Efficacy and Safety of Sitagliptin/Metformin Fixed-Dose Combination (FDC) Compared to Glimepiride in Participants With Type 2 Diabetes Mellitus (MK-0431A-202): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of Sitagliptin/Metformin Fixed-Dose Combination (FDC) Compared to Glimepiride in Participants With Type 2 Diabetes Mellitus (MK-0431A-202) (NCT00993187), a randomized phase 4 study. The primary outcome is Change From Baseline in Hemoglobin A1C (HbA1C) at Week 30.

2. Clinical Question

3. Trial Design

This was a phase 4, multicenter, randomized, double-blind, parallel-group study comparing sitagliptin/metformin fixed-dose combination with glimepiride. The registry lists the study as completed and reports enrollment of 292. The recorded study dates are a start date of 2010-05-04 and a primary completion date of 2013-10-29. These dates describe the registry record and do not, by themselves, explain the timing of individual participants' visits.

Both groups received active study medication alongside matching placebo tablets. This matching-placebo arrangement is consistent with the registry's double-blind designation: participants and study personnel were intended to be masked to treatment assignment, although the details of how masking was maintained are not reported in the registry record. The parallel design means participants were assigned to one of the treatment groups rather than crossing between them, as a general description of that design.

The treatment period was 30 weeks. The combination group started at 50/500 mg twice daily, with an increase to 50/1000 mg twice daily described in the arm record. The glimepiride group started at 1 mg once daily, with further titration described as being based on investigator judgment. The exact timing and rules for these dosing changes are not reported in the registry record.

The primary analysis population was not simply all enrolled participants: the registry defines the Full-Analysis-Set population using randomization, a baseline measurement, exposure to at least one dose, and at least one post-randomization measurement. Primary outcome summaries include 138 participants in the sitagliptin/metformin group and 137 in the glimepiride group. The safety summaries use different denominators, 146 and 144, respectively. The registry does not explain the difference between these analysis populations beyond the primary population definition.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Full-Analysis-Set (FAS) Population included all randomized participants who had a baseline measurement, consumed at least one dose of study medication, and had at least one post-randomization measurement.

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 Change From Baseline in Hemoglobin A1C (HbA1C) at Week 30, measured over Baseline and Week 30.

The primary endpoint was change from baseline in hemoglobin A1c (HbA1c) at Week 30. HbA1c is described in the registry as a blood marker used to report average blood glucose levels over a prolonged period. It is expressed as a percentage of total hemoglobin. Change was defined as the Week 30 value minus the baseline value, so a negative change represents a lower Week 30 measurement than at baseline.

The endpoint is a continuous measure: each participant's result can take a range of values, rather than falling into only a yes-or-no category. The registry reports group-level least-squares means and 95% confidence intervals. Least-squares means are model-based estimates; the registry identifies ANCOVA as the method for the reported comparison. The model's covariates and other implementation details are not reported in the registry record.

Using a change score focuses the comparison on movement from the starting measurement over the treatment period. A group mean gives a concise description of the average change, while the between-group estimate addresses how the model-based mean changes compare. Neither summary describes every participant's response. The confidence intervals convey statistical uncertainty around the estimates, not the range of individual participant outcomes.

The endpoint is relevant to the stated study question because the registry's primary hypothesis concerns the difference between treatments in HbA1c reduction after 30 weeks. That hypothesis concerns a measured laboratory outcome over the specified period. It does not, on its own, answer broader questions about long-term outcomes or other aspects of diabetes care.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous, so its analysis summarizes average change rather than the proportion reaching a yes-or-no outcome. The registry reports least-squares means for the two treatment groups and an ANCOVA comparison of those means. In general, ANCOVA is a regression-based method that can compare group outcomes while accounting for specified covariates. Which covariates were included, how they were coded, and how model assumptions were assessed are not reported in the registry record. The model-based estimates should therefore be read as the results of the posted analysis, rather than as a fully reproducible calculation from the displayed group summaries alone.

A difference in least-squares means is a contrast between model-based group estimates. Its sign depends on the order in which groups are contrasted. The registry reports a difference of -0.8 for the primary outcome, with a 95% confidence interval from -1.0 to -0.6. Because the interval does not cross zero, the reported estimate is statistically distinguishable from no difference at the confidence level stated. In general, the confidence interval also communicates precision: narrower intervals indicate more precise estimates than wider intervals, all else equal. It does not convey the spread of individual outcomes.

The p-value for the primary comparison is reported as p < 0.001. In general, a p-value describes how incompatible the observed data are with a specified null model, under the assumptions of the analysis. It is not a measure of effect size, a measure of clinical importance, or the probability that the result occurred by chance. The registry does not report whether the primary p-value was adjusted for any other hypothesis tests.

How should the secondary comparisons be read?

The secondary outcomes include both continuous and binary measures. A continuous outcome, such as change in body weight, is described with group means and a mean contrast. A binary outcome, such as whether a participant had one or more episodes of hypoglycemia, is described as a percentage and can also be compared using a risk difference. In general, a risk difference expresses the difference between two proportions, with its sign determined by which group is subtracted from which. The registry's reported comparison and the recalculated comparison may use opposite group orderings, so the sign should be interpreted together with the group labels and the stated contrast.

For the hypoglycemia and HbA1c target outcomes, the posted ANCOVA comparisons report differences in percent. The recalculated summaries are simple, unadjusted comparisons based on the posted group counts and percentages. They do not reproduce the sponsor's model-based analysis. The body-weight recalculation is likewise an unadjusted comparison of posted summaries; the standard deviation used in that calculation is identified as approximate. In general, an unadjusted summary can differ from an adjusted model estimate because the methods answer related but not identical questions, and may use different modeling assumptions.

What do the planning illustrations mean?

The planning illustrations use the observed effects as reference points, with alternative rows representing planning for 75% or 50% of the observed effect. They estimate participants per arm for 80% power and participants per arm for 90% power under the stated assumptions: a two-sided significance level of 0.05, equal allocation, and a normal approximation. These are hypothetical design illustrations, not the trial's original sample-size calculation. The sample-size justification is not reported in the registry record. The illustrations are not observed power, do not show the probability that a treatment effect is true, and do not change the uncertainty in the results from this completed trial.

In general, over-dispersion occurs when variability in the data is greater than assumed by a model. If present and not appropriately accounted for, it can make standard errors and confidence intervals misleading. The registry record does not report an over-dispersion assessment; for the continuous primary endpoint, the relevant modeling assumptions and diagnostics are also not reported. No inference about whether such issues occurred can be made from the posted summaries alone.

The reported 95% confidence intervals apply to the specific estimates and analysis methods stated. They do not automatically account for multiple testing, missing outcomes, or model selection. Whether those issues were addressed in the sponsor's broader statistical plan is not reported in the registry record. For this reason, the primary comparison and secondary comparisons should be understood according to their reported analysis descriptions, without assuming unreported adjustments or procedures.

10. Analyses Reported by the Sponsor

The primary result is the registry's reported ANCOVA comparison: a difference in least-squares mean HbA1c change of -0.8, with a 95% confidence interval from -1.0 to -0.6 and p < 0.001. No recalculated primary contrast is reported, so there is no independent summary-statistic calculation to set beside this model-based estimate.

For secondary outcomes, the registry also reports ANCOVA comparisons. The separate recalculations are simple unadjusted comparisons from posted group summaries, and they should not be treated as substitutes for those analyses. Contrast direction can also change with the order of the groups: for example, the registry's body-weight difference is reported as -1.72, while the unadjusted mean difference is expressed in the opposite direction. The posted ANCOVA value and the unadjusted estimate are close in magnitude, but they are not the same estimand or method. For binary secondary outcomes, the reported difference in percent and the unadjusted risk difference likewise use different presentation conventions and may reflect different analysis methods.

The registry record does not flag a discrepancy between the reported analysis and recalculation. The sponsor's own analysis therefore remains the principal result to interpret, while the unadjusted calculations serve as a secondary check on the posted summaries. Their agreement or proximity does not establish that all modeling assumptions hold, nor does it replace missing information about the analysis plan.

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies one primary outcome and lists secondary outcomes, but the multiplicity strategy is not reported in the registry record. It is therefore not possible to determine whether the secondary comparisons were treated as confirmatory, adjusted as a family, or interpreted as exploratory. In general, multiple testing matters because each additional hypothesis test creates another opportunity for a result that appears statistically notable even when the underlying null hypotheses are true. The reported secondary p-values should be interpreted with that limitation in mind.

The primary analysis population is defined as randomized participants who had a baseline measurement, took at least one dose of study medication, and had at least one post-randomization measurement. That definition clarifies which participants could enter the reported primary analysis, but it is not equivalent to an analysis of every randomized participant. The registry does not report how many participants lacked required measurements, why measurements were unavailable, or whether missingness differed by treatment group.

Methods for handling missing data are not reported in the registry record. It is not possible to tell from the posted results whether missing outcome values were imputed, handled through a model-based method, or treated in another way. The registry also does not report sensitivity analyses assessing how conclusions might change under different assumptions about missing outcomes. These details matter because an analysis based only on participants with measured outcomes can be affected if the availability of measurements is related to outcome or treatment.

The reported primary comparison is ANCOVA, but its covariates, model specification, and diagnostic procedures are not reported. Interim analyses and any stopping rules are not reported in the registry record. The sample-size justification is not reported either. The planning illustrations are separate hypothetical calculations based on observed effects and do not document the original design rationale.

These omissions do not change the values posted for the primary comparison, but they limit the ability to reproduce the analysis or assess the full prespecification and robustness of the results. Statistical interpretation should distinguish the reported estimates and uncertainty from procedures that are not documented.

12. 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: duodenal ulcer haemorrhage (0.7%), infective spondylitis (0.7%), influenza (0.7%), contusion (0.7%), overdose (0.7%).

The safety summaries report serious adverse events for 8 of 146 participants in the sitagliptin/metformin group and 9 of 144 in the glimepiride group. Other adverse events are reported for 62 of 146 and 54 of 144 participants, respectively. These are counts alongside the registry's at-risk denominators; the summaries do not by themselves establish that a study treatment caused an event or explain its clinical context.

The registry lists six serious-event terms, each with a reported value of 0.7%: duodenal ulcer haemorrhage, infective spondylitis, influenza, contusion, overdose, and hyperglycaemia. The record does not provide arm-specific counts for these terms in the posted summary, so they should not be attributed to a particular treatment group from this information alone. Deaths are not reported in the registry record.

In general, adverse-event counts are descriptive safety data. Interpretation depends on event definitions, exposure time, severity, attribution, and the completeness of follow-up. Those details are not fully reported in the registry record. The figures therefore describe what was recorded in the safety summaries and do not support a causal conclusion or medical advice.

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

The primary result is a comparison of average HbA1c change over 30 weeks in the study's defined Full-Analysis-Set population. The posted ANCOVA estimate and confidence interval indicate a statistically clear difference between the two groups under the reported model. The interval gives a sense of uncertainty in that group contrast; it does not say that each participant experienced a change near the group average, nor does it describe the full range of individual outcomes.

The result does not establish what would occur after the 30-week treatment period. It does not answer questions about outcomes that were not measured or reported, and it cannot establish whether the estimated contrast applies to people or care settings outside this study. Although the registry names HbA1c as the primary outcome and reports secondary measures, these results should not be expanded into claims about overall health or long-term clinical outcomes.

A statistically small p-value is not a measure of how important a result is to an individual. The registry reports an estimated mean contrast and its precision, but does not provide a threshold for interpreting clinical importance. It also does not report the details needed to assess every feature of the analysis, including the multiplicity strategy and missing-data methods. Those gaps limit how fully the reported estimates can be evaluated.

The secondary outcomes add context but involve separate questions and methods. They do not erase the distinction between the primary outcome and secondary tests, and they do not prove that the treatment groups differ on every outcome. The safety summaries are descriptive and do not demonstrate causation. Overall, the posted data support interpretation of the reported Week 30 comparisons, while leaving broader questions outside the scope of what the registry record establishes.

14. Why This Trial Matters Statistically

This trial illustrates why the estimated treatment contrast, its confidence interval, and the analysis population should be read together. The group means describe average change, while the ANCOVA contrast summarizes the model-based difference between groups. The confidence interval communicates the precision of that contrast. None of these quantities alone describes the spread of individual participant outcomes, and the p-value should not be mistaken for effect size or clinical importance.

It also shows why outcome definitions and analysis methods matter. The primary endpoint is a continuous change score, while several secondary endpoints are binary percentages. A mean difference and a risk difference are not interchangeable summaries. Even when they address related questions, they represent different measures and may use different denominators or models. Contrast direction also depends on group ordering, so estimates should be read alongside the comparison labels.

The registry's primary population definition and its distinct safety denominators highlight that reported results can be based on different participant sets. Knowing who contributed to each analysis helps readers understand what each estimate describes. Missing-data handling and the reasons for differences between analysis populations are not reported, which limits assessment of how robust the estimates are to participant exclusions or unavailable measurements.

Finally, planning calculations based on observed effects are not evidence of observed power and do not explain the original sample-size decision. The original sample-size justification and multiplicity strategy are not reported in the registry record. These distinctions matter when interpreting a completed study: the observed estimates describe the trial results, while design assumptions and analysis procedures require their own documentation.

15. 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 reports the primary ANCOVA result but does not give the model covariates, full model specification, or diagnostic information. That limits independent reproduction of the primary comparison. The primary Full-Analysis-Set definition requires at least one post-randomization measurement, and missing-data methods and sensitivity analyses are not reported in the registry record. The posted primary estimate therefore cannot be assessed against alternative missing-data assumptions from these summaries alone.

The registry's arm descriptions are truncated. They identify starting doses and general titration descriptions, but the exact timing and full rules for dose changes are not reported in the registry record. The record also does not report adherence details or the amount of treatment exposure for individual participants. These gaps restrict interpretation of how treatment implementation may relate to the observed outcomes.

Results for change in fasting plasma glucose are not reported in the registry record, although it is listed as a secondary outcome. The multiplicity approach for the primary and secondary testing is not reported, and neither are interim analyses or the original sample-size justification. Consequently, the secondary comparisons should not be treated as a fully documented confirmatory family of tests.

The unadjusted recalculations use posted group summaries and do not reproduce the sponsor's ANCOVA analyses. The body-weight recalculation uses an approximate standard deviation, as identified in the calculation details. Such calculations provide a limited check, not a substitute for participant-level data or full model results. The registry does not report participant-level distributions, which would help describe variability and outliers beyond group means and confidence intervals.

Finally, the primary follow-up is described at Week 30. The posted results do not establish the persistence of the measured differences beyond that point, and generalization beyond the study population and setting cannot be assessed fully from the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Efficacy and Safety of Sitagliptin/Metformin Fixed-Dose Combination (FDC) Compared to Glimepiride in Participants With Type 2 Diabetes Mellitus (MK-0431A-202) — NCT00993187.
• Registry record last updated 2018-08-22; status last verified 2018-07.
• Results first posted 2014-09-15.
• 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.

19. Record Summary

The registry record describes NCT00993187, a completed phase 4, multicenter, randomized, double-blind, parallel-group study of sitagliptin/metformin fixed-dose combination compared with glimepiride in participants with type 2 diabetes mellitus. Enrollment is listed as 292, and the treatment period was 30 weeks.

The primary endpoint was change from baseline in HbA1c at Week 30. The posted ANCOVA comparison reports a difference in least-squares means of -0.8, with a 95% confidence interval from -1.0 to -0.6 and p < 0.001. Secondary results address hypoglycemia, body weight, and HbA1c target attainment; the listed fasting plasma glucose result is not reported in the registry record.

The registry defines the primary analysis population and reports separate safety denominators. Details of missing-data handling, multiplicity, model specification, interim analyses, and the original sample-size justification are not reported in the registry record. The posted estimates should be interpreted within those limits and within the 30-week study period.