Effect of Pioglitazone and Exenatide on Body Weight and Beta Cell Function: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effect of Pioglitazone and Exenatide on Body Weight and Beta Cell Function (NCT00845182), a randomized phase 4 study. The primary outcome is Effect of Pioglitazone, Exenatide and Combined Pioglitazone and Exenatide on Body Weight.

2. Clinical Question

3. Trial Design

The registry identifies this as a completed phase 4, randomized, single-masked, parallel-group trial sponsored by The University of Texas Health Science Center at San Antonio. Its listed enrollment is 43. The study start is recorded as 2007-06, and primary completion is recorded as 2010-07. The trial compares pioglitazone, exenatide, and the combination of pioglitazone with exenatide.

The arm descriptions say that 15 patients or subjects would be randomized to each treatment group. For the posted primary outcome, however, the registry reports group counts of 10 for pioglitazone, 11 for exenatide, and 9 for the combination. The record does not explain the difference between the arm descriptions and the outcome counts. These values describe different parts of the record and should not be treated as interchangeable.

The study description frames the combination as a way to examine body weight, fat topography, beta cell function, glycemic control, and plasma lipid levels alongside treatment with each drug separately. It also describes assessment of maximal insulin secretory capacity using a maximal hyperglycemic stimulus combined with an intravenous arginine stimulus. These details describe the stated scientific aims; they do not establish that numerical results for every aim were posted.

The registry lists Type 2 Diabetes, Healthy, and Impaired Glucose Tolerance among the conditions. The title and brief summary focus on type 2 diabetes, while the condition list includes the additional labels. The record does not explain how these labels relate to eligibility or to the analyzed population. It also does not identify who was masked, describe the randomization procedure, or specify the planned sample-size justification. Those design details are not reported in the registry record.

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: we analyzed the weight at the end of study completion

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 Effect of Pioglitazone, Exenatide and Combined Pioglitazone and Exenatide on Body Weight, measured over baseline and 6 months.

The primary outcome is body weight, recorded in kilograms. The registry describes the measure as a mean with standard deviation and gives the timeframe as baseline and 6 months. Its population description says that weight at the end of study completion was analyzed. The wording does not specify whether the posted means represent final weight, change from baseline, or another derived weight measure. That distinction is not reported in the registry record, so the values should be described as the posted body-weight outcome rather than assigned a more specific interpretation.

Body weight is used in this trial because the study description asks whether the treatment groups differ on weight while also investigating other metabolic outcomes. A continuous outcome such as weight can be summarized by a group mean and its standard deviation. The mean indicates the group’s average recorded value; the standard deviation describes how dispersed individual values are around that average. These summaries describe the observed groups, not the precision of the estimated difference between groups.

The posted group means are 5.5 kg for pioglitazone, 2.4 kg for exenatide, and 2.7 kg for the combination. Their standard deviations are 7.3 kg, 2.8 kg, and 4.4 kg, respectively. The registry record does not provide group-level confidence intervals for these means. Nor does the outcome description clarify how the baseline measurement was incorporated into the posted outcome values. The reported analyses include ANCOVA and ANOVA, but their estimates and model details are not reported.

7. Results Posted in the Registry

• EXENATIDE: mean 2.40 versus 5.50 on control; difference -3.10 (95% CI -7.92 to 1.72), standardized effect d = -0.57; z-test p = 0.207.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry names ANCOVA and ANOVA as methods for analyses involving the three treatment groups. It does not provide the model specifications, estimated effects, confidence intervals, or details about which comparisons were tested. The methods below explain these statistical approaches in general; they should not be taken as a description of unreported features of this trial’s analysis.

What does an analysis of variance test?

In general, ANOVA compares group means while accounting for variability within groups. An omnibus ANOVA test asks whether the data are compatible with a common mean across the groups included in the analysis. A small p-value can be evidence against that common-mean model, but it does not by itself identify which pair of groups differs, the size of any difference, or whether that difference is important in practice. Follow-up comparisons and their handling of multiple testing require separate reporting.

The registry reports an ANOVA p-value of <0.05 for the analysis involving the three listed groups. Because the record provides no estimate or interval for that analysis and does not state the specific contrast represented by the p-value, its magnitude and detailed interpretation cannot be reconstructed from the record. The reported result should therefore be kept distinct from the recalculated exenatide-versus-pioglitazone contrast.

What does ANCOVA add?

In general, ANCOVA compares group outcomes while adjusting for one or more covariates. When a baseline measurement is included as a covariate, the method can account for baseline differences and estimate group contrasts conditional on that measurement. Whether adjustment is appropriate and how it changes precision depend on the specified model and the data. The registry does not identify the covariates used in this trial’s ANCOVA, so it is not possible to say whether baseline weight was included or how any adjustment affected the result.

The registry reports an ANCOVA p-value of <0.05 for the three groups. It does not provide the adjusted group estimates, adjusted contrasts, confidence intervals, or model assumptions. The p-value therefore cannot tell readers the adjusted difference in kilograms or its precision. It also cannot be directly compared with the unadjusted recalculation as if both analyses estimated precisely the same quantity.

What is the unadjusted recalculation?

The recalculated pairwise comparison uses the posted summary statistics for exenatide and pioglitazone. The estimated mean difference is -3.1 kg, with a 95% confidence interval from -7.918 to 1.718 kg and an unadjusted p-value of 0.207. It is called unadjusted because it does not incorporate covariates such as baseline weight. It is also a simple comparison of two groups, not an attempt to reproduce the registry’s ANCOVA or ANOVA.

A mean difference describes the estimated contrast between the group means, using the comparison’s stated direction. The negative estimate reflects the ordering used in that calculation. The confidence interval expresses uncertainty about the contrast under the calculation’s assumptions. Since it includes zero, the data do not provide a precise estimate that rules out no difference under this unadjusted analysis. The interval also allows a range of potentially meaningful values on either side of zero; the registry does not provide a threshold for clinical importance.

The calculation uses summary statistics rather than the participant-level observations or the sponsor’s fitted models. Consequently, it cannot account for covariate adjustment, missing-data methods, or other modeling choices not captured by the summaries. It should be understood as a check on a particular unadjusted contrast rather than a substitute for the reported trial analysis.

Why can variability affect a comparison?

In general, the standard deviation describes dispersion among observed values, whereas a standard error describes uncertainty in an estimated quantity, such as a mean. Greater within-group variability generally makes a group mean or difference less precise when other factors are held constant. The groups’ posted standard deviations differ: 7.3 kg for pioglitazone, 2.8 kg for exenatide, and 4.4 kg for the combination. The posted counts also differ across groups. These features contribute to the uncertainty of comparisons, but they do not alone determine the results of a model-based analysis.

The comparison’s recalculated pooled standard deviation is 5.419 kg. This is a calculation-specific summary used for the unadjusted effect-size calculation, not a replacement for the separate group standard deviations. The record also gives the recalculated standardized effect as -0.572. A standardized effect expresses a contrast relative to a measure of spread, which can help describe magnitude on a scale independent of the original measurement units. It does not establish clinical importance.

Why might over-dispersion matter?

In general, over-dispersion means observed variability is greater than a statistical model expects. When a model assumes less variability than is present, its standard errors may be too small and its intervals or tests may appear more certain than warranted. Models can address over-dispersion when appropriate, but the needed method depends on the data and outcome. The primary outcome is continuous, and the registry does not report an over-dispersion assessment or a model adjustment for it. This general point is not evidence that over-dispersion occurred in this trial.

More broadly, statistical results depend on assumptions about the outcome distribution, independence of observations, model form, and the way missing measurements are handled. The registry does not report enough detail to evaluate these assumptions for the ANCOVA or ANOVA. The posted means and standard deviations remain useful descriptive summaries, while interpretation of the model-based tests must be limited to what the record reports.

10. Confidence Intervals and Precision

The recalculated exenatide-versus-pioglitazone mean difference is -3.1 kg. Its 95% confidence interval extends from -7.918 to 1.718 kg. A confidence interval is a range of values compatible with the observed data under the calculation’s assumptions; it communicates uncertainty in an estimated contrast rather than the spread of individual weights. The interval is broad relative to the point estimate and includes zero.

Because zero lies within the interval, the unadjusted recalculation does not sharply distinguish a difference from no difference. The limits also include contrasts in either direction. The interval should not be read as a guarantee that the true value lies within these particular limits, nor as evidence that every value in it is equally plausible. It reflects sampling uncertainty under the method used, not uncertainty from every possible source.

The p-value for this recalculation is 0.207. It is unadjusted and belongs to the same pairwise summary-statistic calculation as the confidence interval. The registry’s ANCOVA and ANOVA p-values of <0.05 are separate posted analyses, without reported effect estimates or intervals. Their precision cannot be assessed from the p-values alone.

The group standard deviations are 7.3 kg, 2.8 kg, and 4.4 kg, and the outcome counts are 10, 11, and 9. Variation in the observed outcome and the limited group counts contribute to uncertainty, but these summaries do not justify a claim about what would have been observed with a different design. The registry record does not report confidence intervals for the group means or for the sponsor’s analyses, so their precision is not available for direct assessment.

Finally, the endpoint wording does not resolve whether the posted weight values are final measurements or changes from baseline. Confidence intervals and estimates are only interpretable in relation to the quantity analyzed. The uncertainty applies to the stated recalculated contrast using the posted summary statistics; it does not resolve that ambiguity in the endpoint definition.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are stated as participants per arm, not total participants or total events. They apply to the recalculated exenatide-versus-pioglitazone comparison and are illustrations based on the observed effect, rather than the trial’s original design. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each scenario uses the stated fraction of the observed effect as the effect to detect.

For planning around 100% of the observed effect, the figures are 48 participants per arm for 80% power and 65 participants per arm for 90% power. For 75% of the observed effect, the figures are 86 participants per arm for 80% power and 115 participants per arm for 90% power. For 50% of the observed effect, the figures are 192 participants per arm for 80% power and 257 participants per arm for 90% power.

In plain terms, the 80% power column gives the estimated participants per arm for a design with an 80% chance of detecting the specified effect under the planning assumptions, if that effect is present. The 90% power column uses a higher chance of detection and therefore calls for more participants under the same assumptions. Smaller effects need more participants to detect, which is why the listed requirements rise as the planning effect becomes a smaller fraction of the observed effect.

The corresponding curve should be read as a visual representation of how the planning requirement changes with the effect used for planning. These figures are not observed power; they are planning illustrations and do not describe the trial’s achieved power. The registry record does not report the original sample-size justification or its design assumptions.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -3.100; pooled standard deviation σ = 5.419.
2. n = 2 × (1.96 + 0.842)² × 5.419² / -3.100² = 48.0, rounded up to 48 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -3.1                      # observed difference in means
sigma <- 5.41931244425167                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(845182)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 48 (80% power), 65 (90% power)
 75% of observed effect: n per arm = 86 (80% power), 115 (90% power)
 50% of observed effect: n per arm = 192 (80% power), 257 (90% power)
power.t.test: 49 
Simulated power at n = 48 per arm: 0.783 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Analyses Reported by the Sponsor

The registry reports ANCOVA and ANOVA analyses involving pioglitazone, exenatide, and the combination, each with p < 0.05. These are the trial’s own posted analyses. The record does not provide their effect estimates, confidence intervals, covariate details, or the precise contrasts represented by the tests. Their results therefore cannot be translated from the reported p-values into an estimated difference in kilograms.

The recalculation is a separate, simple, unadjusted comparison of exenatide with pioglitazone. It gives a mean difference of -3.1 kg, a 95% confidence interval from -7.918 to 1.718 kg, and an unadjusted p-value of 0.207. It does not reproduce the ANCOVA or ANOVA and does not include the combination group. The two results can differ because they concern different comparisons and methods; the record lacks details needed to determine the contribution of any particular modeling choice.

The recalculation is useful as a transparent check on the posted summaries, but it is not a replacement for the sponsor’s trial analysis. Conversely, the registry’s reported p-values do not provide the missing effect estimates or precision for the pairwise comparison. The trial’s reported analyses and the recalculation should be read as distinct pieces of statistical evidence, each with its own scope and limitations.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists three outcome measures and two analyses. It reports one secondary outcome. The primary outcome description includes body weight and beta cell function, while the posted primary numerical summaries concern body weight. The record names additional scientific aims and secondary measures, but numerical results for all of them are not included in the posted results.

Multiplicity arises when several outcomes, groups, contrasts, or analyses are tested. Conducting multiple tests can increase the chance that at least one result crosses a significance threshold by chance. Statistical plans may address this through prespecified primary outcomes, testing hierarchies, or adjustment procedures. The registry does not report a multiplicity strategy, so it is not possible to determine whether or how the reported p-values were adjusted for multiple testing. The reported p-values should not be assumed to be adjusted.

The registry reports both ANCOVA and ANOVA for the three groups, but it does not explain why both methods were used, which result was designated primary, or what contrasts were planned. It also does not give the full statistical analysis plan. The trial’s original sample-size justification, interim analyses, and detailed statistical plan are not reported in the registry record.

Missing-data handling is also not reported. The posted outcome counts are 10 for pioglitazone, 11 for exenatide, and 9 for the combination, while the arm descriptions state that 15 subjects or patients would be randomized to each group. The record does not explain the difference, describe missing measurements, or say whether analyses used all randomized participants or a subset. It does not identify an intention-to-treat, per-protocol, or other analysis population. These omissions limit assessment of possible bias and of how the analyzed groups relate to the randomized groups.

The registry identifies the trial as randomized and single-masked, but does not report who was masked or how masking was maintained. It does not describe the randomization sequence or allocation concealment. These details are relevant to evaluating possible bias, but their absence from the registry record does not establish that a particular method was or was not used.

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 non-serious adverse event terms

The serious adverse event summaries report 0 affected participants in each listed treatment group. The corresponding numbers at risk are 12 for pioglitazone, 15 for exenatide, and 15 for the combination. These figures describe the registry’s recorded serious adverse event counts during the reported observation period; the record does not specify that period in the adverse-event entry.

The record also lists other, non-serious adverse events: 2 affected among 12 at risk for pioglitazone, 1 among 15 for exenatide, and 1 among 15 for the combination. These are counts and denominators, not rates calculated for comparison. The serious adverse event figures should not be interpreted as proof that serious events cannot occur, particularly in a study with these recorded group counts.

Deaths are not reported in the adverse-event entry. The record does not provide serious-event terms, with no top serious terms listed, and it does not describe event timing, severity criteria, withdrawals due to adverse events, or statistical comparisons of safety. The adverse-event counts therefore offer limited descriptive information and do not establish comparative safety.

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

The posted results describe body-weight summaries for three treatment groups and report ANCOVA and ANOVA p-values of <0.05. The separate unadjusted recalculation estimates an exenatide-versus-pioglitazone difference of -3.1 kg, with a 95% confidence interval from -7.918 to 1.718 kg and p = 0.207. These are different analyses, and the record does not provide enough detail to equate their findings or identify the exact contrasts underlying the sponsor’s reported tests.

The recalculated interval includes zero and spans values in both directions. It therefore conveys considerable uncertainty about the pairwise contrast. The registry’s reported p-values indicate that the named analyses met the stated threshold, but without effect estimates and confidence intervals they do not show the size or precision of the corresponding result. Neither the p-values alone nor the group means alone establish clinical importance.

The endpoint wording also leaves uncertainty about whether the recorded values represent weight at the end of the study or change from baseline. The timeframe refers to baseline and 6 months, while the population description refers to weight at completion. Without clarification, the numerical summaries should not be assigned a more specific meaning than the posted body-weight outcome.

The results do not establish that one treatment is clinically preferable, nor do they settle the effects on beta cell function, glycemic control, fat topography, or plasma lipids. The brief summary describes these as study aims, but the posted numerical results described in the record do not provide estimates for all of them. The registry’s safety counts are also descriptive and do not provide a definitive comparison of risks.

Overall, the record supports reporting the observed summaries and the stated analyses, while keeping uncertainty and missing information visible. The findings should be interpreted as statistical descriptions of this trial’s posted results rather than as medical advice or a conclusion about what an individual should receive.

17. Why This Trial Matters Statistically

This trial illustrates why a reported p-value and an effect estimate answer different questions. The registry’s ANCOVA and ANOVA results are reported as p < 0.05, but the record does not provide the associated estimates or confidence intervals. A reader can see the reported threshold result, yet cannot assess from those p-values alone the size of the estimated differences or their precision.

The recalculated pairwise contrast makes the importance of scope clear. It compares exenatide with pioglitazone without adjustment, while the registry’s named analyses include all three groups. Its estimate is -3.1 kg, and its confidence interval from -7.918 to 1.718 kg includes zero. This is not a contradiction that can be resolved by choosing one result over the other: the methods and comparisons are not documented as equivalent. Statistical interpretation depends on knowing what quantity each analysis estimates.

The trial also shows how descriptive variability and the analyzed counts matter. Group standard deviations range from 2.8 kg to 7.3 kg, while the posted outcome counts range from 9 to 11. Such variation and limited counts affect the precision available from the outcome summaries. The arm descriptions list 15 subjects or patients per group, and the record does not explain how those planned group sizes relate to the outcome counts.

Finally, an outcome’s label and timing need to be sufficiently clear for its estimates to be interpreted. Here the timeframe includes baseline and 6 months, but the population description refers to weight at completion without specifying whether the outcome is a final value or a change. Statistical reporting is strongest when it makes the estimand, analysis population, model, missing-data handling, and uncertainty explicit. Their absence limits what can be concluded from this record.

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.

The outcome description does not clearly define the numerical quantity summarized. It names body weight in kilograms and gives baseline and 6 months as the timeframe, but also says weight at study completion was analyzed. The record does not specify whether the means represent final weight, change from baseline, or another measure. This ambiguity limits interpretation of the group values and the recalculated contrast.

The planned group descriptions say 15 subjects or patients per arm, whereas the primary outcome summaries report counts of 10, 11, and 9. The overall enrollment is listed as 43. The registry does not explain the relationship among these figures, identify reasons for differences, or state how missing outcome values were handled. The analysis population is not reported.

The registry’s ANCOVA and ANOVA results are each reported only as p < 0.05. Effect estimates, confidence intervals, model covariates, planned contrasts, and analysis assumptions are not reported. Without this information, the sponsor’s analysis cannot be evaluated in detail or directly compared with the unadjusted pairwise recalculation.

The condition list includes Type 2 Diabetes, Healthy, and Impaired Glucose Tolerance, while the title and brief summary focus on type 2 diabetes. The record does not clarify how these labels relate to eligibility or analyzed participants. It also does not report who was masked, how randomization was conducted, the original sample-size rationale, interim analyses, or a multiplicity strategy.

The numerical results for the primary weight outcome do not provide a complete account of the other stated aims, including beta cell function, glycemic control, fat topography, and plasma lipid levels. Safety reporting is limited to counts, denominators, and the absence of listed serious adverse events; deaths and event details are not reported. These limitations restrict conclusions about treatment effects, precision, and safety.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Effect of Pioglitazone and Exenatide on Body Weight and Beta Cell Function — NCT00845182.
• Registry record last updated 2018-01-18; status last verified 2017-05.
• Results first posted 2015-03-17.
• 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, randomized, single-masked, parallel-group phase 4 trial sponsored by The University of Texas Health Science Center at San Antonio. It compares pioglitazone, exenatide, and the combination, with a stated enrollment of 43. The primary outcome is body weight in kilograms, described as a mean with standard deviation and a timeframe of baseline and 6 months.

The posted group means are 5.5 kg, 2.4 kg, and 2.7 kg for pioglitazone, exenatide, and the combination, respectively. The registry reports ANCOVA and ANOVA p-values of <0.05, without estimates or confidence intervals. An unadjusted recalculation of exenatide versus pioglitazone gives a mean difference of -3.1 kg, a 95% confidence interval from -7.918 to 1.718 kg, and p = 0.207.

Interpretation is limited by unclear endpoint wording, unexplained differences between the arm descriptions and analyzed counts, and missing details about analysis methods, missing data, and the analysis population. The registry’s safety entry reports no serious adverse events in the listed groups but provides limited further detail.