The Efficacy and Safety of Liraglutide Adjunct to Insulin Treatment in Type 1 Diabetes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Efficacy and Safety of Liraglutide Adjunct to Insulin Treatment in Type 1 Diabetes (NCT02098395), a randomized phase 3 study. The primary outcome is Change From Baseline in Glycosylated Haemoglobin (HbA1c).

2. Clinical Question

The registry record identifies change in body weight and the number of treatment-emergent symptomatic hypoglycaemic episodes as secondary outcomes. Those outcomes broaden the study question beyond HbA1c, but they do not replace the primary endpoint. The available results do not establish what difference in HbA1c would be clinically important, nor do they report a threshold for judging an individual result as meaningful. The central statistical question is therefore about the direction and estimated size of the group differences, and how precisely those differences are estimated. This is an account of the trial record, not a treatment recommendation.

3. Trial Design

The trial is recorded as Phase 3, randomized, double-blind and parallel-group. Its official title describes a 26-week, insulin-capped, placebo-controlled trial. The registry record lists an enrollment of 835 and gives the study status as completed. The start is recorded as 2014-05 and primary completion as 2015-04. The sponsor was Novo Nordisk A/S.

Three liraglutide dose groups were compared with placebo, with insulin used alongside the assigned study treatment. The registry lists the placebo presentations at different volumes, as well as the active doses. It does not report further details about how the insulin cap was implemented, how randomization was generated or concealed, or whether other aspects of treatment differed by group. Those details are not reported in the registry record.

Double-blinding means the study is described as masking treatment assignment from participants and investigators, but the registry does not specify who else was masked or how masking was maintained. Randomization and blinding are design features intended to reduce certain sources of bias; they do not, by themselves, eliminate every potential source of bias or uncertainty.

The primary outcome was assessed at baseline and Week 26. The record describes the analysis population as a full analysis set of 831 subjects and notes that subjects without qualifying post-randomization data did not contribute to this endpoint analysis. Missing primary endpoint data were imputed using a mixed model for repeated measurements. The registry record does not report the original sample-size justification, planned interim analyses, or a detailed statistical analysis plan. The observed results therefore need to be read as the reported trial analyses, not as evidence about unreported design decisions.

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: Out of the 831 subjects in FAS, 22 subjects in lira 0.6 mg arm, 33 subjects in lira 1.2 mg arm, 35 subjects in lira 1.8 mg arm and 16 in placebo arm did not contribute to this analysis. Missing data imputed from a mixed model for repeated measurements (MMRM) method.

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: Baseline (BL) characteristics were presented for full analysis set (FAS = 831 subjects). Out of 835 randomised subjects, 4 were excluded from FAS:1 subject each in the lira 1.8 mg, 0.6 mg and placebo arm were not exposed to trial treatment and 1 subject in the lira 1.8 mg arm had no post-BL measurements.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Glycosylated Haemoglobin (HbA1c), measured over Week 0, Week 26.

The primary endpoint is change from baseline in glycosylated haemoglobin, or HbA1c, after 26 weeks of treatment. HbA1c is a continuous measure of glycosylated haemoglobin, reported in the registry as percent glycosylated haemoglobin. The outcome represents an average change in the measured value for each treatment group, rather than a count of participants who crossed a particular threshold.

General statistical context: continuous outcomes are often summarized using a group mean because it describes the average measured value, while a measure of spread helps show how variable individual observations are around that average. The registry reports mean change and standard deviation for each group. A mean is useful for comparing average changes between randomized groups, but it does not describe every participant’s experience or the distribution of individual responses in full.

The endpoint is used to assess the trial’s main question about change in HbA1c over the stated treatment period. The registry reports comparisons of each liraglutide dose with placebo, as well as sponsor analyses using a mixed-model approach. The endpoint results do not, on their own, establish whether a difference is clinically important: the registry record does not report a clinical importance threshold. Nor do group averages show whether changes were similar for all participants. Interpretation therefore depends on the estimated between-group differences, their uncertainty, the analysis methods and the limitations of the recorded data.

7. Results Posted in the Registry

• Liraglutide 0.6 mg: mean -0.23 versus 0.01 on control; difference -0.24 (95% CI -0.38 to -0.10), standardized effect d = -0.34; z-test p = <0.001.
• Liraglutide 1.2 mg: mean -0.23 versus 0.01 on control; difference -0.24 (95% CI -0.38 to -0.10), standardized effect d = -0.34; z-test p = 0.001.
• Liraglutide 1.8 mg: mean -0.32 versus 0.01 on control; difference -0.33 (95% CI -0.48 to -0.18), standardized effect d = -0.47; 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 Body Weight

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Liraglutide 0.6 mg: mean -2.37 versus -0.26 on control; difference -2.11 (95% CI -2.71 to -1.51), standardized effect d = -0.73; z-test p = <0.001.
  • Liraglutide 1.2 mg: mean -4.03 versus -0.26 on control; difference -3.77 (95% CI -4.45 to -3.09), standardized effect d = -1.16; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Liraglutide 0.6 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Liraglutide 1.2 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry identifies change in body weight and the number of treatment-emergent symptomatic hypoglycaemic episodes as secondary outcomes. The posted numeric secondary results cover body weight. Across the liraglutide groups, mean body-weight changes are negative, while the placebo group’s mean change is also negative but smaller in magnitude. The posted comparisons describe greater mean weight reduction in the liraglutide groups than in the placebo group. Numerical results for the symptomatic hypoglycaemic episode outcome are not reported in the registry record.

In broad terms, the available weight outcome points in a direction that is distinct from, but can be considered alongside, the primary HbA1c result. It supplies additional information about measured outcomes over the same general treatment period; it does not establish a mechanism or explain the HbA1c findings. The episode outcome would address a different aspect of the trial, but without posted numerical results its pattern and precision cannot be assessed.

Secondary findings generally carry less inferential weight than a prespecified primary result, particularly when several outcomes or comparisons are examined. General statistical context: testing many outcomes increases the chance of finding at least one apparently unusual result by chance if no adjustment or hierarchy is used. The registry record does not report whether a multiplicity strategy was applied to secondary outcomes. Accordingly, the secondary results are best viewed as supportive or exploratory information, rather than independent confirmation of the primary finding. The absence of a posted numerical result for one named secondary outcome should not be mistaken for evidence of no difference.

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

• Change From Baseline in Body Weight
• Number of Treatment-emergent Symptomatic Hypoglycaemic Episodes

9. Statistical Interpretation

10. Statistical Methods Explained

General statistical context: a randomized comparison estimates how outcomes differed between groups assigned to different treatments. Random allocation can help make groups comparable on average at the start of a trial, including for factors that were not measured. It does not ensure that every feature is balanced, and it does not remove uncertainty caused by missing observations, measurement variation or analytic choices.

What does a mean difference represent?

The primary endpoint is continuous, so its group results are reported as means and standard deviations. A mean difference compares the average outcome in one group with the average in another. For the recalculated comparisons, the placebo mean is contrasted with each liraglutide group’s mean change. The negative estimates indicate lower mean change in the liraglutide groups relative to placebo, according to the subtraction direction used in those comparisons. The estimate is expressed in the endpoint’s units: percent glycosylated haemoglobin.

General statistical context: a mean difference is a summary of group averages. It does not say that every participant had a change equal to the mean, and it does not show the full distribution of outcomes. The standard deviation describes spread among observations; it is not the same as the standard error or the confidence interval. Standard errors and confidence intervals instead describe uncertainty in an estimated quantity, given the data and statistical assumptions.

Why are there sponsor and recalculated estimates?

The registry reports sponsor analyses using mixed models, with the parameter described as mean difference in final values. A mixed model for repeated measurements can use outcome measurements across time and account for correlations among repeated observations from the same participant. The primary endpoint description says missing values were imputed using a mixed model for repeated measurements. The registry record does not report enough detail to reconstruct the full model specification, including all covariates, covariance assumptions or sensitivity analyses.

The recalculated comparisons are simple, unadjusted calculations from the posted group summary statistics. They are useful as a transparent reference for the contrast between the reported group means, but they do not reproduce the sponsor’s repeated-measures model. The two approaches may differ because they use different information and analytic assumptions. Since the registry reports no discrepancy flag, there is no stated discrepancy between the recalculations and the sponsor’s analysis. The sponsor’s model-based result remains the trial’s reported analysis; the unadjusted calculation should be read as a supplementary check.

What do p-values say?

General statistical context: a p-value describes how incompatible the observed data are with a specified null model, under the assumptions of the analysis. A small p-value 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 reports p-values for its mixed-model comparisons, while the recalculated results include unadjusted p-values. Because the registry record does not report a multiplicity procedure, p-values across several dose comparisons should not automatically be interpreted as if only one comparison had been made.

A confidence interval and a p-value answer related but different questions. The interval communicates a range of values reasonably compatible with the data and model at its stated confidence level. The p-value summarizes evidence against a null value under the specified test. Both depend on analysis assumptions, and neither resolves the question of clinical importance without context about a meaningful threshold.

Why can over-dispersion matter?

General statistical context: over-dispersion occurs when observed variability in count data exceeds what a statistical model assumes. It can make standard errors too small if the model does not allow for the extra variation, potentially making intervals too narrow and tests too optimistic. The primary endpoint is continuous, not a count outcome, so over-dispersion is not the central issue for interpreting its mean differences. The secondary outcome of symptomatic hypoglycaemic episodes is a count, but the registry record does not provide the detailed analysis model or numerical results needed to assess how that outcome was analyzed. Whether over-dispersion was evaluated is not reported in the registry record.

11. Confidence Intervals and Precision

Confidence intervals help show how precisely a group difference has been estimated. The sponsor reports 95% confidence intervals from mixed-model analyses. For the 1.8 mg group, the sponsor’s mean difference is -0.35, with a 95% confidence interval from -0.5 to -0.2. For the 1.2 mg group, the estimate is -0.23, with a 95% confidence interval from -0.38 to -0.08. For the 0.6 mg group, the estimate is -0.24, with a 95% confidence interval from -0.39 to -0.1.

General statistical context: a 95% confidence interval is produced by a method that, under its assumptions, would contain the target value in 95% of repeated applications. It is not a guarantee about this particular interval, and it does not mean there is a 95% probability that the true value lies inside it. The interval’s width conveys precision: narrower intervals indicate less statistical uncertainty than wider intervals, all else equal.

All three sponsor-reported intervals lie below zero. This is consistent with the direction of the sponsor’s reported differences, but interval position alone does not tell whether the size matters clinically. The registry record does not give a threshold for clinical importance, and the confidence intervals do not describe the range of individual participant outcomes. They describe uncertainty about group-level estimates within the model used.

The recalculated intervals are based on unadjusted summary-statistic comparisons rather than the sponsor’s repeated-measures model. Their precision should therefore be interpreted in light of that difference. As with any model-based interval, the results depend on assumptions and on the observations included or handled through the analysis. The registry record provides a description of missing-data imputation, but not a complete account of model diagnostics or sensitivity analyses.

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 expressed as participants per arm. They are not total event counts. 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 at the stated fraction of the observed effect. The original trial’s own sample-size assumptions and justification are not reported in the registry record.

The 80% power column describes the number of participants per arm that the illustration estimates would be needed to have an 80% chance of detecting the chosen effect under those assumptions, if that effect and model were correct. The 90% power column uses a higher target chance of detecting that effect and therefore generally calls for more participants per arm. These are prospective planning calculations based on observed estimates, not a description of the trial’s original design.

Rows for 100%, 75% and 50% of the observed effect show how the planning requirement changes when the assumed effect is reduced. Smaller effects are harder to distinguish from random variation and therefore need more participants per arm to reach the same target power, under the same assumptions. The sample-size curve should be read as a visual version of that relationship rather than as a new result about the participants who took part.

This is not observed power. Observed power is not a useful substitute for reading the estimated effect and its confidence interval after a trial. The figures are hypothetical planning illustrations, and their assumptions may not match the trial’s actual design or analysis. They should not be interpreted as evidence that the trial was or was not adequately powered.

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 = -0.240; pooled standard deviation σ = 0.710.
2. n = 2 × (1.96 + 0.842)² × 0.710² / -0.240² = 137.3, rounded up to 138 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 <- -0.24                      # observed difference in means
sigma <- 0.709770635795583                     # 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(2098395)
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 = 138 (80% power), 184 (90% power)
 75% of observed effect: n per arm = 245 (80% power), 327 (90% power)
 50% of observed effect: n per arm = 550 (80% power), 736 (90% power)
power.t.test: 138.3 
Simulated power at n = 138 per arm: 0.798 (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 reports mixed-model analyses of mean difference in final values for each liraglutide dose compared with placebo. The reported estimates are -0.35 for 1.8 mg, -0.23 for 1.2 mg and -0.24 for 0.6 mg, with 95% confidence intervals and p-values recorded for each comparison. These are the trial’s model-based analyses and should be distinguished from the recalculated comparisons.

The recalculations use the posted group means and standard deviations to form simple unadjusted comparisons. Their estimates are close in direction and magnitude to the sponsor’s reported results, though the exact estimates differ for some comparisons. The registry record does not flag a discrepancy. The recalculations do not incorporate the sponsor’s repeated-measures model and should not be treated as a replacement for it.

The sponsor’s analysis takes precedence as the reported trial analysis. The record does not fully specify the model, its covariates, or any multiplicity adjustment, so it is not possible to independently reproduce every feature of that analysis from the registry entries alone. The posted p-values and confidence intervals are therefore best read as the sponsor’s inferential results, with the unadjusted calculations serving only as a separate summary-statistic comparison.

15. Multiplicity, Missing Data and Analysis Populations

The primary endpoint includes three dose-versus-placebo comparisons. Multiple comparisons create a multiplicity question: if several hypotheses are tested, the chance of at least one false-positive finding can increase unless the analysis plan accounts for the family of tests. The registry record reports the sponsor’s p-values but does not report whether a multiplicity adjustment, testing hierarchy or other control of family-wise error was used. The recalculated p-values are explicitly unadjusted. The inferential interpretation should therefore distinguish the reported comparisons from a claim that a specific overall error rate was controlled.

The primary analysis population is described as a full analysis set of 831 subjects, consisting of randomized subjects who received at least one dose and had any post-randomization data. The record states that 22 subjects in the 0.6 mg group, 33 in the 1.2 mg group, 35 in the 1.8 mg group and 16 in the placebo group did not contribute to the primary endpoint analysis. These figures are reported as group-specific counts and are not added together.

The registry says missing primary endpoint data were imputed using a mixed model for repeated measurements. General statistical context: model-based handling of missing data relies on assumptions about the relationship between observed and unobserved measurements. The record does not describe those assumptions in detail, nor does it report sensitivity analyses examining alternative missing-data scenarios. It is therefore not possible to assess from the registry record how robust the primary result is to departures from the model’s missing-data assumptions.

Further details about the analysis population, missing-data procedures, multiplicity, interim analyses and statistical analysis plan are not reported in the registry record. Those omissions limit independent assessment of how the analyses were prespecified and how their uncertainty should be interpreted across outcomes and comparisons.

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: hypoglycaemic unconsciousness (2.9%), hypoglycaemia (2.4%), coronary artery disease (1.0%), hyperglycaemia (1.0%), diarrhoea (1.0%).

The registry reports serious adverse events among participants at risk in each group: 20 of 211 in the 0.6 mg group, 21 of 209 in the 1.2 mg group, 14 of 206 in the 1.8 mg group and 14 of 206 in the placebo group. These figures describe the posted serious adverse event counts and the corresponding numbers at risk. They should not be interpreted as proof that treatment caused or prevented an event.

The registry also gives the most frequent serious adverse event terms as hypoglycaemic unconsciousness (2.9%), hypoglycaemia (2.4%), coronary artery disease (1.0%), hyperglycaemia (1.0%), diarrhoea (1.0%) and vomiting (1.0%). The record does not provide group-specific percentages for these terms in the listed information, so the terms should not be assigned to a particular treatment group from these data alone.

The number of deaths is not reported in the registry record. The posted figures are a limited safety summary: they do not describe event timing, severity beyond the serious-event classification, causality assessments or all possible adverse events. Differences in event counts should be interpreted cautiously, particularly without a detailed safety analysis plan or additional context.

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

The results describe mean change in HbA1c over 26 weeks in a randomized, double-blind, placebo-controlled trial of liraglutide added to insulin. The reported sponsor analyses estimate lower mean final values for each liraglutide dose than for placebo, and the confidence intervals quantify uncertainty in those group-level estimates under the model used. The recalculated comparisons from summary statistics point in the same general direction.

These data do not establish that every participant experienced a change in the same direction or by the same amount. Group means and confidence intervals summarize population-level comparisons; they do not show the distribution of individual outcomes. Nor do the results establish whether the estimated differences are clinically important, because a clinical importance threshold is not reported in the registry record.

The findings apply to the recorded study population, treatment conditions, endpoint and follow-up period. The registry describes a multinational trial, but does not provide enough detail to determine how results might differ across participant subgroups or other settings. Results for longer-term outcomes are not reported in the registry record. Likewise, numerical results for the named symptomatic hypoglycaemic episode secondary outcome are not reported in the registry record.

Statistical evidence from a trial is not medical advice and cannot, by itself, determine an individual treatment choice. The record supports description of the measured group outcomes and the uncertainty around them; it does not resolve every question about benefit, harm, clinical importance, durability or generalizability.

18. Why This Trial Matters Statistically

This trial illustrates how a randomized comparison can estimate average differences across several dose groups while retaining a placebo comparator. Its primary endpoint is continuous, making it possible to summarize both average change and variability, and to express between-group uncertainty with confidence intervals. The registry also gives sponsor model-based results alongside the data needed for simple unadjusted re-calculations, offering a useful distinction between a reported analysis and a summary-statistic check.

It also highlights why interpretation involves more than reading a p-value. The estimated differences, their confidence intervals, the use of repeated-measures modeling, the handling of missing observations and the presence of multiple dose comparisons all matter. The registry record does not report several details needed to fully assess those choices, including a multiplicity procedure and the original sample-size justification. Those absences are part of the statistical context rather than evidence that a particular method was or was not used.

Finally, the record shows why primary and secondary outcomes should be kept distinct. The primary HbA1c result is accompanied by sponsor analyses, while secondary information is less complete and includes a named outcome without numerical results. That distinction helps prevent an incomplete set of secondary findings from being treated as equivalent to the principal analysis. The trial is therefore informative not only as a set of estimates, but also as an example of how the completeness and transparency of statistical reporting shape what readers can conclude.

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 provides limited detail about the trial’s statistical plan. The original sample-size assumptions and justification, planned interim analyses, multiplicity procedures and full model specification are not reported in the registry record. Without those details, readers cannot fully assess prespecification, reproduce the sponsor’s analysis or determine how the reported p-values relate to an overall testing strategy.

The primary endpoint analysis excluded subjects without qualifying post-randomization data, and missing values were handled using a mixed model for repeated measurements. The record does not describe the assumptions behind that approach or report sensitivity analyses. Consequently, the robustness of the estimates to different missing-data assumptions cannot be assessed from the listed information.

The listed primary group summaries and sponsor analyses provide information about average HbA1c change, but not a clinical importance threshold or detailed subgroup results. The registry does not report a threshold for judging the clinical relevance of the observed differences. Numerical results for treatment-emergent symptomatic hypoglycaemic episodes are not reported in the registry record, and the number of deaths is also not reported there. The serious adverse event summary does not include enough information to assess event timing, causality or all aspects of safety.

These limitations constrain interpretation of the posted record; they do not establish that unreported analyses were not conducted. They mark what cannot be evaluated using the registry information summarized in the trial results.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: The Efficacy and Safety of Liraglutide Adjunct to Insulin Treatment in Type 1 Diabetes — NCT02098395.
• Registry record last updated 2017-02-09; status last verified 2016-12.
• Results first posted 2016-06-03.
• 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 record describes a completed Phase 3 randomized, double-blind, parallel-group, placebo-controlled trial of liraglutide added to insulin in type 1 diabetes. The official title specifies a 26-week insulin-capped study. The primary endpoint was change from baseline in HbA1c at Week 26, analyzed in a full analysis set of 831 subjects, with missing data handled using a mixed model for repeated measurements.

The sponsor’s mixed-model analyses report mean differences versus placebo for all three liraglutide doses, with 95% confidence intervals and p-values. Simple unadjusted recalculations from the posted summary statistics show estimates in the same general direction. Secondary outcomes include body-weight change and symptomatic hypoglycaemic episodes; numerical results for the latter are not reported in the registry record. Important details, including the original sample-size justification and multiplicity procedures, are also not reported in the registry record.