Pioglitazone/Glimepiride (Sonias) Combination Tablets Special Drug Use Surveillance Survey in Patients With Type 2 Diabetes Mellitus Who Respond Poorly to Pioglitazone: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pioglitazone/Glimepiride (Sonias) Combination Tablets Special Drug Use Surveillance Survey in Patients With Type 2 Diabetes Mellitus Who Respond Poorly to Pioglitazone (NCT02098746), a registered phase not stated study. The primary outcome is Frequency of Adverse Drug Reactions.

2. Clinical Question

3. Trial Design

This is a special drug use surveillance survey of pioglitazone/glimepiride in patients with type 2 diabetes who responded poorly to pioglitazone monotherapy. The stated purpose was to evaluate safety and efficacy during long-term use in routine clinical practice. The registry lists Takeda as the sponsor and marks the study completed.

The registry dates the study start to 2011-06 and primary completion to 2014-07. Enrollment was 294. The intervention listing describes pioglitazone/glimepiride, with the arm description specifying 15 mg/1 mg orally once daily before or after breakfast. No separate comparator arm is described. The allocation method is not stated in the registry record, so the information does not support treating the results as a randomized treatment comparison.

The primary safety population included patients who received at least one dose of the combination. This Safety Analysis Set contained N=289, distinct from the total enrollment figure. The registry does not report why these population counts differ, and no reason should be inferred from the difference alone.

Several outcomes are named, including changes from baseline in glycosylated hemoglobin (HbA1c), fasting blood glucose, and fasting insulin. However, analyses for these outcomes are not posted in the registry record. The record therefore supports describing the surveillance setting, treatment and primary safety outcome, but it does not supply a reported efficacy comparison or a detailed statistical protocol.

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: Participants with a diagnosis of type 2 diabetes mellitus receiving treatment with pioglitazone/glimepiride 15 mg/1 mg once daily in routine clinical practice, because pioglitazone alone was not considered effective, were enrolled in the study.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The Safety Analysis Set (safety assessment population) included all patients who received at least one dose of pioglitazone/glimepiride (N=289).

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: Baseline measures are based on the Safety Analysis Set (safety assessment population) that included all patients who received at least one dose of pioglitazone/glimepiride.

6. Primary Endpoint

The registry lists the primary outcome as Frequency of Adverse Drug Reactions, measured over 12 months.

The primary endpoint is the frequency of adverse drug reactions, defined in the registry as the number of participants who experienced such reactions. The stated timeframe is 12 months. The record describes adverse events as unfavorable and unintended signs, symptoms, or diseases temporally associated with use of a medicinal product during the period from the first dose through the last dose. Events considered possibly associated with the medicinal product are classified as adverse drug reactions.

This kind of endpoint is used to describe how often potentially treatment-associated reactions were recorded among people exposed to the medicine. It is a participant count, not a measure of the number of all events: one participant could potentially have more than one event, while the endpoint counts participants with reactions. The registry says that frequency, seriousness, and time to onset were tabulated by symptom, but detailed onset results are not reported in the registry record.

Because no comparator is described, the endpoint is primarily descriptive in this record. The count can characterize reported reactions in the safety analysis population, but by itself it does not establish how the count would compare under another treatment or without the combination. Nor does the endpoint quantify changes in diabetes measures. Although changes in HbA1c, fasting blood glucose, and fasting insulin are listed as secondary outcomes, their analyses are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a binary participant-level event status: a participant either appears in the count of those with an adverse drug reaction or does not. The reported quantity is the number of participants with the event, not an event rate per unit of person-time. No inferential method for this outcome is reported in the registry record. In particular, no confidence interval, hypothesis test, adjusted estimate, or model specification is reported.

What does a confidence interval add?

As a general statistical concept, a confidence interval expresses uncertainty around an estimate under specified assumptions. It helps distinguish a relatively precise estimate from one that is compatible with a wide range of underlying values. A count alone does not show this uncertainty. No confidence interval for the primary endpoint is reported in the registry record, so the precision of the posted count cannot be assessed from an interval estimate.

Why does the lack of a comparator matter?

As a general principle, a comparison needs groups whose outcomes can be contrasted under a defined design. A treatment-group count without a comparator describes events among people receiving that treatment, but does not estimate a difference between treatments. In this record, no comparator is described and allocation is not stated. It would therefore be inappropriate to interpret the primary count as a comparative effect or as proof that the combination caused every recorded event.

How should the event categories be read?

The registry distinguishes adverse events from adverse drug reactions. Adverse events are described as unfavorable and unintended signs, symptoms, or diseases temporally associated with medicinal product use, from first dose to last dose. Adverse drug reactions are the subset considered possibly associated with the product. The primary endpoint counts participants with reactions. The record says frequency, seriousness, and time to onset were tabulated by symptom, but does not report a detailed analysis method for these summaries.

Binary outcomes can be summarized as counts and proportions. As a general statistical concept, a proportion is often accompanied by an interval estimate, especially when the number of participants or events is limited, because the count alone does not communicate sampling uncertainty. No proportion should be calculated from the posted counts for this explanation, and no interval estimate is reported in the registry record. The listed percentages for serious terms are reported as written.

What do the planning assumptions mean?

The planning illustrations associated with the record use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They define the effect to detect as a stated fraction of the observed effect. These are planning assumptions, not a description of the surveillance survey’s actual allocation or analysis. The illustrations are not observed power and do not show that the study had a particular probability of detecting an effect. The registry record does not report the study’s own sample-size justification.

As a general concept, a sample-size calculation depends on assumptions about the outcome, target effect, variability, significance level, and allocation. Changing those assumptions changes the planning result. Since the survey has no comparator described in the registry record, the two-arm equal-allocation assumptions should not be mistaken for evidence that the study itself used two randomized arms. No calculated planning values are repeated.

For time-to-event planning, the stated convention is to express the requirement as a total number of events using the Schoenfeld method; other planning illustrations are expressed as participants per arm. These conventions describe planning outputs only. They do not establish that a time-to-event analysis was performed for the primary outcome. The registry record does not report a time-to-event statistical model, censoring rules, or an inferential analysis of time to onset.

Over-dispersion is a general concern when count variation exceeds what a simple model assumes; it can make standard errors too small if ignored. It is not possible to determine whether over-dispersion is relevant to an analysis that is not reported. The primary measure is a number of participants with reactions, and no model for counts or repeated events is specified in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The record names five outcome measures and lists changes from baseline in HbA1c, fasting blood glucose, and fasting insulin among the secondary outcomes. It reports zero analyses. This means the registry record does not provide results or statistical tests for the named secondary outcomes. It also does not report a hierarchy of endpoints, a multiplicity adjustment, or a prespecified approach to interpreting several outcome tests.

As a general statistical concept, testing multiple outcomes can increase the chance of obtaining an apparently unusual result when no true effect exists. Prespecifying a primary outcome and a strategy for secondary outcomes can help readers interpret a set of tests. Here, the primary endpoint is identified as frequency of adverse drug reactions, but the approach to multiplicity is not reported in the registry record. No claim about whether adjustments were used can be made.

The record identifies a Safety Analysis Set of N=289, consisting of patients who received at least one dose. It does not report missing-data methods, the amount of missing outcome information, reasons for missingness, or rules for handling participants without complete follow-up. The difference between the total enrollment of 294 and the safety population is not explained. The record also does not describe an intention-to-treat population, a per-protocol population, or sensitivity analyses.

Missingness can affect interpretation if participants with incomplete observations differ from those with complete information. This is a general concern, not a finding about this survey. Without reported missing-data procedures and analysis populations for the secondary measures, the extent to which their results might be affected cannot be assessed. Since analyses for those measures are not posted, their estimates and precision are not available 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 5% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: renal neoplasm (0.3%), atrioventricular block complete (0.3%).

The registry records 2 participants in the serious adverse drug reaction category and 0 in the other adverse drug reaction category among the safety population, which included N=289 patients who received at least one dose. These are participant counts in the posted safety result. The registry lists Renal neoplasm and Atrioventricular block complete as serious terms, each reported as 0.3%.

These figures should be read as surveillance summaries, not as a comparison with another treatment. The record does not describe a comparator, and it does not provide an analysis that estimates how many events would have occurred under a different treatment. The term percentages are quoted as reported; no additional percentages or rates are calculated.

The registry’s definition separates adverse events from adverse drug reactions. The latter are events considered possibly associated with the medicinal product. That classification does not by itself demonstrate causation. The record states that seriousness and time to onset were tabulated by symptom, but it does not report detailed timing, case narratives, or the method used to assess possible association. The counts therefore characterize the posted classifications, while leaving those further details unavailable.

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

The result describes adverse drug reactions recorded among patients in the safety analysis set during a surveillance survey of pioglitazone/glimepiride. The primary measure is a participant count, and the registry reports serious and other reaction categories separately. The serious terms and their percentages add descriptive detail, but the registry record does not provide a comparative safety estimate or uncertainty interval.

The result does not establish that the medicine caused every recorded event. The registry’s own definition identifies adverse drug reactions as events considered possibly associated with the medicinal product, whereas an adverse event is defined more broadly by its temporal association with product use. These categories are useful for describing surveillance findings but do not substitute for a controlled comparison.

The information also does not establish comparative efficacy. Although the record names changes from baseline in HbA1c, fasting blood glucose, and fasting insulin as secondary outcomes, analyses for those outcomes are not reported. The posted primary safety counts cannot answer whether these measures changed, how precisely they changed, or how those changes compare with another treatment.

General statistical interpretation depends on design, completeness of follow-up, and uncertainty as well as the point estimate. The registry record does not report confidence intervals, a detailed analysis plan, missing-data methods, or reasons for the difference between enrollment and the safety analysis population. Those omissions limit assessment of precision and potential bias. The findings should therefore be understood as a description of the recorded surveillance data, not as a broad claim about comparative effects or all possible risks.

13. Why This Trial Matters Statistically

This trial is statistically informative because it illustrates the distinction between a surveillance objective and a comparative treatment question. The record describes routine clinical use and reports adverse-reaction counts in a safety population. Such summaries can document events observed in practice, while a causal or comparative effect requires a design and analysis that support comparison. No comparator or allocation method is described in the registry record.

It also highlights why the definition of an endpoint matters. The primary outcome is not every unfavorable event occurring during treatment; it is the number of participants with events considered possibly associated with the medicinal product. Reading the endpoint definition alongside its count helps prevent the broader adverse-event concept from being confused with the narrower adverse-drug-reaction classification.

The difference between enrollment and the safety analysis population is another important reporting detail. The registry gives both figures but does not explain the difference. Reporting population definitions allows readers to understand which participants contribute to a particular result; when reasons for exclusions or missing information are unavailable, the implications cannot be evaluated.

Finally, named outcomes are not equivalent to reported analyses. The registry lists measures of glycemic status, but posts no analyses for them. This distinguishes an outcome’s presence in the record from an available estimate and its precision. The case underscores the value of complete outcome reporting, clear population accounting, and an explicit statistical plan for interpreting both safety summaries and efficacy measures.

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 registry record leaves important design details unspecified. It does not state the allocation method or describe a comparator, so the safety counts cannot be interpreted as a controlled treatment contrast. It also does not report the study’s statistical plan, interim analyses, or sample-size justification. The planning assumptions associated with illustrative calculations should not be treated as the protocol used for the survey.

The primary result is presented as counts, without a reported confidence interval or statistical test. This limits assessment of precision. The registry describes a 12-month timeframe for the primary endpoint, but detailed follow-up, censoring, and time-to-onset analysis methods are not reported. It says time to onset was tabulated by symptom, but does not provide those results in the record described.

The record gives enrollment of 294 and a Safety Analysis Set of N=289 but does not explain the difference. It does not report missing-data amounts or methods, reasons for incomplete information, or analysis populations for secondary outcomes. These gaps make it difficult to evaluate whether the people contributing to each outcome differ systematically.

Although secondary changes from baseline in HbA1c, fasting blood glucose, and fasting insulin are listed, no analyses are reported. The record therefore does not allow readers to examine estimates, variability, or multiplicity handling for those measures. These limitations are specific to the information available in the registry record and should not be filled by assuming methods or results that are not documented.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Pioglitazone/Glimepiride (Sonias) Combination Tablets Special Drug Use Surveillance Survey in Patients With Type 2 Diabetes Mellitus Who Respond Poorly to Pioglitazone — NCT02098746.
• Registry record last updated 2016-10-18; status last verified 2016-08.
• Results first posted 2016-02-09.
• 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

NCT02098746 is a completed special drug use surveillance survey sponsored by Takeda. It concerns pioglitazone/glimepiride combination tablets in patients with type 2 diabetes who responded poorly to pioglitazone monotherapy. The registry describes routine clinical use and a 12-month primary safety endpoint.

Total enrollment was 294; the Safety Analysis Set included N=289 patients who received at least one dose. The primary outcome is the number of participants with adverse drug reactions. The posted result records 2 participants in the serious category and 0 in the other category; the listed serious terms are each reported as 0.3%.

The record lists secondary changes from baseline in HbA1c, fasting blood glucose, and fasting insulin, but does not report analyses for them. A comparator, allocation method, missing-data strategy, and detailed statistical analysis plan are not reported in the registry record.