This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for MK0431 (Sitagliptin) Pioglitazone Add-on Study for Patients With Type 2 Diabetes Mellitus (0431-055)(COMPLETED) (NCT00372060), a randomized phase 3 study. The primary outcome is Change From Baseline in Hemoglobin A1c (HbA1c ) at Week 12.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00372060 |
| Phase | Phase 3 |
| Condition | Type 2 Diabetes Mellitus |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 134 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Merck Sharp & Dohme LLC |
| Results posted | Yes |
2. Clinical Question
Population
Patients with type 2 diabetes mellitus whose glycemic control was inadequate on pioglitazone monotherapy. The primary analysis used the Full Analysis Set: randomized patients who took at least 1 dose of study medication and had a baseline measurement and at least one post-baseline value. The registry record reports enrollment of 134 participants.
Intervention
Sitagliptin phosphate, also identified as MK0431, added to pioglitazone. The arm descriptions identify one group as MK0431 plus pioglitazone. The registry record does not provide further detail about the medication schedule in the arm descriptions.
Comparator
Placebo with pioglitazone, with the group descriptions also naming sitagliptin. The listed comparison is between the Sitagliptin / Sitagliptin and Placebo / Sitagliptin groups. The registry record does not fully explain the slash-separated labels or give a more detailed treatment schedule.
Primary question
What was the difference between the groups in mean change from baseline in hemoglobin A1c at Week 12? HbA1c is a measure used to assess glycemic control. The registry reports group least-squares means and an ANCOVA estimate of the difference, with a 95% confidence interval and p-value.
The clinical question is about the measured Week 12 change in HbA1c, not about every possible effect of sitagliptin or about longer-term outcomes. The study also listed change from baseline in fasting plasma glucose at Week 12 as a secondary outcome. The registry record describes the study purpose as assessing safety and efficacy in patients whose control on pioglitazone alone was inadequate. That stated purpose gives the clinical context, while the posted primary result addresses a specific laboratory endpoint over a defined follow-up period.
3. Trial Design
The registry describes a completed, randomized, double-blind, parallel-group Phase III study sponsored by Merck Sharp & Dohme LLC. The condition listed is type 2 diabetes mellitus, and the study summary focuses on patients with inadequate glycemic control while taking pioglitazone as monotherapy. Enrollment was 134. The registry record lists the study start as 2006-08-21 and primary completion as 2008-02-05.
The arm descriptions identify MK0431, or sitagliptin, with pioglitazone and a placebo/MK0431 group with pioglitazone. The posted Week 12 outcome groups are titled Sitagliptin / Sitagliptin and Placebo / Sitagliptin. Since the study is described as parallel, the comparison is between randomized groups; the exact meaning and timing of the slash-separated group names are not further explained in the registry record. The intervention list names sitagliptin phosphate, pioglitazone, and placebo.
The primary endpoint was change from baseline in HbA1c at 12 Weeks. The analysis population was the Full Analysis Set with Last Observation Carried Forward. The registry defines this population as randomized patients who took at least 1 dose of study medication and had both a baseline and at least one post-baseline value. Group sizes in the primary outcome are reported as 66.0 and 68.0. The registry also lists four outcome measures and three analyses, but a detailed account of all analysis procedures is not reported in the registry record.
Randomization and double masking are stated design features; the method used to generate the allocation sequence, conceal allocation, or maintain masking is not reported in the registry record. Further detail about stratification, treatment adherence, and any changes to the planned study procedures is also 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.
Before assignment: Following a screening period of at least 4 weeks and a 2 or 8-week observation period, patients who were on pioglitazone monotherapy for at least 8 weeks and met all other entry criteria were randomized to receive: sitagliptin/sitagliptin or placebo/sitagliptin. The starting dose of sitagliptin was 50 mg for all patients.
Period I - Double-blind (Weeks 0-12)
| Stage | Sitagliptin / Sitagliptin | Placebo / Sitagliptin |
|---|---|---|
| STARTED | 66 | 68 |
| COMPLETED | 63 | 67 |
| NOT COMPLETED | 3 | 1 |
Reasons for not completing this period:
| Reason | Sitagliptin / Sitagliptin | Placebo / Sitagliptin |
|---|---|---|
| Adverse Event | 2 | 0 |
| Lack of Efficacy | 0 | 1 |
| No improvement in hyperglycemia | 1 | 0 |
Period II - Open-label (Weeks 12-52)
| Stage | Sitagliptin / Sitagliptin | Placebo / Sitagliptin |
|---|---|---|
| STARTED | 63 | 67 |
| COMPLETED | 49 | 59 |
| NOT COMPLETED | 14 | 8 |
Reasons for not completing this period:
| Reason | Sitagliptin / Sitagliptin | Placebo / Sitagliptin |
|---|---|---|
| Adverse Event | 3 | 0 |
| Lack of Efficacy | 9 | 8 |
| Withdrawal by Subject | 2 | 0 |
Analysis population for the primary outcome, as defined in the registry: Analysis in the Full Analysis Set with Last Observation Carried Forward (The Full Analysis Set population includes all randomized patients who took at least 1 dose of study medication and had both a baseline and at least one post-baseline value.)
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.
| Characteristic | Unit | Sitagliptin / Sitagliptin | Placebo / Sitagliptin | Total |
|---|---|---|---|---|
| Participants analysed | 66 | 68 | 134 | |
| Age, Continuous | years | 57.8 (SD 9.9) | 59.0 (SD 9.2) | 58.4 (SD 9.5) |
| Sex: Female, Male: Female | Participants | 28 | 19 | 47 |
| Sex: Female, Male: Male | Participants | 38 | 49 | 87 |
| Fasting Plasma Glucose | mg/dL | 146.8 (SD 33.1) | 151.3 (SD 34.6) | 149.1 (SD 33.8) |
| Hemoglobin A1c (HbA1c) | Percent | 7.7 (SD 0.9) | 7.6 (SD 0.8) | 7.7 (SD 0.8) |
6. Primary Endpoint
The registry lists the primary outcome as Change From Baseline in Hemoglobin A1c (HbA1c ) at Week 12, measured over 12 Weeks.
| Element | Registry entry |
|---|---|
| Result type | Least Squares Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | Percent |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is change from baseline in hemoglobin A1c (HbA1c) at Week 12, measured as a percent. The baseline measurement was obtained at randomization before study medication. The reported outcome is a continuous measure, and the registry presents each group’s least-squares mean change and its 95% confidence interval.
HbA1c is used in diabetes research as a marker of glycemic control. As a change score, the endpoint expresses how the measured value at follow-up differs from the participant’s baseline value. A group mean summarizes the average change in the analysis population; it does not describe every participant’s individual course. Least-squares means are model-based estimates, rather than simple raw averages, although the specific covariates and full model specification are not reported in the registry record.
Week 12 provides a defined assessment point for comparing the groups over the study’s primary timeframe. The endpoint is therefore informative about the measured short-term change in HbA1c under the study conditions. It does not, on its own, describe other aspects of diabetes management, longer-term clinical outcomes, or the full safety profile. The registry lists change from baseline in fasting plasma glucose at Week 12 as a secondary outcome, but the primary question concerns HbA1c.
Using a continuous endpoint retains information about the magnitude of change rather than reducing the result to a threshold category. General statistical context: a mean can be useful for comparing groups, but it may conceal variation among individual measurements. The confidence intervals and group-level estimates should therefore be understood as summaries of the analysis, not as predictions for every patient.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
At Week 12, the least-squares mean change in HbA1c was -0.4 in the Sitagliptin / Sitagliptin group, with a 95% confidence interval from -0.6 to -0.3. In the Placebo / Sitagliptin group, the reported least-squares mean change was 0.4, with a 95% confidence interval from 0.3 to 0.5. These are changes from baseline, so the signs indicate the direction of the estimated change within each group.
The sponsor-reported ANCOVA estimate of the least-squares mean difference between the groups was -0.8, with a 95% confidence interval from -1.0 to -0.6 and p < 0.001. The confidence interval lies below zero, and the reported p-value indicates that the observed data are inconsistent with a null difference under the assumptions of the stated analysis. This is a statistical description of the posted result, not a determination of clinical importance. A threshold for a clinically meaningful difference is not reported in the registry record.
General statistical context: a confidence interval describes uncertainty around an estimated quantity under the analysis model. A narrower interval indicates greater precision than a wider one, but the interval does not guarantee that the true effect falls within it. The interval also does not account for sources of uncertainty that are not represented in the model, such as problems in measurement or limitations in the design. The reported interval is for the estimated group difference, not a range containing individual participants’ changes.
The groups had 66.0 and 68.0 participants in the primary outcome analysis. Those counts describe the analyzed groups and should not be confused with enrollment alone. The Full Analysis Set definition excludes randomized patients who did not take study medication or did not have the required baseline and post-baseline measurements. Details on the number of randomized participants excluded from the endpoint analysis and the reasons for any missing measurements are not reported in the registry record.
The study also reports adverse-event counts through Week 12 for the two listed groups: 3 and 1 participants with serious adverse events, and 21 and 23 with other adverse events, respectively. The record gives pooled sitagliptin counts through Week 52 of 7 serious and 80 other adverse events. Denominators for these safety counts are not reported in the registry record, so rates cannot be interpreted from those counts alone. Death counts are not reported in the registry record. The listed serious-event terms and percentages are not assigned to a particular group in the record.
Overall, the posted primary result is a statistically distinguishable Week 12 difference in HbA1c change between the analyzed groups under the reported ANCOVA. The registry information does not establish how important that difference is to patients, nor does it answer questions beyond the recorded endpoint, analysis population, and follow-up.
9. Statistical Methods Explained
What does the ANCOVA estimate represent?
The reported comparison used ANCOVA, or analysis of covariance, and is described as a least-squares mean difference. General statistical context: ANCOVA estimates a group contrast while accounting for selected covariates in a statistical model. The resulting least-squares means are model-based group estimates, often described as adjusted means. The specific covariates, model terms, and assumptions are not reported in the registry record, so the adjustment cannot be independently characterized from the record.
The primary endpoint is continuous: each participant has a measured change in HbA1c, and the analysis summarizes the groups using means. The sponsor’s estimate of -0.8 is the reported difference between the least-squares mean changes. The negative sign reflects the direction of the contrast as presented for the Sitagliptin / Sitagliptin group relative to the Placebo / Sitagliptin group. The group-specific means help make that direction interpretable: the reported change is -0.4 in one group and 0.4 in the other.
How should the confidence interval and p-value be read?
General statistical context: a confidence interval combines the point estimate with a range of values compatible with the data and the model at the stated confidence level. The 95% confidence interval for the reported difference is -1.0 to -0.6. Its location below zero is consistent with a negative estimated contrast under the stated group ordering. Confidence intervals are useful for considering precision as well as direction; they are not guarantees about the true value and do not describe individual response variation.
The reported p-value is p < 0.001. General statistical context: a p-value measures how unusual data at least as incompatible with a specified null hypothesis would be if that null hypothesis and the statistical model were correct. It is not the probability that the null hypothesis is true, and it does not measure the probability that the result occurred by chance in an everyday sense. A small p-value alone does not establish clinical importance, eliminate bias, or show that every assumption of the analysis is satisfied.
What does the analysis population mean?
The primary analysis used a Full Analysis Set with Last Observation Carried Forward. The registry defines the Full Analysis Set as including randomized patients who took at least 1 dose of study medication and had both a baseline and at least one post-baseline value. This population is not identical to all randomized participants by definition. The arm counts in the primary outcome are 66.0 and 68.0; the number and characteristics of patients not included in that endpoint analysis are not reported in the registry record.
Last Observation Carried Forward, or LOCF, is a missing-data approach that substitutes a participant’s last recorded measurement for a later missing value. General statistical context: this creates an analysis value without observing the missing endpoint itself. Its validity depends on assumptions about how missingness and the measured outcome relate over time. LOCF can understate or overstate change if the unobserved later value would have differed from the carried-forward value. The registry record identifies LOCF but does not report the amount or pattern of missing data, reasons for missingness, or sensitivity analyses using alternative approaches.
What can be inferred from the posted safety counts?
The registry records counts of participants affected by serious and other adverse events. For the Sitagliptin/Sitagliptin group through Week 12, the counts are 3 for serious adverse events and 21 for other adverse events. For the Placebo/Sitagliptin group through Week 12, the counts are 1 and 23. Pooled sitagliptin data through Week 52 list 7 serious and 80 other adverse events. These are counts, not rates: the relevant denominators are not reported in the registry record. The pooled longer-follow-up group should also be kept distinct from the Week 12 group comparison.
The record lists serious-event terms with percentages, including coronary artery stenosis, cerebral infarction, hypertension, ligament injury, lower limb fracture, and enteritis infectious. It does not identify the group attribution for those terms in the listed information. The percentages should not be treated as group-specific risks or used to compare arms without the corresponding denominators and attribution.
What planning and analysis details are missing?
The registry describes a sample-size planning illustration based on an observed effect, a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. Such planning illustrations are hypothetical design scenarios, not observed trial power. The registry record does not report a sample-size justification for the original trial. Interim analyses, stopping rules, a multiplicity procedure, and a complete statistical analysis plan are also not reported in the registry record. No inference about those procedures should be made from the reported endpoint analysis alone.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| ANCOVA | Least-squares Mean Difference -0.8 | -1.0 to -0.6 | < 0.001 |
The registry reports the sponsor’s ANCOVA analysis as a least-squares mean difference of -0.8, with a 95% confidence interval from -1.0 to -0.6 and p < 0.001. This is the inferential comparison posted for the Week 12 primary endpoint. It is based on the reported analysis method and is distinct from simply subtracting displayed group summaries, because the sponsor identifies the estimate as an ANCOVA least-squares mean difference.
The registry record does not give a separate recalculated estimate or a competing analysis result. It also does not flag a discrepancy between the sponsor’s analysis and a recalculation. The covariates and full model specification behind the ANCOVA are not reported in the registry record, limiting independent reconstruction of that model-based estimate. For interpretation, the posted ANCOVA result is the available analysis, with the group least-squares means and confidence intervals providing additional context.
The reported p-value and confidence interval describe statistical evidence and precision under the sponsor’s analysis. They do not by themselves establish clinical importance, remove concerns about missing data, or extend the result to outcomes and follow-up periods not captured by the primary endpoint.
11. Multiplicity, Missing Data and Analysis Populations
The registry lists the Week 12 change in HbA1c as the primary outcome and change from baseline in fasting plasma glucose at Week 12 as a secondary outcome. It also reports four outcome measures and three analyses. The registry record does not explain whether the secondary outcome and other analyses were part of a prespecified hierarchy, whether any multiplicity adjustment was used, or how the primary endpoint was selected in relation to the other analyses. The reported p-value should therefore be read as the result for the stated comparison, not as evidence that all possible analyses were protected against false-positive findings.
Missing data are relevant because the primary analysis used Last Observation Carried Forward. The registry record defines the Full Analysis Set as randomized patients who took at least 1 dose of study medication and had a baseline and at least one post-baseline value. The number of randomized participants who did not meet that definition, the extent of missing Week 12 measurements, and the reasons for missingness are not reported in the registry record. The record also does not describe whether additional sensitivity analyses were conducted to assess the effect of the LOCF assumption.
General statistical context: multiplicity can arise when several outcomes, time points, or analyses are tested. Without a stated plan for handling multiple comparisons, an individual p-value does not necessarily have the same interpretation as a p-value from a single prespecified test. Missing-data methods likewise depend on assumptions about why values are absent and what those values might have been. Neither issue can be resolved by the reported confidence interval alone.
The detailed statistical analysis plan, interim analyses, and sample-size justification are not reported in the registry record. The registry describes a planning illustration based on the observed effect, but that does not document the original design rationale and must not be interpreted as observed power. These omissions constrain assessment of prespecification, analysis choices, and the robustness of the primary result.
12. What the Result Does, and Does Not, Mean
The primary result describes a difference between randomized groups in average change from baseline in HbA1c at Week 12. The reported least-squares mean difference was -0.8, with a 95% confidence interval from -1.0 to -0.6 and p < 0.001. The confidence interval conveys uncertainty around the estimated group contrast under the ANCOVA model. The reported value is a group-level estimate, not a forecast of the change for a particular participant.
The result does not, by itself, show how participants felt or functioned, establish a threshold for clinical importance, or describe outcomes beyond the endpoint and timeframe reported. A threshold for a clinically meaningful HbA1c difference is not reported in the registry record. The statistical evidence for a group difference should therefore be distinguished from judgments about its clinical relevance.
The finding is limited to the stated study population and analysis. The Full Analysis Set includes randomized participants who took study medication and had the required baseline and post-baseline values, with LOCF used for the primary analysis. The number and characteristics of participants excluded from that analysis, the pattern of missing measurements, and sensitivity analyses are not reported in the registry record. The result also does not resolve the impact of any analysis procedures not described in the record.
Safety information is presented as counts, with Week 12 group counts and pooled sitagliptin counts through Week 52. Denominators for those counts are not reported in the registry record, so the information does not support interpreting them as rates or making a direct risk comparison. The listed serious-event terms are not assigned to a particular group in the posted information, and deaths are not reported in the registry record.
Accordingly, the result supports a statistical comparison of the groups on the recorded primary endpoint under the stated analysis. It does not establish conclusions about every possible benefit or harm, other populations, long-term outcomes, or unreported comparisons. Those questions require information beyond the recorded Week 12 HbA1c analysis.
13. Why This Trial Matters Statistically
This study illustrates why the estimand and analysis population matter when interpreting a clinical-trial result. The primary endpoint is a change from baseline, and the comparison is reported as a least-squares mean difference from ANCOVA. Those details clarify that the headline estimate is a model-based contrast in mean change, rather than a raw count or a simple description of individual outcomes. The group estimates and their confidence intervals give context for the direction and precision of that contrast.
It also highlights the distinction between statistical evidence and clinical interpretation. The reported p-value is small, and the confidence interval for the group difference lies below zero, but the registry does not state a threshold for clinical importance. A statistically distinguishable estimate is not automatically an estimate of meaningful benefit to every participant. The outcome, time point, and analysis population define what the result can address.
LOCF and the Full Analysis Set definition are important to understanding how the endpoint was analyzed. These choices determine which participants contributed and how missing follow-up values were handled. Because the registry does not report the extent or pattern of missingness or sensitivity analyses, readers cannot evaluate how dependent the result may be on those choices using the registry information alone.
The safety record also demonstrates why counts need denominators and clear group attribution. Counts through different follow-up periods cannot be interpreted as comparable rates without the relevant population sizes and a consistent timeframe. The registry provides useful descriptions of the primary analysis and adverse-event counts, while leaving important details about multiplicity, missingness, model specification, and sample-size justification unreported.
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 record gives limited detail about how randomization and masking were implemented. It identifies the study as randomized, double-blind, and parallel, but does not report the allocation-generation method, allocation concealment, or procedures used to preserve masking. Detailed treatment schedules are also not fully clarified by the slash-separated arm labels, so those labels should not be assigned a more specific sequence interpretation than the record states.
The primary result is presented as an ANCOVA least-squares mean difference, but the covariates, model specification, and diagnostic checks are not reported in the registry record. This makes it difficult to reproduce or fully assess the model-based estimate. The analysis used LOCF, but the amount and pattern of missingness, reasons for missing data, and sensitivity analyses are not reported. The Full Analysis Set definition also means the primary analysis does not necessarily include every randomized participant.
The registry lists a secondary fasting plasma glucose outcome, four outcome measures, and three analyses, but does not report a multiplicity strategy or a complete analysis plan. A sample-size justification for the original trial and any interim analyses are not reported in the registry record. The listed planning illustration is not a substitute for the original design rationale.
Safety interpretation is limited by absent denominators for the posted adverse-event counts. The group attribution for listed serious-event terms is not specified, and deaths are not reported in the registry record. Follow-up information also differs between the Week 12 group counts and pooled Week 52 counts. These limitations restrict interpretation of event frequency and comparison across groups or follow-up periods.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: MK0431 (Sitagliptin) Pioglitazone Add-on Study for Patients With Type 2 Diabetes Mellitus (0431-055)(COMPLETED) — NCT00372060.
- Registry record last updated 2017-05-12; status last verified 2017-04.
- Results first posted 2009-03-31.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
This completed Phase III study examined sitagliptin, identified as MK0431, in patients with type 2 diabetes mellitus whose glycemic control was inadequate on pioglitazone monotherapy. The registry describes a randomized, double-blind, parallel-group design and enrollment of 134 participants.
The primary outcome was change from baseline in HbA1c at Week 12, analyzed in the Full Analysis Set with LOCF. The reported ANCOVA least-squares mean difference was -0.8, with a 95% confidence interval from -1.0 to -0.6 and p < 0.001. The registry also lists fasting plasma glucose change at Week 12 as a secondary outcome.
Important details—including the complete statistical plan, multiplicity handling, missing-data pattern, ANCOVA covariates, and original sample-size justification—are not reported in the registry record. Adverse-event counts are posted, but denominators needed to interpret them as rates are not reported.