Comparison of Two Basal Insulins for Patients With Type 2 Diabetes Taking Oral Diabetes Medicines and Exenatide: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparison of Two Basal Insulins for Patients With Type 2 Diabetes Taking Oral Diabetes Medicines and Exenatide (NCT00560417), a randomized phase 3 study. The primary outcome is Change From Baseline in Hemoglobin A1C (HbA1c) at Endpoint (Last Observation Carried Forward [LOCF]).

2. Clinical Question

3. Trial Design

NCT00560417 is a completed phase 3 study sponsored by Eli Lilly and Company. The registry lists a randomized, parallel-group design with no masking. Enrollment was 339. The record lists the study start as 2007-11 and primary completion as 2009-12. These dates describe the registry record; further details about the conduct of the trial are not reported in the registry record.

The two arms received insulin lispro protamine suspension (ILPS) or insulin glargine. The study summary describes both as being used with participants’ oral diabetes medications and exenatide. The comparison therefore concerns two basal insulin regimens in that treatment context; it is not a comparison with placebo or with no insulin. The record does not report the doses, titration schedule, adherence, or other treatment procedures, so those details cannot be used to characterize the contrast further.

The primary outcome was change from baseline in HbA1c at endpoint, with endpoint defined as last observation carried forward (LOCF) up to 24 weeks. The primary analysis population was described as all randomized participants with at least one post-baseline measurement. The registry identifies the endpoint summaries as least-squares means controlled for baseline HbA1c, baseline HbA1c group, and baseline sulfonylurea group.

The record lists secondary outcomes covering HbA1c values and thresholds, self-monitored blood glucose profiles, glycemic variability, self-reported hypoglycemic episodes and rates, and body weight. It lists 11 outcome measures and 43 analyses. The registry record does not report details of the allocation procedure, masking beyond “none,” a sample-size justification, or a prespecified statistical analysis plan.

4. Participant Flow and Accounting

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

Before assignment: Intention to Treat (ITT) Population consisted of all randomized participants who received at least one dose of study drug and had at least one post-baseline measurement.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All randomized participants with at least one post-baseline measurement. Intention to Treat (ITT) population.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Hemoglobin A1C (HbA1c) at Endpoint (Last Observation Carried Forward [LOCF]), measured over Baseline, Endpoint (LOCF) up to 24 weeks.

HbA1c is a continuous measure related to blood glucose over time. In general, it is used in diabetes research to assess glycemic control because it reflects more than a single glucose reading. The primary endpoint was not simply the HbA1c value at endpoint: it was the change from baseline in HbA1c at endpoint, expressed as percent of glycosylated hemoglobin.

The endpoint timeframe was baseline to endpoint, with endpoint defined as up to 24 weeks using last observation carried forward. Change scores express how a measure moved from each participant’s baseline, while a comparison of group-level changes can describe whether the average changes differed between treatment groups. The registry specifies least-squares mean values, adjusted for baseline HbA1c, baseline HbA1c group, and baseline sulfonylurea group. These adjusted summaries aim to account for those listed baseline factors in the group comparison.

The registry record does not report the numerical least-squares means, the between-group estimate, its confidence interval, or a p-value for the primary endpoint. Consequently, the size and precision of the comparison cannot be assessed from the record. The use of HbA1c identifies what was measured; it does not, on its own, establish what the comparison found.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registered primary analysis used least-squares means for a continuous outcome: change in HbA1c from baseline at endpoint. The following general explanations clarify what those terms mean; they do not supply unreported trial results.

What does a least-squares mean represent?

In general, a least-squares mean is a model-based estimate of a group average. Unlike a simple arithmetic mean, it can account for variables included in the statistical model. The registry says the primary summaries were controlled for baseline HbA1c, baseline HbA1c group, and baseline sulfonylurea group. This identifies the factors named for adjustment, but the model form, coefficient estimates, and details of how categories were coded are not reported in the registry record.

An adjusted group summary is not itself the treatment contrast. To interpret the comparison, readers generally need the estimated difference between arms, with a measure of uncertainty such as a confidence interval. A confidence interval describes the range of effect estimates compatible with the data under the model and its assumptions. The registry record does not report a primary confidence interval or adjusted between-group estimate, so neither the magnitude nor precision of the comparison can be assessed.

Why report a standard error?

The registry identifies the dispersion measure for the primary outcome as a standard error. In general, a standard error describes the sampling uncertainty of an estimate: it indicates how much an estimate would tend to vary across repeated samples under a statistical model. It is not the same as the spread of individual participant outcomes. The record does not give the standard errors for the primary group estimates, and the standard error alone would not replace a direct estimate of the between-group difference.

What does LOCF mean?

Last observation carried forward assigns a participant’s last observed measurement to a later endpoint when the endpoint measurement is unavailable. In general, this creates an endpoint dataset in which some values stand in for measurements not observed at the endpoint. It is an assumption-based approach: the carried-forward value is treated as a useful representation of the later outcome, even though the actual later value is unknown. If outcomes change after the last observation, this method may not reflect that change.

The registry specifies an endpoint using LOCF up to 24 weeks, but does not report how many measurements were missing, when observations ended, why data were missing, or whether analyses using alternative missing-data assumptions were conducted. The consequences of LOCF for this trial’s estimates therefore cannot be judged from the record.

What does the analysis population description mean?

The registry describes the primary population as all randomized participants with at least one post-baseline measurement and labels it the intention-to-treat population. That definition makes the post-baseline measurement condition part of the stated population. The record does not report how many randomized participants lacked such a measurement or how they were handled. In general, an analysis population and a missing-data method are related but distinct: the population definition determines who is included, while the missing-data method determines how unavailable outcome values are addressed.

What do the planning illustrations mean?

The sample-size illustrations are described as planning calculations based on the observed effect, not as the trial’s own design justification. The stated assumptions are a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. For outcomes other than time-to-event outcomes, the illustrations are expressed as participants per arm. In general, such calculations illustrate the sample size associated with chosen assumptions; they do not show that the trial had a particular probability of finding a statistically significant result. They are not observed power. The trial’s sample-size justification is not reported in the registry record, and no planning illustration results are available in the record described.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 11 outcome measures and 43 analyses, including a primary HbA1c change outcome and secondary measures spanning HbA1c, glucose monitoring, hypoglycemia, and body weight. Multiple outcomes and analyses create a multiplicity issue: in general, the more statistical comparisons considered, the greater the chance that at least one may appear notable by chance if each is interpreted without accounting for the others. This general concern does not establish that any particular result is false or that an adjustment was required.

The registry record does not report a multiplicity strategy, such as an adjustment to significance thresholds, a hierarchy for testing, or a distinction between confirmatory and exploratory analyses. It also does not report the full statistical analysis plan. The status of any secondary comparison should therefore not be assumed from the number of registered outcomes alone.

For the primary endpoint, the record specifies LOCF at endpoint up to 24 weeks. It does not report the number of participants with missing endpoint data, reasons for missingness, or results from alternative missing-data analyses. In general, conclusions from an imputed endpoint can depend on the assumptions used for missing observations; the record does not provide enough information to evaluate those assumptions empirically.

The stated primary analysis population included randomized participants with at least one post-baseline measurement and was labeled intention-to-treat. The number of randomized participants excluded under that condition is not reported in the registry record. Further details about analysis populations, missing-data sensitivity analyses, and multiplicity procedures are not reported in the registry record.

11. 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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: atrial fibrillation (1.2%), chest pain (0.6%), cholelithiasis (0.6%), ankle fracture (0.6%), drug exposure during pregnancy (0.6%).

The registry reports serious adverse events among participants at risk in each arm: 9 of 170 in the ILPS group and 5 of 167 in the glargine group. It separately reports other adverse events in 96 of 170 participants in the ILPS group and 115 of 167 in the glargine group. These figures are counts with the recorded denominators; they should not be read as a complete description of each participant’s event history or as evidence that an event was caused by a study drug.

The listed serious-event terms and percentages are atrial fibrillation, 1.2%; chest pain, 0.6%; cholelithiasis, 0.6%; ankle fracture, 0.6%; drug exposure during pregnancy, 0.6%; and fall, 0.6%. The registry record does not provide further detail about timing, recurrence, or event-level attribution in this summary. Death counts are not reported in the registry record.

General caution is appropriate when reading safety counts: the number of participants with at least one event is not necessarily the same as the total number of events, and the summary does not describe the full clinical circumstances. The reported counts alone do not establish comparative safety.

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

The record establishes that NCT00560417 was a randomized, parallel-group phase 3 comparison of ILPS and insulin glargine in people with type 2 diabetes taking oral diabetes medications and exenatide. Its primary endpoint was change in HbA1c from baseline at endpoint, with LOCF used through up to 24 weeks. The stated analysis used least-squares means controlled for baseline HbA1c, baseline HbA1c group, and baseline sulfonylurea group.

Those design details explain what the trial set out to compare and how the primary outcome was described. They do not show the outcome. The registry record does not report the primary group estimates, the adjusted treatment difference, a confidence interval, or a p-value. Accordingly, the direction and size of the observed effect, and the uncertainty around it, cannot be stated from this record. A missing numerical comparison should not be replaced with an inference from the study’s title or completion status.

The secondary outcome list and safety summary add context about what else was registered and which adverse-event counts were posted. The list includes HbA1c measures, glucose profiles, glycemic variability, hypoglycemia, and body weight. The listed serious and other adverse-event counts describe events recorded by arm, but do not establish causation or fully characterize the benefit-harm balance.

In general, a treatment comparison is most informative when the estimated effect is considered alongside its uncertainty, the analysis assumptions, and the completeness of outcome reporting. Here, key numerical efficacy results and several methodological details are not reported in the registry record. The trial record can therefore support a careful account of the study design and reported safety counts, but not a conclusion about the comparative HbA1c result.

13. Why This Trial Matters Statistically

This trial illustrates the distinction between a registered clinical question and a fully interpretable statistical result. The record identifies a randomized comparison, a continuous primary endpoint, an analysis population, covariates used for adjustment, and an endpoint-handling method. Those details are important because estimates are meaningful only in relation to the outcome definition and analysis that produced them.

It also shows why the effect estimate and its uncertainty matter. A description of the model and endpoint cannot indicate whether the groups’ changes differed, how large any difference was, or how precise the estimate was. Without the numerical least-squares means or a between-group estimate and confidence interval, readers cannot assess the primary comparison statistically. The absence of those values limits interpretation independently of the trial’s enrollment or phase.

The breadth of registered secondary outcomes further highlights the need to understand multiplicity and analysis plans. The registry lists 11 outcome measures and 43 analyses, but does not report a multiplicity strategy or full analysis plan. In general, a prespecified approach helps distinguish a primary confirmatory question from a wider set of secondary or exploratory comparisons.

Finally, LOCF and the stated analysis population make missing-data reporting relevant. The endpoint definition indicates how an unavailable later measurement could be represented, while the population definition requires a post-baseline measurement. Without counts and sensitivity analyses, the impact of these choices cannot be evaluated. Together, these features make transparent reporting of estimates, uncertainty, missingness, and analysis decisions central to statistical interpretation.

14. 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 most direct limitation is that the registry record does not report numerical results for the primary HbA1c endpoint. It therefore lacks the adjusted group estimates and between-group comparison needed to describe effect direction, magnitude, and precision. Numerical results for the registered secondary outcomes are also not reported in the record described.

The primary endpoint used LOCF up to 24 weeks, but the registry does not report missing endpoint counts, reasons for missingness, or sensitivity analyses. The analysis population was described as randomized participants with at least one post-baseline measurement; the number not meeting that condition is not reported in the registry record. These omissions prevent assessment of how missing observations and population selection may have affected the analysis.

Although adjustment factors are named, the model specification and supporting diagnostics are not reported. The registry record also does not report a multiplicity strategy or full statistical analysis plan, despite listing 11 outcome measures and 43 analyses. It does not report the trial’s sample-size justification. The planning illustrations are not a substitute for that information and are not observed power.

The record identifies a randomized, unmasked, parallel design and total enrollment, but does not report details such as treatment doses, titration, adherence, allocation procedures, or reasons for participant withdrawal. The safety summary supplies group counts and selected serious-event terms, but death counts and further event details are not reported in the registry record. These limitations restrict interpretation of both the efficacy comparison and the safety summary.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Comparison of Two Basal Insulins for Patients With Type 2 Diabetes Taking Oral Diabetes Medicines and Exenatide — NCT00560417.
• Registry record last updated 2011-01-13; status last verified 2011-01.
• Results first posted 2010-12-14.
• 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.

18. Record Summary

NCT00560417 was a completed phase 3 randomized, parallel-group trial sponsored by Eli Lilly and Company, with no masking listed and enrollment of 339. The registry dates the study start as 2007-11 and primary completion as 2009-12. It compared insulin lispro protamine suspension with insulin glargine in participants with type 2 diabetes taking oral diabetes medications and exenatide.

The primary endpoint was change from baseline in HbA1c at endpoint, using LOCF up to 24 weeks. The primary analysis was described as least-squares means controlled for baseline HbA1c, baseline HbA1c group, and baseline sulfonylurea group, in randomized participants with at least one post-baseline measurement. Numerical primary outcome results are not reported in the registry record.

The registry lists 11 outcome measures and 43 analyses. Its safety summary reports serious adverse events in 9 of 170 ILPS participants and 5 of 167 glargine participants; death counts are not reported in the registry record.