Efficacy and Safety of B I1356 (Linagliptin) vs. Placebo Added to Metformin Background Therapy in Patients With Type 2 Diabetes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of B I1356 (Linagliptin) vs. Placebo Added to Metformin Background Therapy in Patients With Type 2 Diabetes (NCT00601250), a randomized phase 3 study. The primary outcome is HbA1c Change From Baseline at Week 24.

2. Clinical Question

The comparison addresses the incremental effect associated with adding linagliptin rather than placebo in the setting of ongoing metformin treatment. It is not a comparison of linagliptin alone with no treatment, nor does the primary outcome describe all aspects of diabetes management or patient experience. HbA1c change is a continuous outcome, so the reported group contrast concerns average change, not the outcome of every individual participant.

The study was randomized, double-blind, and parallel-group. Those features define the trial’s allocation and masking framework, while the posted primary analysis adjusts for baseline HbA1c and previous anti-diabetic medication. The registry record does not report further detail about how randomization was generated or concealed. The main question is limited to the stated week 24 endpoint and the population specified for its analysis.

3. Trial Design

This was a phase 3, randomized, double-blind, placebo-controlled, parallel-group study. It enrolled 701 participants with type 2 diabetes and insufficient glycaemic control despite metformin therapy. The sponsor was Boehringer Ingelheim. The registry lists the study as completed, with a start date of 2008-01 and a primary completion date of 2009-05.

Participants in the active-treatment group received linagliptin 5 mg once daily; participants in the comparator group received matching placebo once daily. Both groups continued metformin background therapy. The stated primary period was 24 weeks. The official title also describes an open-label extension to 80 weeks, with placebo patients switching to BI 1356. The primary result is the week 24 comparison; extension results are not reported in the registry record among the posted outcomes summarized for this analysis.

The primary outcome was mean change in HbA1c from baseline at week 24. The registry describes the reported group means as treatment-adjusted for baseline HbA1c and previous anti-diabetic medication. The analysis population was the Full Analysis Set, defined as treated and randomized patients who had a baseline and at least one on-treatment HbA1c measurement available. Last observation carried forward was used for imputation.

The primary endpoint counts were 175 in the placebo group and 513 in the linagliptin group. These are the counts attached to the posted endpoint analysis, not a restatement of total enrollment. Safety counts use different at-risk group sizes, reported separately in the safety results. The registry record does not report the original sample-size justification, the detailed randomization procedure, or additional operational details about trial conduct. Those features should not be inferred from the design label or enrollment count.

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: The Full Analysis Set (FAS) included all treated and randomised patients with a baseline and at least one on-treatment HbA1c measurement available. Last observation carried forward (LOCF) was used as the imputation rule.

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 HbA1c Change From Baseline at Week 24, measured over Baseline and week 24.

The primary endpoint is change in HbA1c from baseline to week 24, measured in percent. The registry defines the change as the week 24 HbA1c percent minus the baseline HbA1c percent. A negative change therefore represents a lower measured HbA1c at week 24 than at baseline, while a positive change represents a higher measured value. The endpoint is analyzed as a continuous measure, and the arm results are means.

Change from baseline makes the outcome directly interpretable relative to each participant’s starting measurement, while the between-group contrast estimates how the average change differs between randomized groups. The trial’s stated objective concerns glycaemic control, and HbA1c change is the designated primary measure for that objective. The analysis adjusts the treatment comparison for baseline HbA1c and previous anti-diabetic medication, as described in the registry record.

The registry reports an analysis population requiring a baseline HbA1c result and at least one on-treatment HbA1c result. Last observation carried forward was used when an imputed value was needed. The endpoint is therefore not a simple summary of every enrolled participant regardless of available measurements; it reflects the stated Full Analysis Set and imputation rule. It also does not, on its own, describe safety, symptoms, complications, or other outcomes.

As a general statistical point, a continuous endpoint preserves information about the size and direction of changes across the measured scale. An average can summarize a group while still concealing variation between participants, so it should be read together with its uncertainty and the analysis definition. The registry record does not report a threshold for a clinically important HbA1c difference.

7. Results Posted in the Registry

• Linagliptin: mean -0.49 versus 0.15 on control; difference -0.64 (95% CI -0.78 to -0.50), standardized effect d = -0.73; 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.

HbA1c Change From Baseline at Week 6

• 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: mean -0.36 versus 0.07 on control; difference -0.43 (95% CI -0.53 to -0.33), standardized effect d = -0.71; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Linagliptin than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

HbA1c Change From Baseline at Week 12

• 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: mean -0.50 versus 0.10 on control; difference -0.59 (95% CI -0.72 to -0.47), standardized effect d = -0.78; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Linagliptin than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

HbA1c Change From Baseline at Week 18

• 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: mean -0.50 versus 0.15 on control; difference -0.65 (95% CI -0.79 to -0.51), standardized effect d = -0.78; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Linagliptin than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary HbA1c outcomes at weeks 6, 12, and 18 follow the same broad direction as the primary week 24 comparison: the linagliptin group has a lower mean change than the placebo group at each listed visit, and the reported sponsor analyses show adjusted group contrasts below zero. This pattern is consistent across the posted follow-up points, but it does not make the secondary outcomes independent replications of the primary result. They measure the same general outcome at different times in the same study population.

The registry also lists fasting plasma glucose changes at several follow-up visits and the percentage of patients with HbA1c below 7.0% at week 24 among its secondary outcomes. Those outcomes address related but distinct summaries. The direction and magnitude of every secondary contrast should be read from its own analysis, rather than inferred from the primary HbA1c result. A pattern across related measures can add context, but it cannot replace the designated primary comparison.

Unadjusted secondary results carry less weight for confirmatory interpretation because multiple outcomes and follow-up times create multiple opportunities for apparently notable findings. If each test is interpreted separately without accounting for the full testing family, the overall chance of at least one chance finding can rise. As a general statistical principle, multiplicity procedures or a prespecified testing hierarchy can address this issue, but the registry record does not report whether such a plan was used.

The registry lists 15 outcome measures and 14 posted secondary outcomes, while 13 analyses are listed. These counts indicate a broad outcome record, but they do not reveal which results were designated confirmatory, whether all planned outcomes were posted, or how missing results were handled. Such details are not reported in the registry record. The secondary findings should therefore be treated as supporting context and interpreted with attention to their individual analysis definitions and the absence of reported multiplicity handling.

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

• HbA1c Change From Baseline at Week 6
• HbA1c Change From Baseline at Week 12
• HbA1c Change From Baseline at Week 18
• FPG Change From Baseline at Week 24
• FPG Change From Baseline at Week 6
• FPG Change From Baseline at Week 12
• FPG Change From Baseline at Week 18
• Percentage of Patients With HbA1c <7.0% at Week 24.

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is continuous: HbA1c change is recorded as a numeric percentage-point change, and the arm summaries are means. The principal comparison in the registry is an ANCOVA result reported as a least-squares mean difference. ANCOVA is a general regression method for comparing groups while adjusting for specified covariates. In this study, the registry says that treatment means were adjusted for baseline HbA1c and previous anti-diabetic medication. The least-squares mean difference is therefore an adjusted group contrast, not merely the subtraction of the two displayed arm means.

Adjustment can account for measured baseline variables included in the model and can improve the precision of a comparison when the modeling assumptions are suitable. It does not automatically eliminate all bias, correct for unmeasured differences, or guarantee that a model is correctly specified. The registry record does not report the full ANCOVA model, covariate coding, checks of model assumptions, or handling of any additional covariates. Those details should not be inferred from the method label alone.

The posted primary analysis reports a least-squares mean difference of -0.64, a 95% confidence interval from -0.78 to -0.5, and p < 0.001. The direction is relative to placebo: a negative contrast means the adjusted mean change for linagliptin is lower. Because the endpoint is change from baseline, a lower value corresponds to a more negative average change or a smaller positive change. The result is a comparison of group means and does not state the distribution of individual responses.

The registry also allows a simple unadjusted recalculation from the posted arm summaries. That calculation estimates a mean difference of -0.64 and reports p < 0.001. It is a supplementary check, not a reconstruction of the ANCOVA. It does not use the baseline and medication adjustment specified in the sponsor analysis. As a general statistical principle, recalculations based on summary data may not reproduce an analysis that uses participant-level records and a covariate-adjusted model. Here, the recalculated point estimate is close to the sponsor-reported estimate, but the sponsor analysis remains the trial’s reported primary analysis.

The arm summaries include standard errors. A standard error quantifies the sampling uncertainty of an estimated mean; it is not the standard deviation of individual participant outcomes. Standard errors tend to be smaller when estimates are based on more information, but their interpretation depends on the analysis and available observations. The registry does not provide arm-level confidence limits for these means, and the standard errors should not be treated as a description of how widely individual HbA1c changes varied.

The Full Analysis Set includes treated and randomized patients with baseline and at least one on-treatment HbA1c measurement. This definition excludes participants without the required measurement availability from the stated endpoint analysis. Last observation carried forward was the registry’s imputation rule: a participant’s most recent available on-treatment measurement is carried forward to represent a later missing measurement. As a general statistical matter, this approach relies on assumptions about the relationship between the last observed value and the unobserved value. It can understate uncertainty if the imputed value is treated as known and may be sensitive to why measurements are missing. The registry record does not report sensitivity analyses using alternative missing-data methods or details of the missingness pattern.

The posted analysis is one part of a broader outcome set. The registry lists 15 outcome measures, 13 analyses, and 14 posted secondary outcomes. Testing multiple outcomes or time points can increase the chance that at least one result appears unusual by chance when each is examined separately. The registry record does not report whether a multiplicity procedure was used, whether secondary outcomes were organized in a prespecified hierarchy, or how confirmatory claims were allocated across outcomes. Consequently, unadjusted secondary p-values should not be read as if they carried the same evidential weight as a single prespecified primary test.

The planning illustrations are distinct from the trial’s own design calculations. They use the observed effect and stated assumptions of a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. These values are prospective planning illustrations derived after observing the result, not evidence about how the trial was originally planned. The original sample-size assumptions and justification are not reported in the registry record. The planning values are not observed power.

For continuous outcomes, planning values are participants per arm. A target of 80% power means that, under the planning assumptions and a specified true effect, a study of that planned size would have an 80% probability of rejecting the null hypothesis at the stated significance level. The 90% column describes the corresponding probability under its planning scenario. These are design probabilities, not probabilities that the observed result is true. Smaller effects are harder to distinguish from random variation and therefore require more participants under the same assumptions.

In general, randomization supports a causal comparison by making treatment assignment independent of prognostic factors on average, while masking may reduce some differences in care or assessment. Neither feature ensures that all sources of bias are absent. The registry identifies the study as randomized and double-blind but does not report the sequence-generation procedure, concealment details, or the specific people masked. Interpretation should stay within what the posted design description and analysis can support.

11. Confidence Intervals and Precision

The sponsor’s primary estimate is a least-squares mean difference of -0.64 percentage points, with a 95% confidence interval from -0.78 to -0.5. This interval describes a range of values compatible with the observed data under the ANCOVA model and its assumptions. In repeated use of the same method under its assumptions, 95% confidence intervals would cover the target parameter in 95% of repetitions. It is not a statement that there is a 95% probability that this particular interval contains the true effect.

The interval lies below zero, which aligns with the direction of the reported contrast. Its width conveys uncertainty about the size of the mean difference: the estimate is not infinitely precise, even though the reported p-value is small. A confidence interval is generally more informative than a p-value alone because it presents both the direction and a range of plausible magnitudes on the outcome scale.

The simple unadjusted recalculation reports a mean difference of -0.64 and p < 0.001. Its interval, calculated from posted summary statistics, is not the same as the sponsor’s adjusted ANCOVA interval. The recalculation is a separate, unadjusted analysis and should not be substituted for the sponsor’s reported uncertainty. The registry’s sponsor analysis gives the primary inferential result.

Precision is not the same as clinical importance. A relatively narrow interval can still include effects whose practical meaning depends on a clinical threshold, and the registry record does not report such a threshold. Nor does the interval describe individual response variability. It concerns uncertainty around the estimated group contrast in the specified analysis population and model.

As a general statistical point, confidence interval precision depends on factors including the amount of information available, variation in the measured outcome, and the analysis model. The endpoint arm counts differ, and the reported standard errors describe uncertainty in arm means. The registry record does not report further diagnostics that would allow a more detailed assessment of model-based precision.

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 because the outcome is continuous; they are not total event counts. They are illustrations based on the observed effect, not a report of the calculation used to design the trial. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set to a stated fraction of the observed effect.

The 80% power column gives the planned number of participants per arm for an illustrative study to have an 80% probability of rejecting the null hypothesis if the effect being planned for were the true effect and the assumptions held. The 90% column uses the corresponding 90% probability. These probabilities describe performance under a planning scenario; they do not express the probability that the treatment effect is real or that the trial result will be reproduced.

Rows planning for 100% of the observed effect use the full observed estimate as the target. Rows planning for 75% or 50% of the observed effect target smaller effects. Smaller effects require more participants per arm to distinguish them from random variation at the same significance level and power target. For example, for the primary endpoint, the figures are 30 participants per arm for 80% power and 40 for 90% power when planning for 100% of the observed effect; for 75%, they are 53 and 71; for 50%, they are 119 and 159.

These figures are planning illustrations derived from the observed result, not observed power. The original design assumptions and sample-size justification are not reported in the registry record. The illustrations also rely on simplified assumptions and should not be treated as a substitute for a prospective design calculation using a justified target effect and appropriate variance assumptions.

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 = -0.640; pooled standard deviation σ = 0.879.
2. n = 2 × (1.96 + 0.842)² × 0.879² / -0.640² = 29.6, rounded up to 30 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 <- -0.64                      # observed difference in means
sigma <- 0.878865555544778                     # 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(601250)
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 = 30 (80% power), 40 (90% power)
 75% of observed effect: n per arm = 53 (80% power), 71 (90% power)
 50% of observed effect: n per arm = 119 (80% power), 159 (90% power)
power.t.test: 30.6 
Simulated power at n = 30 per arm: 0.796 (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 reported primary analysis is an ANCOVA comparing least-squares means, adjusted for baseline HbA1c and previous anti-diabetic medication. It gives a mean difference of -0.64, a 95% confidence interval from -0.78 to -0.5, and p < 0.001. This is the registry’s stated primary inferential result.

A separate recalculation from posted summary statistics gives an unadjusted mean difference of -0.64 and p < 0.001. The recalculated estimate is close to the sponsor estimate, but the methods are not identical. The sponsor analysis uses the stated covariate adjustment, while the recalculation uses summary statistics and does not reproduce that adjustment or the participant-level model.

The registry indicates no discrepancy between the sponsor’s result and the recalculation. Agreement in the reported point estimate is reassuring as a consistency check, but it does not make the two analyses interchangeable. In particular, precision and inferential results should be interpreted according to the method that produced them. The sponsor’s ANCOVA result is the primary reference for the trial’s reported comparison; the recalculation is supplementary and unadjusted.

Small differences in reported precision can arise from distinct methods, rounding, and the information used in each calculation. The values should not be combined into a new estimate. The sponsor’s confidence interval and p-value describe the sponsor analysis; the recalculated p-value describes the unadjusted summary-statistic comparison.

15. Multiplicity, Missing Data and Analysis Populations

The primary outcome is clearly identified as HbA1c change from baseline at week 24, and the registry separately lists numerous secondary outcomes and follow-up times. Multiple testing matters because evaluating many hypotheses increases the chance that at least one result appears statistically notable by chance if each test is considered in isolation. This concern applies especially when secondary outcomes are used to make confirmatory claims.

The registry record does not report a multiplicity adjustment, a hierarchy for testing the secondary outcomes, or a plan for controlling the overall type I error across endpoints and time points. It is therefore not possible to determine from the posted record whether secondary p-values were adjusted or how confirmatory status was assigned. Unadjusted secondary findings should be regarded as exploratory or supportive unless further documented analysis plans establish otherwise.

For the primary endpoint, the analysis population is the Full Analysis Set: all treated and randomized patients with a baseline and at least one on-treatment HbA1c measurement available. The registry specifies last observation carried forward as the imputation rule. This gives a clear description of the primary analysis population and one approach to missing follow-up values, but it does not disclose how many values were missing, when they were missing, or why participants lacked measurements.

As a general statistical matter, carrying the last observed value forward assumes that this value is a reasonable stand-in for an unobserved later outcome. That assumption may not hold in every situation, and the approach may not fully represent uncertainty about missing outcomes. The registry record does not report sensitivity analyses using other missing-data methods, analyses under alternative assumptions, or a detailed account of missingness by group. No conclusions about the robustness of the result to such choices can be drawn from the imputation label alone.

The record also does not report the complete statistical analysis plan, the original sample-size justification, or whether interim analyses were planned or conducted. These omissions limit assessment of prespecification and the full analytic context. They do not invalidate the posted result, but they constrain what can be established from the registry record.

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: hepatitis viral (0.6%), intervertebral disc protrusion (0.6%), cerebral ischaemia (0.6%), headache (0.6%), rectocele (0.6%).

The registry reports serious adverse events for 4 of 177 participants at risk in the placebo group and 18 of 523 participants at risk in the linagliptin group. These are counts of participants with serious adverse events and the corresponding at-risk denominators reported for safety. They are separate from the primary endpoint analysis counts, which use the Full Analysis Set and a different outcome definition.

The registry also reports other adverse events for 34 of 177 placebo participants and 54 of 523 linagliptin participants. Serious and other adverse event categories should not be combined or treated as interchangeable. The listed serious-event terms include Hepatitis viral, Intervertebral disc protrusion, Cerebral ischaemia, Headache, Rectocele, and Calculus ureteric, with the registry reporting percentages of 0.6% for the first five listed terms and 0.4% for Calculus ureteric.

These counts are descriptive. They do not by themselves establish whether an event was caused by treatment, and the available record does not provide the event-level details needed to assess timing, duration, severity beyond the serious classification, or clinical course. Death counts are not reported in the registry record. The safety figures should be interpreted alongside the posted definitions and available follow-up, not as a complete characterization of safety.

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

The primary result shows a lower adjusted mean change in HbA1c in the linagliptin group than in the placebo group at week 24, on top of metformin background therapy. The sponsor-reported difference is -0.64 percentage points, with a 95% confidence interval from -0.78 to -0.5 and p < 0.001. This is evidence of a between-group difference in the measured endpoint under the stated analysis, not a description of what happened to every participant.

The estimate applies to the defined Full Analysis Set and depends on the analysis model, covariate adjustment, and last-observation-carried-forward imputation specified in the registry. The analysis addresses average change in HbA1c; it does not directly answer questions about all clinical outcomes, long-term effects, or how participants valued the observed difference. The registry record does not report a threshold for a clinically important difference, so statistical evidence and clinical importance should not be treated as identical.

The trial’s placebo comparison isolates the contrast between adding linagliptin and adding matching placebo in the context of metformin treatment, as described in the record. It does not compare linagliptin with every alternative treatment, nor does it establish the outcome for people outside the study population. Generalization depends on how closely other patients and care settings resemble the participants and conditions studied; the posted registry information alone cannot settle that question.

The safety results are a separate part of the evidence. Serious adverse event counts are reported in both groups, but those counts alone do not identify causality or provide a complete account of harms. The primary efficacy estimate and the safety summaries should therefore be considered as distinct outcomes rather than collapsed into a single conclusion.

No treatment recommendation follows from this statistical interpretation. The result describes the posted comparison and its uncertainty. Decisions about the relevance of the measured difference require broader clinical evidence, consideration of potential harms, and values and circumstances not captured by the primary endpoint.

18. Why This Trial Matters Statistically

This trial illustrates how a registry result combines several statistical layers: a randomized group comparison, an adjusted continuous-outcome analysis, a defined analysis population, and an explicit missing-data rule. Reading the estimate together with those details clarifies what the numerical contrast represents. The result is not simply an unqualified difference between all enrolled participants, because the primary endpoint uses the stated Full Analysis Set and adjustment for baseline HbA1c and previous anti-diabetic medication.

The comparison also shows why an estimate and its confidence interval are more informative than a significance label alone. The difference is expressed on the HbA1c percentage scale, while the interval communicates uncertainty around the adjusted group contrast. The result can be statistically clear without answering whether the size is clinically important; that requires a meaningful threshold not reported in the registry record.

The secondary outcomes demonstrate the importance of distinguishing a consistent descriptive pattern from a confirmatory testing strategy. Related measures and repeated visits may provide context, but multiple testing affects how much confidence to place in separate secondary p-values. The record’s lack of reported multiplicity handling limits that assessment. It is also a useful reminder that a result’s evidential role depends on prespecification and analysis plans, not only on its direction.

Finally, planning illustrations based on an observed effect must not be confused with original trial design justification or observed power. They can help readers understand how assumed effect size relates to prospective sample-size needs, but the registry does not report the trial’s original assumptions. Separating the observed analysis, the sponsor’s model, the unadjusted recalculation, and the planning illustration prevents distinct statistical questions from being conflated.

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 primary endpoint analysis uses last observation carried forward for missing measurements. The registry record does not report the amount or pattern of missing data, reasons for missingness, or sensitivity analyses under alternative assumptions. The robustness of the primary estimate to the imputation approach therefore cannot be evaluated from the posted information alone.

The record identifies ANCOVA and the covariates used for adjustment but does not report the full model specification, diagnostics, or the detailed handling of covariates. It also does not report how randomization was generated or concealed, or which parties were masked. Those omissions limit a detailed appraisal of the statistical and operational procedures, although the design is identified as randomized and double-blind.

The registry lists many outcomes and analyses, but does not report a multiplicity strategy or a hierarchy for secondary outcomes. This makes the confirmatory status of secondary findings uncertain. The original sample-size justification and any interim-analysis plan are not reported in the registry record, so the prospective design assumptions cannot be compared with the observed results.

The primary result concerns the stated week 24 endpoint. Although the official title describes an open-label extension to 80 weeks with placebo participants switching to BI 1356, extension outcome results are not reported in the posted results. The available primary comparison therefore does not establish the pattern over that extension period. The record also does not report a clinical-importance threshold for the HbA1c difference, which limits interpretation of practical significance.

Finally, the safety summary gives participant counts and selected serious-event terms, but does not provide enough detail to establish causality, event timing, or a complete comparative safety profile. Death counts are not reported in the registry record. These limitations define what can be concluded from the posted data without implying that unreported information was not collected elsewhere.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Efficacy and Safety of B I1356 (Linagliptin) vs. Placebo Added to Metformin Background Therapy in Patients With Type 2 Diabetes — NCT00601250.
• Registry record last updated 2014-01-28; status last verified 2013-12.
• Results first posted 2011-06-07.
• 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

NCT00601250 is a completed phase 3 randomized, double-blind, parallel-group study of linagliptin 5 mg once daily versus matching placebo, each added to metformin in patients with type 2 diabetes and insufficient glycaemic control. The registry reports enrollment of 701 and a 24-week primary endpoint of HbA1c change from baseline.

The sponsor’s ANCOVA analysis reports a least-squares mean difference of -0.64 percentage points, with a 95% confidence interval from -0.78 to -0.5 and p < 0.001. The primary analysis used the Full Analysis Set and last observation carried forward. The registry also posts secondary outcomes and descriptive safety counts. Multiplicity handling, the original sample-size justification, and sensitivity analyses for missing data are not reported in the registry record.