This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Double-blind Ipragliflozin Add-on Study in Japanese Participants With Type 2 Diabetes Mellitus Who Have Inadequate Glycemic Control on Sitagliptin (MK-0431J-843) (NCT02577003), a randomized phase 3 study. The primary outcome is Change From Baseline in HbA1c at Week 24.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02577003 |
| Phase | Phase 3 |
| Condition | Type 2 Diabetes Mellitus |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 143 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Merck Sharp & Dohme LLC |
| Results posted | Yes |
2. Clinical Question
Population
Japanese participants with type 2 diabetes mellitus whose glycemic control was inadequate on sitagliptin, alongside diet and exercise therapy. The primary analysis population was narrower than everyone randomized: participants had to receive at least one dose of study medication and have an HbA1c measurement at baseline or after baseline.
Intervention
Ipragliflozin once daily for 24 weeks, added to sitagliptin, diet, and exercise.
Comparator
Placebo to ipragliflozin once daily for 24 weeks, with sitagliptin, diet, and exercise continued.
Primary question
Did adding ipragliflozin produce a greater reduction in HbA1c from baseline to Week 24 than adding placebo?
The registry record identifies the primary hypothesis as a between-group comparison of HbA1c change at Week 24. HbA1c is a measure of glycemic status, and the endpoint uses the change from the starting measurement rather than the Week 24 value alone. The posted result is an adjusted contrast from a constrained longitudinal data analysis model, not simply a comparison of two raw averages.
The trial also lists changes in fasting plasma glucose, 2-hour post-meal glucose, glucose total area under the curve after a meal, and body weight as secondary outcomes. These are distinct measures and do not replace the primary question. The record describes seven outcome measures overall, while the secondary-outcome summary names four. Other outcome details are not reported in the registry record.
The question is specific to the studied setting: Japanese participants with inadequate control on sitagliptin, diet, and exercise. The record does not establish how findings would apply to people outside that population or to other treatment settings.
3. Trial Design
This was a Phase III, multicenter, randomized, placebo-controlled, parallel-group, double-blind trial. The registry lists 143 participants as enrolled. The ipragliflozin group received once-daily ipragliflozin in addition to sitagliptin, diet, and exercise; the comparator group received matching placebo to ipragliflozin with the same background care. Treatment was described over 24 weeks.
Random allocation is intended to make treatment groups comparable on average at the point of assignment. A placebo comparator helps distinguish the contrast associated with assignment to ipragliflozin from changes that may occur with background treatment, study participation, or time. Double masking is intended to reduce the influence of treatment knowledge on participant experience and outcome assessment. The registry identifies the study as double-blind but does not specify who was masked or describe procedures for maintaining masking; those details are not reported in the registry record.
The posted primary outcome includes 73 participants in the ipragliflozin plus sitagliptin group and 70 in the placebo plus sitagliptin group. These are the group counts attached to the outcome estimate, not a statement that every randomized participant contributed every measurement. The primary analysis population is defined as randomized participants who received at least one dose of treatment-period study medication and had at least one HbA1c measurement, either at baseline or after baseline.
The registry gives a start date of 2015-11-09, a primary completion date of 2016-11-25, and a status of completed. The rationale for the planned enrollment, details of recruitment, and any protocol amendments are not reported in the registry record. The stated design supports a comparison between the two assigned groups; it does not, by itself, answer questions about other populations, durations, or treatment combinations.
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: Prior to randomization, all participants received placebo for 2 weeks.
Overall Study
| Stage | Ipragliflozin + Sitagliptin | Placebo + Sitagliptin |
|---|---|---|
| STARTED | 73 | 70 |
| COMPLETED | 71 | 65 |
| NOT COMPLETED | 2 | 5 |
Reasons for not completing this period:
| Reason | Ipragliflozin + Sitagliptin | Placebo + Sitagliptin |
|---|---|---|
| Adverse Event | 2 | 4 |
| Withdrawal by Subject | 0 | 1 |
Analysis population for the primary outcome, as defined in the registry: All randomized participants who received at least one dose of treatment period study medication and have at least one measurement of HbA1c (baseline or post-baseline).
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.
| Characteristic | Unit | Ipragliflozin + Sitagliptin | Placebo + Sitagliptin | Total |
|---|---|---|---|---|
| Participants analysed | 73 | 70 | 143 | |
| Age, Continuous | Years | 61.0 (SD 9.1) | 60.0 (SD 10.4) | 60.5 (SD 9.7) |
| Sex: Female, Male: Female | Participants | 19 | 17 | 36 |
| Sex: Female, Male: Male | Participants | 54 | 53 | 107 |
| HbA1c | Percent | 8.05 (SD 0.83) | 7.99 (SD 0.62) | 8.02 (SD 0.73) |
| Estimated Glomerular Filtration Rate (eGFR) | mL/min/1.73m^2 | 82.0 (SD 13.5) | 83.4 (SD 16.7) | 82.7 (SD 15.1) |
| Fasting Plasma Glucose (FPG) | mg/dL | 158.0 (SD 33.2) | 163.0 (SD 26.2) | 160.5 (SD 30.0) |
| 2-hour Post-Meal Glucose (2-hr PMG) | mg/dL | 225.3 (SD 59.9) | 231.5 (SD 48.9) | 228.3 (SD 54.7) |
| Glucose Total Area Under the Plasma Concentration Curve from Hour 0 to Hour 2 (AUC0-2hr) after Meal | mg･hr/dL | 429.4 (SD 86.3) | 443.4 (SD 67.0) | 436.2 (SD 77.5) |
| Body Weight | kg | 69.8 (SD 11.7) | 70.1 (SD 11.1) | 69.9 (SD 11.4) |
| Prior use of other antihyperglycemic agents (AHAs): Yes | Participants | 31 | 30 | 61 |
| Prior use of other antihyperglycemic agents (AHAs): No | Participants | 42 | 40 | 82 |
6. Primary Endpoint
The registry lists the primary outcome as Change From Baseline in HbA1c at Week 24, measured over Baseline and Week 24.
| Element | Registry entry |
|---|---|
| Result type | Least Squares Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | Percent |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is change from baseline in HbA1c at Week 24, measured as percent. The registry defines the change as the Week 24 HbA1c percent minus the Week 0 HbA1c percent. A negative change therefore represents a lower Week 24 value than baseline on this measure. The posted arm estimates are least-squares means, with 95% confidence intervals.
HbA1c is used as a measure of glycemic status over time. Assessing change from baseline focuses the comparison on how the measure moved during the trial rather than only on the final measurement. The trial’s stated primary hypothesis specifically concerns whether adding ipragliflozin produces a greater reduction than adding placebo, so the key statistical result is the between-group contrast in change, not either group’s estimate considered alone.
The endpoint is analyzed with a constrained longitudinal data analysis model. The model includes treatment, time, prior use of AHAs, interactions involving treatment, time and prior use of AHAs, and baseline eGFR. It constrains the mean baseline HbA1c to be the same for both groups. These features mean the posted estimates are model-based, adjusted results. They should not be read as unadjusted group averages or as individual participants’ changes.
The registry identifies HbA1c as the primary outcome and gives the timeframe as baseline and Week 24. Details about laboratory procedures, measurement precision, and any additional endpoint-handling rules are not reported in the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted least-squares mean change was -0.84 in the ipragliflozin plus sitagliptin group, with a 95% confidence interval from -0.99 to -0.69. In the placebo plus sitagliptin group, the estimate was -0.07, with a 95% confidence interval from -0.22 to 0.09. Both are changes on the HbA1c percent scale. The negative estimate in the ipragliflozin group indicates a decrease from baseline in the model-based estimate; the placebo-group interval includes values on either side of zero.
The sponsor-reported difference in least-squares means was -0.77, with a 95% confidence interval from -0.98 to -0.57, and p < 0.001. The registry labels this as a superiority analysis. The contrast is the central result for the stated primary question: it compares the estimated change in the ipragliflozin group with the estimated change in the placebo group, rather than treating the two arm estimates as separate tests.
A confidence interval describes the uncertainty around an estimate under the model and its assumptions. Here, the reported interval for the between-group difference lies below zero, which is consistent with a greater decrease in the ipragliflozin group on the scale used for change. The p-value is evidence against the model’s null hypothesis of no difference, under the analysis used. Neither statistic establishes that the result is free from bias or that the model assumptions are correct.
The group estimates and their intervals describe estimated averages, not the range of changes among individual participants. The registry does not provide individual-level distributions in this summary, so it is not possible from these figures to characterize how consistently individual responses varied. The interval also should not be interpreted as a guarantee that the true effect lies within its bounds; it expresses statistical uncertainty under the specified analysis.
The outcome comparison is based on the defined primary analysis population, not necessarily every person randomized. The record does not give enough detail to assess how all missing measurements were handled, whether adherence differed, or whether the planned analysis was changed. Those uncertainties matter when interpreting any model-based comparison.
These results address the trial’s defined endpoint and timeframe in its study population. They do not alone establish effects on other outcomes, long-term outcomes, or groups that were not represented. The registry also lists secondary glucose and body-weight outcomes, but those results are not included in the primary result. No claim about them follows from the HbA1c contrast.
9. Statistical Methods Explained
The primary result comes from a constrained longitudinal data analysis, or cLDA, model. In general, longitudinal methods analyze measurements collected at multiple times while accounting for the fact that observations from the same participant are related. The registry says the model included treatment and time, along with prior use of AHAs and baseline eGFR. It also included interactions among treatment, time, and prior use of AHAs. The record does not provide further implementation details, such as the covariance structure or the software used.
What does the baseline constraint mean?
The model imposed a constraint that mean baseline HbA1c was the same in both treatment groups. This is a model specification for the baseline means; it does not mean that every participant had the same starting value, or that the observed baseline measurements were identical. In general, a longitudinal model can estimate treatment-related patterns across visits while accounting for baseline structure. The registry does not explain the rationale for this particular constraint beyond stating that it was part of the model.
Terms for treatment and time represent the assigned group and the measurement occasion. Interactions allow the pattern over time to differ by treatment and by prior use of AHAs. Including baseline eGFR makes it one of the model terms, but the registry does not report its coefficient or explain the contribution of that term to the result. The adjusted difference is therefore the model’s estimate under the listed specification, not a simple subtraction of posted arm means.
How should the least-squares means be read?
Least-squares means are model-based group estimates. In general, they summarize the expected outcome for each treatment group after accounting for the terms in the model. They are not necessarily the same as ordinary arithmetic averages of observed measurements. The registry reports the ipragliflozin-group estimate as -0.84 and the placebo-group estimate as -0.07, each with a 95% confidence interval. The separate reported contrast is the difference in least-squares means, -0.77, with its own interval. The contrast is the relevant estimate for the between-group question.
What does the confidence interval communicate?
A confidence interval represents uncertainty in an estimate under the statistical model. In general, a narrower interval indicates greater precision than a wider interval for the same quantity and assumptions. The interval does not show the spread of individual participant outcomes, and it is not a probability statement about the location of a fixed true value. The posted 95% interval for the contrast extends from -0.98 to -0.57. Its position relative to zero is consistent with a negative between-group contrast on the defined scale.
How should the p-value be interpreted?
The reported p-value is p < 0.001 for the superiority analysis. In general, a p-value describes how incompatible the observed data are with a specified null hypothesis, assuming the model and its assumptions. It does not measure the size or clinical importance of an effect, and it is not the probability that the null hypothesis is true. The registry record does not report further details about the testing procedure, such as whether it used a one-sided or two-sided test or how multiplicity was handled.
What does the analysis population include?
The registry defines the primary analysis population as randomized participants who received at least one dose of treatment-period study medication and had at least one HbA1c measurement at baseline or after baseline. This definition includes more than participants with a complete set of measurements, but it is not identical to a statement that every randomized participant was analyzed. In general, analysis-population rules affect which observations contribute to the estimate and how the results should be understood. The record does not report the number excluded from this population or the reasons for any exclusions.
Why can over-dispersion matter?
Over-dispersion is a general issue for count outcomes when observed variability is greater than a model assumes. It is not a reported feature of this primary endpoint, which is continuous HbA1c change. For continuous outcomes, the choice of model and its assumptions about repeated measurements and residual variation still matters. The registry does not provide the model diagnostics needed to assess those assumptions. The statistical result should therefore be read as an estimate under the posted cLDA model, rather than as a model-free summary.
The trial also reports a superiority label and a p-value, but the registry does not describe the full statistical analysis plan. Details such as interim analyses, multiplicity procedures, missing-data assumptions, sensitivity analyses, and model diagnostics are not reported in the registry record. The stated model terms and outcome population are useful for understanding the primary estimate, but they do not supply all information needed for an independent reproduction or full audit of the analysis.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Constrained longitudinal data analysis | Difference in least squares means -0.77 | -0.98 to -0.57 | < 0.001 |
The registry reports the sponsor’s own primary analysis as a constrained longitudinal data analysis. It gives the difference in least-squares means as -0.77, with a 95% confidence interval from -0.98 to -0.57, and p < 0.001. This is the posted comparison for the trial’s primary hypothesis. Merck Sharp & Dohme LLC is identified as the sponsor.
A separate simple, unadjusted re-calculation of the comparison is not reported in the registry record, so there is no independent re-calculated contrast to set against the sponsor’s estimate. The two arm-level estimates and their intervals are also model-based least-squares means; they should not be used to create an unadjusted result. The sponsor’s contrast should be read with its stated model and population definition.
The discrepancy indicator is false, but that does not add an independent validation of the model or its assumptions. It indicates no flagged discrepancy in this record. The sponsor analysis remains the relevant posted result, while a reader’s assessment of its precision and scope depends on the model description, analysis population, and limitations of registry reporting.
11. Multiplicity, Missing Data and Analysis Populations
The registry identifies one primary outcome: change from baseline in HbA1c at Week 24. It also lists secondary outcomes for fasting plasma glucose, 2-hour post-meal glucose, glucose total area under the curve after a meal, and body weight. The record reports seven outcome measures overall, but the summary names four secondary outcomes. The other outcome details are not reported in the registry record.
Multiplicity arises when several outcomes or statistical tests are considered: the chance of at least one apparently notable finding can increase across a collection of tests. In general, a prespecified testing hierarchy or adjustment can address this issue. The registry record does not report whether a multiplicity adjustment, gatekeeping strategy, or other procedure was used. The primary p-value should therefore be understood as the reported result for the primary comparison; the record does not establish how secondary analyses were controlled or prioritized.
The cLDA description specifies model terms and the population definition but does not explain how missing measurements were handled. The defined population required at least one HbA1c measurement at baseline or after baseline, so it does not require a complete series at every measurement time. The record does not report the number of missing values, reasons for missingness, imputation procedures, or sensitivity analyses addressing missing-data assumptions. Those details matter because missingness can affect model-based estimates if the analysis assumptions do not adequately reflect the data.
The record also does not report interim analyses or say whether the analysis plan changed. Randomization is stated, but the summary does not provide a detailed account of all randomized participants, treatment adherence, withdrawals, or analysis exclusions. It is therefore not possible from this record alone to determine how these aspects influenced the result. These omissions do not negate the posted estimate, but they limit how fully its robustness can be evaluated.
12. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Ipragliflozin + Sitagliptin | 2 | 73 | 2.7% |
| Placebo + Sitagliptin | 4 | 70 | 5.7% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Ipragliflozin + Sitagliptin | 12 of 73 |
| Placebo + Sitagliptin | 15 of 70 |
Most frequent serious adverse event terms
| Event term | Organ system | Ipragliflozin + Sitagliptin | Placebo + Sitagliptin |
|---|---|---|---|
| Angina pectoris | Cardiac disorders | 0 of 73 | 1 of 70 |
| Cardiomyopathy | Cardiac disorders | 0 of 73 | 1 of 70 |
| Cerebral infarction | Nervous system disorders | 1 of 73 | 0 of 70 |
| Ureterolithiasis | Renal and urinary disorders | 0 of 73 | 1 of 70 |
| Pulmonary mass | Respiratory, thoracic and mediastinal disorders | 1 of 73 | 0 of 70 |
| Sleep apnoea syndrome | Respiratory, thoracic and mediastinal disorders | 0 of 73 | 1 of 70 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Ipragliflozin + Sitagliptin | Placebo + Sitagliptin |
|---|---|---|---|
| Nasopharyngitis | Infections and infestations | 10 of 73 | 10 of 70 |
| Constipation | Gastrointestinal disorders | 2 of 73 | 4 of 70 |
| Contusion | Injury, poisoning and procedural complications | 0 of 73 | 4 of 70 |
Serious events with the highest arm-level rate: angina pectoris (1.4%), cardiomyopathy (1.4%), ureterolithiasis (1.4%), sleep apnoea syndrome (1.4%), cerebral infarction (1.4%).
The registry reports serious adverse events for the two groups, with participants at risk and participants affected. In the ipragliflozin plus sitagliptin group, 2 of 73 participants were reported as affected by a serious adverse event. In the placebo plus sitagliptin group, 4 of 70 were reported as affected. These are descriptive counts in the posted safety summary; they are not an adjusted comparison or a formal test of a difference.
The record lists serious-event terms including angina pectoris, cardiomyopathy, ureterolithiasis, sleep apnoea syndrome, cerebral infarction, and pulmonary mass, each shown as 1.4%. The summary does not provide enough context to treat these terms as a complete profile of all possible events or to attribute any event to a particular treatment. The listed percentage should be read as posted, without attempting to infer additional counts from it.
Other adverse events are reported for 12 of 73 participants in the ipragliflozin group and 15 of 70 in the placebo group. The registry shows no death value for either group; deaths are not reported in the registry record. Event definitions, severity grading beyond the serious-event distinction, exposure-adjusted comparisons, and detailed event narratives are not reported. These figures describe what appears in the registry summary and do not, by themselves, establish a safety conclusion.
13. What the Result Does, and Does Not, Mean
The posted data show a model-estimated reduction in HbA1c from baseline in the ipragliflozin plus sitagliptin group and a smaller estimated change in the placebo plus sitagliptin group over the 24-week treatment period. The reported difference in least-squares means is -0.77, with a 95% confidence interval from -0.98 to -0.57. The registry reports p < 0.001 for the superiority analysis. These numbers describe the primary endpoint in the trial’s defined analysis population.
The result does not establish the size of change for every participant. Least-squares means summarize group-level estimates under a model, while individuals may have had different measurements and changes. The registry summary does not show the individual-level distribution needed to describe that variation. The confidence interval quantifies uncertainty around the estimated contrast under the model; it does not remove uncertainty about the model, missing data, or the conduct of the trial.
The result also does not answer every question about the treatment combination. The primary endpoint is HbA1c change at Week 24, not a long-term outcome or a direct measure of every aspect of health. The listed secondary outcomes are separate questions, and their results are not included in the primary comparison. The registry record does not report the full statistical plan, details of missing-data handling, or all sensitivity analyses.
Finally, the study population and treatment context matter. The record concerns Japanese participants with type 2 diabetes and inadequate control on sitagliptin, diet, and exercise. It does not establish that the same estimate applies in different populations, with different background treatments, or over a different timeframe. The statistical result is evidence about the specific randomized comparison and endpoint reported, not a universal estimate for all settings.
14. Why This Trial Matters Statistically
This trial is statistically informative because it reports a randomized comparison against placebo while both groups continued sitagliptin, diet, and exercise. That design focuses the primary contrast on adding ipragliflozin rather than comparing entirely different background treatment settings. Randomization and double masking can reduce some sources of systematic difference, although the registry summary does not provide enough detail to assess every aspect of implementation.
The record also makes the form of the primary analysis visible. It reports least-squares means, a between-group contrast, confidence intervals, a p-value, and the cLDA model terms. This supports a more careful reading than looking only at whether either group changed from baseline. The between-group estimate is the relevant quantity for the stated superiority question, and its interval conveys uncertainty as well as direction.
At the same time, the trial illustrates why statistical significance is only part of interpretation. A small p-value does not describe clinical importance, the variation among individuals, or the reliability of results under alternative assumptions. Details of missing-data handling, multiplicity, interim analyses, and sensitivity checks are not reported in the registry record. Those omissions limit independent assessment, even when the primary estimate and interval are clearly posted.
The outcome set also shows the distinction between a designated primary endpoint and other outcomes. HbA1c change is the stated primary question; glucose and body-weight measures are listed as secondary outcomes. Without the complete analysis plan and secondary results, readers should not infer that the primary result answers those other questions. Overall, the record is useful for understanding a specific model-based contrast, while also demonstrating how registry reporting defines the boundaries of statistical interpretation.
15. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
Several trial-specific details needed for a fuller appraisal are not reported in the registry record. The record identifies randomization, double masking, and a parallel-group design, but does not describe the randomization procedure, allocation concealment, who was masked, or how masking was maintained. These details would help assess the possibility of selection or measurement bias.
The primary model is described by its terms and baseline constraint, but technical details such as the covariance structure, estimation procedure, model diagnostics, and sensitivity analyses are not reported. The posted least-squares means and contrast can be interpreted as results under the stated model, but the record does not allow an independent check of its assumptions or full reproduction of the analysis.
The population definition requires at least one HbA1c measurement at baseline or after baseline, but the record does not state how many randomized participants did not enter this population, how many measurements were missing, or why. The method for handling missing data and its assumptions are not reported. Likewise, details of adherence, withdrawals, and treatment exposure are not provided in this summary.
The registry lists four secondary outcomes and seven outcome measures overall, but does not include all outcome details. Multiplicity procedures, interim analyses, and any changes to the statistical plan are not reported in the registry record. Safety reporting is also limited to the summary counts and listed terms; death values are absent. These gaps mean that the posted primary result can be described, but not exhaustively evaluated from the registry summary alone.
Finally, the findings concern the stated population, background therapy, and 24-week endpoint. The registry does not support extending the estimate beyond those boundaries without additional evidence.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Double-blind Ipragliflozin Add-on Study in Japanese Participants With Type 2 Diabetes Mellitus Who Have Inadequate Glycemic Control on Sitagliptin (MK-0431J-843) — NCT02577003.
- Registry record last updated 2018-09-04; status last verified 2018-08.
- Results first posted 2018-03-07.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
19. Record Summary
This completed Phase III study examined once-daily ipragliflozin added to sitagliptin, diet, and exercise, compared with placebo added to the same background care, in Japanese participants with type 2 diabetes and inadequate glycemic control. The registry lists 143 enrolled participants and a 24-week treatment period.
The primary endpoint was change from baseline in HbA1c at Week 24. The sponsor-reported cLDA estimate for the difference in least-squares means was -0.77, with a 95% confidence interval from -0.98 to -0.57 and p < 0.001. The analysis population required at least one dose of treatment-period medication and an HbA1c measurement at baseline or after baseline.
The registry also lists secondary glucose and body-weight outcomes and reports safety counts. Details concerning multiplicity, missing-data handling, and other statistical-plan elements are not reported in the registry record. The result should be interpreted as a model-based comparison for this trial population, treatment context, and timeframe.