Combined Liraglutide and Metformin Therapy in Women With Previous Gestational Diabetes Mellitus (GDM): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Combined Liraglutide and Metformin Therapy in Women With Previous Gestational Diabetes Mellitus (GDM) (NCT01234649), a randomized phase 3 study. The primary outcome is Insulin Secretion-Sensitivity Index (IS-SI).

2. Clinical Question

3. Trial Design

This was a phase 3, randomized, parallel-group trial with triple masking, listed as completed. The registry lists enrollment as 153. The comparison was between metformin XR plus liraglutide and metformin XR plus placebo; both groups received metformin, so the contrast addresses the addition of liraglutide rather than metformin versus no metformin.

The intervention descriptions specify gradual metformin dose escalation and a daily injection, active liraglutide in one group and placebo in the other. The liraglutide dose was increased as tolerated up to a maximum of 1.8 mg daily. The treatment timeframe for the primary endpoint was 84 weeks. The registry lists the study start as 2011-08-11 and primary completion as 2019-04-24.

The primary outcome is a continuous measure, the insulin secretion-sensitivity index, reported as a mean with standard deviation. The posted outcome summaries include 35 participants in the liraglutide-metformin group and 37 in the placebo-metformin group. These outcome-specific counts are not the same as the registry enrollment figure; the record does not explain the difference in the posted primary-outcome information.

Randomization and masking describe important design features, but the registry record does not report the allocation method in more detail, how masking was implemented, or whether masking was assessed. It also does not report the original sample-size justification, the planned analysis population, or the trial's detailed statistical analysis plan. Those omissions limit how fully the design and analysis can be reconstructed from the registry record.

4. Participant Flow and Accounting

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

Before assignment: 166 were consented but 153 met criteria

Overall Study

Reasons for not completing this period:

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 Insulin Secretion-Sensitivity Index (IS-SI), measured over 84 weeks of treatment.

The primary endpoint is the insulin secretion-sensitivity index (IS-SI), measured after 84 weeks of treatment. It is a continuous index intended to represent insulin secretion in relation to insulin sensitivity. The trial description connects the measure to the study's aim of assessing whether adding liraglutide to metformin changes insulin sensitivity and normalizes insulin secretion in women with prior GDM and metabolic defects.

Insulin secretion and insulin sensitivity are related physiological features. A measure that considers them together can be useful because insulin secretion is interpreted in the context of the body's sensitivity to insulin, rather than as an isolated quantity. This is a general explanation of the concept; the registry record does not report the formula used to calculate IS-SI, its component measurements, or a threshold for a clinically meaningful change.

The endpoint is reported as a mean and standard deviation in each group. The posted summaries give a mean of 418.4 for the metformin-plus-liraglutide group and 333.0 for the metformin-plus-placebo group. The unadjusted recalculation estimates a mean difference of 85.4, with a 95% confidence interval from -58.97 to 229.8. This interval reflects substantial uncertainty about the size and direction of the difference. The registry record does not report how a change in the index should be translated into a clinical outcome, nor does it identify a threshold that would indicate an important benefit to participants.

7. Results Posted in the Registry

• Metformin XR Plus Liraglutide: mean 418.40 versus 333.00 on control; difference 85.40 (95% CI -58.97 to 229.77), standardized effect d = 0.28; z-test p = 0.246.

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.

Fasting Blood Glucose (FBG)

• 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:
  • Metformin XR Plus Liraglutide: mean 90.00 versus 91.70 on control; difference -1.70 (95% CI -5.14 to 1.74), standardized effect d = -0.23; z-test p = 0.333.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Metformin XR Plus Liraglutide and the control. The interval gives the range of effects that remain compatible with the data.

Mean Glucose During OGTT (MBG)

• 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:
  • Metformin XR Plus Liraglutide: mean 121.60 versus 118.80 on control; difference 2.80 (95% CI -6.40 to 12.00), standardized effect d = 0.14; z-test p = 0.551.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Metformin XR Plus Liraglutide and the control. The interval gives the range of effects that remain compatible with the data.

Homeostasis Model Assessment of Insulin Resistance (HOMA-IR)

• 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:
  • Metformin XR Plus Liraglutide: mean 2.20 versus 2.45 on control; difference -0.25 (95% CI -0.92 to 0.42), standardized effect d = -0.17; z-test p = 0.464.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Metformin XR Plus Liraglutide and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary outcomes cover glucose measurements, insulin sensitivity and secretion measures, and body-size measures. The subset of reported secondary results includes outcomes whose unadjusted recalculated directions do not all align. This mixed pattern does not form a single, uniformly consistent signal across the different measures. The registry lists 21 posted secondary outcomes, while detailed inferential information is not presented for every listed outcome in the record.

Secondary outcomes can add context to the primary endpoint by showing whether other metabolic measures move in a compatible direction. They are not interchangeable with IS-SI: each represents a different measurement and can have different variability, meaning, and analysis assumptions. A difference in one secondary measure does not by itself establish a corresponding change in the primary endpoint or a later clinical outcome.

The registry reports sponsor analyses for some secondary measures, while the recalculated comparisons are simple unadjusted analyses. Where these differ, the sponsor's own analysis takes precedence as the reported trial result, but the registry record does not provide enough detail to reconcile each discrepancy. The secondary results should therefore be read as supporting context rather than as a substitute for the primary outcome.

Testing many outcomes creates a multiplicity concern: even when no underlying effects exist, repeated testing can make chance findings more likely. This is a general statistical issue. The registry record does not report a multiplicity adjustment or identify a hierarchy for interpreting secondary outcomes. Without that information, unadjusted secondary p-values carry less weight than a clearly prespecified primary analysis, and isolated nominal findings should not be treated as confirmatory evidence. The overall pattern is most appropriately described as exploratory context with limits on how strongly it can support the primary conclusion.

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

• Fasting Blood Glucose (FBG)
• Mean Glucose During OGTT (MBG)
• Homeostasis Model Assessment of Insulin Resistance (HOMA-IR)
• Matsuda Insulin Sensitivity Index Derived From OGTT
• Insulinogenic Index (IGI) /HOMA-IR
• Absolute Body Weight
• Change in Body Weight From Baseline to End of Study (Expressed as % Compared to Baseline)
• Body Mass Index (BMI)

9. Statistical Interpretation

10. Statistical Methods Explained

What does the unadjusted recalculation represent?

The recalculation compares the posted group means using a simple, unadjusted difference in means. For the primary endpoint, the estimated difference is 85.4, defined in the direction of metformin XR plus liraglutide relative to metformin XR plus placebo. It uses the posted group sizes and standard deviations rather than individual participant data. It does not adjust for baseline characteristics, account for any additional design features, or reproduce an analysis that might have been prespecified by the investigators.

“Unadjusted” is important: the recalculation is a transparent comparison of summary statistics, not a substitute for the sponsor's analysis. The registry reports that the sponsor used ANOVA for the primary comparison, but does not give the model details or adjusted estimate. When the recalculation and the sponsor's result differ, the sponsor's own analysis takes precedence as the trial's reported analysis. The recalculation remains useful for understanding the posted group summaries and the uncertainty that accompanies a simple comparison.

What does the confidence interval express?

A confidence interval gives a range of effect estimates compatible with the data under the assumptions of the analysis. In general, a wider interval indicates less precision, while a narrower interval indicates greater precision. For the primary recalculation, the interval extends from -58.97 to 229.8. Since it includes zero, the summary-statistic recalculation is compatible with no mean difference as well as with differences in either direction. It is not a probability statement that the true effect has a particular chance of lying within the displayed interval.

The reported ANOVA p-value and the recalculated interval come from different analyses, and the registry does not give an ANOVA confidence interval. They should not be combined as though one directly describes the other. The record does not explain whether the ANOVA included covariates, used a different analysis population, or handled missing observations differently.

Why can reported and recalculated p-values differ?

A p-value depends on both the estimated contrast and how its uncertainty is calculated. Different models, analysis populations, assumptions, or handling of observations can yield different uncertainty estimates and therefore different p-values. These are general statistical possibilities, not descriptions of what happened in this trial. The registry record does not report enough detail to identify the reason for the discrepancy between the primary ANOVA result and the unadjusted recalculation.

The sponsor's primary result is reported as p <0.04, while the unadjusted recalculation is p = 0.246. A p-value measures how incompatible the observed data are with a specified null model under the analysis assumptions; it is not the probability that the treatment has no effect, and it does not measure clinical importance. The sponsor's p-value is a thresholded result, not an exact value, and the record does not pair it with an effect estimate or confidence interval.

Why does variability matter?

The primary outcome has substantial spread within each group: the reported standard deviations are 387.0 and 206.0. Standard deviation describes variation among observations around a group mean. When individual measurements vary widely relative to the difference between group means, it is harder to estimate the group contrast precisely. This general relationship helps explain why the recalculated interval is broad; it does not establish the cause of the observed variability.

Some continuous outcomes can have distributions with more variability than a simple model anticipates, a feature often called over-dispersion in applicable settings. Over-dispersion can make standard errors too small if the analysis model fails to allow for the extra variability. It is a general modeling concern, not a finding reported for this trial. The registry record does not report distributional checks or any special variance model for the primary endpoint.

What does the design contribute?

Random allocation is intended to make comparison groups similar on average with respect to factors that could affect outcomes. Masking can reduce the influence of expectations on treatment delivery, outcome assessment, or reporting, depending on how it is implemented. These are general design principles. The registry identifies the trial as randomized and triple masked, but the implementation details and the people included in the masking designation are not reported in the registry record.

A parallel-group design compares participants assigned to one treatment group with those assigned to another over the same study period. Here, both groups received metformin and differed by liraglutide versus placebo. This design supports a between-group comparison of the combination therapy with metformin alone, subject to the trial's analysis methods and the completeness of the outcome data.

What is not established by the posted analysis?

The registry record does not report the original sample-size assumptions, a detailed statistical analysis plan, the primary analysis population, missing-data methods, or whether interim analyses were conducted. It also does not report an adjustment for multiple outcome comparisons. These details matter because they clarify which analysis was planned, how uncertainty was handled, and whether the reported result accounts for testing several outcomes. Their absence does not show that a procedure was or was not used; it means the procedure is not reported in the registry record.

11. Confidence Intervals and Precision

The unadjusted primary estimate is a mean difference of 85.4, with a 95% confidence interval from -58.97 to 229.8. The interval is wide in relation to the point estimate and crosses zero. It therefore does not pin down a single direction or a narrowly defined magnitude for the difference. The observed group means favor a higher IS-SI mean in the liraglutide-metformin group, but the recalculation remains imprecise.

A confidence interval describes uncertainty in an estimate under the model and assumptions used to calculate it. In general, intervals that include zero for a difference between means are compatible with no difference at the confidence level used. They may also include potentially meaningful differences in either direction; deciding clinical meaning requires context and a meaningful-effect threshold, neither of which is reported for IS-SI in the registry record.

The confidence interval is from the unadjusted recalculation, not the sponsor's ANOVA. The sponsor's result is reported as p <0.04, but its effect estimate and confidence interval are not reported in the registry record. Consequently, the precision of the sponsor's estimated effect cannot be evaluated from that p-value alone. A significance threshold does not communicate how large or stable an estimate is.

Group standard deviations of 387.0 and 206.0 also show that individual IS-SI values varied within the groups. The registry record does not explain the source of that variability or provide further distributional information. The outcome-specific group sizes are 35 and 37, and the record does not report how missing outcome data, if any, were handled. These factors make it important to distinguish the point estimate from the range of values compatible with the recalculation, and to avoid treating the estimate as a precise measure of a clinical effect.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total event counts. They are illustrations based on the observed effect and the stated planning assumptions: a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The original trial's own sample-size assumptions are not reported in the registry record, so these figures should not be read as the trial's enrollment target or as a reconstruction of its design calculation.

The 80% power column describes the number of participants per arm an illustrative future study would require, under those assumptions, to have an 80% chance of detecting the chosen effect if that effect were the true one. The 90% power column describes the corresponding planning target for a 90% chance. These are planning concepts, not guarantees about what a study will conclude.

The rows consider planning for 100%, 75%, or 50% of the observed effect. For the primary outcome, the illustrations range from 204 participants per arm for 80% power and 273 for 90% power when planning for 100% of the observed effect, to 815 per arm for 80% power and 1,090 for 90% power when planning for 50% of the observed effect. The row for 75% of the observed effect gives 362 per arm for 80% power and 485 for 90% power.

Smaller effects are harder to distinguish from random variation, so detecting a smaller effect generally requires more participants. That relationship is reflected in the increasing planning figures as the assumed fraction of the observed effect decreases. These calculations are not observed power; they are prospective sample-size illustrations based on the stated assumptions, not a measure of the trial's achieved probability of detecting an effect.

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 = 85.400; pooled standard deviation σ = 307.521.
2. n = 2 × (1.96 + 0.842)² × 307.521² / 85.400² = 203.6, rounded up to 204 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 <- 85.4                      # observed difference in means
sigma <- 307.521009735224                     # 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(1234649)
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 = 204 (80% power), 273 (90% power)
 75% of observed effect: n per arm = 362 (80% power), 485 (90% power)
 50% of observed effect: n per arm = 815 (80% power), 1,090 (90% power)
power.t.test: 204.5 
Simulated power at n = 204 per arm: 0.802 (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 registry reports the sponsor's primary analysis as an ANOVA superiority comparison with p <0.04. It does not report the corresponding effect estimate, confidence interval, model details, or exact p-value. The sponsor's reported analysis therefore supplies the trial's own inferential result, but the posted information is not sufficient to evaluate the size and precision of that result fully.

The unadjusted recalculation from the posted means and standard deviations estimates a mean difference of 85.4, with a 95% confidence interval from -58.97 to 229.8 and p = 0.246. This is not the same analysis as the sponsor's ANOVA, and the registry record does not explain the discrepancy. Possible reasons for different results in general include differences in modeling, analysis population, or handling of observations, but none can be identified as the reason in this trial from the registry information.

Because the sponsor's own analysis takes precedence when it differs from a simple recalculation, the ANOVA result should be identified as the trial's reported analysis. At the same time, its thresholded p-value should not be mistaken for an effect estimate or a measure of precision. The recalculation is best understood as an unadjusted check on the posted summary statistics, not as a replacement for the sponsor's analysis or an explanation of why the results differ.

15. Multiplicity, Missing Data and Analysis Populations

The registry record identifies a primary outcome and lists secondary outcomes, but it does not report a detailed statistical analysis plan or an approach for controlling multiplicity. When several outcomes are tested, the chance of at least one apparently notable result arising by chance can increase. This is a general statistical concern, not evidence that any particular posted result is false. Without a reported adjustment, testing hierarchy, or other prespecified strategy, secondary p-values should be interpreted cautiously.

The record also does not report the primary analysis population, whether the analysis followed an intention-to-treat principle, or how participants with missing outcome measurements were handled. It does not state whether missing values were imputed, whether a complete-case analysis was used, or whether sensitivity analyses examined the impact of missing data. These are not reported in the registry record; their absence from the record does not establish which procedures were used or whether data were missing.

Those omissions matter because analysis population and missing-data assumptions can influence both the estimated group contrast and its uncertainty. A complete description would clarify who contributed to each analysis and how incomplete follow-up was treated. The registry lists outcome-specific primary group sizes of 35 and 37, but does not explain how those counts relate to enrollment or whether any participants lacked endpoint data.

The sponsor's primary result is identified as ANOVA, but further model details are not reported. There is no reported account of covariate adjustment, model diagnostics, interim analyses, or the original sample-size justification. These gaps limit independent assessment of whether the reported analysis followed a prespecified plan and how robust its conclusions are to alternative reasonable choices. They should be treated as limits of the information in the registry record, not as proof that an analysis procedure was omitted.

16. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse events affecting 0 of 78 at risk in the metformin XR plus liraglutide group and 0 of 75 at risk in the metformin XR plus placebo group. Deaths are reported as 0 in each group. These counts describe the safety information posted in the registry; they do not establish that no adverse experiences of any kind occurred.

For other adverse events, the registry reports 30 affected among 78 at risk in the liraglutide group and 14 affected among 75 at risk in the placebo group. These are counts of participants reported as affected, not a comparison adjusted for follow-up or other factors. The registry record does not provide event-level detail in the information used here, and no serious adverse-event terms are listed.

The safety denominators differ from the outcome-specific primary group sizes of 35 and 37. The registry record does not explain this difference, so the figures should be interpreted using their own reported denominators rather than assumed to refer to the same analysis population. Safety counts alone do not establish causation or quantify risk for an individual.

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

The posted primary summaries show a higher mean IS-SI in the metformin-plus-liraglutide group than in the metformin-plus-placebo group. The simple unadjusted recalculation estimates a mean difference of 85.4, but its 95% confidence interval, from -58.97 to 229.8, is wide and includes zero. The sponsor's ANOVA is reported with p <0.04, and that sponsor analysis takes precedence over the recalculation as the trial's reported inferential result. Its effect estimate and confidence interval are not reported in the registry record.

The result does not, by itself, establish the size or clinical importance of a difference in IS-SI. The record does not report a clinically meaningful threshold for the index, nor does it link the measured difference to later development of type 2 diabetes or another clinical outcome. The study's rationale concerns metabolic risk after GDM, but rationale is not evidence that a particular long-term outcome changed in this trial.

The comparison is specific to the studied intervention and population: adding liraglutide to metformin XR versus metformin XR plus placebo in women described as overweight or obese with metabolic defects and recent GDM history. It is not a comparison of liraglutide alone with no treatment, and the posted results should not be generalized automatically to other populations or treatment settings.

The secondary outcomes offer additional metabolic measurements, but their interpretation is limited by the number of outcomes and the absence of reported multiplicity procedures. Safety counts are also descriptive and have limitations in their denominators and event detail. Taken together, the registry record supports a careful account of the reported statistical findings, not medical advice or a definitive conclusion about treatment effects on health outcomes.

18. Why This Trial Matters Statistically

This trial addresses a clinically relevant period after gestational diabetes, when future metabolic health is a concern. Its randomized comparison is focused: both groups received metformin, while the intervention group also received liraglutide and the comparator group received placebo. That structure makes the central question the incremental comparison of adding liraglutide to metformin in the stated population.

Statistically, the record illustrates why a p-value should not be read in isolation. The sponsor reports a superiority analysis with p <0.04, while an unadjusted recalculation from posted group summaries gives a much wider, zero-crossing confidence interval and p = 0.246. The distinction between a sponsor analysis and a simple summary-statistic recalculation matters: they are not interchangeable, and the registry does not supply enough detail to reconcile them. The sponsor analysis takes precedence as the trial's own reported analysis, while the recalculation makes the uncertainty in a basic comparison visible.

The primary endpoint is a physiological index rather than a direct clinical event. It can be relevant to a metabolic research question, but its interpretation requires information about the measure's clinical meaning and a threshold for important change; neither is reported in the registry record. The secondary outcome set broadens the metabolic picture, while also raising the general issue of multiple testing and the need for a clear analysis hierarchy.

The trial is also a useful reminder that trial registration may leave important analytical details unreported. The original sample-size rationale, analysis population, missing-data strategy, and multiplicity plan are not reported in the registry record. Those details are important for judging precision and robustness, and their absence constrains what can be concluded from the posted results.

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 recalculation relies on group-level means, standard deviations, and outcome-specific group sizes. It cannot reproduce an analysis using individual participant data, adjust for baseline values, or assess the distribution of observations. The recalculated interval is broad and includes zero, while the sponsor's ANOVA is reported only as p <0.04. Without the sponsor's estimate, interval, model specification, or analysis population, the discrepancy cannot be resolved from the registry record.

The registry lists enrollment as 153, whereas the posted primary-outcome summaries give group sizes of 35 and 37. The record does not explain how these figures relate, and no inference should be made about the reason for the difference. Safety denominators are also reported separately. The record does not clarify whether these different counts reflect eligibility, follow-up, outcome availability, or another factor.

The original sample-size justification, statistical analysis plan, missing-data methods, multiplicity strategy, and primary analysis population are not reported in the registry record. There is also no reported clinically meaningful threshold for IS-SI, and the posted primary outcome is not a direct long-term clinical outcome. Although the trial description identifies randomized allocation and triple masking, implementation details are not reported.

These limitations constrain interpretation rather than demonstrate that the study was improperly conducted. The trial is listed as completed, but the registry information alone does not establish the robustness of the reported result, the clinical importance of the endpoint difference, or how the findings relate to outcomes beyond the measured treatment period.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Combined Liraglutide and Metformin Therapy in Women With Previous Gestational Diabetes Mellitus (GDM) — NCT01234649.
• Registry record last updated 2019-07-26; status last verified 2019-07.
• Results first posted 2019-07-26.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

NCT01234649 is a completed phase 3, randomized, triple-masked, parallel-group trial comparing metformin XR plus liraglutide with metformin XR plus placebo in women with metabolic defects and a recent history of GDM. The primary outcome was mean IS-SI after 84 weeks of treatment.

The sponsor reports an ANOVA superiority result of p <0.04. An unadjusted recalculation estimates a mean difference of 85.4, with a 95% confidence interval from -58.97 to 229.8 and p = 0.246. The sponsor's own analysis takes precedence, but its effect estimate and confidence interval are not reported in the registry record, and the discrepancy cannot be reconciled from the posted information.

The record does not report the original sample-size justification, missing-data strategy, analysis population, or multiplicity plan. The primary result concerns a physiological index; its clinical importance and relationship to longer-term outcomes are not established by the posted information.