Linagliptin Versus Placebo in Type 2 Diabetic Patients With Inadequate Glycaemic Control on Metformin in Combination With Pioglitazone: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Linagliptin Versus Placebo in Type 2 Diabetic Patients With Inadequate Glycaemic Control on Metformin in Combination With Pioglitazone (NCT00996658), a randomized phase 3 study. The primary outcome is Change From Baseline in HbA1c (Glycosylated Hemoglobin) After 24 Weeks.

2. Clinical Question

The central question is whether the change in HbA1c over 24 weeks differed between participants assigned to linagliptin and those assigned to placebo, with both groups continuing metformin in combination with pioglitazone. HbA1c is a measure of glycosylated hemoglobin, reported as a percentage of total hemoglobin. The primary outcome compares change from baseline rather than only the final measurement.

3. Trial Design

This was a Phase III, randomized, double-blind, placebo-controlled, parallel-group study. It evaluated once-daily linagliptin 5 mg as add-on therapy to metformin in combination with pioglitazone, compared with placebo added to the same background therapy. The stated study duration was 24 weeks. The registry lists an enrollment of 278 participants, and the study status is completed.

The primary endpoint was change from baseline in HbA1c after 24 weeks. The Full Analysis Set included randomized patients who received study medication and had both a baseline HbA1c value and an on-treatment HbA1c value. The registry states that three subjects in each arm were excluded for site non-compliance. The endpoint summaries list 89 participants in the placebo arm and 179 in the linagliptin arm. These analysis counts describe the posted endpoint results and should not be assumed to represent every participant enrolled.

The registry lists the study start as 2009-10 and primary completion as 2012-03. The named sponsor is Boehringer Ingelheim. The study also listed secondary outcomes involving HbA1c at earlier time points, HbA1c response thresholds, and fasting plasma glucose. The registry record does not report the original sample-size justification, the allocation procedure beyond randomized assignment, or further operational details of masking. Details of the statistical plan beyond the posted analyses are not reported in the registry record.

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) includes all randomized patients who received study medication and had both a baseline HbA1c value and an on-treatment HbA1c value. Three subjects in each arm excluded for site non-compliance.

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.

Population: Six participants (three participants per treatment) were excluded from all analyses due to serious site non-compliance. Thus, the total participants in Baseline Measures (272) is six less than total in Participant Flow (278).

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in HbA1c (Glycosylated Hemoglobin) After 24 Weeks, measured over baseline, 24 weeks.

The primary endpoint is change from baseline in HbA1c after 24 weeks, measured as a percentage of total hemoglobin. HbA1c reflects glycosylated hemoglobin and is used as a continuous measure of glycaemic control. An analysis of change asks how the measured value moved from each participant’s baseline, then compares the groups’ average changes.

The registry reports least-squares mean changes, with standard errors, for the Full Analysis Set. The placebo group’s reported change was -0.27, and the linagliptin group’s was -0.84. The sponsor-reported ANCOVA comparison gives a mean difference of -0.57, with a 95% confidence interval from -0.83 to -0.31. The negative direction indicates a more negative reported change in the linagliptin group than in the placebo group.

As a general statistical point, a continuous endpoint retains information about the amount of change rather than reducing participants to categories such as response or non-response. That can make it useful for comparing average changes between groups. The endpoint alone does not establish how meaningful a given change is to patients; the registry record does not report a threshold for clinical importance.

7. Results Posted in the Registry

• Linagliptin 5 mg Tablet: mean -0.84 versus -0.27 on control; difference -0.57 (95% CI -0.90 to -0.24), standardized effect d = -0.41; z-test p = <0.001.

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 HbA1c (Glycosylated Hemoglobin) After 6 Weeks

• 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 errors.
• Result:
  • Linagliptin 5 mg Tablet: mean -0.60 versus -0.19 on control; difference -0.41 (95% CI -0.69 to -0.13), standardized effect d = -0.36; z-test p = 0.004.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Linagliptin 5 mg Tablet than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in HbA1c (Glycosylated Hemoglobin) After 12 Weeks

• 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 errors.
• Result:
  • Linagliptin 5 mg Tablet: mean -0.82 versus -0.28 on control; difference -0.54 (95% CI -0.85 to -0.23), standardized effect d = -0.42; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Linagliptin 5 mg Tablet than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in HbA1c (Glycosylated Hemoglobin) After 18 Weeks

• 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 errors.
• Result:
  • Linagliptin 5 mg Tablet: mean -0.91 versus -0.37 on control; difference -0.54 (95% CI -0.85 to -0.23), standardized effect d = -0.42; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Linagliptin 5 mg Tablet than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary HbA1c changes at 6, 12, and 18 weeks follow the same direction as the 24-week primary result: the linagliptin group’s reported average change is more negative than the placebo group’s at each of those time points. This offers a consistent descriptive pattern across the reported HbA1c measurements. These time points are repeated assessments within one trial, however, and are not independent replications of the primary comparison.

The registry also lists secondary outcomes for HbA1c response thresholds and fasting plasma glucose. Detailed results for those outcomes are not reported in the registry record, so no conclusion about their estimates or uncertainty can be drawn from the listed outcome titles alone. The relationship between the secondary findings and the primary result is therefore clearest for the posted earlier HbA1c changes, rather than for every listed secondary outcome.

As a general statistical point, testing multiple outcomes and time points creates more opportunities for apparently notable findings to arise by chance. Unadjusted secondary p-values do not account for that broader set of comparisons. The registry record does not report a multiplicity adjustment or a prespecified hierarchy for interpreting secondary outcomes. Consequently, secondary results carry less inferential weight than the designated primary endpoint, especially when their multiplicity handling is unclear. They can describe the pattern in the measurements, but they should not be treated as definitive confirmation on their own.

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

• Change From Baseline in HbA1c (Glycosylated Hemoglobin) After 6 Weeks
• Change From Baseline in HbA1c (Glycosylated Hemoglobin) After 12 Weeks
• Change From Baseline in HbA1c (Glycosylated Hemoglobin) After 18 Weeks
• Occurrence of Absolute Efficacy Response (HbA1c < 7%) After 24 Weeks
• Occurrence of Absolute Efficacy Response (HbA1c < 6.5%) After 24 Weeks
• Occurrence of Relative Efficacy Response (Reduction in HbA1c >= 0.5%) After 24 Weeks
• Change From Baseline in Fasting Plasma Glucose (FPG) After 24 Weeks
• Change From Baseline in Fasting Plasma Glucose (FPG) After 6 Weeks

9. Statistical Interpretation

10. Statistical Methods Explained

The sponsor reports an analysis of covariance (ANCOVA) for the primary endpoint. ANCOVA is a general statistical method for comparing outcome values between groups while accounting for specified covariates. In a clinical trial, adjustment can improve precision when the model and covariates are appropriate. The registry identifies the method and reports the final-value mean difference, but does not report the full model specification, the covariates used, or other implementation details. Those details are not reported in the registry record.

What does a least-squares mean describe?

A least-squares mean is a model-based group mean rather than necessarily the simple arithmetic average of the observed values. It is estimated from the fitted statistical model. The registry labels the primary group summaries as least-squares means and gives standard errors for them. The sponsor’s separate ANCOVA result is a mean difference between groups. Without the complete model details, it is not possible to reconstruct the sponsor’s adjusted estimate from the registry summaries alone.

What does the standard error describe?

A standard error describes the uncertainty in an estimated mean due to sampling variability under the statistical framework used. It is not the same as the spread of individual participants’ measurements. The primary arm summaries report standard errors of 0.13 for placebo and 0.11 for linagliptin. The registry does not report a standard deviation for those arm summaries. The simple re-calculation estimates a pooled standard deviation from the standard errors and sample sizes, so that value is approximate rather than a directly reported measure of individual variability.

How should the re-calculation be distinguished from ANCOVA?

The re-calculated comparison uses the posted group summaries to produce a simple, unadjusted estimate and interval. It reports a mean difference of -0.57, an unadjusted interval from -0.904 to -0.236, and an unadjusted p-value of <0.001. This does not reproduce covariate adjustment or any other features of the sponsor’s ANCOVA. It is a check based on summary statistics, not a substitute for the sponsor’s model-based analysis. The sponsor reported -0.57 with a 95% confidence interval from -0.83 to -0.31 and p < 0.001.

What does a p-value say?

As a general explanation, a p-value measures how unusual data at least as different as those observed would be under a specified null hypothesis and analysis model. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. The sponsor’s reported p-value and the re-calculated unadjusted p-value arise from different analyses, so they should not be treated as interchangeable.

Why can model choice matter?

As a general statistical point, adjustment can affect both an estimate’s precision and its uncertainty. The sponsor’s ANCOVA may use information not available in the arm-level summary statistics. Conversely, a simple unadjusted re-calculation has fewer model assumptions but also cannot account for covariates. The registry does not report enough detail to assess every model assumption or reproduce the ANCOVA. It also does not report whether diagnostics, sensitivity analyses, or alternative models were used.

What does the Full Analysis Set mean?

The registry defines the Full Analysis Set as randomized patients who received study medication and had both a baseline HbA1c value and an on-treatment HbA1c value. Three subjects in each arm were excluded for site non-compliance. This definition describes who contributed to the posted primary analysis, but it is not equivalent to a complete account of every randomized participant’s follow-up. The handling of missing endpoint measurements beyond these stated criteria, including any imputation or sensitivity analyses, is not reported in the registry record.

As a general matter, missing outcomes can affect an estimate if participants without measurements differ systematically from those with measurements. A clear account of exclusions, missingness, and sensitivity analyses helps readers assess that possibility. The available record gives the Full Analysis Set definition and site non-compliance exclusions, but further missing-data methods are not reported in the registry record.

11. Confidence Intervals and Precision

The sponsor’s primary estimate is a mean difference of -0.57, with a 95% confidence interval from -0.83 to -0.31. The interval gives a range of values compatible with the observed data and the sponsor’s analysis model, under repeated use of the same method in comparable circumstances. It describes uncertainty in the estimated group difference; it is not a range containing a stated proportion of individual participants’ outcomes.

Every value in the sponsor-reported interval is negative. In the context of the endpoint, that is consistent with a more negative average change in the linagliptin group than in the placebo group. The interval conveys the direction and precision of the estimate, while the point estimate supplies its central value. Statistical precision is distinct from clinical importance, for which the registry record does not report a threshold.

The summary-statistic re-calculation gives an unadjusted 95% interval from -0.904 to -0.236. It has the same point estimate, -0.57, but a wider interval than the sponsor’s ANCOVA interval. The difference is not evidence of an inconsistency by itself: the analyses use different approaches, and the simple re-calculation does not reproduce the sponsor’s adjustment. The sponsor’s analysis is the registry-reported inferential result.

As a general explanation, confidence intervals depend on the estimate, variability, sample size, and analysis method. A wider interval represents greater uncertainty about the size of the group difference. A narrower interval represents greater precision under the specified method, but does not establish that an effect is important to patients or that it will persist outside the measured period.

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 sample-size figures are planning illustrations based on the observed effect, not the trial’s original design calculations. For this continuous HbA1c endpoint, the figures are participants per arm, not total events. The accompanying curve should be read as a visual representation of how the planning requirement changes as the effect to detect becomes smaller or larger; it is not an additional observed trial result.

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 as a stated fraction of the observed effect. For each assumed effect, the 80% power column gives the estimated participants per arm for a design with an 80% chance of detecting an effect of that size under those assumptions, if the effect and model assumptions hold. The 90% power column describes the corresponding planning target for a 90% chance.

Planning for 100% of the observed effect uses the full observed estimate. Planning for 75% or 50% of that effect assumes a smaller difference to detect. Smaller effects need more participants to achieve the same power, which is why the planning requirements rise as the assumed fraction falls. The listed values are useful for illustrating the relationship between effect size and sample size, not for inferring the design choices made for this completed trial.

These are not observed power estimates. The original sample-size justification and its assumptions are not reported in the registry record.

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 = -0.570; pooled standard deviation σ = 1.395.
2. n = 2 × (1.96 + 0.842)² × 1.395² / -0.570² = 94.1, rounded up to 95 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 <- -0.57                      # observed difference in means
sigma <- 1.39533460023628                     # 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(996658)
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 = 95 (80% power), 126 (90% power)
 75% of observed effect: n per arm = 168 (80% power), 224 (90% power)
 50% of observed effect: n per arm = 377 (80% power), 504 (90% power)
power.t.test: 95 
Simulated power at n = 95 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.

14. Analyses Reported by the Sponsor

The sponsor’s posted primary analysis is an ANCOVA comparison of final values. It reports a mean difference of -0.57, a 95% confidence interval from -0.83 to -0.31, and p < 0.001. The simple unadjusted re-calculation from the arm summaries also gives a point estimate of -0.57, with a confidence interval from -0.904 to -0.236 and p <0.001.

The matching point estimates are reassuring as a descriptive check, but the intervals and p-values are not identical because the re-calculation does not reproduce the sponsor’s ANCOVA. The sponsor’s result is the study’s reported model-based analysis and should be used as the principal inferential result. The re-calculation is supplementary and relies on summary statistics, including an approximate pooled standard deviation.

The registry record does not provide the full ANCOVA model specification, so the adjusted analysis cannot be independently reconstructed from the posted group summaries alone. The record also does not report additional details that would establish how the model was selected or validated. The agreement in direction and point estimate does not remove those limits on what can be checked from the posted information.

15. Multiplicity, Missing Data and Analysis Populations

The registry identifies one primary endpoint: change from baseline in HbA1c after 24 weeks. It also lists secondary outcomes that include HbA1c changes at earlier visits, response thresholds, and fasting plasma glucose. Multiple outcomes and repeated assessment times raise a multiplicity question: when many hypotheses are examined, the chance of at least one apparently unusual result can increase. As a general statistical approach, a study may define a hierarchy or adjust inference across tests, but the registry record does not report such a plan or adjustment.

The record lists 11 outcome measures and 11 analyses, but those counts alone do not explain how the analyses were prioritized or whether all planned comparisons were reported. The registry record does not report a complete statistical analysis plan, an interim-analysis plan, or the original sample-size justification. It is therefore not possible to determine from the record whether additional prespecified procedures governed multiplicity or other analysis choices.

The primary Full Analysis Set includes randomized participants who received study medication and had both baseline and on-treatment HbA1c values. The record states that three subjects in each arm were excluded for site non-compliance. It does not report a detailed accounting of missing measurements by reason or arm, how missing values were handled beyond the stated population definition, or sensitivity analyses addressing missing outcomes. These limitations matter because excluding or lacking outcome measurements can affect the interpretation if the observed participants differ from those not included.

The registry also does not report details of the randomization sequence generation, concealment, or the exact masking procedures. The study is described as randomized and double blind, but the available description does not establish those operational details. These are reporting gaps rather than evidence that a particular procedure was or was not followed.

16. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: acute myocardial infarction (1.1%), hepatitis e (1.1%), skin ulcer (1.1%), anaemia (0.5%), cataract (0.5%).

The registry reports serious adverse events in 3 of 89 participants at risk in the placebo group and 4 of 183 at risk in the linagliptin group. Other adverse events are reported for 16 of 89 and 31 of 183, respectively. These are descriptive counts with the recorded at-risk denominators; they should not be interpreted as evidence that an event was caused by the assigned study medication. The at-risk counts are not identical to the primary endpoint analysis counts in the linagliptin arm, so the populations used for safety and efficacy summaries should not be conflated.

The registry lists serious-event terms including acute myocardial infarction, hepatitis E, and skin ulcer at 1.1%, and anaemia, cataract, and upper gastrointestinal haemorrhage at 0.5%. The listed terms do not establish a causal relationship, and the registry summary does not identify a comparative estimate for each term. Death counts are not reported in the registry record. Safety interpretation is limited to these reported counts and event descriptions.

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

The registry’s primary result describes a difference between the average HbA1c changes recorded for the two study groups over 24 weeks. The sponsor-reported ANCOVA estimate is -0.57, with a 95% confidence interval from -0.83 to -0.31. The point estimate is negative, and the interval is entirely below zero. This supports a statistical distinction between the group averages under the reported analysis.

The result does not, on its own, state whether the size of the difference is important to patients. The registry record does not report a clinical-importance threshold for the endpoint. Nor does the result establish what happened after the stated observation period, whether the same pattern would appear in other populations or settings, or how the outcome relates to outcomes not measured in this analysis.

The comparison concerns linagliptin or placebo added to metformin in combination with pioglitazone among people described as having insufficient glycaemic control on that therapy. It does not compare linagliptin with other active medicines, and it does not evaluate linagliptin without the stated background therapy. The findings should be interpreted within that population, treatment context, endpoint, and follow-up period.

The primary endpoint is an efficacy measure, not a complete assessment of safety. The adverse-event counts are descriptive and do not establish causation. The registry record does not report death counts or detailed methods for handling missing primary outcome values beyond the Full Analysis Set definition. These limits mean the posted result answers a specific group-comparison question; it does not resolve every question about benefit, risk, or long-term outcomes.

18. Why This Trial Matters Statistically

This trial illustrates several central issues in interpreting clinical-trial statistics. First, a group difference needs both an estimate and an uncertainty interval. The sponsor’s mean difference of -0.57 conveys the estimated contrast, while the 95% confidence interval from -0.83 to -0.31 indicates uncertainty around that contrast under the reported ANCOVA analysis. A p-value alone would not convey the size or precision of the estimated difference.

Second, the distinction between a sponsor-reported model and a re-calculation from summary statistics matters. Both give the same point estimate, but the re-calculation is unadjusted and has a different interval. Understanding what an analysis includes—and what cannot be reproduced from registry summaries—helps prevent a simple check from being mistaken for the original inferential analysis.

Third, repeated outcomes and time points raise questions about multiplicity. Earlier HbA1c measurements show a pattern in the same direction, but they are not independent trials, and adjustment procedures are not reported in the registry record. Statistical consistency across related measurements can be informative descriptively, while still requiring caution in formal inference.

Finally, planning calculations should not be confused with trial design history or observed power. The illustrated sample-size requirements depend on specified planning assumptions and effect-size choices; the original design justification is not reported in the registry record. These distinctions make the endpoint estimate easier to interpret without overextending what the posted analysis establishes.

19. 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 leaves several trial-specific uncertainties. Although the sponsor reports an ANCOVA analysis, the full model specification and covariates are not reported. This prevents exact reproduction of the adjusted analysis from the group summaries. The simple re-calculation uses approximate variability derived from standard errors, so it should not be treated as a direct replication of the sponsor’s model.

The primary analysis population excludes participants without both baseline and on-treatment HbA1c values, as well as three subjects in each arm excluded for site non-compliance. Further details about missingness, reasons for missing outcomes, and sensitivity analyses are not reported in the registry record. The displayed efficacy sample sizes also differ from the linagliptin safety at-risk count, so readers should keep the outcome-specific populations distinct rather than assume one denominator applies throughout.

The registry lists multiple secondary outcomes but does not report a multiplicity strategy. It does not report the original sample-size justification or interim-analysis procedures. Detailed results for several listed secondary outcomes are not reported in the registry record, limiting assessment of the full outcome pattern. The trial is described as randomized and double blind, but the registry description does not give the operational details of sequence generation, allocation concealment, or masking.

Finally, the primary result covers the stated 24-week timeframe and HbA1c endpoint. The record does not report a threshold for clinical importance, and the adverse-event summary does not establish event causation. These reporting limits do not negate the posted estimate, but they constrain how broadly it can be interpreted.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Linagliptin Versus Placebo in Type 2 Diabetic Patients With Inadequate Glycaemic Control on Metformin in Combination With Pioglitazone — NCT00996658.
• Registry record last updated 2014-03-26; status last verified 2014-02.
• Results first posted 2013-03-26.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

This completed Phase III randomized, double-blind, placebo-controlled parallel-group study evaluated linagliptin 5 mg once daily added to metformin plus pioglitazone over 24 weeks in people with type 2 diabetes and insufficient glycaemic control. Registry enrollment was 278. The primary endpoint was change from baseline in HbA1c in the Full Analysis Set.

The posted least-squares mean changes were -0.27 with placebo and -0.84 with linagliptin. The sponsor’s ANCOVA analysis reported a mean difference of -0.57, with a 95% confidence interval from -0.83 to -0.31 and p < 0.001. The simple unadjusted re-calculation has the same point estimate but a different interval and p-value.

The registry also reports earlier HbA1c outcomes and adverse-event counts. Details of multiplicity handling, the full adjusted model, the original sample-size justification, and several missing-data procedures are not reported in the registry record.