Efficacy and Safety Study of Linagliptin (5 mg Administered Orally Once Daily) Over 24 Weeks in Type 2 Diabetic Patients With Insufficient Glycaemic Control Despite Metformin Therapy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety Study of Linagliptin (5 mg Administered Orally Once Daily) Over 24 Weeks in Type 2 Diabetic Patients With Insufficient Glycaemic Control Despite Metformin Therapy (NCT01215097), a randomized phase 3 study. The primary outcome is HbA1c Change From Baseline at Week 24.

2. Clinical Question

The central comparison is the difference in mean HbA1c change at week 24. A negative change represents a decrease from baseline; the comparison asks whether the estimated decrease differed between the randomized groups. The primary outcome is continuous, so the key summary is the estimated difference in means, accompanied by a confidence interval and a significance test.

The trial also assessed safety and listed secondary outcomes involving HbA1c at other time points and fasting plasma glucose. The clinical question is specific to the study population and treatment context: linagliptin or placebo added to ongoing metformin treatment. The registry record does not report details such as participant age, disease duration, or eligibility thresholds beyond the stated diagnosis and insufficient control despite metformin.

3. Trial Design

This was a Phase 3, randomized, double-blind, placebo-controlled, parallel-group study. The registry lists an enrollment of 306. Participants were assigned to linagliptin 5 mg once daily or placebo once daily, with both regimens described as add-on therapy to metformin. The planned treatment period in the record was 24 weeks.

Random allocation is intended to make the treatment groups comparable, on average, with respect to both measured and unmeasured characteristics. Double masking is intended to reduce the influence of treatment knowledge on participant reporting, clinical management, and outcome assessment. A parallel-group design compares groups receiving their assigned regimens during the same study period rather than switching participants between treatments.

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

The registry record identifies Boehringer Ingelheim as sponsor and lists the study as completed. It does not report the original sample-size justification, the full randomization procedure, or details of the masking procedures. Further protocol-level details should not be inferred from the design labels alone.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: 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 at week 24.

The primary endpoint was change in HbA1c from baseline at week 24, measured in percent. HbA1c reflects glycaemic exposure over time and is commonly used in diabetes trials as a continuous measure of glycaemic control. Change from baseline expresses the follow-up measurement relative to each participant’s starting value, while comparing changes between randomized groups helps distinguish the treatment-group contrast from changes that may occur over time in both groups.

The reported arm summaries are least-squares means, adjusted for baseline HbA1c and previous anti-diabetic medication. This adjustment accounts for those specified variables in the analysis model; it does not mean that every possible difference between participants has been removed. The primary comparison is the adjusted mean difference between linagliptin and placebo.

Change in a continuous measure retains information about the magnitude of shifts across the study groups, rather than reducing outcomes to categories such as improved or not improved. The standard error describes uncertainty around an estimated arm mean, while the confidence interval for the between-group contrast conveys a range of values compatible with the data and analysis. The registry record specifies baseline and week 24 as the primary timeframe. It does not report a minimum clinically important difference for interpreting the size of the contrast.

7. Results Posted in the Registry

• Linagliptin 5mg: mean -0.66 versus -0.14 on control; difference -0.52 (95% CI -0.69 to -0.35), standardized effect d = -0.74; 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 5mg: mean -0.46 versus -0.02 on control; difference -0.43 (95% CI -0.56 to -0.31), standardized effect d = -0.83; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Linagliptin 5mg 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 5mg: mean -0.65 versus -0.06 on control; difference -0.59 (95% CI -0.75 to -0.43), standardized effect d = -0.87; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Linagliptin 5mg 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 5mg: mean -0.65 versus -0.12 on control; difference -0.53 (95% CI -0.70 to -0.35), standardized effect d = -0.72; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Linagliptin 5mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry lists secondary outcomes that include HbA1c change at several follow-up times and fasting plasma glucose change at several time points. The posted secondary HbA1c results at weeks 6, 12, and 18 show a more negative estimated mean change in the linagliptin group than in the placebo group. Their sponsor-reported ANCOVA comparisons are also negative, with confidence intervals that exclude zero and reported p-values below < 0.001. This broad pattern is consistent with the direction of the primary week 24 result.

These secondary results help describe how the measured HbA1c pattern appeared across the reported follow-up times, but they do not replace the prespecified primary endpoint. The registry record also lists fasting plasma glucose outcomes and a week 24 HbA1c outcome for Chinese participants. Results for those listed outcomes are not reported in the registry record. The record lists 13 secondary outcomes as posted, but the detailed secondary results available for this analysis are limited to the reported HbA1c follow-up comparisons.

Multiple secondary comparisons raise a multiplicity issue: the more hypotheses tested, the greater the chance that at least one result may appear statistically notable by chance if each test is considered separately. Unadjusted secondary p-values do not account for that accumulation of testing. The registry record does not report a multiplicity adjustment or a hierarchy for interpreting secondary outcomes. Accordingly, the secondary results are best treated as supportive and descriptive rather than as independent confirmatory evidence. Their consistency with the primary result adds context, but does not erase the limitations of multiple testing or missing details about the analysis plan.

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
• HbA1c Change From Baseline at Week 24(Chinese Only)
• 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

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is a continuous outcome: change in HbA1c from baseline at week 24. For a continuous outcome, the analysis can summarize each group with a mean and compare those means. The registry gives least-squares means and standard errors. A least-squares mean is a model-based mean adjusted for specified variables; here, the record names baseline HbA1c and previous anti-diabetic medication as adjustment variables. The estimate is consequently not simply an unadjusted difference between raw group averages.

What does the adjusted contrast represent?

The reported adjusted mean difference is the estimated contrast between the groups after accounting for the specified adjustment variables in the ANCOVA analysis. ANCOVA combines a comparison of group means with regression adjustment for selected covariates. Adjustment can improve precision when covariates are related to the outcome and can account for chance imbalances in those measured covariates. It does not guarantee balance on unmeasured factors, and the registry record does not name additional covariates beyond baseline HbA1c and previous anti-diabetic medication for the primary means.

The arm means were -0.14 for placebo and -0.66 for linagliptin 5mg; the reported adjusted contrast was -0.52. The arm estimates and the adjusted contrast are related, but an adjusted contrast need not be obtained by subtracting the displayed arm means because the adjustment and model structure affect the estimates. The reported ANCOVA result is therefore the relevant sponsor analysis rather than a contrast reconstructed from rounded arm summaries.

How should the standard error and confidence interval be understood?

A standard error describes the sampling uncertainty of an estimate: under repeated studies conducted under comparable conditions, estimates would vary. It is not the spread of individual participant values. A confidence interval uses the estimate and its uncertainty to give a range of values compatible with the data under the analysis assumptions. The reported primary contrast has a 95% confidence interval from -0.70 to -0.34. Because the interval does not include zero, it aligns with the reported small p-value for a test of no difference.

The p-value is calculated relative to a null hypothesis and a statistical model. It does not express the probability that the treatment effect is zero, and it does not measure clinical importance. The registry reports p < 0.001 for the ANCOVA comparison. That reported test is distinct from the unadjusted p-value in the recalculation, which is based on summary statistics and has p < 0.001.

What is the role of the recalculation?

The posted summary statistics permit a simple, unadjusted comparison as an additional statistical reference. Its confidence interval is approximate because the recalculation uses summary data rather than the participant-level ANCOVA model. The pooled standard deviation used for that reconstruction is itself marked approximate in the calculations. Such a reconstruction can help readers understand the scale of the contrast and its uncertainty, but it cannot reproduce the adjustments, model details, or exact standard errors of the sponsor’s analysis.

For this trial, the recalculated estimate is -0.52, with an approximate 95% confidence interval from -0.689 to -0.351. The sponsor reported an adjusted estimate of -0.52, with a 95% confidence interval from -0.70 to -0.34. Similarity between those results supports a consistent statistical reading, but does not make the methods identical. Where the methods differ, the sponsor’s own analysis takes precedence for the trial’s reported primary inference.

What do the analysis population and imputation mean?

The Full Analysis Set included patients who were randomized and treated and who had both a baseline and at least one on-treatment HbA1c measurement. This definition means that the primary analysis was not described as including every enrolled participant regardless of available outcome data. The registry specifies last observation carried forward, or LOCF, as the imputation rule. In general, LOCF substitutes a participant’s last observed outcome for a later missing measurement. This can preserve participants in an analysis but relies on an assumption about what the missing value would have been; the validity of that assumption depends on why data are missing and how outcomes evolve.

The registry record does not report the extent or reasons for missing outcome data, or sensitivity analyses using other missing-data assumptions. Those omissions limit assessment of how robust the result is to missingness. They do not by themselves show that the reported estimate is biased; they mean that the record does not allow that concern to be evaluated fully.

Why can over-dispersion matter?

For continuous outcomes, variability among participant measurements affects the standard error and the width of confidence intervals. Greater variability generally makes an estimated group contrast less precise for a fixed amount of information. General statistical explanation: over-dispersion refers to variability greater than a model assumes. If a model understates variability, standard errors can be too small and intervals too narrow. The primary endpoint is continuous, and the reported summaries include standard errors; the registry record does not identify an over-dispersion issue or describe a special correction for it.

The trial’s interpretation also depends on the planned analysis choices, including outcome definition, adjustment variables, analysis population, and missing-data handling. The registry record states some of these elements, but a complete statistical analysis plan is not reported in the registry record. Therefore, the available material supports interpretation of the reported ANCOVA result, but not a full audit of every analytical decision.

11. Confidence Intervals and Precision

The sponsor-reported primary estimate is an adjusted mean difference of -0.52, with a 95% confidence interval from -0.70 to -0.34. The interval is wholly below zero, indicating that the reported data and model are more compatible with a negative difference than with no difference. Its endpoints also show that the estimate is not known exactly: a range of contrasts remains compatible with the analysis.

A confidence interval is not a guarantee that the true effect lies within its endpoints, nor does it assign a probability to each possible effect after the data are observed. In general, a 95% confidence-interval procedure would capture the true parameter in 95% of repeated studies if its assumptions held. In this trial, the interval reflects the sponsor’s ANCOVA analysis and its assumptions, including the specified adjustment and analysis population.

The recalculated interval, from -0.689 to -0.351, is approximate and comes from an unadjusted reconstruction using posted summary statistics. It should not be read as more authoritative because it has more displayed digits. The sponsor’s rounded interval is the result of the reported adjusted analysis. The similarity in direction and scale between the intervals is reassuring about the broad statistical pattern, but the intervals are not interchangeable.

Statistical precision is distinct from clinical importance. The interval indicates uncertainty in the mean contrast, but the registry record does not report a threshold for a clinically meaningful difference. The interval also addresses the group-level endpoint under the analysis model; it does not describe the spread of individual responses or guarantee that every participant experienced a change similar to the group mean.

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 endpoint is continuous and the listed assumptions specify equal allocation between the two arms. They are not total event counts. The original trial’s own sample-size assumptions and justification are not reported in the registry record; the displayed planning exercise is a separate illustration based on the observed effect.

The calculation assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It considers planning for 100%, 75%, or 50% of the observed effect. The 80% power column gives the illustrative number of participants per arm associated with an 80% chance of detecting the stated effect under those assumptions, if that effect were the true effect. The 90% power column gives the corresponding planning figure for a 90% chance under the same assumptions. These are design probabilities, not guarantees for a future study.

The rows that plan for smaller effects require more participants per arm than those planning for larger effects. That is a general property of statistical design: smaller differences are harder to distinguish from random variation, so more information is needed to detect them with a chosen power and significance level. For example, the primary-outcome planning row for 50% of the observed effect lists 116 participants per arm for 80% power and 155 participants per arm for 90% power; the row for 100% lists 29 and 39, respectively.

These figures are planning illustrations, not observed power. They do not show the probability that the trial’s result is true or the probability that a treatment effect exists. They also do not replace the sponsor’s design rationale, which is 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 = -0.520; pooled standard deviation σ = 0.705.
2. n = 2 × (1.96 + 0.842)² × 0.705² / -0.520² = 28.9, rounded up to 29 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 <- -0.52                      # observed difference in means
sigma <- 0.7050722272488                     # 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(1215097)
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 = 29 (80% power), 39 (90% power)
 75% of observed effect: n per arm = 52 (80% power), 69 (90% power)
 50% of observed effect: n per arm = 116 (80% power), 155 (90% power)
power.t.test: 29.9 
Simulated power at n = 29 per arm: 0.789 (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-reported primary analysis used ANCOVA and reported an adjusted mean difference of -0.52, with a 95% confidence interval from -0.70 to -0.34 and p < 0.001. This is the trial’s reported primary comparison. The registry describes the arm means as adjusted for baseline HbA1c and previous anti-diabetic medication.

The recalculation is a simple, unadjusted reconstruction from posted summary statistics. It gives an estimate of -0.52, an approximate 95% confidence interval from -0.689 to -0.351, and p < 0.001. It does not reproduce the sponsor’s ANCOVA model or its covariate adjustment. The recalculated interval is therefore a supplementary check, not a replacement for the reported analysis.

The direction and general scale of the sponsor result and reconstruction are consistent, and the discrepancy flag is false. Small differences in interval endpoints and p-value format reflect the different analysis approaches and reporting precision. The sponsor’s own analysis takes precedence for the trial’s primary inference. The reconstruction is useful as a transparent, simple comparison of the posted summary statistics, but should not be treated as a second independent trial result.

15. Multiplicity, Missing Data and Analysis Populations

The registry record identifies one primary endpoint, HbA1c change from baseline at week 24, alongside multiple secondary outcomes. It does not report whether the trial used a formal multiplicity procedure, a hierarchical testing strategy, or another approach to control the family-wise chance of false-positive findings across outcomes and time points. Without that information, secondary p-values should be interpreted as unadjusted evidence for their individual comparisons, not as confirmatory tests that account for all the analyses.

Multiplicity matters because repeated testing creates more opportunities for a result to cross a conventional significance threshold by chance. The primary comparison remains the clearest inferential focus because it is identified as primary in the registry. Secondary outcomes can illuminate consistency and related measures, but their evidential weight is lower when the handling of multiple comparisons is not reported.

The primary analysis population was the Full Analysis Set, defined as randomized and treated patients with baseline and at least one on-treatment HbA1c measurement. The record states that LOCF was used as the imputation rule. This gives readers some information about how incomplete measurements were handled, but the extent and pattern of missingness are not reported in the registry record. Reasons for missing data, numbers contributing at each time point, and sensitivity analyses under alternative assumptions are also not reported in the registry record.

LOCF carries an assumption that the last observed measurement is a reasonable substitute for a later missing measurement. General statistical explanation: if outcomes change after the last observation, or if missingness is related to participants’ outcomes, this substitution may not represent the missing values well. Alternative approaches can make different assumptions and may yield different estimates. The information available does not establish whether missingness materially affected the estimate; it limits independent evaluation of that possibility.

The registry record names baseline HbA1c and previous anti-diabetic medication as adjustment variables for the primary means, but the complete statistical analysis plan is not reported in the registry record. The timing and conduct of interim analyses, further analysis-population definitions, and sensitivity-analysis plans are also not reported in 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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: lacunar infarction (1.0%), cholecystitis (1.0%), gastric cancer (0.5%), benign prostatic hyperplasia (0.5%), alanine aminotransferase increased (0.5%).

The registry reports serious adverse events affecting 2 participants in the placebo group, with 100 at risk, and 4 participants in the linagliptin 5mg group, with 205 at risk. For other adverse events, it reports 13 affected participants in the placebo group, with 100 at risk, and 29 in the linagliptin group, with 205 at risk. These are separate categories as recorded; the counts should not be added together because the record does not establish whether participants or events overlap between categories.

The serious-event terms listed include lacunar infarction and cholecystitis at 1.0% each, and gastric cancer, benign prostatic hyperplasia, alanine aminotransferase increased, and comminuted fracture at 0.5% each. These registry entries describe reported events and do not establish that a treatment caused them. The record does not provide enough detail to infer comparative risk from the terms alone.

Deaths are not reported in the registry record. The safety counts and at-risk denominators differ from the primary analysis counts, and the record does not explain those differences. Adverse-event data should be read as descriptive registry reporting, not as a complete assessment of all possible harms or as proof of equivalence between groups.

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

The primary statistical result is a lower mean HbA1c change in the linagliptin group than in the placebo group over the 24-week study period, with a sponsor-reported adjusted mean difference of -0.52. The 95% confidence interval from -0.70 to -0.34 indicates that the estimated contrast is reasonably separated from zero under the reported ANCOVA model. The reported p-value is p < 0.001.

This result answers a specific group-level question in patients described as having type 2 diabetes with insufficient glycaemic control despite metformin: how did mean HbA1c change differ when linagliptin or placebo was added? It does not show that every participant had the same response, nor does it provide a prediction for an individual. The endpoint is a measure of glycaemic control, not a direct measure of symptoms, complications, quality of life, or long-term clinical events.

The statistical evidence should not be confused with evidence about clinical importance. The registry record does not report a threshold for a clinically meaningful HbA1c difference, so the size of the estimate cannot be classified against such a threshold from this record alone. Nor does a small p-value establish the probability that the treatment effect is true; it describes the data in relation to a null hypothesis and model.

The result is limited to the study treatment context and follow-up period. The record describes add-on therapy to metformin and a 24-week comparison. It does not establish long-term effects or findings in populations not represented by the stated eligibility context. The safety information is also limited to the reported adverse-event entries, and the record does not report deaths. Thus, the primary data support a clear between-group difference in the measured HbA1c endpoint, while leaving clinical importance, individual response, longer-term outcomes, and a complete benefit-harm assessment unresolved.

18. Why This Trial Matters Statistically

This trial illustrates several important distinctions in clinical-trial interpretation. First, the primary estimate is an adjusted contrast from ANCOVA, whereas a reconstruction from posted arm summaries is unadjusted and approximate. Similarity between them can provide a useful consistency check, but they answer through different calculations and the sponsor’s analysis remains the primary reported result.

Second, the endpoint is continuous. Reporting a mean change and a confidence interval conveys both the direction of the group contrast and its precision. A p-value alone would not show the range of effects compatible with the data or whether the estimate is clinically important. The absence of a reported clinical-importance threshold makes that distinction especially relevant.

Third, analysis population and missing-data choices affect what the estimate represents. The Full Analysis Set required baseline and on-treatment HbA1c data, and LOCF was used for imputation. Those details are essential for understanding the scope of the primary analysis, while missingness patterns and alternative-assumption analyses are not reported in the registry record.

Finally, the secondary outcomes provide context but involve multiple comparisons. Without a reported multiplicity strategy, those findings carry less confirmatory weight than the primary endpoint. The example demonstrates why statistical significance, precision, analysis choices, clinical relevance, and the completeness of registry reporting need to be considered together rather than treated as interchangeable questions.

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 result is based on the Full Analysis Set rather than an analysis described as including every enrolled participant irrespective of outcome availability. The registry definition requires a baseline and at least one on-treatment HbA1c measurement. LOCF imputation is specified, but the extent and reasons for missing data and sensitivity analyses are not reported in the registry record. Those omissions make it difficult to evaluate the robustness of the estimate to alternative assumptions about missing outcomes.

The registry reports an ANCOVA adjusted for baseline HbA1c and previous anti-diabetic medication, but the full statistical analysis plan is not reported in the registry record. Details of the randomization procedure, the allocation concealment process, and the masking procedures are also not reported. These details should not be inferred from the design labels.

The follow-up described for the primary endpoint is 24 weeks. The record does not establish how the observed contrast relates to longer-term outcomes. It also does not report a threshold for clinical importance, limiting interpretation of whether the estimated difference is meaningful in clinical terms.

The listed secondary outcomes are numerous, while a multiplicity strategy is not reported. This limits the confirmatory interpretation of secondary p-values. The available posted secondary results include HbA1c measurements at weeks 6, 12, and 18, but results for other listed outcomes are not reported in the registry record.

Safety information is reported as event counts and at-risk groups, with no deaths reported in the registry record. The record does not provide a full explanation of the difference between safety denominators and primary-analysis counts, or enough detail to establish causal attribution for listed serious events. These are limitations of the available record, not evidence that a particular result is invalid.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Efficacy and Safety Study of Linagliptin (5 mg Administered Orally Once Daily) Over 24 Weeks in Type 2 Diabetic Patients With Insufficient Glycaemic Control Despite Metformin Therapy — NCT01215097.
• Registry record last updated 2016-08-25; status last verified 2016-07.
• Results first posted 2013-06-03.
• 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

The registry describes a completed, randomized, double-blind, parallel-group Phase 3 trial of linagliptin 5 mg once daily versus placebo, each added to metformin, over 24 weeks. Enrollment was 306. The primary endpoint was change in HbA1c from baseline at week 24 in the Full Analysis Set, with LOCF imputation and adjustment for baseline HbA1c and previous anti-diabetic medication.

The reported ANCOVA estimate was an adjusted mean difference of -0.52, with a 95% confidence interval from -0.70 to -0.34 and p < 0.001. A simple unadjusted recalculation was similar in direction and scale, but the sponsor’s adjusted analysis takes precedence. Interpretation is limited by unreported details about missingness, multiplicity, and the complete statistical analysis plan, as well as by the 24-week outcome period and absence of a reported clinical-importance threshold.