Study of Human Regular U-500 Insulin in Adult Participants With Type 2 Diabetes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Human Regular U-500 Insulin in Adult Participants With Type 2 Diabetes (NCT01774968), a randomized phase 4 study. The primary outcome is Change From Baseline to Week 24 in Glycated Hemoglobin A1c (HbA1c).

2. Clinical Question

The clinical question concerns two schedules for the same insulin, rather than comparison of insulin with a different drug or with no treatment. The registry describes the study as a randomized, open-label, parallel clinical trial in people with type 2 diabetes whose control was inadequate on high-dose U-100 insulin, with or without oral agents. It lists 325 participants enrolled.

HbA1c is the primary outcome measure. The registry reports adjusted mean changes for the two groups and a between-group estimate. The three-times-daily group had a least-squares mean change of -1.12 percentage points of HbA1c, with a standard error of 0.08; the twice-daily group had a change of -1.22, also with a standard error of 0.08. The reported between-group estimate is the difference for twice daily relative to three times daily.

The trial record does not by itself establish which schedule should be used for an individual. It records a comparison in a particular study population over a specified period, using a specified outcome and analysis. The results describe that comparison and its precision; they do not answer every question about other populations, longer follow-up, or outcomes beyond those recorded.

3. Trial Design

The registry classifies this as a Phase 4, randomized, open-label, parallel clinical trial. The two groups received Human Regular U-500 Insulin with different dosing schedules: three times a day or two times a day. Both regimens were titrated based on blood glucose readings and continued for 24 weeks. The recorded enrollment was 325.

Open-label means that the study was not masked according to the registry record. Participants and study personnel therefore were not described as blinded to treatment schedule. Randomization is intended, in general, to allocate participants to comparison groups by chance and thereby reduce systematic differences at the point of assignment. It does not guarantee that the groups will be identical in every characteristic, and the registry record does not report the randomization procedure in further detail.

The primary outcome was change in HbA1c from baseline to Week 24. Its listed analysis used least-squares means from a mixed-effects model for repeated measures. The analysis population included randomized participants who received at least one dose of U-500R and had evaluable HbA1c data. The registry does not report the number excluded from this analysis or the reasons for any exclusions.

The trial also listed secondary measures, including HbA1c target attainment, hypoglycemic events, body weight, insulin dose, fasting plasma glucose, and time to reach HbA1c targets. The record lists 14 secondary outcome measures. The posted results include a subset of secondary continuous-outcome estimates; results for other listed secondary measures are not reported in the registry record.

The registry records a start in 2013-02, primary completion in 2014-05, and status as completed. More detail about site conduct, allocation concealment, adherence, and the original statistical analysis plan is not reported in 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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Randomized participants who received at least 1 dose of U-500R with evaluable HbA1c 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: All randomized participants.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline to Week 24 in Glycated Hemoglobin A1c (HbA1c), measured over Baseline, Week 24.

The primary endpoint measures change in glycated hemoglobin A1c (HbA1c) from baseline to Week 24. HbA1c is a continuous measure expressed as a percentage of HbA1c. Change describes how the measured value at follow-up differs from the baseline value, while the between-group contrast addresses whether the average change differed between the dosing schedules.

In general, a continuous endpoint preserves information across the range of measured values rather than reducing each participant to a category such as meeting or not meeting a target. The registry reports least-squares mean changes, which are model-based adjusted averages, rather than simple raw group averages. The listed model includes investigator, baseline total daily dose category, treatment, visit, and treatment-by-visit interaction as fixed effects, with baseline HbA1c as a covariate.

HbA1c is used in this trial as the primary measure of glycemic control. The endpoint therefore directly addresses the stated comparison of two U-500R dosing schedules. The recorded analysis compares change over the period from baseline to Week 24; it does not describe the longer-term course or all other aspects of treatment experience.

The primary analysis population is defined as randomized participants who received at least one dose of U-500R and had evaluable HbA1c data. The registry reports the number analyzed in each group, but does not provide a detailed account of missing measurements or the assumptions used to handle them.

7. Results Posted in the Registry

• Human Regular U-500 Insulin BID: mean -1.22 versus -1.12 on control; difference -0.10 (95% CI -0.32 to 0.12), standardized effect d = -0.10; z-test p = 0.377.

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 Week 24 in Body Weight

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Human Regular U-500 Insulin BID: mean 4.88 versus 5.44 on control; difference -0.56 (95% CI -1.70 to 0.58), standardized effect d = -0.11; z-test p = 0.334.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Human Regular U-500 Insulin BID and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline to Week 24 in Total Daily Dose (TDD; Units) of Insulin

• 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:
  • Human Regular U-500 Insulin BID: mean 51.39 versus 55.19 on control; difference -3.80 (95% CI -32.07 to 24.47), standardized effect d = -0.03; z-test p = 0.792.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Human Regular U-500 Insulin BID and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline to Week 24 in Total Daily Dose (TDD; Units/kg) of Insulin

• 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:
  • Human Regular U-500 Insulin BID: mean 0.32 versus 0.32 on control; difference 0.00 (95% CI -0.22 to 0.22), standardized effect d = 0.00; z-test p = 1.000.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Human Regular U-500 Insulin BID and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists a broad set of secondary outcomes alongside the primary HbA1c comparison. These include attainment of several HbA1c targets, hypoglycemic event measures, body weight, total daily insulin dose, fasting plasma glucose, and time to reach HbA1c targets. They address different features of the treatment experience and are not interchangeable with the primary continuous HbA1c endpoint.

The posted continuous secondary estimates do not provide a single, consistent basis for extending the primary conclusion to every outcome. Some have estimates close to zero, while others have wider uncertainty. Sponsor-reported analyses for these continuous secondary contrasts are not reported in the registry record. Results for many of the other listed secondary measures are also not reported in the registry record, so a complete pattern across all secondary endpoints cannot be assessed from the recorded results.

In general, secondary outcomes can help describe whether a primary finding is accompanied by changes in other measures, but they often involve multiple comparisons. When many outcomes are tested, the chance of observing at least one apparently notable result by chance can increase. If no adjustment for multiplicity is made, individual unadjusted p-values should be interpreted cautiously and should not be treated as confirmatory evidence on their own.

The registry record does not state whether a multiplicity adjustment was planned or applied to secondary analyses. Nor does it provide a full set of secondary results and estimates that would allow an assessment of consistency across outcomes. The secondary outcomes can therefore be viewed as additional descriptive information, but they do not overturn or strengthen the primary result without fuller reporting and a stated analysis plan.

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

• Percentage of Participants Achieving HbA1c of ≤6.5%, <7.0%, <7.5%, and <8.0% at Week 24
• 30-Day Adjusted Rate of Hypoglycemic Events
• Change From Baseline to Week 24 in Body Weight
• Change From Baseline to Week 24 in Total Daily Dose (TDD; Units) of Insulin
• Change From Baseline to Week 24 in Total Daily Dose (TDD; Units/kg) of Insulin
• Change From Baseline to Week 24 in Fasting Plasma Glucose (FPG) Levels
• Time to Reach HbA1c Target Values
• Percentage of Participants With Hypoglycemic Events

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome was analyzed using a mixed-effects model for repeated measures (MMRM). In general, repeated-measures models are used when an outcome is collected at more than one time point for the same participants. They can estimate how the outcome changes over time while accounting for the fact that observations from one person are related. The registry identifies baseline and Week 24 as the primary endpoint timeframe and describes the analysis as a repeated-measures model.

What does a least-squares mean represent?

A least-squares mean is a model-based group average adjusted for the factors included in the model. It is not necessarily the same as the unadjusted arithmetic average of the observed values. Here, the registry says that the model included investigator, baseline total daily dose category (≤300 or >300 units), treatment, visit, and treatment-by-visit interaction as fixed effects. Baseline HbA1c was included as a covariate. The posted group changes of -1.12 and -1.22 are reported as least-squares means, each with a standard error of 0.08.

Why include these model terms?

In general, including baseline measures and design factors can account for known sources of variation and can make the adjusted comparison more precise, depending on the model and data. A treatment-by-visit interaction allows the model to represent different treatment patterns across visits. The registry gives the terms used, but does not report the full model specification, covariance structure, estimation details, or diagnostic checks. Those details cannot be inferred from the brief description.

What is the between-group contrast?

The sponsor-reported least-squares mean difference is -0.10 percentage points for twice daily relative to three times daily. A difference is an estimate of the contrast between group means; the sign depends on the order in which groups are subtracted. The negative sign indicates that the estimated change in the twice-daily group was lower than the estimated change in the three-times-daily group under that stated ordering. Since the endpoint is a change score and both changes are negative, the contrast should be read in context rather than as a standalone measure of improvement.

The registry reports a 95% confidence interval from -0.33 to 0.12 and p = 0.371. In general, a confidence interval describes a range of values compatible with the data and analysis under the interval's statistical assumptions. A p-value summarizes how inconsistent the observed data are with a specified null hypothesis, under the analysis assumptions. Neither quantity establishes the probability that a hypothesis is true, and neither on its own determines clinical importance.

What does the simple re-calculation add?

The unadjusted re-calculation uses the posted group estimates and standard errors to make a simple comparison. Its estimate is -0.10 and its unadjusted p-value is 0.377. The calculation is explicitly approximate because it uses summary statistics and an approximate standard deviation rather than participant-level data. It does not reproduce the sponsor's MMRM, including its covariates and repeated-measure structure. It serves as a check on the direction and broad scale of the reported contrast, not as an alternative primary analysis.

For the secondary continuous outcomes, the same kind of simple unadjusted re-calculation is posted for body weight and total daily insulin dose in units. The units-per-kilogram dose comparison has an estimated difference of 0.0 and a confidence interval from -0.222 to 0.222, without a sponsor-reported analysis. Standard errors are listed for the arm estimates. In general, a standard error describes the estimated sampling variability of an estimate; it is not the spread of individual participant measurements.

Why can over-dispersion matter?

In general, over-dispersion occurs when the observed variability in data is greater than a model assumes. It can make standard errors too small if it is not addressed, leading to confidence intervals that are too narrow and p-values that appear more persuasive than warranted. This is particularly relevant to count outcomes such as event rates. The registry lists a 30-day adjusted rate of hypoglycemic events as a secondary measure, but the detailed analysis model and results for that outcome are not reported in the registry record. There is therefore no basis to determine whether or how over-dispersion was assessed for that outcome.

The registry describes the primary population as randomized participants who received at least one dose and had evaluable HbA1c data. In general, conditioning an analysis on treatment received and evaluable outcomes can mean that the analyzed group differs from all randomized participants. The record does not give a detailed flow of participants, explain missing-data assumptions, or state whether sensitivity analyses were conducted. Those omissions limit independent assessment of the analysis.

11. Confidence Intervals and Precision

The sponsor-reported primary contrast is -0.10 percentage points of HbA1c, with a 95% confidence interval from -0.33 to 0.12. The interval extends on both sides of zero. In general, that means the data and model are compatible with a range of possible between-group differences, including a small difference favoring either schedule and values near no difference.

The interval is more informative than the p-value alone because it displays the estimated range of uncertainty on the outcome scale. It also makes clear that the data do not pinpoint a single exact difference. The reported p-value of 0.371 does not establish that the true difference is zero; it indicates that the observed data are not unusual under the null hypothesis used for that test, given the model assumptions.

The group estimates each have a standard error of 0.08. A standard error describes uncertainty in an estimated average, not the range of individual HbA1c changes. The primary comparison is model-adjusted, and the confidence interval reported by the sponsor should be used in preference to a simplified interval derived from summaries.

The simple recalculation produces a similar point estimate and an unadjusted p-value of 0.377. Since it is based on summary statistics and an approximate standard deviation, it does not capture every feature of the sponsor's repeated-measures model. Its role is limited to a basic check rather than a more precise or authoritative estimate.

The registry does not report a clinically justified difference threshold or the margin for the non-inferiority-or-equivalence analysis. Without that information, the interval cannot establish whether its full range is clinically negligible or meets a formal equivalence criterion. Precision is therefore best described in statistical terms: the central estimate is close to zero, but the interval permits differences in either direction.

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 use the observed primary effect as a reference and show hypothetical participants per arm, not the actual number enrolled or analyzed. They assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each row considers detecting the stated fraction of the observed effect. The trial's own original design assumptions are not reported in the registry record, so these figures should not be read as a reconstruction of its planned sample size.

The 80% power column gives the hypothetical number of participants per arm that would yield an 80% chance of detecting the specified effect under the listed assumptions, if that effect and assumptions were correct. The 90% power column uses a higher target chance of detecting the effect and therefore calls for a larger hypothetical sample under the same assumptions. These are planning illustrations, not observed power.

The illustrations consider 100%, 75%, and 50% of the observed effect. As the effect to detect becomes smaller, more participants are needed to distinguish it from random variation under the same assumptions. For the primary endpoint, the listed figures range from 1,623 participants per arm at 100% of the observed effect for 80% power to 8,689 participants per arm at 50% of the observed effect for 90% power. These values are derived from the observed contrast, which is uncertain; they should not be interpreted as evidence that the trial was designed for, or had, those sample sizes.

In general, power is a property of a planned test under specified assumptions, not a measure of whether a completed trial has proved an effect. The observed trial enrollment was 325, and the primary analysis included 162 participants in the three-times-daily group and 161 in the twice-daily group. Those observed counts are distinct from the hypothetical planning values.

13. How the Sample Size Is Derived

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

Formula

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

Worked calculation for this trial

1. Observed difference Δ = -0.100; pooled standard deviation σ = 1.017.
2. n = 2 × (1.96 + 0.842)² × 1.017² / -0.100² = 1,622.5, rounded up to 1,623 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -0.0999999999999999                      # observed difference in means
sigma <- 1.01666610489745                     # 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(1774968)
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 = 1,623 (80% power), 2,173 (90% power)
 75% of observed effect: n per arm = 2,885 (80% power), 3,862 (90% power)
 50% of observed effect: n per arm = 6,491 (80% power), 8,689 (90% power)
power.t.test: 1623.5 
Simulated power at n = 1,623 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 comparison as a mixed models analysis of the least-squares mean difference. The estimate is -0.10, the 95% confidence interval runs from -0.33 to 0.12, and p = 0.371. It is labeled as a non-inferiority-or-equivalence analysis. The record does not report the margin or other decision criteria needed to determine what formal conclusion the analysis was intended to support.

The simple re-calculation from posted summary statistics gives an estimate of -0.10 and an unadjusted p-value of 0.377. The close point estimates and p-values indicate that the simplified comparison is broadly aligned with the sponsor-reported result. However, the re-calculation is not the same analysis: it does not reproduce the repeated-measures model or its adjustment for investigator, baseline dose category, treatment, visit, treatment-by-visit interaction, and baseline HbA1c.

The sponsor's own analysis takes precedence for interpretation of the trial. The recalculation is a secondary statistical check, and its interval and p-value should not displace the reported model result. Neither the reported label nor a non-significant p-value alone establishes equivalence or non-inferiority; the necessary margin and rule are not reported in the registry record.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome and 14 secondary outcome measures. Testing multiple endpoints creates a general multiplicity concern: the more statistical comparisons made, the greater the chance that at least one result may appear notable because of random variation. Prespecified testing hierarchies or other adjustments can address this issue, but whether such procedures were planned or used is not reported in the registry record.

The record reports one sponsor analysis for the primary endpoint, labeled as non-inferiority or equivalence. The margin, hypothesis structure, and decision rule are not reported in the registry record. Without those details, it is not possible to verify how the analysis label should be interpreted or whether it followed a prespecified criterion. The unadjusted re-calculation is not a substitute for those missing design details.

For the primary endpoint, the stated analysis population consists of randomized participants who received at least one dose of U-500R and had evaluable HbA1c data. The registry does not provide a participant flow showing the number randomized, treated, and included at each visit, or reasons for absent data. It also does not report the assumptions used for missing outcomes, the extent of missingness, or sensitivity analyses examining alternative assumptions.

MMRM can use available repeated measurements under specified assumptions about missingness. In general, the validity of an analysis using such a model depends in part on whether its assumptions are reasonable for the way data are missing. The registry description does not give enough detail to assess that issue. The exact analysis population and model are reported, but a complete evaluation of missing-data handling is not possible.

The randomization method, allocation concealment, and any prespecified subgroup analyses are not reported in the registry record. These gaps do not erase the posted result, but they limit assessment of the analysis plan, the potential for selective emphasis, and the robustness of the conclusions beyond the reported primary model.

16. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hypoglycaemia (3.7%), acute myocardial infarction (1.9%), cardiac failure congestive (1.8%), pneumonia (1.8%), chest pain (1.2%).

The registry reports serious adverse events for the two treatment groups using the number at risk and the number affected. For the three-times-daily group, 28 of 162 participants were reported as affected by serious adverse events. For the twice-daily group, 27 of 163 were reported as affected. These are counts of participants affected, not counts of all individual events.

For other adverse events, the registry reports 80 of 162 participants in the three-times-daily group and 94 of 163 in the twice-daily group as affected. The categories are reported separately from serious adverse events. The serious-event terms listed include hypoglycaemia at 3.7%, acute myocardial infarction at 1.9%, cardiac failure congestive at 1.8%, pneumonia at 1.8%, chest pain at 1.2%, and cellulitis at 1.2%. The registry record does not identify the treatment-group allocation for these listed percentages.

In general, adverse-event counts describe events recorded during a study; they do not by themselves establish that an event was caused by the treatment. Differences in event counts also require context, including definitions, exposure, follow-up, and the prespecified safety analysis. The registry record does not report deaths for these groups, and detailed methods for classifying or comparing adverse events are not reported.

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

The primary result describes the estimated difference in HbA1c change between two schedules of U-500 regular insulin over 24 weeks in the trial population. The sponsor-reported estimate is -0.10 percentage points for twice daily relative to three times daily. Its 95% confidence interval, from -0.33 to 0.12, includes zero and spans possible differences in either direction. The result does not provide strong evidence of a difference under the reported model, but it does not prove that the schedules have identical effects.

The registry labels the analysis as non-inferiority or equivalence. A formal conclusion of non-inferiority or equivalence depends on a prespecified margin and decision rule. Those are not reported in the registry record, so the label cannot be used on its own to establish either conclusion. The nonsignificant p-value likewise should not be read as proof of sameness.

The findings apply to the participants and study conditions described in the record: adults with type 2 diabetes inadequately controlled on high-dose U-100 insulin therapy, with or without oral agents, assigned to one of two dosing schedules and followed for the primary endpoint through Week 24. They do not establish what would occur in other populations, with other regimens, over other periods, or for outcomes that were not reported.

The registry also lists secondary outcomes, but full results are not reported for many of them. The primary endpoint should therefore remain the main basis for interpreting the posted comparison. This statistical account does not determine treatment choice or provide individual medical advice; it explains the estimate, uncertainty, and limits of what the registry record can establish.

18. Why This Trial Matters Statistically

This trial illustrates how the same drug can be studied through a comparison of dosing schedules, with randomization intended to support a fair between-group comparison. The primary analysis used a model-based estimate that adjusted for stated study factors and baseline HbA1c, rather than relying only on raw change scores. That distinction matters when interpreting what the reported group values represent.

It also illustrates why estimates and confidence intervals should be read together. The point estimate is close to zero, but the confidence interval allows effects in either direction. A p-value that is not statistically significant does not prove that two regimens are equivalent. The registry's non-inferiority-or-equivalence label makes the missing margin especially important: without it, readers cannot judge the result against a formal threshold.

The gap between the sponsor analysis and a simple summary-statistic re-calculation is small in this case, but the methods differ. The re-calculation is useful as a check, while the adjusted repeated-measures analysis remains the primary reported result. This distinction is useful in applied biostatistics: a close numerical check can support confidence in basic arithmetic without replacing the model specified for the trial analysis.

Finally, the registry demonstrates the limits of interpreting secondary outcomes when results and multiplicity procedures are incomplete. A list of planned or recorded measures is not the same as a fully reported set of estimates. Statistical conclusions depend not only on calculations, but also on transparent reporting of populations, assumptions, analysis rules, and uncertainty.

19. Limitations

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

The registry record leaves important details unavailable for independent evaluation. Although the primary outcome is labeled as a non-inferiority-or-equivalence analysis, the margin and decision criteria are not reported. The statistical conclusion intended by that analysis therefore cannot be independently assessed from the record.

The primary analysis population excludes randomized participants without at least one dose or without evaluable HbA1c data. The record does not report a detailed flow of participants, reasons for missing measurements, or sensitivity analyses. As a result, the possible influence of missing data and post-randomization exclusions cannot be evaluated fully.

The study was open label. The registry does not describe masking, and none was recorded. While the primary endpoint is a laboratory measure, open-label conduct may be relevant to treatment behavior, co-interventions, or reporting of other outcomes in general. The record does not provide information to assess those possibilities in this trial.

The registry provides only limited secondary-outcome results in the statistical summaries. It lists a range of additional measures, but many results and sponsor analyses are not reported in the registry record. Multiplicity procedures are also not reported, limiting the weight that can be placed on isolated secondary comparisons.

The primary endpoint covers baseline to Week 24. Longer-term results are not reported in the registry record. Detailed information on adherence, allocation concealment, site conduct, the original sample-size justification, and the full statistical analysis plan is also not reported in the registry record. These limitations constrain interpretation of robustness and generalizability without negating the result that is reported.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study of Human Regular U-500 Insulin in Adult Participants With Type 2 Diabetes — NCT01774968.
• Registry record last updated 2015-09-22; status last verified 2015-09.
• Results first posted 2015-05-12.
• 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

NCT01774968 is a completed Phase 4 randomized, open-label, parallel trial comparing Human Regular U-500 Insulin given three times a day with the same insulin given two times a day. Both regimens were titrated using blood glucose readings and continued for 24 weeks. The record lists 325 participants enrolled.

The primary outcome was change in HbA1c from baseline to Week 24. The sponsor-reported difference for twice daily relative to three times daily was -0.10 percentage points, with a 95% confidence interval from -0.33 to 0.12 and p = 0.371. The interval includes zero; the registry's non-inferiority-or-equivalence label cannot be fully evaluated because its margin and decision rule are not reported.

The registry lists 14 secondary measures and reports some secondary continuous estimates. Full results for many listed secondary outcomes, multiplicity procedures, and detailed missing-data methods are not reported in the registry record. Interpretation is consequently centered on the stated primary analysis and its uncertainty.