by at least 2 points, the reduction occurs across at least 2 different histological categories, and fibrosis does not worsen. This is a combined histology-based definition: improvement in the activity score alone is not sufficient if the fibrosis condition is not met.

The registry describes NAS as the sum of scores for steatosis, hepatocellular ballooning, and lobular inflammation, with a range of 0-8. It separately describes fibrosis grading from no fibrosis through cirrhosis. The activity score captures several aspects of liver-tissue appearance, while the no-worsening condition prevents a participant from qualifying solely on the basis of improvement in the activity components if the fibrosis criterion is breached.

Histology is used in this endpoint because the trial's stated question concerns changes in liver tissue, rather than only a blood test or a metabolic measurement. The registry gives scoring criteria and an assessment time, but the posted primary result is a count of participants meeting the response definition; it does not include participant-level score changes or a distribution of baseline and follow-up histology. The posted counts therefore indicate how many participants met the registry's composite response rule, not the magnitude of change for every participant.

7. Results Posted in the Registry

• Pioglitazone: 58.0% (29/50) versus 17.6% (9/51) on control; risk difference 40.4 percentage points (95% CI 23.1 to 57.6), risk ratio 3.29 (1.74–6.22); unadjusted two-proportion 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.

Number of Participants With Resolution of NASH

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Pioglitazone: 52.0% (26/50) versus 19.6% (10/51) on control; risk difference 32.4 percentage points (95% CI 14.8 to 50.0), risk ratio 2.65 (1.43–4.91); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Pioglitazone than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Liver Fat by Magnetic Resonance and Spectroscopy (MRS).

• 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:
  • Pioglitazone: mean 7.00 versus 11.00 on control; difference -4.00 (95% CI -6.61 to -1.39), standardized effect d = -0.66; z-test p = 0.003.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Pioglitazone than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Total Body Fat

• 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:
  • Pioglitazone: mean 36.00 versus 36.00 on control; difference 0.00 (95% CI -3.23 to 3.23), standardized effect d = 0.00; z-test p = 1.000.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Pioglitazone and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes extend the primary histology question in several directions: whether NASH resolved, how liver fat and body fat were measured, and additional metabolic, biochemical, and histological measures listed in the registry. The posted results do not form a single interchangeable measure of benefit or response. Each endpoint reflects a different construct and may use different participants or measurement procedures.

The posted secondary outcomes for NASH resolution and liver fat point in a direction consistent with the primary comparison's group pattern, while the total-body-fat summary does not show a mean difference. These are distinct outcomes and should be interpreted on their own definitions and scales. Their agreement or lack of agreement with the primary result does not by itself confirm or invalidate the primary endpoint. The registry also lists secondary measures for which no numerical result is included in the posted record.

Unadjusted secondary comparisons carry less weight when many outcomes are considered. In general, testing multiple outcomes increases the opportunity for some apparently notable results to arise by chance, even if there is no true effect for those outcomes. No multiplicity strategy is reported in the registry record, so the secondary p-values should be regarded as exploratory descriptions rather than as a family of results with a stated overall error control. The record does not report a hierarchy of secondary endpoints or identify which analyses were designated confirmatory.

The number of posted secondary outcomes and the absence of a reported analysis plan also limit cross-outcome synthesis. Differences in denominators may reflect outcome availability or assessment, but the registry record does not explain them. The safest overall reading is that selected secondary summaries provide additional descriptive context around the primary histology result, while their unadjusted nature and incomplete reporting leave uncertainty about how robustly they support broader conclusions.

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

• Number of Participants With Resolution of NASH
• Mean Individual Histological Scores
• Individual Histological Scores
• Mean Individual Histological Scores
• Liver Transaminases (AST and ALT).
• Liver Fat by Magnetic Resonance and Spectroscopy (MRS).
• Homeostatic Model Assessment of Insulin Resistance (HOMA-IR)
• Hepatic Insulin Sensitivity

9. Statistical Interpretation

10. Statistical Methods Explained

What does a participant count mean? The primary outcome is binary at the participant level: each person either meets the histological response definition or does not. The registry posts the number meeting the definition and the number assessed in each randomized group. This makes the outcome interpretable as a frequency of response under the stated criteria, rather than as an average change in NAS points.

What is a risk ratio? A risk ratio compares the proportion meeting an outcome in one group with the proportion in another. A value of 1 corresponds to equal proportions under this measure; values above 1 indicate a higher observed proportion in the first named group, and values below 1 indicate a lower one. For the primary comparison, the posted risk ratio is 3.287, with a 95% confidence interval from 1.736 to 6.221. This relative measure does not give the absolute difference between groups. It also does not show how much an individual participant's histology changed.

What is a confidence interval? A confidence interval represents uncertainty around an estimate under the statistical method used to calculate it. In repeated applications of a method with the same coverage properties, intervals would capture the underlying quantity at the stated rate. For an individual trial, the interval gives a range of values reasonably compatible with the data and method; it is not a guarantee that the true value lies within it, nor a probability statement about the particular interval. Wider intervals indicate less precision than narrower intervals on the same scale.

What does the p-value add? The primary comparison has an unadjusted p-value of <0.001. In general, a p-value measures the compatibility of the observed data with a specified null model, assuming that model and the test assumptions. It is not an effect-size measure, and it does not address whether the outcome definition captures outcomes important to participants. Because the reported value is unadjusted, it should not be treated as if it accounts for testing many outcomes or comparisons.

How do continuous outcomes differ? The posted liver-fat result is a continuous measure, reported as a group mean with a standard deviation. The pioglitazone group mean is 7.0 percentage of fat in liver, with a standard deviation of 5.0, and the placebo group mean is 11.0, with a standard deviation of 7.0. The posted mean difference is -4.0, with a 95% confidence interval from -6.612 to -1.388 and an unadjusted p-value of 0.003. A mean difference describes a contrast in group averages, while the standard deviations describe variation among observations within each group. They answer different questions.

For a continuous outcome, a confidence interval for the mean difference indicates the range of contrasts compatible with the data and the calculation method. The interval for liver fat does not include 0, the value that would represent no mean difference under this contrast. That observation is a feature of the unadjusted result, not proof that every participant experienced a change in the same direction. Group summaries do not show the distributions, outliers, or individual trajectories.

Why can over-dispersion matter? As a general statistical concept, over-dispersion occurs when the variability in data is greater than a model assumes. If a model underestimates variability, standard errors can be too small and confidence intervals too narrow, making results appear more precise than they are. The primary endpoint is a participant count, but the registry record does not describe a model for repeated counts or any over-dispersion assessment. The concept is therefore general context, not a claim about an analysis performed in this trial.

What does randomization contribute? In general, random assignment aims to make treatment groups comparable in expectation with respect to both measured and unmeasured baseline characteristics. This supports interpretation of between-group contrasts, particularly when assignment and follow-up are implemented as intended. It does not ensure exact balance in a particular trial, resolve missing outcome data, or substitute for information about adherence and analysis populations. The record identifies the trial as randomized, but the method of sequence generation, concealment, adherence, and participant flow is not reported in the registry record.

Why distinguish posted comparisons from sponsor analyses? The registry posts group summaries and unadjusted comparisons for several outcomes, but the reported analyses list is empty. These comparison statistics do not include a described sponsor analysis plan or adjustment strategy. The record does not report whether any models adjusted for baseline values or other covariates. Interpret the posted estimates as the comparisons they state, not as evidence of unreported modeling decisions.

General statistical interpretation also depends on the outcome definition, measurement quality, follow-up, and assumptions used for calculation. For the histological endpoint, the registry states the criteria but does not report central reading procedures, reader agreement, or blinding of pathology assessment. For each outcome, the available group summaries provide less information than participant-level records. The posted statistics can quantify selected contrasts, but they cannot answer questions requiring individual data or methods that are not reported in the registry record.

11. Confidence Intervals and Precision

Precision is about how tightly the data constrain an estimate. For the primary endpoint, the posted risk ratio is 3.287 and its 95% confidence interval runs from 1.736 to 6.221. The interval excludes 1, the value corresponding to equal group proportions on the risk-ratio scale, but it spans a range of relative contrasts. Thus the registry presents evidence of a group difference under the unadjusted comparison while leaving uncertainty about the exact relative size.

The interval's scale matters. A risk ratio expresses a relative comparison, not a percentage-point change. The primary result also gives participant counts—29 of 50 and 9 of 51—which retain the underlying group context. The confidence interval does not remove uncertainty about outcome assessment, missing data, or the applicability of the participants to populations not described in the record.

The liver-fat comparison is reported on an absolute scale as a mean difference of -4.0, with a 95% confidence interval from -6.612 to -1.388. The total-body-fat mean difference is 0.0, with a 95% confidence interval from -3.232 to 3.232. These intervals illustrate that precision is outcome-specific; a result for one measure does not set the precision of another. Standard deviations, sample sizes, measurement scales, and analysis choices all affect uncertainty.

Confidence intervals describe sampling uncertainty under their calculation method. They do not include every source of uncertainty, such as possible measurement variation, departures from the intended design, or limitations in the registry's reporting. The registry record does not report enough analysis detail to assess all such sources. Accordingly, the intervals are useful measures of statistical precision, but not a complete account of certainty about the trial's broader interpretation.

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 sample-size figures are planning illustrations based on the observed effect, not the trial's original sample-size justification. For the binary primary endpoint, the entries are participants per arm. The assumptions stated for these illustrations are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Each row considers detecting a stated fraction of the observed effect.

The 80% power column describes the planned number of participants per arm that would, under those assumptions and for the effect size used in that row, give an 80% chance of a statistically significant result if that effect were the true one. The 90% column uses a higher target chance of detecting that effect, and consequently calls for more participants in the corresponding planning scenario. These are prospective design calculations, not results from the trial.

Rows that plan for a smaller fraction of the observed effect require more participants per arm than rows based on a larger fraction. This is a general consequence of needing more information to distinguish a smaller group contrast from random variation. The illustrations should not be read as a recommended design for a new study without considering its population, outcome, analysis plan, and other design choices.

This is not observed power. It is not a measure of the probability that the trial's finding is true, and it does not establish that the original study was adequately sized. The trial's own sample-size assumptions and justification are not reported in the registry record. For time-to-event outcomes, the stated convention is total events rather than participants per arm, but the primary endpoint is a participant-count outcome, so its planning figures are per arm.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.176; treatment proportion p₁ = 0.580; average p̄ = 0.378.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6858 = 1.3442.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6236 = 0.5249.
4. n = (1.3442 + 0.5249)² / (0.4035)² = 21.5, rounded up to 22 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
p0 <- 0.176470588235294                          # control proportion
p1 <- 0.58                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(994682)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 22 (80% power), 29 (90% power)
 75% of observed effect: n per arm = 37 (80% power), 49 (90% power)
 50% of observed effect: n per arm = 77 (80% power), 102 (90% power)
power.prop.test: 21.5 
Simulated power at n = 22 per arm: 0.805 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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

Multiplicity arises when a study evaluates several outcomes, time points, subgroups, or comparisons. As a general statistical issue, the more unadjusted tests considered, the greater the chance that at least one result will appear statistically notable by chance. Researchers may address this through a prespecified hierarchy, adjustment procedure, or by labeling secondary analyses exploratory. The registry record does not report a multiplicity plan, an adjustment method, or a hierarchy governing the primary and secondary outcomes. The posted p-values are therefore unadjusted and should not be interpreted as controlling the error rate across all outcomes.

The record lists 19 outcome measures and 18 posted secondary outcomes. The reported analyses list is empty. It is not clear from the registry record which endpoints were included in any complete analysis plan, whether analyses were prespecified, or how any unreported outcomes were handled. This absence does not establish that analyses were not conducted; it means that their methods are not reported in the registry record.

Missing-data methods are also not reported in the registry record. The primary outcome summaries use 51 placebo participants and 50 pioglitazone participants, while the registry enrollment is 176. Without a participant-flow account and missing-outcome rule, the reason for the smaller endpoint denominators is not established. It is not possible to determine from the posted summaries whether missingness was related to treatment, outcome status, or other factors.

The registry record does not specify whether the primary analysis followed an intention-to-treat principle, used a per-protocol population, or applied another analysis population. It also does not report sensitivity analyses for missing data, an adjusted analysis, or a detailed statistical analysis plan. These omissions matter because analysis population and missing-data assumptions can affect how a randomized comparison is interpreted. The group counts and their unadjusted comparisons remain descriptive of the posted outcome denominators, but the record does not support claims about unreported analytical choices.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: biopsy-related (5.9%), diverticulitis (5.6%), urine retention (5.6%), atypical chest pain (5.0%), colelithiasis (5.0%).

For the first 18 months, the registry records serious adverse events in 5 of 51 participants in the placebo group and 5 of 50 participants in the pioglitazone group. Other, non-serious adverse events are recorded in 7 of 51 placebo participants and 15 of 50 pioglitazone participants. These are participant counts in the stated groups and period; the record does not give a comparative statistical analysis of these counts.

The later safety categories are reported separately: placebo during the pioglitazone open-label period has 10 serious and 12 other adverse events among 36 participants; pioglitazone during months 18-36 has 9 serious and 10 other adverse events among 40 participants. These periods and groups should not be combined into the initial randomized comparison because the record does not describe a common randomized contrast for them.

The registry lists serious-event terms with percentages, including biopsy-related events at 5.9%, diverticulitis at 5.6%, urine retention at 5.6%, atypical chest pain at 5.0%, colelithiasis at 5.0%, and kidney stones at 5.0%. Death counts are not reported in the registry record. Counts alone do not establish causation, comparative risk, severity patterns, or whether an event was related to treatment; detailed event-level descriptions and an analysis plan are not reported.

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

The primary result describes a difference between randomized groups in the number of participants meeting a specific liver-histology response definition at 18 months. The counts and unadjusted comparison are the central evidence posted for that endpoint. They speak to the registry's operational definition, not to every possible interpretation of NASH, liver function, symptoms, or long-term health.

The result should not be treated as a complete account of benefit and harm. The registry also posts selected secondary measurements and adverse-event counts, but the secondary comparisons are unadjusted and the safety entries are not accompanied by a comparative analysis. No single endpoint can stand in for all outcomes. The record does not establish whether histological changes corresponded to later clinical outcomes or whether the observed group pattern persisted after 18 months.

Randomization supports a comparison between assigned groups in principle, but the registry record does not report the randomization procedure, analysis population, adherence, missing-data handling, or participant flow. Those omissions constrain assessment of how closely the posted comparison reflects the intended randomized contrast. The stated enrollment is 176, while the primary endpoint summaries use 51 and 50 participants; the record does not account for that difference. It would be inappropriate to infer the reason or treat it as a particular kind of loss to follow-up.

The findings also do not establish how effects vary by glucose status or ethnicity. The summary describes those characteristics as part of the study's aims, but subgroup outcome estimates and actual demographic distributions are not reported in the registry record. Likewise, the randomized comparison should not be projected to the later open-label and months 18-36 periods, for which the record gives separate safety categories but not a corresponding randomized primary-outcome comparison.

In short, the registry offers an interpretable 18-month group contrast for a defined histological endpoint, along with selected secondary and safety summaries. The degree of uncertainty is reflected partly in the confidence interval, but statistical precision cannot resolve limitations in reporting, study conduct details, or generalizability. The result is evidence about the posted data and endpoint—not a definitive statement about every outcome or population of interest.

17. Why This Trial Matters Statistically

This trial illustrates why the endpoint definition matters as much as the headline count. The primary response requires improvement across multiple histological categories and no worsening of fibrosis. A simpler count of NAS reduction alone would answer a different question. Clear criteria allow readers to understand what is classified as a response and to distinguish that classification from a general judgment about disease status.

It also demonstrates the value of presenting both group denominators and an uncertainty interval. Counts show the observed data on which the comparison rests; the risk ratio and confidence interval express the contrast and its statistical precision. A p-value alone would not convey either the magnitude of the relative comparison or the range of values compatible with the data. Different effect measures answer different questions, so the scale should always be explicit.

The record offers a useful contrast between an efficacy-oriented endpoint and multiple secondary measures, including liver fat, body fat, and other listed outcomes. Their distinct patterns emphasize that a trial can address several related but nonidentical biological questions. Without a reported multiplicity plan, these secondary results are best considered contextual rather than as a set of independent confirmations.

Finally, the record shows the limits of registry summaries. It gives trial status, broad design features, selected outcomes, and adverse-event counts, but not the complete analysis plan, detailed participant flow, or missing-data procedures. Statistical interpretation depends not only on estimates, but also on knowing how they were generated and which participants contributed. Transparent reporting of those details is essential for evaluating precision, potential bias, and applicability.

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.

A major trial-specific limitation is the gap between the registry enrollment of 176 and the primary-outcome denominators of 51 and 50. The record does not provide a participant-flow explanation, so the reasons for the denominators and the handling of participants without a posted endpoint result are not reported in the registry record. This prevents assessment of possible selection or missing-data effects from the summaries alone.

The registry states that the study aimed to examine a multiethnic cohort and response across glucose-tolerance categories, but actual demographic distributions and subgroup results are not reported in the registry record. The applicability of the overall summaries to particular ethnic or glycemic groups therefore cannot be evaluated from the posted results.

Details of random-sequence generation, allocation concealment, masking procedures, histology reading, adherence, and analysis populations are not reported in the registry record. The arm descriptions are incomplete, including a pioglitazone dosing description that ends after stating 30 mg/day. These gaps make it difficult to reconstruct treatment exposure or assess implementation.

Several secondary outcomes are listed, but only selected results are posted, and the reported analyses list is empty. No multiplicity approach or detailed analysis plan is reported. The primary comparison is unadjusted in the posted statistics, and the record does not describe baseline-adjusted analyses or sensitivity analyses.

Later periods include open-label safety categories, but a later randomized comparison for the primary histology endpoint is not reported. The adverse-event counts also lack event-level context and a comparative analysis. These reporting limitations do not negate the posted counts, but they restrict conclusions about later outcomes, causal attribution of adverse events, and the robustness of the broader interpretation.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: University of Texas H.S.C. San Antonio Pioglitazone in Non-Alcoholic Steatohepatitis Trial (UTHSCSA NASH Trial) — NCT00994682.
• Registry record last updated 2017-04-05; status last verified 2017-02.
• Results first posted 2017-04-05.
• 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 describes a completed, phase 4, randomized, quadruple-masked, parallel-group trial of pioglitazone and placebo in NASH. It lists enrollment of 176, a study start of 2008-12, and primary completion in 2014-12. The primary outcome is a histological response at 18 months, defined by a reduction of at least 2 NAS points across at least 2 categories without worsening fibrosis.

The posted primary counts are 29 of 50 in the pioglitazone group and 9 of 51 in the placebo group. The unadjusted risk ratio is 3.287, with a 95% confidence interval from 1.736 to 6.221 and p < 0.001. The record also includes selected secondary results, later-period adverse-event counts, and planning illustrations based on observed effects.

Important details—including the original sample-size justification, analysis plan, missing-data handling, analysis populations, participant flow, and subgroup results—are not reported in the registry record. The estimates should be interpreted as summaries of the posted outcomes and denominators.