Sitagliptin in Combination With Metformin and Sulfonylurea: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Sitagliptin in Combination With Metformin and Sulfonylurea (NCT00686634), a na phase 4 study. The primary outcome is Hemoglobin A1c (HbA1c) Change From Baseline.

2. Clinical Question

3. Trial Design

The registry identifies this as a phase 4, completed study sponsored by Charles Drew University of Medicine and Science. The listed start is 2008-01, and primary completion is recorded as 2010-03. Enrollment is 108 participants with type 2 diabetes mellitus.

The structured design describes a single-group study, with allocation recorded as NA and masking recorded as none. The only arm described is sitagliptin 100 mg once daily. These details do not describe a randomized, masked comparison. Although the brief summary says the study will compare sitagliptin with thiazolidinediones as a third-line oral agent, the structured record does not list a thiazolidinedione arm or a like-for-like control group. The comparator and the method, if any, for making that stated comparison are not reported in the registry record.

The primary outcome is HbA1c change from baseline, assessed at baseline and 4 months. The listed analysis population applies last observation carried forward to subjects with paradoxical worsening of control who switched to alternate treatment before 4 months. The registry does not describe the full analysis population or explain how other missing measurements were handled.

Secondary outcomes include the number of subjects with HbA1c 7.5% or less at 4 months, the number maintaining HbA1c 7.5% or less by 1 year, and the number of adverse events. The record identifies these outcomes but does not report their results in the information available for this analysis. It also lists no analyses. Thus the registry supports description of the intended treatment and outcome framework, not an assessment of a comparative treatment effect.

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: Last Observation Carried Forward for subjects with paradoxical worsening of control (switched to alternate treatment before 4 months)

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 Hemoglobin A1c (HbA1c) Change From Baseline, measured over Baseline, 4 months.

The primary endpoint is mean change in hemoglobin A1c (HbA1c) from baseline to 4 months, measured in percentage units. HbA1c reflects average blood glucose over time and is commonly used in diabetes studies as a measure of glycemic control. A change score expresses how the measured value at follow-up differs from the participant’s baseline value; the mean summarizes those changes across the analysis population.

The registry specifies standard deviation as the measure of dispersion. Standard deviation describes how spread out individual changes are around their mean. It is not the same as a confidence interval: the former describes variation among observations, while the latter, when calculated, describes uncertainty about an estimated mean or contrast.

The endpoint can describe within-group change over the stated follow-up, but a single-group result cannot by itself separate a treatment-associated change from other explanations, such as changes over time or co-interventions. A comparison group and an appropriate comparative analysis would be needed to estimate a relative effect. The primary mean change and its standard deviation are not reported in the registry record, so the magnitude and precision of the observed endpoint cannot be assessed.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a continuous measure summarized as a mean. In general, a mean change is calculated by taking each participant’s follow-up value relative to baseline and averaging those changes in the analysis population. A mean is a compact description of the group’s average observed trajectory; it does not imply that every participant changed by that amount. The registry gives the outcome unit as percentage and specifies standard deviation as the dispersion measure. The standard deviation describes variability among participant-level observations, not uncertainty in the estimated mean itself.

In general, uncertainty about an estimate is often expressed with a confidence interval. Such an interval conveys a range of values compatible with the data and the calculation method, subject to the method’s assumptions. A confidence interval for a group’s mean change would address uncertainty about that change; a confidence interval for a contrast would address uncertainty about the difference between groups. The registry record does not report either interval for the primary outcome, and no primary numerical estimate is posted in the record.

The design determines which effects can be estimated. A single-group study can describe change within the group over time. It does not, on its own, provide a contemporaneous counterfactual showing what would have happened without the intervention or under another treatment. Random allocation, when present and implemented appropriately, can support a comparison by reducing systematic baseline differences between groups. Here, the structured design is single-group, allocation is NA, and masking is none. The registry does not list a comparison arm. The summary’s stated comparison with thiazolidinediones is therefore not matched by the structured arm information.

The primary endpoint is identified as a mean, but the registry record does not specify a statistical test, model, covariate adjustment, or method for calculating uncertainty. Those details are not reported in the registry record, so they should not be inferred from the endpoint type. In general, a model or test should match the design, outcome scale, and data structure. For repeated measurements, for example, analysis choices may account for the relationship between observations from the same participant; the registry does not say whether such an approach was used.

The stated analysis population uses last observation carried forward for subjects with paradoxical worsening of control who switched to alternate treatment before 4 months. This method carries a participant’s last recorded measurement forward to represent a later time point. As a general matter, it assumes that the carried-forward value is a reasonable stand-in for the unobserved value. That assumption may be unrealistic when the outcome is changing, when switching is related to the outcome, or when missingness is informative. Carrying values forward can also make uncertainty appear smaller than it is if the imputation is treated as though it were observed data. The record does not describe sensitivity analyses or other approaches to missing measurements.

Over-dispersion is a general issue for count outcomes when variation exceeds what a chosen count model expects. If present, it can make standard errors too small and tests too optimistic unless the analysis accounts for it. The listed adverse-event outcome is a count, but the registry record does not report an analysis model or event-count results suitable for evaluating over-dispersion. No particular count model should be attributed to this study.

The sample-size assumptions describe a planning illustration based on a fraction of the observed effect, with two-sided significance level 0.05, equal allocation between two arms, and a normal approximation. Except for time-to-event outcomes, those illustrations are stated as participants per arm. These are planning assumptions, not a description of the trial’s actual design or an observed-power calculation. The record does not report a sample-size justification for the study itself, and the primary observed effect is not reported in the registry record, so the planning illustration cannot be interpreted as evidence about the precision achieved in this trial.

Finally, the registry lists no analyses. That absence means the record does not describe which statistical analyses were conducted or how they relate to the registered outcome. It does not establish that no analyses occurred; the analysis details are simply not reported in the registry record.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Sitagliptin")
n   <- c(102)
m   <- c(-1.3)
sp  <- c(1.8)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Sitagliptin  102  -1.300  1.800  -1.654  -0.946

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary HbA1c change outcome and secondary outcomes concerning HbA1c thresholds and adverse events. When a study examines several outcomes or several comparisons, the chance of at least one apparently unusual result can increase. General approaches to this issue include prespecifying a primary outcome, adjusting inference across multiple tests, or treating secondary results as exploratory. The registry record does not report a multiplicity strategy or the analyses performed, so it is not possible to determine how multiplicity was handled.

The primary outcome’s stated population uses last observation carried forward for subjects with paradoxical worsening of control who switched to alternate treatment before 4 months. The record does not report the number of participants affected by this rule or describe missing-data handling for other circumstances. It also does not report sensitivity analyses to assess whether conclusions depend on that imputation choice. In general, missing-data methods rely on assumptions about why measurements are absent; different assumptions can produce different estimates.

The analysis populations beyond this specified rule are not reported in the registry record. It is therefore unclear from the registry information whether participants were analyzed according to an intention-to-treat principle, according to treatment received, or using another population definition. The listed endpoint and population note should not be treated as a complete statistical analysis plan. These gaps, together with the absence of posted primary and secondary outcome estimates, limit assessment of robustness and selective reporting.

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.

The registry lists 108 participants at risk for serious adverse events and records 0 affected participants. For other adverse events, it also lists 108 at risk and 0 affected participants. These figures describe the posted registry entries; they should be read as counts of affected participants in the specified categories, not as a complete account of event types, timing, severity, or exposure.

The secondary outcome list includes the number of adverse events, but a result for that broader outcome is not reported in the registry record. The serious adverse-event terms list contains no terms. The death count is not reported in the registry record. A zero in the affected-participant field does not show whether events were actively sought, how long safety was observed, or how fully the data were captured. The record does not give those details.

Accordingly, the figures support only a narrow description of what the registry reports for these categories. They do not establish that no adverse events occurred in any sense, nor do they permit a detailed safety comparison: the structured record identifies no comparator arm, and the broader adverse-event results are not reported.

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

The record documents a completed phase 4 study of sitagliptin 100 mg once daily, with mean HbA1c change from baseline to 4 months designated as the primary outcome. That information clarifies what the study set out to measure and the treatment regimen listed. It does not reveal the size or direction of the observed HbA1c change, because the primary estimate and its dispersion are not reported in the registry record.

The intended comparison described in the brief summary is not supported by the structured design details, which identify a single group and list no comparator arm. A within-group change, if available, would describe what happened over time among participants in that group. It would not by itself establish what would have happened under a thiazolidinedione or without the study treatment. The record does not report a comparative estimate.

The listed secondary outcomes signal further topics of interest, including reaching or maintaining HbA1c 7.5% or less and adverse events. Their results are not reported in the registry record. The adverse-event fields list zero affected participants for serious and other events, but those entries do not provide a full safety profile, and the death count is not reported.

The sample-size material is described as a planning illustration based on an observed effect and specified assumptions. Such calculations are planning tools, not observed power and not evidence that the study achieved a particular level of precision. With the primary result absent, the illustration cannot resolve the uncertainty about the actual treatment-associated change. The appropriate conclusion is therefore limited: the registry reports the study framework and selected safety counts, but does not provide enough outcome information to judge efficacy, comparative benefit, or the completeness of safety reporting.

14. Why This Trial Matters Statistically

This trial record illustrates why design labels and narrative summaries need to be read together. The brief summary describes a comparison between sitagliptin and thiazolidinediones, while the structured design lists a single group, allocation as NA, and one sitagliptin arm. Without a listed comparator and comparative result, the stated comparison cannot be evaluated from the registry information.

It also illustrates the distinction between defining an endpoint and reporting its result. The primary endpoint is clearly named as mean HbA1c change from baseline to 4 months, with standard deviation specified for dispersion. Yet the estimate and its uncertainty are not reported in the registry record. An outcome definition alone cannot establish the direction, magnitude, or precision of a finding.

The stated last-observation-carried-forward rule is another useful statistical detail to notice. It signals that at least one kind of treatment switch and missing follow-up was anticipated in defining the analysis population, while leaving other missing-data procedures and sensitivity checks unreported. Those choices matter because conclusions can depend on how incomplete measurements are handled.

Finally, safety counts require careful scope. The posted serious and other adverse-event fields list zero affected participants, but the broader adverse-event result, safety follow-up details, and death count are not reported. This record is therefore a practical example of how registry design information, analysis definitions, outcome results, and safety reporting each answer different questions—and how missing results constrain interpretation.

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 central limitation is that the registry record does not report the primary HbA1c change estimate or its standard deviation. Without those values, neither the observed change nor its precision can be described. No confidence interval or test result for the primary endpoint is reported.

A second limitation is the mismatch between the brief summary’s stated comparison with thiazolidinediones and the structured design, which lists a single group receiving sitagliptin and no comparator. The allocation is recorded as NA and masking as none. The registry record therefore does not establish a comparative design or provide the information needed to assess a relative effect.

The record lists secondary HbA1c threshold outcomes and adverse events, but does not report their results in the information available for this analysis. It does not report the death count, and the serious adverse-event terms list is empty. Safety follow-up duration, ascertainment procedures, and event detail are not reported in the registry record.

Some analysis information is present: the primary outcome uses last observation carried forward for subjects with paradoxical worsening of control who switched to alternate treatment before 4 months. However, the record does not explain missing-data handling in other situations, give a complete analysis-population definition, or report sensitivity analyses. Statistical tests, models, covariate adjustments, and a multiplicity strategy are also not reported. The registry’s completion status and enrollment do not remedy these gaps in outcome reporting.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Sitagliptin in Combination With Metformin and Sulfonylurea — NCT00686634.
• Registry record last updated 2014-11-26; status last verified 2014-11.
• Results first posted 2014-11-26.
• 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

NCT00686634 is a completed phase 4 registry record sponsored by Charles Drew University of Medicine and Science. It concerns type 2 diabetes mellitus, lists enrollment of 108, and describes sitagliptin 100 mg once daily. The registered primary outcome is mean HbA1c change from baseline to 4 months, with standard deviation as the dispersion measure. The stated analysis population uses last observation carried forward for subjects with paradoxical worsening of control who switched to alternate treatment before 4 months.

The structured design is single-group, allocation is NA, and masking is none. Although the brief summary describes a comparison with thiazolidinediones, no comparator arm is listed. The primary and secondary outcome estimates are not reported in the registry record. Serious and other adverse-event entries each list 0 affected participants among 108 at risk; the death count is not reported. The record does not provide enough result information to assess the primary outcome or a comparative effect.