Efficacy and Safety of Oral Semaglutide Versus Placebo in Subjects With Type 2 Diabetes Mellitus Treated With Diet and Exercise Only: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of Oral Semaglutide Versus Placebo in Subjects With Type 2 Diabetes Mellitus Treated With Diet and Exercise Only (NCT02906930), a randomized phase 3 study. The primary outcome is Change in HbA1c.

2. Clinical Question

3. Trial Design

This was a phase 3, randomized, double-masked, parallel-group trial. It compared oral semaglutide with placebo in subjects with type 2 diabetes mellitus treated with diet and exercise only. The registry describes the trial as global. Enrollment was 703, and the record marks the study as completed.

The registry lists three oral semaglutide groups, at 3 mg, 7 mg, and 14 mg, plus a placebo group. The arm-description entries contain no further details. The adverse-event record lists 175 participants at risk for the serious-event summary in each semaglutide dose group and 178 in the placebo group. These safety denominators should not be treated as the number analyzed for the primary endpoint: the registry says the primary endpoint's analyzed number is the number of participants with available data.

The trial started on 2016-09-20, and primary completion was recorded as 2017-10-30. The primary endpoint description specifies change from baseline at week 0 to week 26. It distinguishes an in-trial observation period, which includes time after rescue medication or premature discontinuation of trial product, from an on-treatment period without rescue medication, which excludes time after rescue medication began.

The registry does not report the randomization procedure, allocation concealment, or the details of how masking was maintained. It also does not report the planned sample-size justification, a full statistical analysis plan, or interim-analysis procedures. These omissions mean the design can be described from the recorded features, but further procedural details cannot be established from the registry record.

4. Participant Flow and Accounting

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

Before assignment: Data presented in "participant flow" is based on the in-trial observation period, which was the time period from when a participant was randomised until the final scheduled visit, including any period after initiation of rescue medication and/or premature discontinuation of trial product.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Overall number of participants analyzed = full analysis set (FAS) which comprised all randomised participants. Number Analyzed = number of 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.

6. Primary Endpoint

The registry lists the primary outcome as Change in HbA1c, measured over Week 0, week 26.

HbA1c, or glycosylated haemoglobin, is a continuous measure. The primary endpoint was the change in HbA1c from baseline at week 0 to week 26, expressed as a percentage of HbA1c. The registry specifies that the endpoint is summarized as a mean with standard deviation. It also states that the in-trial observation period was the basis for evaluation and that an additional analysis was made for the on-treatment period without rescue medication.

In general, HbA1c is used in diabetes trials because it provides a measure related to blood glucose over time, rather than a reading from one moment. A change-from-baseline endpoint asks how the measured value at follow-up differs from a participant's starting value. Comparing average changes between randomized groups can describe the contrast associated with assignment to the study groups, subject to the analysis methods and data available.

The endpoint definition alone does not show the size or precision of any group contrast. The registry record does not give a single reduced primary comparison, so the mean changes, their uncertainty, and the result of a test for the main comparison are not reported in the registry record. The endpoint's description should therefore not be taken as evidence about the direction or magnitude of a treatment effect.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary endpoint as a mean of a continuous measure: change in HbA1c from week 0 to week 26. It specifies standard deviation as the measure of dispersion. The record does not report the numerical group means, a between-group estimate, or a confidence interval for the primary comparison. The methods below explain general statistical concepts relevant to this type of endpoint; they are not claims about procedures used in this trial unless explicitly identified as registry entries.

What does a mean change describe?

For a continuous endpoint, a mean change summarizes the average movement from baseline among participants with data included in an analysis. A comparison of mean changes between randomized groups can express the difference in average change. It is important to distinguish each group's change from the contrast between groups: a change within one group does not, on its own, estimate the randomized comparison.

The standard deviation describes variation among individual measurements around the group mean. It is not the same as the standard error or a confidence interval. The registry states that the primary endpoint is reported as a mean with standard deviation, but does not give those values. It also does not report the exact model or calculation used to estimate the treatment contrast.

Why do observation periods matter?

The record describes an in-trial observation period and an on-treatment observation period without rescue medication. The in-trial period runs from randomization until the final scheduled visit and includes periods after rescue medication or premature discontinuation of trial product. The on-treatment period without rescue medication includes time while a participant was on trial product and excludes time after rescue medication began.

In general, these definitions answer different questions about follow-up. An analysis that includes post-rescue observations retains data collected after rescue medication begins; an analysis that excludes such observations focuses on a more restricted treatment period. Which estimand or analysis is most relevant depends on the scientific question and prespecified analysis plan. The registry provides the period definitions but does not report the numerical results for either analysis or enough detail to establish how missing or post-discontinuation observations were handled.

What does a confidence interval contribute?

In general, a confidence interval communicates uncertainty around an estimated effect. It helps readers assess which effect sizes are compatible with the data under the analysis assumptions. An estimate without an interval gives less information about precision; an interval does not guarantee that the true value lies within its limits. No confidence interval for the primary comparison is reported in the registry record.

Likewise, a p-value summarizes how unusual data at least as incompatible with a null hypothesis would be if the null hypothesis and model assumptions held. It does not measure the probability that a treatment effect is real, and it does not quantify clinical importance. The registry record does not report a p-value for a single reduced primary comparison.

What do the planning assumptions mean?

The record describes a sample-size planning illustration based on the observed effect, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It describes planning for a stated fraction of the observed effect. Such an illustration is a hypothetical design calculation: it asks what sample size would be needed under the stated assumptions to detect a chosen effect size with a specified power. It is not observed power and does not establish the trial's original sample-size justification.

The record also distinguishes participant counts per arm from total event counts for time-to-event settings. Because the primary outcome is continuous, a participant-per-arm planning interpretation would be the relevant general form if such an illustration were reported. The record does not provide the resulting planning values in the registry record recorded for this analysis, so no required sample size can be stated. Nor does the planning description establish that the original trial used these assumptions.

Why can variability affect planning?

In general, greater variability in a continuous outcome makes a given average contrast harder to estimate precisely. Planning calculations therefore depend not only on the target effect, but also on assumptions about outcome variability, the significance threshold, allocation, and the desired probability of detecting that effect. Departures from assumptions can make a planning illustration less representative of a future study. The registry record does not report the variance assumptions underlying the original trial design.

These general concepts clarify what the recorded endpoint and planning description mean, but they do not fill in missing trial results. The trial's observed primary estimate, uncertainty interval, test result, and full statistical model are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates multiple endpoints, time points, dose groups, or statistical comparisons. In general, making many inferences can increase the chance of at least one apparently notable result when each is judged separately. Prespecified procedures can address this, for example by defining a hierarchy or adjusting the decision threshold, but the appropriate approach depends on the trial's analysis plan.

The registry lists 48 outcome measures and 18 analyses. It also lists secondary measures including body weight, fasting plasma glucose, seven-point self-monitored plasma glucose profiles, postprandial glucose increments, fasting insulin, pro-insulin and glucagon ratios, and HOMA-IR. These entries indicate a broad set of recorded outcomes; they do not establish which comparisons were confirmatory, how they were ordered, or whether multiplicity adjustments were used. Those procedures are not reported in the registry record.

For the primary endpoint, the registry states that the outcome has several categories or time points and is not reduced to one comparison. The record does not provide a single primary contrast that would permit a direct assessment of its estimate, uncertainty, or statistical significance. It also does not report whether the in-trial and on-treatment analyses were treated as co-primary, supportive, or otherwise related in the testing strategy.

The primary analysis population is described as the full analysis set, comprising all randomized participants. The number analyzed is defined as the number with available data. This definition does not tell readers how many participants had endpoint data, how missing observations were handled, or whether sensitivity analyses were conducted. These missing-data procedures are not reported in the registry record. In general, missing-data assumptions can influence continuous-outcome estimates, especially when the probability of missingness is related to outcomes or group assignment. Without the analysis details and data completeness information, the robustness of the primary result cannot be assessed from the registry record.

11. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: abdominal pain (0.6%), acute respiratory failure (0.6%), bacterial pyelonephritis (0.6%), cholecystitis acute (0.6%), cholelithiasis (0.6%).

The serious adverse-event counts are reported with the number at risk in each group: 5 of 175 for oral semaglutide 3 mg, 3 of 175 for oral semaglutide 7 mg, 2 of 175 for oral semaglutide 14 mg, and 8 of 178 for placebo. The registry also reports other adverse events affecting 48, 37, and 45 participants in the respective semaglutide groups, and 26 in the placebo group. These categories are presented separately in the registry record.

These counts are descriptive. In general, a count does not show the timing, duration, severity distribution, or causal relationship of events to study treatment. A group comparison also needs attention to follow-up time, event definitions, and uncertainty. The registry record does not report adjusted safety comparisons, confidence intervals, or statistical tests for these counts. The listed at-risk denominators are safety figures and should not be assumed to equal the number with primary endpoint data.

The record lists one death in the 14 mg group and none in the other listed groups. It lists abdominal pain, acute respiratory failure, bacterial pyelonephritis, acute cholecystitis, cholelithiasis, and visual hallucination among serious-event terms, each at 0.6%. Group attribution for these terms is not reported in the registry record. The counts and terms describe reported entries; they do not by themselves establish whether an event was caused by treatment or indicate a comparative safety conclusion.

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

The registry record documents a randomized, double-masked, parallel-group phase 3 comparison of oral semaglutide and placebo in subjects with type 2 diabetes treated with diet and exercise only. It identifies change in HbA1c from week 0 to week 26 as the primary endpoint and describes two observation periods. It also records safety-event counts for the three semaglutide dose groups and placebo.

However, the primary outcome is not reduced to one comparison in the registry record. A reader cannot determine from the recorded information the numerical difference in mean HbA1c change between any semaglutide dose and placebo, how precise that difference is, or what significance test result was obtained. Those efficacy results are not reported in the registry record. The endpoint description and study design do not establish the size, direction, or practical importance of an effect.

In general, statistical evidence should be interpreted using an effect estimate, its uncertainty, and the analysis population and methods. A result may be estimated imprecisely even when the point estimate appears notable, or precisely without being important in practice. Those general principles cannot be applied to a specific primary treatment contrast because the relevant estimate and interval are not reported.

The adverse-event counts likewise do not establish causality or comparative safety. They do not include, in the record, adjusted analyses or uncertainty intervals. The listed serious-event terms have no reported group attribution, and the single death listed for the 14 mg group does not by itself identify a cause or a treatment relationship.

Accordingly, the recorded information supports a description of the trial's question, design, endpoint, analysis-population definition, and selected safety summaries. It does not support a conclusion about whether oral semaglutide did or did not change HbA1c relative to placebo, nor a conclusion about overall safety or clinical benefit. Such conclusions would require outcome estimates and fuller analysis details that are not reported in the registry record.

13. Why This Trial Matters Statistically

This trial is statistically notable in the way its registry record separates the endpoint definition from a reduced treatment comparison. A clear endpoint definition is necessary for interpreting a trial, but it is not enough to evaluate the estimated effect. Here, the endpoint is change in HbA1c from week 0 to week 26, summarized as a mean with standard deviation, while the primary outcome is not reduced to one comparison. That gap limits direct assessment of effect size and precision.

The record also distinguishes an in-trial observation period from an on-treatment period without rescue medication. In general, defining how post-rescue and post-discontinuation observations are treated is important because the analysis period shapes the question being answered. The registry gives the period definitions, but does not report the numerical results or enough analytic detail to assess how the two approaches affect interpretation.

Multiple doses, a placebo comparator, numerous outcome measures, and multiple analyses create a setting where the planned hierarchy and handling of multiplicity matter. The registry lists 48 outcome measures and 18 analyses, but does not report the testing strategy or multiplicity procedures. Without those details, readers cannot determine which findings were intended as confirmatory or how the overall risk of chance findings was addressed.

The safety summaries also illustrate why counts need context. Counts are useful descriptions, but comparisons require attention to denominators, event definitions, follow-up, uncertainty, and potential attribution. The registry lists denominators for the serious-event summaries, but does not report adjusted comparisons or uncertainty intervals. The broader statistical lesson is that transparent estimates and analysis methods are essential for separating a trial's design and recorded observations from conclusions about effects.

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 main limitation of the registry information for interpreting efficacy is that the primary outcome is not reduced to one comparison. The record identifies the endpoint and observation periods, but does not report a single numerical estimate of mean change by group, a between-group contrast, an uncertainty interval, or a significance result. The magnitude and precision of the primary outcome therefore cannot be assessed from the registry record.

The record does not report the detailed statistical analysis plan, the model used for the primary endpoint, or how the two observation periods relate to the primary inference. Although the full analysis set is defined as all randomized participants, the number analyzed is the number with available data; the actual endpoint-data counts and missing-data methods are not reported in the registry record. The effect of missing observations therefore cannot be assessed.

The registry lists many outcome measures and analyses but does not report the multiplicity strategy or identify the confirmatory hierarchy. This limits interpretation of any individual secondary result. Arm descriptions are blank, and details of randomization procedures and masking implementation are not reported in the registry record.

Safety information is also limited to recorded counts and selected serious-event terms. Adjusted comparisons, confidence intervals, event timing, and group attribution for the listed terms are not reported in the registry record. These limitations do not imply that the trial lacked such procedures or data; they mean that the record does not describe them. The conclusions that can be drawn from the registry information are consequently narrower than conclusions based on full results and methods documentation.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Efficacy and Safety of Oral Semaglutide Versus Placebo in Subjects With Type 2 Diabetes Mellitus Treated With Diet and Exercise Only — NCT02906930.
• Registry record last updated 2022-07-20; status last verified 2022-07.
• Results first posted 2020-02-17.
• 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

The registry record describes a completed, global, randomized, double-masked, parallel-group phase 3 trial of oral semaglutide versus placebo in subjects with type 2 diabetes treated with diet and exercise only. Enrollment was 703. The listed semaglutide groups were 3 mg, 7 mg, and 14 mg, alongside placebo.

The primary endpoint was change in HbA1c from week 0 to week 26, expressed as a percentage and summarized as a mean with standard deviation. The registry describes both an in-trial observation period and an on-treatment period without rescue medication. The full analysis set comprised all randomized participants, and the number analyzed was defined as those with available data.

The primary outcome is not reduced to one comparison, and a single numerical efficacy estimate is not reported in the registry record. Safety summaries include serious adverse events, other adverse events, deaths, and selected serious-event terms. The record does not report the detailed statistical plan, missing-data methods, or multiplicity procedures.