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 Incretins in Impaired Fasting Glucose (NCT00364377), a randomized phase 4 study. The primary outcome is Lowering of Fasting Glucose.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00364377 |
| Phase | Phase 4 |
| Condition | Pre-diabetes |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 22 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Mayo Clinic |
| Results posted | Yes |
2. Clinical Question
Population
People with impaired fasting glucose, described in the registry record as people with pre-diabetes. The brief summary frames the question around elevated fasting glucose in people who do not yet have diabetes.
Intervention
Sitagliptin 100 mg once daily. The registry summary describes sitagliptin as raising levels of glucagon-like peptide-1, a hormone released by the intestine in response to food.
Comparator
Placebo once daily. Participants were randomized to sitagliptin or placebo, and the trial is described as double-masked and parallel-group.
Primary question
Whether fasting glucose differed between the sitagliptin and placebo groups after 8 weeks. The primary outcome is a mean fasting-glucose measure in mmol/l, based on measurements before inpatient meal studies.
The study asks a focused physiological question: among people with impaired fasting glucose, how does fasting glucose over the study period compare between those assigned sitagliptin and those assigned placebo? It does not ask whether the treatment prevents future diabetes, changes long-term health outcomes, or has effects beyond the study period. Those questions are not answered by the posted primary outcome.
The registry describes a phase 4 randomized study sponsored by Mayo Clinic, with 22 participants enrolled. The posted primary analysis includes 11 participants in each group and is identified as per protocol, with all participants completing the intervention. The endpoint description indicates that fasting glucose was calculated as the mean of blood glucose measurements taken at -30, -20, -10 and 0 minutes before each inpatient meal study. The outcome timeframe is 8 weeks.
The result should therefore be read as a comparison of average fasting-glucose values in the two analyzed groups at the specified study assessment. The registry record does not report a broader clinical outcome or an assessment of whether any difference would persist after treatment ended.
3. Trial Design
The registry record describes a randomized, double-masked, parallel-group phase 4 study. Participants with impaired fasting glucose were assigned to sitagliptin or placebo. The sitagliptin regimen was 100 mg once daily; placebo was also taken once daily. The registry lists Mayo Clinic as the sponsor and gives an enrollment of 22. The study status is completed.
The primary outcome was assessed over 8 weeks. The analysis population is described as per protocol, and the record says that all participants completed the intervention. The posted arm results include 11 participants assigned to sitagliptin and 11 assigned to placebo. The outcome is continuous: the registry reports the mean fasting-glucose value in each group, with standard errors as the measure of dispersion.
The record identifies a two-sided t-test for the reported comparison. Its posted p-value is >0.05. The record does not report a test statistic, degrees of freedom, or the precise test implementation. A separate unadjusted re-calculation based on the posted summary statistics gives a mean difference of -0.05 mmol/l and p = 0.768. Because the registry record flags a discrepancy between the sponsor’s analysis and this re-calculation, the sponsor’s own analysis takes precedence when describing the trial result.
The registry record does not report the trial’s original sample-size justification or its original design assumptions. The planning illustrations elsewhere on this page are based on the observed effect and stated general assumptions, not on the study’s original design. The record also does not report details of random-sequence generation, allocation concealment, masking procedures, or interim analyses. These omissions do not establish that such procedures were absent; they mean their details are not reported in the registry record.
There are no posted secondary outcome measures in the record. The result is therefore centered on the single posted primary outcome and the comparison between the two assigned groups.
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
| Stage | Sitagliptin | Placebo |
|---|---|---|
| STARTED | 11 | 11 |
| COMPLETED | 11 | 11 |
| NOT COMPLETED | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: Analysis was per protocol - all participants completed the intervention
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 | Placebo | Total |
|---|---|---|---|---|
| Participants analysed | 11 | 11 | 22 | |
| Age Continuous | years | 55.2 (SD 2.5) | 54.1 (SD 2.4) | 54.7 (SD 1.8) |
| Sex: Female, Male: Female | Participants | 10 | 8 | 18 |
| Sex: Female, Male: Male | Participants | 1 | 3 | 4 |
| Region of Enrollment: United States | participants | 11 | 11 | 22 |
6. Primary Endpoint
The registry lists the primary outcome as Lowering of Fasting Glucose, measured over 8 weeks.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | Standard Error |
| Unit | mmol/l |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is the lowering of fasting glucose, expressed as a mean in mmol/l. The registry defines the measure as the mean of blood glucose readings taken at -30, -20, -10 and 0 minutes before each inpatient meal study. Its timeframe is 8 weeks. The posted arm summaries give a mean and standard error for each group.
Fasting glucose is used to represent glucose levels before a meal, rather than the response after eating. A mean combines the specified readings into one summary for the assessment, making it possible to compare the groups using a continuous outcome. In general, a continuous outcome can retain information about the measured values rather than classifying each participant only by whether a threshold was crossed. This is a general statistical feature, not evidence about the clinical importance of any particular difference in this study.
The endpoint is narrower than the broader questions mentioned in the study summary. It measures fasting glucose at the stated assessment, not the development of diabetes over time, postprandial glucose as a separately posted primary result, or longer-term outcomes. Those outcomes are not reported in the registry record as posted results for this analysis.
The reported group means are 5.78 mmol/l for sitagliptin and 5.83 mmol/l for placebo. The unadjusted mean difference is -0.05 mmol/l, calculated as the sitagliptin mean minus the placebo mean in the posted comparison. The direction is toward a lower mean in the sitagliptin group, but the comparison’s uncertainty and the sponsor’s reported analysis need to be considered before interpreting that direction.
7. Results Posted in the Registry
- Sitagliptin: mean 5.78 versus 5.83 on control; difference -0.05 (95% CI -0.38 to 0.28), standardized effect d = -0.13; z-test p = 0.768.
Mean difference: Sitagliptin
95% CI: -0.38 to 0.28
Recalculated from the posted summary statistics
8. Statistical Interpretation
The posted mean fasting-glucose values are 5.78 mmol/l for sitagliptin and 5.83 mmol/l for placebo. The unadjusted re-calculation gives a mean difference of -0.05 mmol/l, where a negative value reflects a lower mean in the sitagliptin group under the subtraction order shown in the comparison. The calculated p-value is 0.768. The sponsor’s reported analysis is a two-sided t-test with p >0.05.
These analyses do not provide evidence of a statistically detectable difference between the groups in the posted primary outcome. That statement describes the analyses; it does not establish that the groups are identical or that the true difference is zero. A non-significant result can occur when the difference is small, when the estimate is imprecise, or both. The posted confidence interval extends on both sides of zero, so it includes values consistent with either a lower or a higher fasting-glucose mean for sitagliptin relative to placebo.
The interval is broad relative to the observed mean difference, and the analyzed groups are small, with 11 participants in each arm. This means the point estimate should not be read in isolation. The observed direction is only a description of the sample means; it is not enough to establish the direction or size of an underlying group difference with confidence.
The registry record flags a discrepancy between the sponsor’s reported analysis and the unadjusted re-calculation from the posted summary statistics. The sponsor’s own analysis takes precedence. The re-calculation is useful as a check on the numerical summaries, but it does not replace the study’s reported analysis or reveal why the two results differ. The registry record does not report enough detail to explain the discrepancy.
The result is limited to the specified mean fasting-glucose endpoint at 8 weeks in the per-protocol analysis. It does not establish effects on diabetes development, post-meal glucose, longer-term glucose control, or outcomes after the study period. The registry record does not post secondary outcomes for this trial. The findings should therefore be described narrowly: the posted group means were close, the sponsor reported p >0.05, and the confidence interval leaves uncertainty about the size and direction of the underlying difference.
Statistical non-significance is not a demonstration of equivalence. An equivalence conclusion would require a prespecified equivalence framework and margins, neither of which is reported in the registry record. The record also does not report a prespecified threshold for a clinically important difference. Without those details, the observed estimates cannot be used to decide whether a possible difference is clinically meaningful.
9. Statistical Methods Explained
The primary outcome is a continuous measurement, reported as a group mean. The comparison is between two randomized parallel groups: sitagliptin and placebo. The registry identifies the sponsor’s method as a two-sided t-test. A two-sided test evaluates whether the data are inconsistent with a difference in either direction, rather than testing only for a difference in one chosen direction. This is a general description of the method.
What the mean difference describes
The posted comparison expresses the contrast as sitagliptin minus placebo. The group means are 5.78 mmol/l and 5.83 mmol/l, respectively, and the unadjusted mean difference is -0.05 mmol/l. The negative sign indicates that the sitagliptin sample mean is lower under this subtraction order. A difference in sample means is an estimate from the participants included in the analysis; it is not automatically the exact difference that would be seen in a wider population.
The record gives a standard error of 0.12 for each group mean. In general, a standard error describes the precision of an estimate across repeated samples under a statistical model. It is different from the spread of individual participant values: it describes uncertainty in the estimated mean, rather than the range of glucose readings among participants. The registry reports standard errors, not standard deviations, for the arm summaries.
How the test and interval fit together
A t-test uses the observed difference and its uncertainty to assess how compatible the data are with a null hypothesis of no group difference. The registry reports the sponsor’s result as p >0.05. The unadjusted re-calculation gives p = 0.768. A p-value is not the probability that the null hypothesis is true, and it does not measure the probability that the result will replicate. In general, it describes how unusual data at least as discrepant as those observed would be under a specified null model and its assumptions.
A confidence interval gives a range of values compatible with the observed data under the method used to form the interval. The posted interval for the mean difference extends below and above zero. That means the interval includes both a negative contrast and a positive contrast. It does not show that every value in the interval is equally plausible, nor does it provide a probability that the true value lies in the interval after the data have been observed. The registry record does not report the confidence level in the comparison details.
The re-calculation is identified as simple and unadjusted. It is based on posted summary statistics and is not an alternative sponsor analysis. Since the registry record flags a discrepancy, the sponsor’s own analysis takes precedence. The record does not report the details needed to determine the source of the discrepancy, such as differences in the analysis set, model specification, or other implementation choices. Those possibilities should not be treated as explanations without documentation.
Why variation matters
In general, variability affects how precisely a group mean and a difference between group means can be estimated. Greater variability tends to make it harder to distinguish a modest difference from sampling noise, while more precise measurements or more participants can narrow uncertainty under otherwise comparable conditions. This is a general statistical explanation; the registry record does not provide a detailed variance model or the original assumptions used to design this study.
Over-dispersion is a general concern when a chosen statistical model understates the actual variation in data. If the true variation is larger than the model assumes, standard errors may be too small and tests may appear more decisive than they should. The registry record does not state that over-dispersion was assessed or observed here, so it should not be presented as a feature of this trial’s results.
What the reported analysis does not establish
The registry identifies a per-protocol analysis and says all participants completed the intervention. Completion does not by itself document every detail of outcome availability, protocol adherence, or analysis handling. The record does not report an intention-to-treat analysis, a missing-data method, covariate adjustment, or a multiplicity procedure. It also does not report interim analyses. These details are not reported in the registry record, so the posted result should not be described as adjusted for them.
The primary comparison is the only posted outcome analysis identified. No secondary outcomes are posted in the record. The absence of posted secondary results does not establish that no other measurements were collected; it means they are not available as registry results in this record. The interpretation should remain tied to the reported fasting-glucose comparison and its stated timeframe.
10. Confidence Intervals and Precision
The confidence interval is important because it shows the uncertainty around the estimated mean difference, not just the direction of the sample means. The posted interval extends below and above zero. Values below zero correspond to lower fasting glucose in the sitagliptin group under the reported contrast; values above zero correspond to higher fasting glucose in that group. Thus, the interval does not pin down the direction of the underlying difference.
The interval also extends beyond the observed mean difference of -0.05 mmol/l. This indicates that the estimate is not highly precise in relation to the range of values compatible with the posted data. The interval should not be read as a collection of equally likely values, and it is not a guarantee that the true difference lies within it. In general, a confidence interval is a method-based range that reflects sampling uncertainty under its assumptions.
The trial’s analyzed groups each contain 11 participants. With groups of this size, an estimate can be sensitive to the observations included and to the assumptions used in the analysis. The registry reports standard errors of 0.12 for both group means. The comparison’s interval is available in the posted results, but the registry record does not identify its confidence level in the comparison details.
The sponsor’s reported p-value is >0.05, while the unadjusted re-calculation gives p = 0.768. Both are described as non-significant at the stated conventional threshold used in the planning assumptions, but the discrepancy means the sponsor’s analysis should lead the interpretation. A non-significant test does not make the interval narrow, prove there is no effect, or establish equivalence. The interval’s span is a direct reminder that the estimated difference remains uncertain.
Precision is also distinct from clinical importance. The registry record does not report a threshold for a clinically important difference, so the range cannot be classified as clinically negligible or clinically important. The most defensible reading is that the posted estimate is small in magnitude, but the interval leaves uncertainty about both direction and size.
11. Planning a Similar Trial: Sample Size
This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.
| Effect to detect | Participants per arm, 80% power | Participants per arm, 90% power |
|---|---|---|
| 100% of the observed effect | 995 | 1,332 |
| 75% of the observed effect | 1,769 | 2,368 |
| 50% of the observed effect | 3,979 | 5,327 |
The sample-size table and curve are planning illustrations based on the observed effect, not the trial’s original design assumptions. The table gives participants per arm for 80% power and participants per arm for 90% power. These are not total participant counts or total events: for this continuous outcome, the figures are participants per arm.
The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to a stated fraction of the observed effect. The rows show planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. The registry record does not report the trial’s own original sample-size assumptions or justification.
In plain terms, the 80% power column gives the number of participants per arm in the illustration for a design with an 80% chance of detecting the specified effect under its assumptions. The 90% power column gives the corresponding figure for a design with a 90% chance. Power is a property of a planned design under assumed conditions; it does not describe the probability that the trial’s observed conclusion is correct.
The table makes clear that planning for a smaller effect requires more participants. For example, planning for 75% of the observed effect gives 1,769 participants per arm for 80% power and 2,368 for 90% power. For 50% of the observed effect, the figures are 3,979 and 5,327 per arm, respectively. Planning for 100% of the observed effect gives 995 per arm for 80% power and 1,332 for 90% power. These are illustrations rather than a reconstruction of the original design. This is not observed power.
12. How the Sample Size Is Derived
Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z1−α/2 = 1.960) and the desired power (z1−β = 0.842 for 80% and 1.282 for 90%).
Formula
n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.
Worked calculation for this trial
- Observed difference Δ = -0.050; pooled standard deviation σ = 0.398.
- n = 2 × (1.96 + 0.842)2 × 0.3982 / -0.0502 = 994.6, rounded up to 995 participants per arm for 80% power.
The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).
R code
alpha <- 0.05
z_a <- qnorm(1 - alpha / 2) # two-sided
delta <- -0.0499999999999998 # observed difference in means
sigma <- 0.397994974842648 # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))
# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")
# Simulation: exact two-sample t-test using simulated means and variances
set.seed(364377)
n <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n)) # difference in sample means
s2 <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1)) # pooled sample variance
tstat <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))
Output from running the code
100% of observed effect: n per arm = 995 (80% power), 1,332 (90% power)
75% of observed effect: n per arm = 1,769 (80% power), 2,368 (90% power)
50% of observed effect: n per arm = 3,979 (80% power), 5,327 (90% power)
power.t.test: 995.6
Simulated power at n = 995 per arm: 0.799 (20000 replicates)The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.
13. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| t-test, 2 sided | not reported | > 0.050 |
The registry reports the sponsor’s comparison as a two-sided t-test with p >0.05. The unadjusted re-calculation from the posted summary statistics gives a mean difference of -0.05 mmol/l, a confidence interval extending below and above zero, and p = 0.768. The registry record flags a discrepancy between the sponsor’s own analysis and the re-calculation.
The sponsor’s own analysis takes precedence. The re-calculation is a check based on the posted group summaries, not a replacement for the reported analysis. It is useful to show that a straightforward calculation from the arm means and standard errors also does not indicate a statistically detectable difference, but it cannot establish why the results differ or which analysis details account for the discrepancy.
The registry record does not report the information needed to resolve that difference. It does not explain whether the analyses used different data, analysis populations, models, or other choices. These are examples of details that could matter in general, not explanations confirmed for this study. The sponsor’s reported p-value is given only as >0.05, without a more precise value or test statistic.
Accordingly, the interpretation should retain the sponsor’s result as the primary reported analysis and identify the re-calculation as unadjusted. Neither result supports a claim of equivalence, and neither establishes that the groups had exactly the same fasting-glucose values.
14. Multiplicity, Missing Data and Analysis Populations
The registry record lists one primary outcome measure and one posted analysis, with no secondary outcomes posted. This provides a narrow basis for interpreting the results: the focus is the fasting-glucose comparison at 8 weeks. The record does not report a multiplicity adjustment or a plan for controlling the chance of false-positive findings across multiple tests. Because only one outcome analysis is posted, it is not possible to infer from the posted results how any broader family of tests may have been handled.
The record does not report whether additional outcomes or analyses were collected but not posted. The absence of secondary results in the registry should therefore be described as an absence of posted secondary outcomes, not proof that no other measurements were made. The statistical analysis plan and any prespecification details beyond the listed primary outcome are not reported in the registry record.
The primary result is described as per protocol, and the registry says all participants completed the intervention. This gives some information about completion, but does not explain how the analysis population was defined in detail, whether all randomized participants contributed outcome data, or how any unavailable measurements were handled. A missing-data method is not reported in the registry record. The record also does not report an intention-to-treat analysis.
Random allocation is listed, but the methods used to generate the allocation sequence and conceal assignment are not reported in the registry record. Double masking is listed, but the procedures used to maintain masking and whether masking was assessed are not reported. These omissions limit what can be judged about potential sources of bias from the registry description; they do not show that the procedures were inadequate.
Interim analyses and stopping rules are not reported in the registry record. The original sample-size justification and design assumptions are also not reported. Consequently, the planning table on this page should not be treated as evidence of the study’s intended enrollment target or as a description of the trial’s original design.
The record identifies a sponsor-reported two-sided t-test and flags a discrepancy with the summary-statistic re-calculation, but it does not explain the discrepancy. The record also does not provide enough detail to assess whether covariate adjustment or other analysis choices were used. These gaps make a narrow reading of the posted primary comparison more appropriate than broader conclusions about the analysis plan.
15. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Sitagliptin | 0 | 11 | 0.0% |
| Placebo | 0 | 11 | 0.0% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Sitagliptin | 0 of 11 |
| Placebo | 0 of 11 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Sitagliptin | Placebo |
|---|---|---|---|
| Rash | Skin and subcutaneous tissue disorders | 0 of 11 | 0 of 11 |
The registry lists 11 participants at risk for serious adverse events in each group, with 0 affected in the sitagliptin group and 0 affected in the placebo group. It also lists 11 participants at risk for other adverse events in each group, with 0 affected in each group. These figures describe what is posted in the registry record for this study.
The listed figures should not be expanded into a claim that no harms occurred in any context. They refer to the reported adverse-event categories and the participants shown as at risk. The registry record does not list any top serious adverse-event terms. Death counts are not reported in the registry record.
These counts are small and the study is focused on a short, 8-week intervention. The registry record does not provide detail about event collection procedures, severity grading for other adverse events, or follow-up beyond the study timeframe. The figures therefore summarize the posted safety entries; they do not establish the absence of uncommon or longer-term events.
The appropriate reading is descriptive: the registry reports no affected participants in either posted adverse-event category for either group, among 11 at risk per group. This is not a basis for medical advice or a definitive comparison of safety between treatments.
16. What the Result Does, and Does Not, Mean
The posted data show similar fasting-glucose group means at the 8-week assessment: 5.78 mmol/l with sitagliptin and 5.83 mmol/l with placebo. The unadjusted difference is -0.05 mmol/l, in the direction of a lower sample mean with sitagliptin. The sponsor’s reported two-sided t-test gives p >0.05, and the re-calculation gives p = 0.768. The registry record flags a discrepancy, so the sponsor’s own analysis takes precedence.
The result does not establish that sitagliptin and placebo have identical effects. The confidence interval extends on both sides of zero, which leaves uncertainty about whether the underlying group difference is negative or positive. A non-significant test is not proof of no effect, and a small observed difference does not by itself establish that any underlying effect would be clinically unimportant. The registry record does not report a threshold for clinical importance or an equivalence analysis.
The endpoint addresses mean fasting glucose, measured from specified pre-meal readings, at the stated assessment. It does not answer whether the intervention changes post-meal glucose as a distinct posted outcome, prevents later diabetes, or affects outcomes beyond the intervention period. Those results are not reported in the registry record.
The analysis is per protocol, and the record states that all participants completed the intervention. The posted results include 11 participants per arm. The result should not be generalized beyond the population and assessment represented by this small trial without evidence from other data. The registry record does not report the original sample-size justification, details of the statistical plan, or a clinically important difference threshold.
In short, the posted primary result is a non-significant comparison with uncertainty around the size and direction of the difference. It is informative about the reported fasting-glucose measure in this study, but it does not establish equivalence, long-term benefit, or the absence of treatment effects on outcomes that were not posted.
17. Why This Trial Matters Statistically
This trial is statistically instructive because a small observed difference and a non-significant test can coexist with substantial uncertainty. The means are 5.78 mmol/l and 5.83 mmol/l, and the unadjusted difference is -0.05 mmol/l. Yet the confidence interval extends across zero, so the data do not resolve the direction of the underlying difference. Reading the point estimate without the interval would give an incomplete account.
The record also illustrates why a p-value should not stand alone. The sponsor reports p >0.05, while a simple unadjusted re-calculation gives p = 0.768. The registry flags a discrepancy, and the sponsor’s analysis takes precedence. This makes clear that summary-statistic calculations can be useful checks, but they cannot substitute for a reported analysis whose details are not fully available.
Another statistical lesson is the distinction between a trial’s observed sample size and a planning illustration. The trial enrolled 22 participants, while the planning figures based on the observed effect are much larger. Those illustrations depend on assumptions and do not reconstruct the original design. They also show that designing to detect smaller effects requires more participants.
Finally, the endpoint and timeframe matter. The result concerns mean fasting glucose over 8 weeks, not every aspect of glucose metabolism or longer-term clinical outcomes. Keeping the conclusion aligned with the measured outcome is essential: statistical interpretation should describe both what the data address and what the registry record leaves unanswered.
18. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
The primary limitation apparent from the posted results is the small analyzed group size: 11 participants per arm. The confidence interval extends on both sides of zero, indicating that the estimate does not determine the direction of the underlying difference. The registry record does not report the original sample-size justification, so the study’s intended ability to detect a particular effect cannot be assessed from this record.
The outcome is narrowly defined and assessed over 8 weeks. It is a mean fasting-glucose measure based on pre-meal readings, not a posted long-term outcome. The registry record does not post secondary outcomes, diabetes development, or follow-up results after the intervention. Conclusions should remain restricted to the primary measure and timeframe described.
The record identifies a per-protocol analysis and states that all participants completed the intervention. It does not report a missing-data method, an intention-to-treat analysis, or detailed criteria for the analysis population. The randomization method, allocation concealment, and procedures used to maintain double masking are also not reported in the registry record.
The sponsor’s analysis and the unadjusted re-calculation are flagged as discrepant. The sponsor’s analysis takes precedence, but the record does not explain the reason for the discrepancy. The sponsor’s p-value is reported only as >0.05, and the statistical plan details needed to understand the analysis fully are not reported.
The record does not report an equivalence framework, a clinical-importance threshold, multiplicity handling, interim analyses, or original design assumptions. Safety entries show 0 affected among 11 at risk for both serious and other adverse events in each group, but the registry record does not provide detail that would support conclusions about uncommon or longer-term events. These limitations affect how broadly the findings can be interpreted.
19. Related Tutorials
20. Related Statistical Calculators
21. Sources
- ClinicalTrials.gov: Incretins in Impaired Fasting Glucose — NCT00364377.
- Registry record last updated 2011-12-06; status last verified 2011-11.
- Results first posted 2011-12-06.
- 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.
22. Record Summary
NCT00364377 is a completed randomized, double-masked, parallel-group phase 4 study sponsored by Mayo Clinic. It enrolled 22 people with impaired fasting glucose. Participants were assigned to sitagliptin 100 mg once daily or placebo once daily, with 11 participants per group in the posted primary analysis.
The primary outcome was mean fasting glucose in mmol/l at 8 weeks, calculated from readings taken at -30, -20, -10 and 0 minutes before inpatient meal studies. The reported group means were 5.78 for sitagliptin and 5.83 for placebo. The sponsor reported a two-sided t-test with p >0.05. An unadjusted re-calculation gives a mean difference of -0.05 mmol/l and p = 0.768; the registry flags a discrepancy, and the sponsor’s analysis takes precedence.
The result does not establish equivalence or address unposted outcomes. The record reports 0 affected participants among 11 at risk in each group for serious and other adverse events. Death counts are not reported in the registry record.