ALPHA Sitagliptin Add on to Metformin (0431-103): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for ALPHA Sitagliptin Add on to Metformin (0431-103) (NCT00833027), a non-randomized phase 4 study. The primary outcome is Change From Baseline in HbA1c at Week 24.

2. Clinical Question

The study addressed whether adding sitagliptin to ongoing metformin therapy was followed by a change in HbA1c over the study period among patients whose glycemic levels were inadequately controlled on metformin. The registry identifies the primary outcome as change from baseline in HbA1c at Week 24, measured as a percentage-point value at Week 24 minus the value at Week 0. The primary analysis population was defined as patients who received at least one dose of study medication and had a Week 24 measure.

This question is descriptive rather than comparative. The study is recorded as a single-group, non-randomized, open-label study, with sitagliptin as the intervention and no control arm described. Consequently, its outcome can describe the observed change among the specified patients, but the design does not directly estimate how that change differs from an untreated group or another treatment group. It also cannot separate a change associated with adding sitagliptin from changes that might occur over time for other reasons.

The registry describes the study as intended to let Canadian physicians experience the efficacy and tolerability of adding sitagliptin in real-life clinical practice. That context helps explain the practical focus, but it does not change the limits of a single-group result. A numerical primary outcome estimate, its variability, and a confidence interval are not reported in the registry record. The record therefore does not establish the size or precision of the observed HbA1c change.

3. Trial Design

The registry classifies NCT00833027 as a completed Phase 4 study with non-randomized allocation, no masking, and a single-group model. The study title describes a 24 Week, multicenter, open-label evaluation of adding sitagliptin 100 mg once daily in patients with type 2 diabetes whose glycemic control was inadequate on metformin. The intervention is sitagliptin phosphate, and the arm description identifies the group as sitagliptin. Enrollment was 608.

Because the study was open-label and had no randomized control group, participants and investigators were not masked to the intervention, and the design does not provide a concurrent control for interpreting changes over time. The registry explicitly characterizes this as not a randomized comparison and cautions that arms are not a like-for-like control comparison. In this record, only a single group is described, so a between-arm treatment contrast is not available.

The registry lists the study as completed. Its recorded start date is 2008-03-01 and its primary completion date is 2009-10-30. The primary outcome timeframe is baseline and Week 24. A secondary outcome is change from baseline in HbA1c at Week 12. The record lists two outcome measures and one analysis, but does not give a numerical primary or secondary outcome result.

The enrolled count should not be confused with the number contributing to the primary outcome. The primary population required both receipt of at least one dose and a Week 24 measure, and the number meeting that definition is not reported in the registry record. The record also does not describe the recruitment process, site count, treatment adherence, or the duration of individual follow-up beyond the outcome timeframe stated for HbA1c. Those details should not be inferred from the multicenter and 24 Week wording alone.

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: Includes all patients who received at least one dose of study medication and have a Week 24 measure.

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 HbA1c at Week 24, measured over Baseline and Week 24.

The primary endpoint is change from baseline in HbA1c at Week 24. HbA1c is recorded as a percentage, and the registry defines the change as the Week 24 HbA1c percentage minus the Week 0 HbA1c percentage. The outcome is therefore a continuous measure, with a mean and standard deviation specified as the summary form.

Change from baseline expresses how a measurement differs from its starting value for the same participant. Summarizing those changes with a mean describes the average direction and size of change in the analysis population. A standard deviation describes how dispersed individual changes are around that mean; it is not the uncertainty of the mean itself. An interval estimate for the mean would be needed to communicate uncertainty about the average more directly, but such an estimate is not reported in the registry record.

HbA1c is used in diabetes studies as a measure of glycemic status over time. Measuring it at a stated follow-up point allows investigators to describe whether the group’s measured glycemic status changed during the study period. The Week 24 endpoint is the primary outcome named in the registry, while change from baseline at Week 12 is listed as a secondary outcome.

The numerical mean change, standard deviation, and number of participants contributing to the Week 24 endpoint are not reported in the registry record. Without those results, the endpoint definition can be explained, but the observed average change and its precision cannot be evaluated.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Interpreting a continuous endpoint begins with its definition. The primary outcome is change from baseline in HbA1c at Week 24, defined as the Week 24 value minus the Week 0 value. This direction matters: a negative change would mean the later measured percentage was lower than the baseline percentage, while a positive change would mean it was higher. The registry does not report the actual mean change, so neither direction can be assigned to the study result.

What does a mean change describe?

A mean change combines participant-level changes into a group average. It is a concise description of the center of the observed change distribution, but it does not show whether most participants changed by a similar amount or whether individual changes varied widely. The registry identifies the endpoint summary as a mean and the dispersion as a standard deviation. The standard deviation describes variation among individual measurements around the mean. It is not interchangeable with a confidence interval, which expresses uncertainty in an estimated quantity.

Because the endpoint is defined as a within-participant change, paired baseline and follow-up measurements are relevant to its construction. The registry gives the analysis population as patients who received at least one dose and had a Week 24 measure. It does not report the number of paired measurements available, the handling of patients without a Week 24 measure, or the specific statistical procedure used to estimate the mean and standard deviation. A reader should not infer an imputation method or assume that missing measurements were handled in a particular way.

What would a confidence interval add?

In general, a confidence interval gives a range of values compatible with the data and the estimation procedure, conveying uncertainty about an estimated mean. A narrow interval indicates greater precision than a wide interval, under the assumptions of the method. A confidence interval does not remove design limitations, and it cannot correct for the lack of a control group. No confidence interval for the primary outcome is reported in the registry record, so precision cannot be assessed from the record.

What can a p-value answer?

In general, a p-value assesses how compatible the observed data are with a specified null hypothesis under a statistical model. It does not measure the size or clinical importance of an effect, and it does not establish causation. Interpretation requires knowing the hypothesis, model, analysis population, and relevant assumptions. The registry record does not report a p-value or enough detail to identify the inferential procedure. No conclusion about statistical significance can be drawn from the listed outcome definition alone.

Why does the design matter to the analysis?

A single-group analysis can estimate the group’s change over time, if outcome results are available. It cannot directly estimate a treatment contrast against a concurrent control group. Random allocation is commonly used to balance measured and unmeasured factors between groups on average; this study is explicitly non-randomized and single-group, so that comparative protection is absent. The posted design does not support interpreting a before-and-after change as a causal treatment effect without additional assumptions and evidence.

The record’s planning illustration is distinct from the study’s actual analysis. It is described as a planning illustration based on the observed effect, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is stated as a fraction of the observed effect. These are planning assumptions, not evidence that this study had two arms or used those procedures in its primary analysis. The registry describes only a single group. Sample-size calculations under hypothetical assumptions are planning illustrations, not observed power, and no planning sample-size values or observed effect estimate are reported in the registry record.

In general, normal approximations are convenient when the sampling distribution of an estimate is adequately represented by a normal distribution. Whether that approximation is suitable depends on the data and analysis conditions. The registry does not report diagnostic checks or model details. Over-dispersion is a concern in settings where observed variability exceeds what a model assumes; in general, ignoring it can make uncertainty estimates too small. This primary endpoint is continuous, and the record does not report a count model or an over-dispersion assessment, so that concept should not be mistaken for a documented issue in this analysis.

The registry identifies the primary outcome as mean change with standard deviation, but does not provide the result or the inferential details. The most defensible statistical reading is therefore limited to the outcome definition, timing, and analysis-population rule. A full assessment of estimation, uncertainty, assumptions, and reproducibility is not possible from 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 100mg")
n   <- c(375)
m   <- c(-0.6)
sp  <- c(0.95)
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 100mg  375  -0.600  0.950  -0.696  -0.504

11. Multiplicity, Missing Data and Analysis Populations

The registry lists the primary outcome as change from baseline in HbA1c at Week 24 and a secondary outcome as change from baseline in HbA1c at Week 12. It also states that there are two outcome measures and one analysis. These entries identify planned or recorded outcomes, but do not explain how the outcomes were handled statistically or whether any adjustment for multiple testing was used. A multiplicity strategy is not reported in the registry record.

Multiplicity matters when several outcomes, time points, or analyses are tested. In general, conducting multiple tests can increase the chance of at least one apparently notable result under a null model. Adjustment methods or a clearly defined hierarchy can address that concern, but the record does not describe either. Since numerical outcome results and p-values are also not reported, the practical effect of any multiplicity approach cannot be assessed.

The primary analysis population includes patients who received at least one dose and had a Week 24 measure. This is an outcome-available population definition, not a complete description of how missing measurements were handled. Enrollment was 608, but the number with a Week 24 measure is not reported in the registry record. The record does not state whether missing data were imputed, excluded under additional rules, or addressed through sensitivity analyses.

It also does not report an intention-to-treat definition, a per-protocol population, or other analysis populations. Those labels should not be assumed from the fact that one analysis is listed. Without participant flow, missing-data methods, and population counts, readers cannot determine how sensitive the estimate might be to missing outcomes or exclusions.

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

Serious events with the highest arm-level rate: chest pain (0.5%), depression (0.5%), acute coronary syndrome (0.3%), cardiac failure congestive (0.2%), cardiac tamponade (0.2%).

The safety record lists 608 patients at risk in the sitagliptin 100mg group. It reports 17 patients affected by serious adverse events and 0 affected in the other adverse-event category. Those figures describe recorded event categories in this registry entry; they do not establish whether an event was caused by the study medication. The record does not give a comparative group for assessing relative safety.

The listed serious adverse-event terms and percentages are Chest pain, 0.5%; Depression, 0.5%; Acute coronary syndrome, 0.3%; Cardiac failure congestive, 0.2%; Cardiac tamponade, 0.2%; and Myocardial infarction, 0.2%. These percentages are presented as listed for the terms. The record does not provide event counts for each term, so counts should not be inferred from the percentages. The terms are not described as necessarily exclusive of one another.

The deaths value is not reported in the registry record. The reported 0 in the other adverse-event category should not be taken to mean that no adverse events of any kind occurred; it is the value recorded for that category. The registry entry does not provide further detail about event definitions, ascertainment, exposure duration, or causal assessment. These limitations constrain interpretation of the safety figures.

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

The record establishes the study question and design: a completed, open-label, non-randomized, single-group study evaluated sitagliptin 100 mg once daily added to metformin in patients with type 2 diabetes and inadequate glycemic control on metformin. Its primary endpoint was mean change from baseline in HbA1c at Week 24, among patients who received at least one dose and had a Week 24 measure. Enrollment was 608.

Those details do not establish the numerical outcome. The mean change, standard deviation, confidence interval, p-value, and primary analysis count are not reported in the registry record. As a result, readers cannot determine the direction or magnitude of the HbA1c change, its precision, or how many enrolled patients contributed to the primary endpoint. The secondary Week 12 change is also listed without a numerical result.

Even if a within-group change were available, this design would not show how outcomes would have differed under a concurrent alternative. There is no randomized control group described, and the registry expressly warns that this is not a randomized comparison. A before-and-after difference may reflect the intervention, changes over time, other care, or other factors. A single-group result alone cannot distinguish among these explanations.

The study’s practical, real-life context is relevant to the kind of experience it sought to describe, but it does not remove uncertainty about causation, missing outcomes, or generalizability. Details about participant selection, missing-data handling, and the analysis procedure are not reported in the registry record. The serious adverse-event figures describe recorded events in the study group, not comparative risk or confirmed medication causation.

Accordingly, the registry record supports a description of the study’s purpose, endpoint, population rule, and safety entries. It does not support a claim that HbA1c improved or worsened, a claim about the size or statistical significance of any change, or a conclusion that sitagliptin caused the reported outcomes.

14. Why This Trial Matters Statistically

This study illustrates an important distinction between measuring change and estimating a treatment effect. A single-group study can record HbA1c at baseline and follow-up and define a mean change. That statistic describes the observed participants over time, but without a control group it does not establish what would have happened in the absence of the added intervention. The registry explicitly identifies the design as non-randomized and single-group.

It also demonstrates why outcome definitions and outcome results should be read separately. The record precisely identifies the primary endpoint, its timing, its units, and the population required for inclusion. Yet the numerical result and analysis count are not reported in the registry record. Clear endpoint specification is useful, but it cannot substitute for an estimate and an account of its uncertainty.

The population rule highlights another recurring statistical issue: enrollment is not necessarily the endpoint sample. Although enrollment was 608, a Week 24 measurement was required for the primary analysis population, and its size is not reported. Without participant counts and a missing-data description, readers cannot evaluate how much information contributes to the endpoint or whether exclusions could affect the interpretation.

Finally, the planning assumptions listed for sample-size illustrations should be kept separate from the actual study design. A hypothetical calculation using equal allocation between two arms does not turn a single-group study into a comparative trial. Such illustrations can show how planning depends on assumptions, but they are not observed power and do not supply missing outcome evidence. The record is therefore a useful example of how design, analysis population, reporting completeness, and interpretation jointly determine what a clinical result can support.

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 most direct limitation is that numerical efficacy results are not reported in the registry record. The primary mean change in HbA1c, its standard deviation, an interval estimate, and a p-value are absent. The secondary HbA1c outcome at Week 12 is also listed without a numerical result. The observed direction, magnitude, and precision of change therefore remain unknown.

The study has no comparator arm described and is classified as non-randomized, open-label, and single-group. Consequently, a within-group change cannot provide a direct contrast with another treatment or with no added study medication. The design leaves alternative explanations for changes over time unresolved. The registry itself cautions that this is not a randomized comparison.

The primary population required at least one dose and a Week 24 measure, but the number meeting that definition is not reported. The record does not specify how missing measurements were handled, whether additional exclusions were applied, or whether sensitivity analyses were conducted. It also does not report the actual statistical procedure, multiplicity handling, or detailed analysis populations.

For safety, 17 patients are listed as affected by serious adverse events, while the record gives 0 for the other adverse-event category. Deaths are not reported in the registry record. Event-specific counts, ascertainment details, causal assessments, and comparative safety data are not given. The listed percentages for serious terms should therefore be read as registry entries, not as evidence of comparative risk or cause.

Finally, the registry description of real-life practice intent does not supply information about participant selection or make the findings representative of all patients with type 2 diabetes. The record does not provide enough detail to assess the extent of generalizability or the influence of site-level practice.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: ALPHA Sitagliptin Add on to Metformin (0431-103) — NCT00833027.
• Registry record last updated 2017-05-15; status last verified 2017-04.
• Results first posted 2010-09-21.
• 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

NCT00833027 is a completed Phase 4, multicenter, open-label, non-randomized, single-group study. It evaluated sitagliptin 100 mg once daily added to metformin in patients with type 2 diabetes and inadequate glycemic control on metformin. Enrollment was 608.

The primary outcome was mean change from baseline in HbA1c at Week 24, defined as Week 24 minus Week 0, in patients who received at least one dose and had a Week 24 measure. Change from baseline in HbA1c at Week 12 was listed as a secondary outcome. Numerical results for these outcomes, the primary analysis count, and the statistical analysis details are not reported in the registry record.

The safety entry lists 17 patients affected by serious adverse events, 0 in the other adverse-event category, and deaths not reported in the registry record. The single-group design does not provide a randomized comparator for estimating a treatment contrast.