A Study of the Effect of LY2189265 on Blood Pressure and Heart Rate in Type 2 Diabetes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of the Effect of LY2189265 on Blood Pressure and Heart Rate in Type 2 Diabetes (NCT01149421), a randomized phase 2 study. The primary outcome is Change From Baseline to 16 Weeks in Mean 24-hour Systolic Blood Pressure (SBP).

2. Clinical Question

This question concerns an average change in a continuous measurement, not the number of participants who crossed a blood-pressure threshold. The comparison therefore describes group-level changes and their statistical uncertainty. The result should be read in the context of the endpoint definition, analysis population, and the sponsor's reported analysis, rather than as a complete account of every blood pressure or heart-rate outcome in the trial.

3. Trial Design

The registry record describes a completed Phase 2, randomized, quadruple-masked, parallel-group study sponsored by Eli Lilly and Company. Enrollment was 755. The recorded start was 2010-06, and primary completion was 2011-09. The population had type 2 diabetes mellitus. The study compared LY2189265 at two doses with placebo; all three regimens were described as subcutaneous once-weekly treatment.

The primary endpoint was change from baseline to 16 weeks in mean 24-hour systolic blood pressure. Blood pressure was measured with an ambulatory monitoring device attached to the participant's nondominant arm. The record describes a measurement schedule of every 20 minutes in daytime hours and every 30 minutes at night, as programmed into the device. The analysis defined diurnal hours as 0800 to 2100 and nocturnal hours as 0000 to 0600.

The primary analysis used least-squares means from a mixed model for repeated measurements, with adjustment for baseline, pooled sites, treatment, visit, treatment-by-visit interaction, and the stratified variable of diagnosis of hypertension. The stated analysis population consisted of participants who received at least one dose of study treatment and had evaluable blood pressure data. The registry record does not report the trial's original sample-size justification or the full statistical plan, so the enrollment cannot be interpreted as evidence for a particular power target or decision rule.

Secondary outcomes included further 24-hour, daytime, nighttime, and clinic blood pressure measures, heart rate, pulse pressure, and mean arterial pressure. The registry record also includes other secondary measures. Results for each endpoint should be considered in light of its own measurement, analysis population, and uncertainty; they do not all answer the primary question.

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: Participants who received at least one dose of LY2189265 or Placebo with evaluable blood pressure data.

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: Participants who received at least one dose of LY2189265 or Placebo.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline to 16 Weeks in Mean 24-hour Systolic Blood Pressure (SBP), measured over Baseline, 16 weeks.

The primary endpoint was the change from baseline to 16 weeks in mean 24-hour systolic blood pressure, measured in millimeters of mercury (mmHg). Systolic blood pressure is the higher pressure recorded during a heartbeat. Using a 24-hour ambulatory monitor allows repeated measurements during ordinary daytime and nighttime periods, rather than relying only on a single clinic reading. That approach describes blood pressure across the monitoring period and allows separate summaries for times of day.

The record specifies that measurements were taken every 20 minutes during daytime and every 30 minutes during nighttime, with the analysis defining diurnal and nocturnal periods separately from the device's daytime and nighttime recording schedule. The endpoint is a continuous measure, and the reported treatment summaries are least-squares means from a repeated-measures model. These adjusted means are not simply raw averages: the model accounts for the listed baseline, site, treatment, visit, interaction, and hypertension-diagnosis terms.

Change from baseline focuses the comparison on how the average measurement shifted over the specified follow-up period. Placebo provides a concurrent reference for changes that may occur over time regardless of assigned active treatment. The endpoint does not by itself describe individual responses, the frequency of particular blood pressure levels, or longer-term clinical outcomes. The registry record does not report a threshold defining a clinically important difference for this endpoint, so statistical estimates should not be equated with a clinical threshold.

7. Results Posted in the Registry

• 1.5 Milligram (mg) LY2189265: mean -3.41 versus -0.63 on control; difference -2.78 (95% CI -4.44 to -1.12), standardized effect d = -0.29; z-test p = 0.001.
• 0.75 Milligram (mg) LY2189265: mean -1.70 versus -0.63 on control; difference -1.07 (95% CI -2.69 to 0.55), standardized effect d = -0.12; z-test p = 0.196.

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 to 26 Weeks in Mean 24-hour Systolic Blood Pressure (SBP)

• 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:
  • 1.5 Milligram (mg) LY2189265: mean -2.51 versus 0.15 on control; difference -2.66 (95% CI -4.39 to -0.93), standardized effect d = -0.27; z-test p = 0.003.
  • 0.75 Milligram (mg) LY2189265: mean -1.56 versus 0.15 on control; difference -1.71 (95% CI -3.40 to -0.02), standardized effect d = -0.18; z-test p = 0.047.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for 1.5 Milligram (mg) LY2189265 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 0.75 Milligram (mg) LY2189265 than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline to 16 Weeks in High-Sensitivity C-Reactive Protein (Hs-CRP)

• 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:
  • 1.5 Milligram (mg) LY2189265: mean -0.79 versus 0.29 on control; difference -1.08 (95% CI -2.18 to 0.02), standardized effect d = -0.18; z-test p = 0.054.
  • 0.75 Milligram (mg) LY2189265: mean -0.20 versus 0.29 on control; difference -0.49 (95% CI -1.85 to 0.87), standardized effect d = -0.07; z-test p = 0.479.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between 1.5 Milligram (mg) LY2189265 and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between 0.75 Milligram (mg) LY2189265 and the control. The interval gives the range of effects that remain compatible with the data.

Measurement of LY2189265 Drug Concentration for Pharmacokinetics - Area Under the Concentration Time Curve (AUC)

• 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:
  • 0.75 Milligram (mg) LY2189265: mean 5,998 versus 11,208 on control; difference -5,210 (95% CI -5,760 to -4,660), standardized effect d = -1.74; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for 0.75 Milligram (mg) LY2189265 than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes broaden the picture beyond the 16-week primary systolic blood-pressure endpoint. They include systolic and diastolic pressure, heart rate, pulse pressure, and mean arterial pressure, with measurements at later visits and across 24-hour, daytime, nighttime, and clinic settings. The registry record also contains secondary results for a biomarker and a pharmacokinetic measure. These outcomes address different questions and may have different analysis populations and measurement properties.

The posted pattern across the secondary material should be treated as complementary rather than as a replacement for the primary result. A later systolic blood-pressure assessment provides information at another visit, while heart-rate and other pressure measures describe related but distinct physiological measurements. Biomarker and drug-concentration results do not directly answer the primary blood-pressure comparison. The registry record includes 24 posted secondary outcomes, but the existence of many outcomes does not mean that every result has the same evidential weight.

Secondary analyses often involve multiple outcomes, visits, doses, and measurement settings. General statistical explanation: When many hypotheses are tested, the chance of at least one apparently notable result arising by chance can increase unless the analysis plan accounts for multiplicity. The registry record does not report a multiplicity adjustment or enough detail to determine which secondary tests were prespecified as confirmatory. Accordingly, unadjusted secondary p-values carry less weight than a clearly specified primary test, and should be interpreted as exploratory or supporting evidence unless the record establishes otherwise.

The overall interpretation should focus on coherence, precision, and relevance to the primary question rather than selecting isolated secondary findings. Differences across visits or measures can arise from their distinct definitions, missing data, and statistical uncertainty. The secondary results may help characterize the broader set of measurements collected, but they do not independently establish the primary comparison or resolve the limitations in the reported analysis plan.

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

• Change From Baseline to 26 Weeks in Mean 24-hour Systolic Blood Pressure (SBP)
• Change From Baseline to 16 and 26 Weeks in Mean 24-hour Diastolic Blood Pressure (DBP)
• Change From Baseline to 16 and 26 Weeks in Mean 24-Hour Heart Rate (HR)
• Change From Baseline to 16 and 26 Weeks in Mean 24-hour Pulse Pressure
• Change From Baseline to 16 and 26 Weeks in Mean 24-hour Mean Arterial Pressure (MAP)
• Change From Baseline to 16 and 26 Weeks in Mean Daytime, Nighttime, and Clinic Systolic Blood Pressure (SBP)
• Change From Baseline to 16 and 26 Weeks in Mean Daytime, Nighttime, and Clinic Diastolic Blood Pressure (DBP)
• Change From Baseline to 16 and 26 Weeks in Mean Daytime, Nighttime, and Clinic Heart Rate (HR)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: each participant's blood pressure measurement can take values along a scale, and the analysis compares average changes between assigned groups. The registry reports least-squares means and standard errors. A least-squares mean is a model-based group estimate adjusted for the variables included in the model. It is therefore distinct from a simple arithmetic mean, even when both describe an average.

Why use a repeated-measures model?

General statistical explanation: When an outcome is measured at more than one visit, measurements from the same participant are related rather than independent. A mixed model for repeated measures can represent that within-participant relationship and estimate treatment-group patterns over time. It can also use observations from available visits under its modeling assumptions, rather than treating each visit as an unrelated analysis. The registry record states that the primary least-squares means came from this type of model. It does not report all model specification details, such as the covariance structure or the exact approach to missing observations.

For this analysis, the listed adjustment terms were baseline, pooled sites, treatment, visit, treatment-by-visit interaction, and diagnosis of hypertension as the stratified variable. Baseline adjustment accounts statistically for the starting measurement. Treatment and visit terms describe group and time patterns; their interaction allows the model's treatment contrast to vary by visit. Adjustment can improve comparability and precision when the model is appropriate, but it does not remove the need to examine the underlying design, missing data, or assumptions. The registry record does not report a full diagnostic assessment of those assumptions.

What does the treatment contrast express?

The sponsor reports least-squares mean differences between each LY2189265 group and placebo. A negative difference in this setting means the estimated change in the active-treatment group is more negative than the estimated change in placebo. It is an average contrast, not an individual response. The sponsor's reported contrast is based on the mixed model. By contrast, the calculations labelled recalculated use the posted group summary statistics to make simple unadjusted comparisons. Such calculations do not incorporate the complete model adjustment listed for the sponsor's analysis and can produce different uncertainty estimates or p-values.

The recalculated standard deviations for the primary comparisons are approximate because the posted group dispersion is a standard error, not a standard deviation. That conversion relies on the posted group sizes and summary information. General statistical context: a standard error describes uncertainty in an estimated mean, while a standard deviation describes variability among observations. They are not interchangeable. Reconstructing a standard deviation from a standard error and group size is an approximation when the original participant-level data and analysis details are unavailable. This is one reason the recalculated results should not displace the sponsor's model-based analysis.

What do p-values and intervals contribute?

General statistical explanation: A p-value describes the compatibility of observed data with a null hypothesis under a particular model and testing procedure. It is not the probability that the null hypothesis is true, and it does not quantify the probability that an observed effect will repeat. The sponsor's superiority-or-other tests report p <0.001 for the 1.5 mg comparison and p = 0.149 for the 0.75 mg comparison. The record also contains a different set of reported analyses labelled non-inferiority or equivalence, including p <0.001 for both comparisons. The registry record does not report the hypotheses, margins, or detailed testing plan needed to reconcile those labels and results. They should therefore be quoted in their stated context, not treated as interchangeable tests.

A confidence interval gives a range of estimates compatible with the observed data and the specified method, under its assumptions. Its width conveys precision: narrower intervals indicate less statistical uncertainty than wider intervals, all else being equal. For the sponsor's primary comparisons, the record labels the intervals as 97.3%. The recalculated intervals come from an unadjusted method and should not be mistaken for the sponsor's model-based intervals. Neither an interval nor a p-value establishes clinical importance; the registry record does not state a clinically important threshold for this outcome.

Why can repeated measurements and dispersion matter?

General statistical explanation: Repeated measurements may be correlated, and a model that accounts for that correlation can differ from a calculation treating observations as independent. The between-participant variability also affects how precisely a mean difference can be estimated. If data are more variable than a model assumes, uncertainty can be understated; this is one reason analysts assess distributional and model assumptions. For the primary measure, the record presents standard errors for the adjusted means, but it does not provide participant-level observations or a complete set of model diagnostics. A reviewer therefore cannot independently verify every modeling choice from the summary information alone.

The reported analysis population required at least one dose and evaluable blood pressure data. That is a stated population definition, but it is not the same as an account of how many participants contributed to every visit-specific model estimate. Visit-level missingness and the details of its handling are not reported in the registry record. These gaps limit independent reconstruction and should be kept in mind when reading precision estimates.

11. Confidence Intervals and Precision

The sponsor's primary mixed-model comparison for 1.5 mg versus placebo was -2.79 mmHg, with a reported 97.3% confidence interval from -4.58 to -1.00. The interval lies on the negative side of zero, which is consistent with the sponsor's reported superiority-or-other p-value of <0.001. For 0.75 mg versus placebo, the reported difference was -1.07 mmHg, with a 97.3% confidence interval from -2.83 to 0.68. This interval includes zero, and the sponsor's superiority-or-other result was p = 0.149. These are the sponsor's adjusted estimates and intervals.

The simple unadjusted re-calculation for the higher dose gives -2.78 mmHg, with an interval from -4.443 to -1.117. For the lower dose it gives -1.07 mmHg, with an interval from -2.692 to 0.552. The small differences between some estimates and interval limits reflect that these calculations use posted summary statistics rather than the sponsor's full mixed model. The sponsor's own analysis takes precedence because the record flags a discrepancy.

General statistical explanation: A confidence interval is not a guarantee that the true effect lies within its endpoints. It describes uncertainty under a specified method and its assumptions. A wider interval allows a broader range of effect sizes; a narrower interval indicates greater precision. Whether an interval crosses zero is relevant to a conventional test of no difference, but it does not by itself establish whether a difference is clinically important. The registry record does not report a clinical-importance threshold.

Precision also depends on the data contributing to an estimate, variability, and modeling choices. The registry reports the primary analysis population, but does not report full visit-specific missingness or all details required to reproduce the adjusted intervals. The sponsor's interval and the recalculated interval answer related but not identical statistical questions.

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 illustrations express the number of participants per arm for the continuous-outcome comparisons. They are not total event counts. They use the observed effect as a starting point and show planning scenarios for detecting the full observed effect, 75% of the observed effect, or 50% of the observed effect. The assumptions listed for these illustrations are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect.

The 80% power column means the number of participants per arm in a planning scenario designed to have an 80% chance of detecting the specified effect under those assumptions, if that effect and the model assumptions are correct. The 90% power column uses a higher target chance of detection, and therefore generally calls for more participants for the same assumed effect. Smaller effects need more participants because they are harder to distinguish from random variation. The illustrations are for planning, not a reconstruction of the trial's original design; the registry record does not report the original sample-size assumptions or justification.

These values are not observed power. Observed power is not a useful substitute for examining the estimated effect and its uncertainty after a trial. The figures instead show how planning requirements change when the assumed target effect is reduced. They should not be read as proof that the trial was adequately or inadequately powered, since the original design assumptions and full statistical plan are not reported in the registry record.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -2.780; pooled standard deviation σ = 9.449.
2. n = 2 × (1.96 + 0.842)² × 9.449² / -2.780² = 181.3, rounded up to 182 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 <- -2.78                      # observed difference in means
sigma <- 9.44892635762079                     # 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(1149421)
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 = 182 (80% power), 243 (90% power)
 75% of observed effect: n per arm = 323 (80% power), 432 (90% power)
 50% of observed effect: n per arm = 726 (80% power), 972 (90% power)
power.t.test: 182.3 
Simulated power at n = 182 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 sponsor's analysis uses a mixed model for repeated measures and reports adjusted least-squares mean differences. For the primary 16-week endpoint, it reports a difference of -2.79 mmHg for 1.5 mg versus placebo, with a 97.3% confidence interval from -4.58 to -1.00. Its superiority-or-other result is p <0.001. For 0.75 mg versus placebo, the sponsor reports -1.07 mmHg, with a 97.3% confidence interval from -2.83 to 0.68 and a superiority-or-other p-value of 0.149.

The simple recalculations use posted group summaries and yield similar point estimates, but they are unadjusted and do not reproduce the sponsor's model. The record flags a discrepancy, so the sponsor's own analysis takes precedence. The recalculations are best treated as a check on the broad direction and scale of the contrast, not as an alternative primary analysis.

The record also lists analyses labelled non-inferiority or equivalence, with p-values that do not align straightforwardly with the separate superiority-or-other results. The registry record does not report the hypotheses, margins, or full statistical testing strategy needed to interpret that apparent inconsistency. The reported values should be kept tied to their stated analysis labels; the record does not support a reconstruction of the sponsor's decision framework.

15. Multiplicity, Missing Data and Analysis Populations

The registry record describes a randomized trial with a defined primary endpoint and multiple secondary endpoints. It does not report a multiplicity adjustment, a hierarchy of hypotheses, or a complete prespecified testing sequence. Without those details, it is not possible to determine from the registry record whether secondary tests were intended to be confirmatory, how their error rates were controlled, or how the differently labelled sponsor analyses fit together. This is especially relevant when results span doses, visits, and several types of blood pressure and heart-rate measurements.

The primary analysis population is described as participants who received at least one dose of LY2189265 or placebo and had evaluable blood pressure data. The record does not report a full participant flow for the endpoint or the amount and pattern of missing measurements at each visit. It also does not provide enough information to independently establish the assumptions used for handling missing repeated measurements. Missingness can affect both the estimated group contrast and its precision, particularly if the availability of measurements is related to participant characteristics or outcomes.

The record gives the adjustment terms for the primary mixed model, but not a complete statistical analysis plan, all model details, or diagnostic results. It also does not report the original sample-size justification or interim-analysis plan. These omissions mean that the trial's enrollment and the reported p-values cannot be evaluated against an unreported target or complete testing strategy.

These reporting gaps do not negate the posted estimates; they limit how far readers can independently reconstruct the analysis and assess safeguards against selective inference. The clearest approach is to distinguish the sponsor's primary adjusted analysis from the unadjusted recalculations, keep secondary results in their exploratory context, and avoid treating the absence of a reported procedure as proof that no procedure was used.

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 4% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: uterine dilation and curettage (0.8%), pregnancy (0.8%), cholelithiasis (0.8%), benign prostatic hyperplasia (0.8%), coronary artery disease (0.4%).

The registry lists serious adverse events for each treatment group. In the 1.5 mg LY2189265 group, 12 participants were reported as affected by serious adverse events among 251 at risk. In the 0.75 mg group, 8 were affected among 254 at risk. In the placebo group, 8 were affected among 250 at risk. These counts describe participants affected, as recorded, and should not be interpreted as a comparison adjusted for differences in exposure, follow-up, or event characteristics.

The listed serious-event terms include uterine dilation and curettage, pregnancy, cholelithiasis, and benign prostatic hyperplasia, each reported as 0.8%; coronary artery disease and ventricular tachycardia are each listed as 0.4%. The registry record does not provide the group-specific counts for each of these terms in the material. The number of deaths is not reported in the registry record.

Serious adverse-event figures are not a complete account of all harms, and these counts alone do not establish whether a difference between groups is statistically reliable. They also do not describe severity, timing, duration, or causal attribution for each event. The registry record's adverse-event summary should therefore be read as a descriptive report, not as a basis for an individual medical decision or a definitive comparative safety conclusion.

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

The primary endpoint estimates average change in mean 24-hour systolic blood pressure from baseline to 16 weeks. The sponsor's adjusted analysis reports a larger reduction relative to placebo for the 1.5 mg group, with an interval that remains below zero. For the 0.75 mg comparison, the sponsor's interval spans zero and its superiority-or-other test reports p = 0.149. The posted group means and contrasts describe averages; participants within a group may have had different trajectories and measurements.

This result does not establish that every participant's blood pressure changed in the same direction, nor does it quantify the probability that an individual would experience a particular outcome. A group mean can conceal variation among participants. The endpoint also does not, by itself, establish effects on clinical events, long-term outcomes, or measures not included in the reported comparison. The registry record does not provide a threshold for clinical importance, so statistical evidence should not be treated as an automatic judgment about the practical meaning of the estimated difference.

The trial compared two doses with placebo in a specific study population: people with type 2 diabetes treated with oral antihyperglycemic medications. The result should not be generalized beyond that population or the stated measurement period without additional evidence. The record's secondary outcomes add other measurements and time points, but their multiplicity and reporting limitations mean that they should not be treated as equivalent to the primary endpoint.

The analysis also has reporting limitations. The sponsor's model-based estimates and the simple recalculations are not identical analyses, and the registry record flags a discrepancy. The sponsor's own analysis takes precedence. The record does not report full missing-data details, the original sample-size assumptions, or a complete multiplicity strategy. Consequently, the most defensible reading is a description of the reported group-level primary comparison and its uncertainty, with careful attention to the limits of the registry information.

18. Why This Trial Matters Statistically

This trial illustrates why a clinical result is more than a comparison of group means. The endpoint relies on repeated ambulatory measurements, and its adjusted estimate comes from a repeated-measures model. The statistical contrast therefore depends on how visits, within-participant measurements, baseline values, sites, treatment, and the hypertension stratification variable are represented. Reading the estimate alongside its interval helps keep the magnitude and precision in view.

It also shows why the analysis used for interpretation matters. The sponsor's model-based results and simple unadjusted recalculations give similar broad point estimates but are not interchangeable. The record flags a discrepancy, so the sponsor's analysis takes precedence. A transparent recalculation can be informative as a check, but cannot replace an analysis that uses the stated model and its adjustments.

Multiple doses, visits, and secondary outcomes create a further statistical challenge. If many comparisons are examined, isolated results can appear notable by chance unless the testing plan addresses multiplicity. The registry record does not report the full plan needed to determine how those analyses were prioritized. That limitation makes it especially important not to elevate a secondary result above the primary question without supporting design information.

Finally, planning illustrations are not retrospective proof of power. They show how required participants per arm vary with assumed effect size under stated assumptions, while the trial's own design assumptions are not reported. Separating such planning concepts from observed evidence, and separating statistical significance from clinical importance, supports a more disciplined interpretation of this randomized comparison.

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.

Several limitations are specific to the information recorded for this trial. The primary endpoint is an adjusted least-squares mean from a mixed model, but the registry record does not provide every model specification or diagnostic needed to reproduce that analysis. The posted summary statistics permit simple recalculations, but these are unadjusted and use approximate variability for the primary comparisons. The record flags a discrepancy between those calculations and the sponsor analysis; the sponsor's analysis takes precedence.

The record does not report the original sample-size justification, the full statistical analysis plan, interim analyses, or a multiplicity strategy. These omissions limit interpretation of the enrollment, the range of reported hypothesis tests, and the evidential status of secondary outcomes. The record also does not provide full visit-level missingness or the detailed procedure used for missing repeated measurements. The stated primary analysis population does not, by itself, resolve those questions.

Many related endpoints were measured, including blood pressure across time periods and settings, heart rate, and other measures. The secondary results therefore need cautious interpretation, particularly where multiple tests may be involved. The registry record does not report a clinical-importance threshold for the primary difference, so clinical relevance cannot be inferred solely from the statistical results.

Finally, the safety summary is descriptive and does not include all detail needed to compare event timing, exposure, severity, or attribution. The number of deaths is not reported in the registry record. These points constrain what can be concluded from the posted summary, without implying that unreported procedures or data were necessarily absent from the study.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study of the Effect of LY2189265 on Blood Pressure and Heart Rate in Type 2 Diabetes — NCT01149421.
• Registry record last updated 2015-02-02; status last verified 2015-01.
• Results first posted 2015-02-02.
• 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

NCT01149421 is a completed Phase 2 randomized, quadruple-masked, parallel-group study sponsored by Eli Lilly and Company. It enrolled 755 participants with type 2 diabetes mellitus and compared once-weekly subcutaneous LY2189265 at 1.5 mg and 0.75 mg with placebo.

The primary endpoint was change from baseline to 16 weeks in mean 24-hour systolic blood pressure measured by ambulatory monitoring. The sponsor's mixed-model analysis reported a difference of -2.79 mmHg for 1.5 mg versus placebo and -1.07 mmHg for 0.75 mg versus placebo. The record flags a discrepancy with simple unadjusted recalculations, so the sponsor's analysis takes precedence.

The record includes secondary outcomes covering blood pressure, heart rate, and other measures, as well as descriptive serious adverse-event counts. Important details—including the original sample-size rationale, full multiplicity plan, and visit-level missing-data information—are not reported in the registry record.