Long-term Use of Sonias Combination Tablets in Patients With Type 2 Diabetes Mellitus: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Long-term Use of Sonias Combination Tablets in Patients With Type 2 Diabetes Mellitus (NCT02098733), a registered phase not stated study. The primary outcome is Number of Participants With Adverse Drug Reactions.

2. Clinical Question

In practical terms, the registry describes a question about safety during use in routine care, not a question about comparative efficacy. HbA1c, fasting blood glucose and fasting insulin are listed among the secondary outcomes, including changes from baseline and measured levels. However, numeric results for those outcomes are not reported in the registry record. The available information therefore identifies what investigators intended to assess, but does not support conclusions about the size or precision of changes in these measures.

3. Trial Design

The record identifies this study as a special drug use surveillance survey of long-term pioglitazone/glimepiride use in people with type 2 diabetes mellitus. Its stated purpose is to evaluate safety and efficacy in the routine clinical setting. The listed intervention is pioglitazone/glimepiride at 15 mg/1 mg or 30 mg/3 mg, orally once daily before or after breakfast. The study is marked completed, and the sponsor is Takeda.

The recorded start is 2011-06 and the primary completion is 2014-05. Enrollment is listed as 1168. Those details describe the registry record, but they do not establish how participants were recruited, how long each participant was followed, or whether treatment decisions were made according to a prespecified protocol. Such details are not reported in the registry record.

No comparator treatment is listed. The record specifically notes that this is not a randomized comparison and that allocation is not stated. As a result, the design information does not support interpreting outcomes as an effect relative to another treatment. In routine-care surveillance, observed outcomes can reflect the medicine, participants’ underlying health, other care, and the circumstances that led to treatment; without a comparable control group, these influences cannot be separated by a treatment contrast.

The primary outcome is the number of participants with adverse drug reactions over 12 months. The record also names secondary outcomes for HbA1c, fasting blood glucose and fasting insulin, including baseline changes and measured levels. Although seven outcome measures are listed, the record reports no analyses. It does not describe the statistical analysis plan, sample-size justification, interim analyses, or procedures for handling missing observations. These omissions limit what can be inferred about how the planned outcomes were evaluated.

The adverse-event summary uses 1158 participants as the number at risk. The registry does not reconcile this figure with the enrollment of 1168, nor does it explain whether the populations differ because of eligibility, follow-up, or data availability. The primary endpoint population is described only as all enrolled participants with available data. Accordingly, enrollment should not be assumed to be the denominator for every outcome.

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 historical diagnosis of type 2 diabetes mellitus for whom therapy with pioglitazone combined with glimepiride is suitable and long-term treatment is considered necessary were enrolled in one treatment group to receive pioglitazone/glimepiride 15 mg/1 mg or 30 mg/3 mg, orally once daily before or after breakfast.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All enrolled participants with available 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 enrolled participants with available data.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With Adverse Drug Reactions, measured over For 12 months.

The primary endpoint is the number of participants with adverse drug reactions over the stated timeframe of 12 months. The record defines an adverse event as an unfavorable and unintended sign, symptom or disease temporally associated with use of a medicinal product, from the first dose through the last dose. Among these events, those considered possibly associated with a medicinal product are defined as adverse drug reactions. This distinction matters: the endpoint is not described as a count of every unfavorable event, nor does the definition establish that an event was caused by the medicine.

Counting participants with at least one event is a way to describe how many people experienced an event, rather than how many event episodes occurred. It can be useful in safety surveillance because it gives a participant-level measure of occurrence. However, it does not convey the number of repeated events per participant, their duration, severity, or clinical consequences. Those details require separate outcome reporting.

The registry record does not report an observed value for this primary endpoint. The serious and other adverse-event figures in the safety summary are not identified as the primary adverse drug reaction count. They should not be substituted for it: a serious event is not automatically a suspected drug reaction, and an adverse drug reaction need not be serious. With no primary count or corresponding uncertainty estimate reported, the endpoint’s frequency and precision cannot be assessed from the record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the primary endpoint count?

In general, a participant-level event endpoint records whether each participant experienced the defined event during the specified observation period. The registry’s primary endpoint is the number of participants with adverse drug reactions, with a stated timeframe of 12 months. This differs from counting event episodes: a participant with repeated events may still contribute as a participant with an event rather than as several participants. The endpoint description does not specify how repeated events, event onset, or participants with incomplete follow-up were handled.

The definition also separates adverse events from adverse drug reactions. An adverse event is defined by its unfavorable and unintended nature and its temporal association with medicinal-product use. The subset considered possibly associated with a medicinal product is called an adverse drug reaction. This is a classification of association, not proof of causation. Judgments about possible association can depend on clinical review and available information; the registry does not describe the assessment process.

How should counts and uncertainty be understood?

In general, a count describes the observed number of participants meeting an endpoint definition in the population being summarized. A count alone does not show how precisely the underlying event frequency is known. A confidence interval expresses the uncertainty around an estimate under a statistical model: wider intervals indicate less precision, while narrower intervals indicate greater precision. No confidence interval for the primary endpoint is reported in the registry record, and no primary count is posted from which to assess its precision.

For safety outcomes, the denominator matters. A count among participants at risk is interpretable only with a clear account of who was considered at risk and whether all had comparable opportunity for observation. The adverse-event summary lists 1158 at risk, while enrollment is 1168. The registry does not explain this difference or provide outcome-specific follow-up information. It would therefore be inappropriate to replace the reported risk denominator with the enrollment figure or to infer a reason for the discrepancy.

Why does the comparison structure matter?

In general, a randomized controlled comparison can support a treatment contrast when allocation, follow-up and analysis are appropriately conducted. This record says the study is not a randomized comparison and does not state allocation. It lists pioglitazone/glimepiride without a control treatment. There is therefore no direct between-group estimate of treatment effect to interpret. A descriptive event count can characterize recorded experience, but it cannot separate treatment effects from participant characteristics, clinical decisions or other influences.

The record’s planning assumptions mention a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation and an effect expressed as a fraction of an observed effect. These are assumptions for a planning illustration, not evidence that the surveillance survey used those design features. They do not establish an actual comparator, the study’s original sample-size calculation, or observed statistical power. The registry does not report its sample-size justification.

What can be inferred from the safety summary?

The adverse-event summary records serious-event and other-event participant counts. In general, serious adverse events and adverse drug reactions are different classifications: seriousness concerns the event’s status under safety reporting criteria, whereas a suspected drug reaction concerns its possible association with a medicine. The registry’s 23 serious-event participants and 0 participants with other events do not, on their own, provide the primary adverse drug reaction count. Nor do the listed event terms establish causality or comparative risk.

Event summaries can be affected by how events are collected, coded, reviewed and followed. The record does not report those procedures in sufficient detail to assess their influence. It also does not describe whether event categories overlap, how repeated events were handled, or whether the displayed term percentages use a particular denominator. Accordingly, the term-level percentages should be quoted as registry descriptions rather than reinterpreted as newly calculated rates.

What is missing about the analysis?

The record lists seven outcome measures and reports no analyses. It does not report a statistical analysis plan, confidence intervals, hypothesis tests, covariate adjustment, interim analyses, missing-data methods, or sensitivity analyses. Without such information, there is no basis to attribute an unreported model or statistical procedure to the study. General statistical considerations can explain why these details matter, but they cannot fill in what the registry does not report.

Multiplicity is a general concern when many outcomes are examined: repeated testing can increase the chance of at least one apparently unusual result if tests are interpreted without regard to the number of comparisons. The listed outcomes include several measures of glycemic control, but no analysis plan or multiplicity procedure is described. Since numerical results and analyses are not reported, the record does not allow an assessment of whether multiplicity affected interpretation.

Finally, planning calculations are not findings about the completed survey. Sample-size illustrations based on an observed effect are hypothetical planning exercises, not observed power and not proof that a study was adequately powered. No numerical planning result is discussed in the registry record excerpt, and no inference about the design’s statistical power is warranted.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists seven outcome measures. Named secondary outcomes include HbA1c, fasting blood glucose and fasting insulin, with both change-from-baseline and level measures. The primary outcome is the number of participants with adverse drug reactions. The record reports no analyses and does not describe a prespecified hierarchy, multiplicity adjustment, or strategy for distinguishing confirmatory from exploratory outcomes. It is therefore not possible to determine how multiple outcomes were handled.

In general, testing or interpreting many outcomes can make isolated findings harder to distinguish from chance variation. A prespecified analysis plan and clear reporting of all planned outcomes help readers understand which findings were primary and how uncertainty was managed. Those details are not reported in the registry record. No numeric secondary outcome results are posted, so there are no estimates or tests from which to assess this issue directly.

The primary outcome population is described as all enrolled participants with available data. This wording indicates that the endpoint is not necessarily based on everyone enrolled, but the record does not state how many had available data for the primary outcome, why data might be unavailable, or how missing observations were handled. It does not report imputation, exclusion rules, sensitivity analyses, or participant flow by outcome. The safety summary separately lists 1158 participants at risk, while enrollment is 1168; the reason for the difference is not reported in the registry record.

Without outcome-specific denominators and missing-data methods, readers cannot assess whether missing observations could affect the results. The available record also does not define analysis populations beyond the primary endpoint wording. Any assumptions about intention-to-treat, per-protocol, or other analysis sets would be unsupported.

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: road traffic accident (0.2%), oesophageal carcinoma (0.1%), pancreatic carcinoma (0.1%), lung neoplasm malignant (0.1%), prostate cancer (0.1%).

The adverse-event summary lists 23 participants with serious adverse events among 1158 at risk and 0 participants with other adverse events. These are participant counts in the registry summary. They are not identified as the primary endpoint count of participants with adverse drug reactions. Seriousness and suspected association with a medicine are separate concepts, so the serious-event count should not be read as a count of reactions attributed to pioglitazone/glimepiride.

The listed serious-event terms are Road traffic accident at 0.2%, Oesophageal carcinoma at 0.1%, Pancreatic carcinoma at 0.1%, Lung neoplasm malignant at 0.1%, Prostate cancer at 0.1%, and Anaemia at 0.1%. These are reported term-level percentages. The registry does not give enough information to reinterpret them using a different denominator, infer event timing, or determine whether events were considered related to treatment. The terms describe reported events, not causal conclusions.

The record does not explain why the adverse-event summary’s at-risk count differs from the enrollment of 1168. It also does not provide details about event ascertainment, follow-up completeness, repeated events, or uncertainty intervals. The figures therefore offer a limited descriptive account of reported safety events, rather than a complete estimate of treatment-attributable risk or a comparison with another treatment.

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

The registry record establishes that a long-term surveillance survey was conducted to evaluate pioglitazone/glimepiride in routine clinical care for people with type 2 diabetes mellitus. It identifies the regimen, the primary safety endpoint and several secondary measures of glycemic control. It also records selected adverse-event counts and serious-event terms. Those details are useful for understanding the intended scope of the survey and the nature of the available safety summary.

However, the record does not report an observed value for the primary endpoint, the number of participants with adverse drug reactions over 12 months. It also does not report numeric results for HbA1c, fasting blood glucose or fasting insulin. The safety summary’s serious-event participants are not a substitute for the primary reaction count, and the listed event terms do not establish a causal relationship with treatment. The available information therefore does not quantify the primary outcome or the secondary efficacy outcomes.

The record describes no comparator and states that the survey is not a randomized comparison. Accordingly, it cannot support a direct claim about how the recorded outcomes differ from outcomes under another treatment. In general, outcomes observed in routine care may reflect the treatment as well as patient characteristics, treatment selection and other aspects of care. The registry does not provide details that would allow those influences to be separated.

Absence of an outcome result in the registry record should not be taken as evidence that the outcome was absent or unchanged. It means that the corresponding result is not reported in the registry record. Likewise, the presence of an event term is evidence that the term appears in the safety summary, not that the medicine caused the event.

The most defensible reading is therefore bounded: the record describes a completed surveillance survey and provides a limited adverse-event summary, but it does not provide enough outcome or analysis information to estimate the primary adverse drug reaction frequency, establish efficacy changes, or make comparative causal conclusions.

13. Why This Trial Matters Statistically

This record illustrates a central statistical distinction in clinical research: describing outcomes in a treated group is not the same as estimating what treatment caused. The study’s routine-care surveillance setting is relevant to safety description, but the registry identifies no comparator and states that the study is not randomized. Without a credible comparison, outcomes cannot be interpreted as a treatment contrast.

It also demonstrates why endpoint definitions and endpoint results must be distinguished. The primary endpoint is clearly named and defined as the number of participants with adverse drug reactions over 12 months, yet the registry does not report its observed count. Readers can understand the intended measure without being able to assess its frequency or precision. This is a reporting limitation, not a statistical result.

Safety summaries require careful separation of event categories. The adverse-event summary includes serious events and other events, while the primary endpoint concerns suspected drug reactions. Treating these categories as interchangeable would change the question being answered. The reported event terms and percentages also need to be interpreted descriptively; they do not, by themselves, identify causation.

Finally, the difference between enrollment and the at-risk count underscores the importance of clear denominators and analysis populations. The registry does not explain that difference or detail missing outcome data. Such information helps readers judge how completely outcomes were captured and what population an estimate represents. In this record, the study purpose and outcome definitions are more fully described than the numerical results and analytical methods, so conclusions must remain correspondingly limited.

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 important limitation of the posted record is that it does not report a numeric result for the primary endpoint. The number of participants with adverse drug reactions over 12 months is not available, and no uncertainty estimate is reported. The serious-event count cannot replace that endpoint because seriousness does not establish suspected drug association.

The adverse-event summary lists 1158 participants at risk, while enrollment is 1168. The registry record does not explain this difference. It also does not give outcome-specific denominators or enough information to determine how much follow-up participants contributed. The primary endpoint population is described as enrolled participants with available data, but the number with available data and the handling of unavailable data are not reported.

No comparator treatment is listed, and the registry states that this is not a randomized comparison; allocation is not stated. Therefore, the design information does not support a like-for-like treatment contrast. The record also does not report numeric outcomes for HbA1c, fasting blood glucose or fasting insulin, despite listing these as secondary outcomes.

Further limitations are the absence of reported analyses and statistical methods. The registry does not describe the analysis plan, missing-data procedures, interim analyses, multiplicity handling, or sample-size justification. It does not provide details about adverse-event collection, review of possible drug association, or management of repeated events. These gaps prevent assessment of how the observed safety summary was assembled and how precisely it reflects the underlying experience.

These limitations concern what can be concluded from the registry record; they do not establish that the unreported procedures were or were not performed. Where details are absent, they are not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Long-term Use of Sonias Combination Tablets in Patients With Type 2 Diabetes Mellitus — NCT02098733.
• Registry record last updated 2016-04-15; status last verified 2016-03.
• Results first posted 2016-04-15.
• 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

NCT02098733 is a completed special drug use surveillance survey of pioglitazone/glimepiride (Sonias) for long-term use in people with type 2 diabetes mellitus. The registry lists Takeda as sponsor, enrollment of 1168, a start of 2011-06 and primary completion of 2014-05. The regimen is 15 mg/1 mg or 30 mg/3 mg, orally once daily before or after breakfast.

The primary outcome is the number of participants with adverse drug reactions over 12 months, among enrolled participants with available data. An observed result for this endpoint is not reported in the registry record. The safety summary lists 23 participants with serious adverse events and 0 with other adverse events among 1158 at risk; the record does not explain the difference between that at-risk count and enrollment. HbA1c, fasting blood glucose and fasting insulin are listed as secondary outcomes, but numeric results and analyses are not reported. No comparator is listed, and the record states that this is not a randomized comparison.