An Efficacy and Safety Study of Fixed-dose Rosiglitazone/Glimepiride to Treat Chinese Type 2 Diabetes Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for An Efficacy and Safety Study of Fixed-dose Rosiglitazone/Glimepiride to Treat Chinese Type 2 Diabetes Patients (NCT01453049), a randomized phase 3 study. The primary outcome is Change From Baseline in Glycosylated Hemoglobin (HbA1c) at Week 24.

2. Clinical Question

The trial record describes a 24-week comparison in drug-naive participants with type 2 diabetes. The primary measure, HbA1c, reflects average plasma glucose over a period of about 2 to 3 months, according to the registry description. The outcome was defined as the Week 24 value minus the baseline value, so a negative change represents a decrease from baseline.

The record lists 86 participants enrolled overall, while the primary outcome analysis includes 35 participants assigned to the fixed-dose combination and 38 assigned to glimepiride. The registry record does not report why these counts differ or provide a full participant-flow account. The endpoint population definition and the available group counts therefore need to be kept distinct from the overall enrollment figure.

3. Trial Design

This was a phase 3, randomized, double-blind, parallel-group study sponsored by GlaxoSmithKline. Its official title describes a multicenter comparison of fixed-dose rosiglitazone/glimepiride combination therapy with glimepiride monotherapy for 24 weeks in drug-naive participants with type 2 diabetes. The record lists 86 participants enrolled and a terminated study status. The registry record does not report a reason for termination.

The intervention group received fixed-dose rosiglitazone/glimepiride, with listed combinations of 4mg/1mg, 4mg/2mg, and 4mg/4mg. The comparison group received glimepiride alone, with listed doses of 1mg, 2mg, and 4mg. Randomization is intended to distribute participant characteristics across groups by chance, while double masking is intended to limit the influence of treatment knowledge on participants and study personnel. The registry record does not describe the randomization procedure, masking procedures, or how dose levels were selected during follow-up.

The primary endpoint was change from baseline in HbA1c at Week 24. The outcome description specifies the Full Analysis Set and says missing values were imputed using Last Observation Carried Forward. The complete population-description text is not available in the registry record. The record lists 33 outcome measures, with 32 secondary outcomes posted, but no reported analyses are listed for the posted secondary outcomes.

The study record lists start and primary-completion months of 2010-04 and 2010-10, respectively. It does not provide the trial's original sample-size justification, target enrollment rationale, detailed statistical analysis plan, or the reason the study was terminated. Those omissions limit what can be inferred about the intended analysis and how the reported outcomes relate to the study's original design.

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: Full Analysis Set (FAS): all randomized participants who received \>=1 dose of study medication and had \>=1 post-Baseline efficacy assessment. Missing values were imputed using Last Observation Carried Forward (used to estimate subsequent missing data points). Only those participants contributing data at the indicated time points were analyzed.

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 Glycosylated Hemoglobin (HbA1c) at Week 24, measured over Baseline (Week 0) and Week 24.

The primary endpoint is change from baseline in glycosylated hemoglobin, or HbA1c, at Week 24. HbA1c is a form of hemoglobin used to assess average plasma glucose over a period of about 2 to 3 months. The registry describes the endpoint in percent of total hemoglobin and defines change as the Week 24 value minus the baseline value. A negative mean change therefore indicates that the average measured HbA1c was lower at Week 24 than at baseline.

HbA1c is useful in a trial with a 24-week follow-up because it reflects glucose exposure over a period that is relevant to ongoing glycemic control, rather than a single glucose measurement. The registry description notes that the American Diabetes Association has recommended a value below 53 millimoles per mole (7.0%) for most participants. That general reference does not determine whether an individual participant met a target in this trial; the posted primary result is a comparison of mean changes between treatment groups.

The registry reports a mean and standard deviation for each group. The primary analysis population is described as randomized participants who received at least one dose of study medication and had at least one post-baseline efficacy assessment, with missing values imputed using Last Observation Carried Forward. This definition matters because it is not identical to the total number enrolled. The registry record does not provide a complete account of missing primary outcome measurements or further detail about the truncated population description.

7. Results Posted in the Registry

• Glimepiride: mean -1.47 versus -1.75 on control; difference 0.28 (95% CI -0.25 to 0.81), standardized effect d = 0.24; z-test p = 0.299.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change From Baseline in Fasting Plasma Glucose (FPG) at Week 24

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Glimepiride: mean -0.88 versus -1.91 on control; difference 1.03 (95% CI -0.15 to 2.21), standardized effect d = 0.41; z-test p = 0.087.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Glimepiride and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in Homeostasis Model Assessment Sensitivity (HOMA-S) at Week 24/EW

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Glimepiride: mean -2.92 versus 1.72 on control; difference -4.64 (95% CI -8.84 to -0.44), standardized effect d = -0.79; z-test p = 0.030.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Glimepiride than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Homeostasis Model Assessment Beta-cell Function (HOMA-B) at Week 24/EW

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Glimepiride: mean 1.01 versus 1.05 on control; difference -0.04 (95% CI -0.89 to 0.81), standardized effect d = -0.03; z-test p = 0.929.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Glimepiride and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary outcomes cover several kinds of measures: fasting plasma glucose, responder classifications, HbA1c thresholds, fasting proinsulin and insulin, HOMA sensitivity and beta-cell function, and dose levels. The record lists 32 secondary outcomes posted, while the overall record lists 33 outcome measures. The available secondary results do not form a single uniform pattern across measures, and their sample sizes vary by endpoint.

These results provide context around the primary HbA1c comparison, but they do not replace it. Some secondary measures can reflect related aspects of glycemic control or metabolic response, while others classify participants using thresholds or assess different physiological measures. A result on one of these outcomes does not automatically validate or contradict the primary result, because each measure has its own definition, variability, analysis population, and uncertainty. The primary comparison remains the prespecified focus described in the registry record.

The registry lists no reported analyses for the posted secondary outcomes. The displayed comparisons are therefore unadjusted recalculations rather than reported sponsor analyses. Interpreting multiple secondary outcomes creates a multiplicity concern: when many hypotheses are examined, some apparently notable results may occur by chance even if there are no corresponding underlying effects. Without a reported multiplicity strategy, individual unadjusted findings carry less weight than a result supported by a prespecified analysis that accounts for the set of comparisons.

The secondary outcomes should consequently be read as descriptive and exploratory evidence. They may help identify questions for further study, but they do not establish a consistent overall treatment effect or resolve the uncertainty in the primary estimate. The registry record does not report a hierarchy for secondary outcomes, a family-wise error approach, or a plan for choosing among them after results were observed.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Change From Baseline in Fasting Plasma Glucose (FPG) at Week 24
• Number of HbA1c Responders and Non-responders
• Number of FPG Responders and Non-responders
• Number of Participants Who Achieved HbA1c <7%, HbA1c <=6.5%, or Who Achieved a Decrease of >=0.7% From Baseline
• Change From Baseline in Fasting Proinsulin and Insulin at Week 24/Early Withdrawal (EW)
• Change From Baseline in Homeostasis Model Assessment Sensitivity (HOMA-S) at Week 24/EW
• Change From Baseline in Homeostasis Model Assessment Beta-cell Function (HOMA-B) at Week 24/EW
• Number of Participants at Various Dose Levels at Week 24/EW

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it records a numerical change in HbA1c, rather than classifying participants only as responders or non-responders. The registry reports each group's mean change and standard deviation. A mean describes the average observed change in the analyzed group; a standard deviation describes how spread out individual changes are around that group average. A group mean is therefore not a description of every participant's outcome.

What does the difference represent?

The unadjusted mean difference compares the glimepiride group's mean change with the fixed-dose combination group's mean change. The reported estimate is 0.28, with the comparison defined as glimepiride minus the combination. Because negative changes indicate decreases from baseline, a positive difference in this direction means the average decrease was larger in the combination group. The estimate is a contrast between group averages, not a measure of individual benefit or a causal effect adjusted for other variables.

Random assignment supports a treatment-group comparison by aiming to balance participant characteristics across groups on average. However, randomization does not guarantee that every characteristic will be balanced in a particular trial, especially when the analyzed groups are modest in size. The registry record does not report baseline characteristics in the registry record for this analysis, nor does it describe adjusted modeling of the primary endpoint. The recalculation is therefore an unadjusted summary-statistic comparison.

What does the confidence interval add?

A confidence interval expresses uncertainty around an estimated group difference. In general, a wider interval indicates less precision, while a narrower interval indicates greater precision for the same type of estimate. The primary interval extends across zero, so the data are compatible with a range that includes a difference favoring either group as well as no difference. It should not be interpreted as proving that the treatments are identical. The registry record lists the interval and estimate but does not list a sponsor-reported primary analysis.

The reported unadjusted p-value is 0.299. In general, a p-value describes how unusual data at least as discrepant as those observed would be under a specified null model. It is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of an effect. The registry record does not report the sponsor's test, model, or adjustment strategy, so this p-value belongs to the simple unadjusted recalculation and should not be attributed to a reported sponsor analysis.

Why do standard deviations matter?

The observed means must be interpreted alongside the spread of participant outcomes. When outcomes vary substantially within groups, the group averages can be estimated less precisely, particularly when the number of analyzed participants is limited. The primary standard deviations are 1.078 and 1.227, with 35 and 38 participants contributing to the two groups. These values describe variability in the measured change; they are not standard errors or confidence limits.

Standard errors quantify uncertainty in an estimated mean rather than variability among individual participants. The recalculated comparison uses the posted group sizes and standard deviations to form an unadjusted estimate and interval. This is a general statistical distinction: a standard deviation describes the distribution of observations, while a standard error describes the precision of an estimate. Neither statistic removes limitations arising from missing data, trial conduct, or unreported analysis choices.

How should the missing-data method be read?

The primary outcome description says that missing values were imputed using Last Observation Carried Forward, a method that carries a participant's last observed value forward to represent later missing measurements. In general, this method is easy to describe but relies on an assumption about what would have happened after the last observed value. If outcomes would have changed after that point, carrying the last value forward may not represent the unobserved outcome well. The registry record does not report sensitivity analyses using alternative missing-data assumptions, the number of imputed values, or the timing and reasons for missing measurements.

The Full Analysis Set is defined as randomized participants who received at least one dose and had at least one post-baseline efficacy assessment. This is the population associated with the posted primary result. It differs from an intention-to-treat definition that would include every randomized participant regardless of treatment receipt or availability of a post-baseline measurement. The record's wording does not support conclusions about the handling of participants who lacked all post-baseline assessments.

Why does over-dispersion matter in general?

Over-dispersion is a general issue for count or event outcomes when observed variation exceeds what a simple model assumes. It can make standard errors too small if the extra variability is ignored, which in turn can make intervals too narrow and tests too optimistic. The primary endpoint is continuous, and no over-dispersion analysis is reported or needed to describe the posted mean comparison. This concept is relevant to statistical interpretation generally, not a claim about the primary outcome in this trial.

Overall, the statistical evidence in the registry supports describing the group means and an imprecise unadjusted contrast. It does not permit reconstruction of the original statistical plan, a sponsor-adjusted estimate, or a definitive account of how all missing information was handled beyond the stated Last Observation Carried Forward method.

11. Confidence Intervals and Precision

The primary mean difference is 0.28, and its confidence interval spans zero. That interval is the most useful summary of the precision of the unadjusted between-group estimate: it makes clear that the observed contrast is not tightly pinned down and that the data are compatible with effects in either direction. The unadjusted p-value of 0.299 is consistent with that uncertainty, but the interval gives more information about the range of differences compatible with the data than a thresholded test result alone.

The group standard deviations, 1.078 and 1.227, are appreciable in relation to the observed difference, and the primary endpoint groups include 35 and 38 participants. This combination means that the estimated group contrast is not highly precise. The interval should not be read as a set of equally likely values or as a statement that all values within it are clinically equivalent. Clinical importance cannot be determined from statistical uncertainty alone, and the registry record does not provide a threshold for a clinically important between-group difference.

The posted confidence interval is an unadjusted recalculation from the summary statistics. No sponsor-reported analysis is listed in the registry record, and the trial's analysis plan is not reported. If an adjusted or otherwise prespecified sponsor analysis exists outside this record, it cannot be evaluated from these registry entries. There is no discrepancy flagged between the summary-statistic calculation and the posted arm values, but that does not establish that the recalculation matches an analysis that was never reported.

Precision also depends on what population and missing-data assumptions the estimate represents. The outcome description specifies the Full Analysis Set and Last Observation Carried Forward, but the record does not give the number of imputed observations or sensitivity analyses. Thus, the interval communicates sampling uncertainty under the unadjusted calculation; it does not encompass uncertainty about the missing-data assumptions, possible protocol deviations, or unreported analysis choices.

12. 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.

The planning figures are participants per arm, not total participants or total events. The stated illustration assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It uses the observed effect as its reference and considers planning for 100%, 75%, and 50% of that observed effect. These are planning assumptions for the illustration; the trial's own original sample-size assumptions and justification are not reported in the registry record.

For the primary HbA1c comparison, planning for 100% of the observed effect corresponds to 269 participants per arm for 80% power and 360 participants per arm for 90% power. Planning for 75% of the observed effect corresponds to 478 participants per arm for 80% power and 640 participants per arm for 90% power. Planning for 50% of the observed effect corresponds to 1,075 participants per arm for 80% power and 1,438 participants per arm for 90% power.

In plain language, power is the chance that a study designed under specified assumptions will detect an effect of the assumed size using the chosen test and significance level. The 80% and 90% columns represent different planning targets for that chance. Smaller effects require more participants to distinguish them from random variation with the same planning assumptions. The figures are not observed power and should not be interpreted as a retrospective measure of whether this trial succeeded. They are illustrations based on the observed effect, not evidence about the trial's original design or what its investigators intended to detect.

13. 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 z_1−α/2 = 1.960) and the desired power (z_1−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

1. Observed difference Δ = 0.280; pooled standard deviation σ = 1.158.
2. n = 2 × (1.96 + 0.842)² × 1.158² / 0.280² = 268.5, rounded up to 269 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 <- 0.28                      # observed difference in means
sigma <- 1.1580424777536                     # 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(1453049)
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 = 269 (80% power), 360 (90% power)
 75% of observed effect: n per arm = 478 (80% power), 640 (90% power)
 50% of observed effect: n per arm = 1,075 (80% power), 1,438 (90% power)
power.t.test: 269.5 
Simulated power at n = 269 per arm: 0.797 (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.

14. Multiplicity, Missing Data and Analysis Populations

The registry record describes a primary HbA1c endpoint and a broad collection of secondary outcomes, including glucose measurements, responder categories, insulin-related measures, HOMA measures, and dose levels. It does not report how the investigators planned to control the chance of false-positive findings across multiple outcomes, subgroups, or comparisons. There is no reported testing hierarchy, multiplicity adjustment, or distinction between confirmatory and exploratory secondary analyses in the record.

This omission matters because a collection of unadjusted tests does not have the same error properties as a single prespecified test. In general, the more hypotheses examined, the greater the opportunity for a seemingly notable result to arise by chance. A finding from an unadjusted secondary comparison should therefore be treated cautiously, especially when no sponsor analysis or prespecified interpretation framework is reported. The primary result also needs to be interpreted in light of the absent analysis details, but the multiplicity issue is particularly important for the numerous secondary measures.

The primary outcome description does state that missing values were imputed using Last Observation Carried Forward. The registry record does not report the number of missing values, the reasons for missingness, or sensitivity analyses under other assumptions. It also does not report how participants without any post-baseline efficacy assessment were handled beyond defining the Full Analysis Set as requiring at least one such assessment. These gaps make it difficult to assess the robustness of the outcome to missing data.

The record lists 86 enrolled participants, while the primary outcome reports 35 and 38 participants in the treatment groups. It does not provide a complete participant-flow explanation that reconciles these counts. Nor does it describe a per-protocol analysis, a safety analysis population definition beyond the group denominators for adverse events, or an intention-to-treat analysis including all randomized participants. Those analysis-population details are not reported in the registry record.

The registry does not report the original sample-size justification, interim analyses, stopping rules, or the detailed statistical analysis plan. These cannot be inferred from the retrospective planning illustration or from the trial's terminated status. Without that information, the planned handling of missing data, multiplicity, and analysis populations remains uncertain.

15. 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 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports zero participants with serious adverse events in each group: 0 among 40 participants at risk in the fixed-dose combination group and 0 among 43 at risk in the glimepiride group. These are participant counts, not event rates. No serious adverse-event terms are listed in the record.

Other adverse events were reported for 6 of 40 participants at risk in the fixed-dose combination group and 10 of 43 at risk in the glimepiride group. These figures describe the posted counts and denominators; they do not provide information about event severity, duration, causal attribution, or the number of events per participant. They should not be treated as a complete safety comparison without further detail on definitions and follow-up.

Death counts are not reported in the registry record. The absence of a listed serious adverse event in these figures does not establish that no unrecorded events occurred, and a trial of this size cannot by itself characterize uncommon harms precisely. The registry record is marked terminated, but it does not report a reason for termination or connect that status to safety findings.

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

The posted primary estimates show decreases in mean HbA1c from baseline in both treatment groups over the study period. The unadjusted group difference is 0.28 in the direction of a larger average reduction in the fixed-dose combination group. However, the interval spans zero and the unadjusted p-value is 0.299, so the posted summary statistics do not establish a precise between-group difference.

This is not the same as demonstrating that the treatments have identical effects. A non-significant unadjusted comparison can reflect uncertainty, and the interval is compatible with differences in either direction. Equivalence or non-inferiority would require an appropriate prespecified margin and analysis; neither is reported in the registry record. The trial's stated primary objective was superiority, and the available recalculation does not establish that objective.

The results also do not establish that either treatment is effective or ineffective for an individual participant, nor do they give medical advice. They describe average changes in a defined trial population, under the endpoint and missing-data approach stated in the record. Individual responses may vary, as reflected by the reported standard deviations. The trial comparison does not supply a complete picture of longer-term outcomes, outcomes beyond those listed, or treatment decisions for people outside the study.

Interpretation is limited by the absence of a reported sponsor analysis, detailed statistical plan, original sample-size justification, participant-flow explanation, and missing-data sensitivity analyses. The registry's terminated status is relevant context, but its reason is not reported. The secondary outcomes span multiple measures and lack a reported multiplicity plan, so they should not be used as if they were independent confirmatory evidence.

In short, the registry supports a description of the observed group means and their uncertain unadjusted contrast. It does not support a definitive claim about superiority, equivalence, clinical importance, or the reason for termination. Those conclusions would require information not reported in the registry record.

17. Why This Trial Matters Statistically

This trial illustrates why a difference between group averages should be interpreted together with its precision. Both groups had mean HbA1c reductions, and the unadjusted estimate favors a larger reduction in the combination group. Yet the confidence interval crosses zero and the p-value is 0.299. Reporting the direction of the point estimate without its uncertainty could make the comparison seem more decisive than the data support.

It also shows the distinction between an observed outcome analysis and planning illustrations based on observed effects. The sample-size figures describe how many participants per arm would be required under stated assumptions to plan for effects of different sizes. They do not reveal the study's original sample-size rationale and are not observed power. The registry record does not report the original justification, so the illustration cannot be used to claim that the trial was designed or powered in a particular way.

The record includes a broad set of secondary measures but no reported secondary analyses or multiplicity strategy. This is a common interpretive challenge in clinical research: a collection of results may seem to offer multiple opportunities for confirmation, but unadjusted comparisons do not necessarily provide independent evidence. Clear prespecification and transparent reporting are important to distinguish confirmatory analyses from exploratory observations.

Finally, the primary outcome population, total enrollment, and adverse-event denominators are not interchangeable. The registry reports 86 enrolled, primary group counts of 35 and 38, and safety denominators of 40 and 43. Keeping these analysis-specific counts separate helps prevent misstatements about how many participants contributed to each type of result. The record's missing participant-flow and termination details further show why registry entries should be interpreted within the limits of what they actually document.

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 registry record is marked terminated, but the reason for termination is not reported. Without that context, the status cannot be connected to the outcome estimates, study conduct, safety, or recruitment. The record lists 86 enrolled participants, while the primary endpoint includes 35 and 38 participants in the two groups; it does not explain the difference or provide a full account of withdrawals, exclusions, and missing measurements.

The primary result available for comparison is an unadjusted recalculation from posted group summary statistics. No sponsor-reported primary analysis is listed in the record. The original statistical analysis plan, sample-size justification, analysis model, and any prespecified covariate adjustment are not reported. The recalculation should therefore not be mistaken for a reported sponsor conclusion.

Missing primary values were handled using Last Observation Carried Forward, according to the outcome description. The registry does not state how many values were imputed, why they were missing, or whether results were tested under alternative missing-data assumptions. The primary population definition also requires at least one post-baseline efficacy assessment, so it does not describe all randomized participants.

The trial compares groups over 24 weeks and the registry record does not provide information on outcomes beyond the posted measures and timeframe. The group counts are limited, and the primary confidence interval indicates substantial uncertainty about the group contrast. Safety figures are participant counts with denominators, but detailed event descriptions and death counts are not reported. The reason for termination and details of protocol implementation are also unavailable in the record.

Finally, the large set of secondary outcomes is not accompanied by a reported multiplicity plan or sponsor analyses. Those results should be interpreted cautiously and cannot be treated as independent confirmation of the primary comparison. These limitations do not invalidate the posted descriptive data; they define what can and cannot be concluded from this registry record.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: An Efficacy and Safety Study of Fixed-dose Rosiglitazone/Glimepiride to Treat Chinese Type 2 Diabetes Patients — NCT01453049.
• Registry record last updated 2012-03-27; status last verified 2011-12.
• Results first posted 2012-01-13.
• 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.

22. Record Summary

NCT01453049 is a phase 3 randomized, double-blind, parallel-group study comparing fixed-dose rosiglitazone/glimepiride with glimepiride alone in drug-naive participants with type 2 diabetes. The planned treatment comparison was 24 weeks, and the listed primary endpoint was change from baseline in HbA1c at Week 24. The registry reports 86 participants enrolled and a terminated status.

The posted primary group means were -1.75 and -1.47, with an unadjusted mean difference of 0.28. Its confidence interval spans zero, and the unadjusted p-value is 0.299. The record lists secondary outcomes and adverse-event counts, but no sponsor-reported analyses are listed for the primary comparison or secondary outcomes. The original sample-size rationale, detailed statistical plan, reason for termination, and complete participant-flow account are not reported in the registry record.