A Study of Dulaglutide (LY2189265) in Children and Adolescents With Type 2 Diabetes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Dulaglutide (LY2189265) in Children and Adolescents With Type 2 Diabetes (NCT02963766), a randomized phase 3 study. The primary outcome is Change From Baseline in Hemoglobin A1c (HbA1c) (Pooled Doses) at Week 26.

2. Clinical Question

3. Trial Design

This was a Phase 3 randomized, double-blind, parallel-group study with an open-label extension. The registry reports enrollment of 154 participants and a completed study. Its stated purpose was to evaluate the safety, efficacy, pharmacokinetics, and pharmacodynamics of dulaglutide compared with placebo in pediatric participants with type 2 diabetes. The study duration was described as approximately 60 weeks.

During the double-blind period, participants received placebo, 0.75 mg/week dulaglutide, or 1.5 mg/week dulaglutide for 26 weeks. In the extension, the placebo group changed to open-label 0.75 mg/week dulaglutide, while participants in the two dulaglutide groups continued their respective doses for 26 weeks. The primary endpoint was assessed at Week 26, so its principal comparison concerns the blinded treatment period rather than the subsequent extension.

The primary result pools the two dulaglutide doses, whereas secondary HbA1c results are also reported separately by dose. The primary analysis population was not simply every enrolled participant: it included randomized participants who received at least one dose and had evaluable baseline and post-baseline HbA1c. The registry record does not report the group allocation sequence, the detailed eligibility criteria, or additional operational details of randomization and masking.

The design supports a randomized comparison of assigned treatments over the double-blind period. The open-label extension is a distinct follow-up phase in which treatment assignment and exposure differ from the initial comparison. Its outcomes should therefore not be treated as another blinded placebo comparison.

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.

Double-Blind Period

Reasons for not completing this period:

Open Label Extension (OLE)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All randomized participants who received at least one dose of study drug and had evaluable baseline and post-baseline HbA1c.

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: All randomized participants.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Hemoglobin A1c (HbA1c) (Pooled Doses) at Week 26, measured over Baseline, Week 26.

The primary endpoint was change from baseline in hemoglobin A1c (HbA1c), pooled dulaglutide doses, at Week 26. HbA1c is a form of hemoglobin that is measured to identify average plasma glucose concentration over a prolonged period. It provides a measure of glycemic status that reflects more than a single glucose reading, which makes it useful for assessing change over a treatment period.

The endpoint is expressed as percentage of HbA1c. The registry reports least-squares means and standard errors for the groups, along with a model-based difference between pooled dulaglutide and placebo. A negative between-group estimate indicates that the estimated change was lower in the dulaglutide group than in the placebo group under the stated contrast. The individual group estimates were -0.7 for pooled dulaglutide and 0.5 for placebo; the reported adjusted difference was -1.3.

The primary endpoint analysis used baseline and Week 26 measurements and adjusted for baseline HbA1c, insulin use, metformin use, treatment, time, and treatment-by-time. These adjustments define how the reported least-squares means and contrast were estimated. They do not change the endpoint’s basic interpretation as a comparison of mean change in HbA1c at the specified time point.

Using a continuous outcome preserves information about the magnitude of change across participants. It differs from a threshold outcome such as the percentage reaching HbA1c ≤7.0%, which classifies participants according to whether they crossed a specified cutoff.

7. Results Posted in the Registry

• Pooled Dulaglutide: mean -0.70 versus 0.50 on control; difference -1.20 (95% CI -1.77 to -0.63), standardized effect d = -0.75; z-test p = <0.001.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change From Baseline in HbA1c (Individual Doses) at Week 26

• 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:
  • 0.75 mg Dulaglutide: mean -0.50 versus 0.50 on control; difference -1.00 (95% CI -1.64 to -0.36), standardized effect d = -0.63; z-test p = 0.002.
  • 1.5 mg Dulaglutide: mean -1.00 versus 0.50 on control; difference -1.50 (95% CI -2.14 to -0.86), standardized effect d = -0.95; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for 0.75 mg Dulaglutide 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 1.5 mg Dulaglutide than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Fasting Blood Glucose (FBG) at Week 26

• 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:
  • 0.75 mg Dulaglutide: mean -0.47 versus 0.96 on control; difference -1.43 (95% CI -2.62 to -0.24), standardized effect d = -0.49; z-test p = 0.019.
  • 1.5 mg Dulaglutide: mean -1.54 versus 0.96 on control; difference -2.50 (95% CI -3.69 to -1.31), standardized effect d = -0.86; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for 0.75 mg Dulaglutide 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 1.5 mg Dulaglutide than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants With HbA1c ≤7.0%

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • 0.75 mg Dulaglutide: 60.0% (30/50) versus 17.8% (8/45) on control; risk difference 42.2 percentage points (95% CI 24.6 to 59.8), risk ratio 3.37 (1.73–6.58); unadjusted two-proportion z-test p = <0.001.
  • 1.5 mg Dulaglutide: 54.0% (27/50) versus 17.8% (8/45) on control; risk difference 36.2 percentage points (95% CI 18.5 to 54.0), risk ratio 3.04 (1.54–5.99); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for 0.75 mg Dulaglutide 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 proportion with the event is statistically higher for 1.5 mg Dulaglutide than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary results cover several kinds of outcomes, including individual-dose HbA1c change, fasting blood glucose, the percentage meeting an HbA1c threshold, body mass index, hypoglycemia, rescue for severe persistent hyperglycemia, pancreatitis, and pancreatic enzymes. Only some of the secondary outcome results are detailed in the record summarized for analysis. Results for the other listed outcomes are not reported in the registry record.

Among the posted results, the dose-specific HbA1c comparisons and fasting blood glucose comparisons have negative estimates relative to placebo, while the threshold outcome has reported logistic-regression comparisons against placebo. These outcomes are directionally related to glycemic measures but are not interchangeable: a continuous change, a threshold percentage, and a fasting glucose change answer different questions. Their overall pattern is broadly consistent with the primary endpoint’s direction, but they do not replace the pooled primary result.

Secondary findings generally carry less inferential weight when many outcomes or comparisons are considered and an adjustment strategy is not reported. Each additional test creates another opportunity for a result to appear statistically notable by chance. The registry lists multiple outcomes and analyses, but a multiplicity plan is not reported in the registry record. Accordingly, individual secondary p-values should be treated as supporting or exploratory evidence rather than as a set of independent confirmations.

The secondary outcomes also differ in analysis population, available observations, and measurement scale. The record gives different group sizes for some endpoints, so the number contributing to one result should not be assumed to apply to another. Without complete outcome results, analysis details, and multiplicity information, the secondary pattern is informative but not a substitute for a prespecified hierarchy or a comprehensive assessment across all measured outcomes.

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

• Change From Baseline in HbA1c (Individual Doses) at Week 26
• Change From Baseline in Fasting Blood Glucose (FBG) at Week 26
• Percentage of Participants With HbA1c ≤7.0%
• Change From Baseline in Body Mass Index (BMI) at Week 26
• Percentage of Participants With Self-Reported Events of Hypoglycemia
• Percentage of Participants Requiring Rescue for Severe, Persistent Hyperglycemia
• Number of Participants With Adjudicated Pancreatitis
• Change From Baseline in Pancreatic Enzymes at Week 26

9. Statistical Interpretation

10. Statistical Methods Explained

The registry reports a least-squares mean analysis for the primary continuous endpoint. A least-squares mean is a model-based group estimate adjusted for specified terms in the analysis. It is useful when the comparison is intended to account for baseline values and other covariates, rather than merely subtracting raw group averages. The reported contrast is the adjusted estimate for pooled dulaglutide relative to placebo.

What does the repeated-measures model do?

The primary analysis used a restricted maximum likelihood based mixed-effects model for repeated measures, or MMRM. General explanation: a repeated-measures model analyzes measurements collected from participants over time while accounting for the fact that observations from the same person are related. It can use the pattern of repeated measurements to estimate treatment and time effects together. The registry describes adjustment for baseline, insulin use, metformin use, treatment, time, and treatment-by-time, with Type III sums of squares.

Restricted maximum likelihood is an estimation approach used in mixed models to estimate variance components while accounting for fixed effects in the model. The registry says the model used unstructured variance-covariance structures for actual values and for change from baseline. General explanation: a covariance structure represents how measurements vary and relate across time. An unstructured choice allows the model to estimate the distinct variances and covariances rather than imposing a simpler pattern. The registry does not report further diagnostics or sensitivity analyses for this structure.

How should the standard errors be read?

The group summaries report standard errors, not standard deviations. General explanation: a standard error describes the uncertainty in an estimated mean, while a standard deviation describes variability among individual observations. They answer different questions and should not be interchanged. The standard errors accompany the least-squares means; the between-group confidence interval communicates uncertainty in the estimated contrast.

The posted recalculation is a simple unadjusted comparison reconstructed from summary statistics. Its pooled standard deviation is approximate because it is derived from the reported standard errors. It is not the same analysis as the registry-reported mixed model, which adjusts for the listed covariates and repeated measurements. The reported model result is therefore the appropriate source for interpreting the sponsor’s primary inferential analysis; the recalculation is a supplementary comparison and may differ because it uses a different method.

What does the p-value address?

General explanation: a p-value measures how incompatible the observed data, or more extreme data, are with a specified null hypothesis under the statistical model. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. The primary registry-reported p-value is <0.001. It should be interpreted alongside the estimated difference and confidence interval, not alone.

The secondary continuous endpoints also report least-squares means and standard errors. The binary threshold outcome reports percentages by group and registry-reported logistic-regression odds ratios. General explanation: continuous outcomes retain magnitude information, whereas a binary outcome groups observations into categories based on a cutoff. Odds ratios compare odds, not percentages or mean changes, so the effect scale differs from that of HbA1c change.

The registry reports 16 outcome measures and 12 analyses, while 15 secondary outcomes are posted. Those counts do not by themselves establish which outcomes were prespecified, how many statistical tests were planned, or whether a multiplicity strategy was used. The statistical plan beyond the endpoint descriptions and reported methods is not reported in the registry record. In particular, the record does not report the original sample-size justification, interim analyses, or a correction strategy for multiple secondary comparisons.

11. Confidence Intervals and Precision

The primary estimate is -1.3, with a 95% confidence interval from -1.8 to -0.7. General explanation: a confidence interval describes a range of estimates compatible with the data and analysis procedure at the stated confidence level. It conveys uncertainty around the estimated group contrast; it is not a range containing every participant’s individual response.

Because the primary interval is entirely below zero, the registry-reported analysis is consistent with a negative mean difference for pooled dulaglutide relative to placebo. Its endpoints also indicate that the estimate is not exact: values across the interval remain compatible with the data under the model. The p-value of <0.001 addresses the null comparison, while the interval gives more direct information about the range of plausible contrasts.

The registry-reported dose-specific HbA1c intervals are -1.7 to -0.4 for 0.75 mg dulaglutide and -2.2 to -0.9 for 1.5 mg dulaglutide. Their estimates and intervals are secondary results and should be interpreted in their own analysis context. They do not constitute a formal comparison between the two doses.

Confidence intervals reflect sampling uncertainty under the analysis and its assumptions. They do not account automatically for every possible source of uncertainty, including limitations in outcome reporting, missing data, model choice, or generalizability to other populations. The registry record does not report all information needed to evaluate those issues. The intervals are therefore best read as precision summaries for the stated analyses, rather than as complete measures of uncertainty about every aspect of the trial.

12. Planning a Similar Trial: Sample Size

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

The sample-size figures are planning illustrations based on the observed effect, not the trial’s original design calculations. For the continuous outcomes and the other non-time-to-event outcomes, the figures are participants per arm; time-to-event calculations, if present, are expressed as total events. The illustrations assume a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They use the stated fraction of the observed effect as the effect to detect.

The 80% power column means the number of participants per arm in an illustrative design intended to have an 80% chance of detecting the specified effect under those assumptions, if that effect and the planning model were correct. The 90% column describes the corresponding target with a 90% chance. These are design-planning quantities, not results about the probability that this completed trial detected an effect.

The rows distinguish planning for 100%, 75%, and 50% of the observed effect. A smaller assumed effect is harder to distinguish from random variation, so the illustration requires more participants per arm as the target effect becomes smaller. For the primary pooled comparison, the listed figures are 29 and 38 participants per arm for planning at 100% of the observed effect, 50 and 67 at 75%, and 113 and 151 at 50%, for the 80% and 90% targets respectively.

These calculations use an observed result as a planning reference and rely on simplified assumptions. They do not reproduce the trial’s actual design rationale or account for every feature of its adjusted repeated-measures analysis. The original sample-size justification and design assumptions are not reported in the registry record. This is not observed power.

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 = -1.200; pooled standard deviation σ = 1.603.
2. n = 2 × (1.96 + 0.842)² × 1.603² / -1.200² = 28.0, 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 <- -1.2                      # observed difference in means
sigma <- 1.60307397017298                     # 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(2963766)
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), 38 (90% power)
 75% of observed effect: n per arm = 50 (80% power), 67 (90% power)
 50% of observed effect: n per arm = 113 (80% power), 151 (90% power)
power.t.test: 29 
Simulated power at n = 29 per arm: 0.799 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary analysis as a mixed-model least-squares mean difference in final values of -1.3, with a 95% confidence interval from -1.8 to -0.7 and p < 0.001. The simple recalculation from posted summary statistics estimates a mean difference of -1.2. These estimates are close in direction and magnitude, but they are not identical because they arise from different analyses.

The recalculation is an unadjusted comparison based on group summaries and an approximate pooled standard deviation. The registry-reported analysis uses the stated repeated-measures model and adjusts for baseline HbA1c, insulin use, metformin use, treatment, time, and treatment-by-time. The sponsor’s model-based result is therefore the main result for interpreting the trial’s reported primary analysis. The recalculation provides a secondary check, not an alternative adjusted estimate.

The registry record does not flag a discrepancy between the posted recalculation and the sponsor analysis. Neither estimate should be interpreted without its method: the sponsor’s model addresses the planned adjusted comparison as reported, while the recalculation shows what a simpler summary-statistic comparison yields.

15. Multiplicity, Missing Data and Analysis Populations

The registry reports a primary endpoint and multiple secondary outcomes, including dose-specific HbA1c, fasting blood glucose, threshold HbA1c, body mass index, hypoglycemia, rescue for hyperglycemia, pancreatitis, and pancreatic enzymes. It also reports multiple analyses. When several hypotheses are tested, the chance of at least one apparently unusual result can increase. General explanation: multiplicity procedures address this issue by controlling an error rate or by organizing confirmatory and exploratory questions.

A multiplicity adjustment or testing hierarchy is not reported in the registry record. The record therefore does not establish whether secondary analyses were adjusted, which outcomes were considered confirmatory, or how interpretation was intended to proceed across the full set of tests. This matters because a nominal p-value for an individual secondary endpoint does not necessarily have the same evidential status as a primary result assessed within a prespecified plan.

The primary endpoint description specifies an analysis population of randomized participants who received at least one dose and had evaluable baseline and post-baseline HbA1c. It also describes the mixed-effects repeated-measures model, its covariates, and its covariance structure. The registry record does not report the full approach to missing observations, the assumptions used for incomplete outcome data, or sensitivity analyses examining departures from those assumptions. MMRM is the stated method, but its mention alone does not answer every question about how missingness affected the estimate.

The original sample-size justification, interim analyses, and broader statistical analysis plan are not reported in the registry record. Their absence limits assessment of whether the posted analyses fully match the original design and how the number of planned tests was handled. These are reporting limits; they do not, by themselves, establish that the analyses were inappropriate.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pilonidal cyst (2.0%), pyelonephritis (2.0%), carbon monoxide poisoning (2.0%), right ventricular failure (2.0%), nonalcoholic fatty liver disease (2.0%).

During the double-blind period, serious adverse events were reported for 3 of 51 participants in the placebo group, 1 of 51 in the 0.75 mg dulaglutide group, and 1 of 52 in the 1.5 mg dulaglutide group. During the open-label extension, the corresponding reported counts were 0 of 47 for participants who had received placebo during the double-blind period and then received 0.75 mg dulaglutide, 1 of 49 in the 0.75 mg dulaglutide group, and 2 of 50 in the 1.5 mg dulaglutide group. No deaths were reported in any of these groups.

The registry also lists serious adverse-event terms each reported as 2.0%: pilonidal cyst, pyelonephritis, carbon monoxide poisoning, right ventricular failure, nonalcoholic fatty liver disease, and genital herpes. These counts and terms describe events recorded in the trial; they do not, by themselves, establish whether treatment caused an event.

The extension is open-label and follows a change in treatment for the original placebo group, so its group counts should be read separately from the blinded comparison. Adverse-event counts are not a complete safety assessment, and the record’s counts alone do not establish comparative risk or rule out uncommon events.

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

The primary result indicates that the registry-reported adjusted mean HbA1c change at Week 26 was lower in the pooled dulaglutide group than in the placebo group, with an estimated difference of -1.3 and a 95% confidence interval from -1.8 to -0.7. The p-value was <0.001. These statistics describe the trial’s model-based group comparison among participants in the stated analysis population.

The result does not mean that every participant had the same change, or that the estimate will apply unchanged to all children and adolescents with type 2 diabetes. It does not establish a clinically important difference, because a threshold for clinical importance is not reported in the registry record. Statistical evidence against a no-difference hypothesis and clinical importance are related but distinct questions.

The primary comparison concerns the 26-week double-blind period and the pooled dose groups. It does not establish a direct contrast between the two dulaglutide doses, nor does the open-label extension provide a continued placebo comparison. The secondary endpoints offer additional information on other measurements, but their multiplicity and incomplete result reporting limit how strongly they can be treated as independent confirmation.

The result is not a complete assessment of long-term outcomes or safety. The record describes an approximately 60-week study with a 26-week open-label extension, but the primary endpoint is at Week 26. The serious adverse-event counts are descriptive and should not be used alone to infer causation or comparative safety. The registry results are statistical evidence from this trial, not individualized guidance or a guarantee of outcome.

18. Why This Trial Matters Statistically

This trial illustrates how a clinically recognizable endpoint can be analyzed in more than one way. HbA1c change is reported as a continuous outcome, while the secondary threshold endpoint classifies participants according to whether HbA1c was ≤7.0%. The continuous analysis estimates average change; the threshold analysis concerns the proportion meeting a cutoff. Their scales and interpretations differ even when they relate to the same clinical measure.

The primary result also shows why adjusted and unadjusted estimates should be distinguished. The registry’s repeated-measures model accounts for specified baseline and treatment terms, while a summary-statistic recalculation is a simpler comparison. Both can help readers understand the data, but the adjusted registry analysis is the direct account of the sponsor’s reported primary method.

Precision is as important as statistical significance. The confidence interval communicates a range of uncertainty around the estimated contrast, while the p-value addresses compatibility with a null comparison under the analysis. Neither statistic describes the distribution of individual responses or determines clinical importance.

Finally, the registry record shows why interpretation depends on reporting beyond a headline result. Analysis populations, repeated-measures methods, secondary outcomes, missing-data information, and multiplicity plans all shape what conclusions can reasonably be drawn. Where those details are not reported, uncertainty about the analysis context should remain visible rather than being filled in by assumption.

19. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The primary endpoint is assessed at Week 26, during the double-blind period. The trial also included an open-label extension, but that phase does not preserve the original placebo comparison because the placebo group changed to dulaglutide. The primary result therefore should not be treated as evidence from a blinded comparison across the full study duration.

The primary estimate pools the 0.75 mg and 1.5 mg dulaglutide groups. Dose-specific HbA1c results are available as secondary analyses, but the registry record does not report a formal direct test comparing the doses. A difference between separate estimates against placebo is not itself evidence of a statistically established difference between doses.

The primary population requires receipt of at least one dose and evaluable baseline and post-baseline HbA1c. The record does not report enough detail to determine how participants outside that analysis population were handled, or whether sensitivity analyses addressed missing outcome measurements. It also does not report the original sample-size justification or a multiplicity strategy for the range of endpoints and analyses.

Several secondary outcomes are listed, but results for all of them are not reported in the registry record. The study is described as involving pediatric participants with type 2 diabetes; the available record does not provide further details about the age distribution or other characteristics needed to judge applicability to particular subgroups. The sample-size illustrations are based on observed effects and simplified assumptions, not the trial’s original design calculations.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study of Dulaglutide (LY2189265) in Children and Adolescents With Type 2 Diabetes — NCT02963766.
• Registry record last updated 2022-07-01; status last verified 2022-06-01.
• Results first posted 2022-07-01.
• 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

NCT02963766 is a completed Phase 3 randomized, double-blind, parallel-group study of once-weekly dulaglutide compared with placebo in children and adolescents with type 2 diabetes. The registry reports enrollment of 154 participants and an approximately 60-week study duration, including a 26-week double-blind period and a 26-week open-label extension.

The primary endpoint was change from baseline in HbA1c at Week 26, pooling dulaglutide doses. The sponsor-reported mixed-model analysis estimated a difference of -1.3, with a 95% confidence interval from -1.8 to -0.7 and p < 0.001. The analysis adjusted for baseline HbA1c, insulin use, metformin use, treatment, time, and treatment-by-time.

The registry also lists secondary efficacy and safety outcomes. Some secondary results are posted, while results for other listed outcomes are not reported in the registry record. Multiplicity procedures, the original sample-size rationale, and detailed missing-data sensitivity analyses are not reported in the registry record.