Research Study to Investigate How Well Semaglutide Works Compared to Liraglutide in People Living With Overweight or Obesity: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Research Study to Investigate How Well Semaglutide Works Compared to Liraglutide in People Living With Overweight or Obesity (NCT04074161), a randomized phase 3 study. The primary outcome is Change From Baseline (Week 0) to Week 68 in Body Weight (%) (Semaglutide 2.4 mg Versus Liraglutide 3.0 mg).

2. Clinical Question

3. Trial Design

This was a phase 3, randomized, parallel-group study. The registry records quadruple masking, meaning that the study was described as masked across multiple roles; the specific roles and how masking was implemented are not reported in the registry record. Random assignment was used to allocate treatment. The study enrolled 338 participants and is recorded as completed.

The interventions included semaglutide, placebo for semaglutide, liraglutide, and placebo for liraglutide. The active treatments were given by subcutaneous injection, with semaglutide scheduled once weekly and liraglutide once daily. All arm descriptions include a reduced-calorie diet and increased physical activity. The registry describes clinic visits and phone calls with the study doctor, as well as counseling on food and activity.

The primary endpoint compared the two active treatments, semaglutide 2.4 mg and liraglutide 3.0 mg. The registry states that data were planned for those arms only for this endpoint. The posted primary outcome summaries include 117 participants per active-treatment group. Other posted outcomes include pooled placebo comparisons, with different numbers analyzed depending on the endpoint. These analysis counts should not be taken as the enrollment count or as evidence that every randomized participant had an available measurement.

The study began on 2019-09-11, and primary completion was recorded as 2021-03-27. The registry describes a study lasting about 1½ years. Details of the original allocation ratio, the full protocol, and the rationale for the planned sample size are not reported in 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: Total 338 adults with obesity (BMI greater than or equal to (\>=30.0) kilograms per square meter (kg/m\^2)) or overweight (BMI \>= 27.0 kg/m\^2) and at least one weight-related comorbidity were randomized in a 3:1:3:1 manner to receive treatment with either semaglutide subcutaneously (s.c.) 2.4 milligram (mg) once weekly or semaglutide placebo once weekly or liraglutide s.c. 3.0 mg once daily or liraglutide placebo once daily as an adjunct to a reduced-calorie diet and increased physical activity.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The full analysis set (FAS) included all randomized participants. Overall Number of Participants Analyzed = participants with available data for this outcome measure. It was planned to report data only for arms 'semaglutide 2.4 mg and liraglutide 3.0 mg' for this outcome measure.

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: The full analysis set (FAS) included all randomized participants.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline (Week 0) to Week 68 in Body Weight (%) (Semaglutide 2.4 mg Versus Liraglutide 3.0 mg), measured over Baseline (week 0), week 68.

The primary endpoint is change in body weight, expressed as a percentage, from baseline at week 0 to week 68. It is a continuous outcome: each participant’s weight change can take a range of values, and the reported group summaries are means with standard deviations. The endpoint is not the percentage of participants who reached a specific weight-loss threshold; those threshold outcomes are listed separately among the secondary outcomes.

A percentage change expresses the change relative to a participant’s baseline body weight, allowing the outcome to be described on a common scale rather than only in kilograms. A negative value represents a decrease from baseline. Comparing group means gives an estimate of the average difference between randomized groups over the specified period. The registry describes the primary result as applying to the in-trial period, defined as the uninterrupted interval from randomization to the last contact with the trial site.

This endpoint is useful for comparing the average weight trajectory between treatment groups, but it does not describe every participant’s response. The standard deviation gives information about variation among participants, while the treatment-difference confidence interval describes uncertainty around the group comparison. Neither group mean nor comparison alone indicates how many individuals reached a particular threshold; that is addressed by separate categorical outcomes.

7. Results Posted in the Registry

• Liraglutide 3.0 mg: mean -6.40 versus -16.40 on control; difference 10.00 (95% CI 7.64 to 12.36), standardized effect d = 1.09; z-test p = <0.001.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change From Baseline (Week 0) to Week 68 in Waist Circumference

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Semaglutide 2.4 mg: mean -13.60 versus -2.00 on control; difference -11.60 (95% CI -14.05 to -9.15), standardized effect d = -1.29; z-test p = <0.001.
  • Liraglutide 3.0 mg: mean -6.80 versus -2.00 on control; difference -4.80 (95% CI -7.04 to -2.56), standardized effect d = -0.60; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Semaglutide 2.4 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Liraglutide 3.0 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline (Week 0) to Week 68 in Body Weight (Kilograms (kg))

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Semaglutide 2.4 mg: mean -15.80 versus -1.40 on control; difference -14.40 (95% CI -17.22 to -11.58), standardized effect d = -1.45; z-test p = <0.001.
  • Liraglutide 3.0 mg: mean -6.80 versus -1.40 on control; difference -5.40 (95% CI -8.14 to -2.66), standardized effect d = -0.57; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Semaglutide 2.4 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Liraglutide 3.0 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline (Week 0) to Week 68 in Body Weight (%) (Semaglutide 2.4 mg Versus Pooled Placebo and Liraglutide 3.0 mg Versus Pooled Placebo)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Semaglutide 2.4 mg: mean -16.40 versus -1.60 on control; difference -14.80 (95% CI -17.49 to -12.11), standardized effect d = -1.51; z-test p = <0.001.
  • Liraglutide 3.0 mg: mean -6.40 versus -1.60 on control; difference -4.80 (95% CI -7.16 to -2.44), standardized effect d = -0.59; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Semaglutide 2.4 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Liraglutide 3.0 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary summaries cover waist circumference, body weight in kilograms, and percentage change in body weight for comparisons with pooled placebo. Across these outcomes, the recalculated comparisons show a consistent direction: the active-treatment group means are more negative than the pooled-placebo means. This pattern is broadly in keeping with the primary finding that the active-treatment groups had decreases in body weight, while addressing different comparisons and, in the case of waist circumference, a different measure.

These secondary results provide supporting context about weight and waist measurements, but they should not be treated as equivalent to the planned primary comparison between semaglutide and liraglutide. The placebo comparisons answer whether each active group differs from pooled placebo under the displayed calculations; the primary question directly compares the active treatments. The secondary analyses also use different analyzed group counts, and the registry outcomes include measures beyond the three continuous summaries.

Unadjusted secondary results carry less inferential weight when many outcomes or contrasts are considered. Testing multiple outcomes increases the chance that some apparently strong findings arise by chance, unless the analysis plan accounts for multiplicity. The registry record does not report a multiplicity adjustment for these recalculated comparisons. Their p-values should therefore be read as unadjusted evidence for the individual contrasts, not as confirmatory results across the full set of secondary outcomes.

The registry also lists threshold outcomes for weight reduction and changes in blood pressure. The secondary findings should be interpreted according to the endpoint and comparison actually reported; they do not independently establish a broad effect across every outcome listed. Detailed analysis plans and multiplicity procedures are not reported in the registry record.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Number of Participants Who From Baseline (Week 0) to Week 68 Achieved Body Weight Reduction Greater Than or Equal to (>=) 10% (Yes/no)
• Number of Participants Who From Baseline (Week 0) to Week 68 Achieved Body Weight Reduction >=15% (Yes/no)
• Number of Participants Who From Baseline (Week 0) to Week 68 Achieved Body Weight Reduction >=20% (Yes/no)
• Change From Baseline (Week 0) to Week 68 in Waist Circumference
• Change From Baseline (Week 0) to Week 68 in Body Weight (Kilograms (kg))
• Change From Baseline (Week 0) to Week 68 in Body Weight (%) (Semaglutide 2.4 mg Versus Pooled Placebo and Liraglutide 3.0 mg Versus Pooled Placebo)
• Change From Baseline (Week 0) to Week 68 in Systolic Blood Pressure
• Change From Baseline (Week 0) to Week 68 in Diastolic Blood Pressure

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous, and the registry reports mean change in percentage of body weight. A mean is the arithmetic average of the participant-level measurements included in a group summary. The standard deviation describes how spread out those measurements are around the mean. It reflects participant-to-participant variation; it is not the uncertainty in the estimated mean or treatment difference.

What does the adjusted analysis estimate?

The reported primary analysis is an analysis of covariance, or ANCOVA. In general, ANCOVA estimates a difference between treatment groups while accounting for covariates included in the model. Adjustment can improve precision or account for baseline differences, depending on the specified model. The covariates, model details, and missing-data handling for the reported ANCOVA are not reported in the registry record. Therefore, the adjusted result should be described using the reported estimate and interval without inferring which variables were included or how they affected the estimate.

The registry reports a treatment difference of -9.38, a 95% confidence interval from -11.97 to -6.80, and p < 0.001. The treatment difference is a contrast between groups, rather than a mean change within one group. The negative sign depends on the order used for the contrast. The primary group means themselves are negative because each represents a decrease from baseline. A contrast may have the opposite sign if it subtracts the group means in the opposite order.

What is the unadjusted recalculation?

The recalculated comparison uses the posted group means and standard deviations to estimate an unadjusted difference between the two active-treatment groups. For the primary outcome, the recalculated contrast is expressed as liraglutide minus semaglutide. Its estimate is 10.0, with a 95% confidence interval from 7.641 to 12.36 and p < 0.001. This calculation does not reproduce the ANCOVA model. It is a simple comparison based on group-level summaries and does not account for model covariates or any other features that may have been part of the sponsor’s analysis.

The recalculated confidence interval is an estimate of the range of values compatible with the unadjusted comparison under the method used. It is not a range containing a known proportion of individual participant outcomes. The unadjusted p-value assesses the compatibility of the data with a null difference under that calculation’s assumptions; it is distinct from the p-value reported for the ANCOVA. The primary interpretation should give priority to the registry’s reported analysis rather than treating the two calculations as interchangeable.

Why report a standard deviation?

The standard deviations for the primary group summaries are 10.5 for semaglutide and 7.7 for liraglutide. They describe the dispersion of participant-level percentage changes around the corresponding group means. A larger standard deviation indicates greater spread in the observed measurements, but does not by itself show whether the group mean is estimated precisely. Precision of a group comparison depends on both variability and the number of participants analyzed, as well as the statistical method used.

For context, the recalculated pooled standard deviation is 9.207. This is used in the accompanying standardized-effect and planning calculations, not as the ANCOVA model’s residual standard deviation. Standardized effect sizes express a mean contrast relative to a measure of spread. Such values can help describe scale, but do not change the original percentage-point scale or replace the treatment difference and its confidence interval.

What can be concluded from the reported significance test?

The registry labels the primary analysis as superiority and reports p < 0.001. In general, a p-value is calculated under a null hypothesis and the assumptions of the specified analysis. It measures how incompatible the observed data are with that null model; it is not the probability that either treatment is effective, nor a measure of clinical importance. The confidence interval is more informative about the range and direction of the estimated treatment difference. Statistical significance does not show how outcomes vary between individuals or resolve questions not measured by the endpoint.

Over-dispersion is a concern primarily for outcomes whose variability exceeds what a chosen probability model expects, particularly count outcomes. In general, ignoring over-dispersion can make standard errors too small and inference too optimistic. The primary endpoint is continuous and is summarized using means and standard deviations; the registry identifies ANCOVA as the reported method. No specific over-dispersion analysis is reported for this endpoint, and it would not be appropriate to infer one from these summaries.

The registry identifies the full analysis set as including all randomized participants, but also defines the number analyzed for this measure as participants with available data. That distinction means the analysis count for the endpoint can be smaller than the number randomized. The missing-data procedures and any sensitivity analyses are not reported in the registry record.

11. Confidence Intervals and Precision

The registry’s primary ANCOVA estimate is -9.38, with a 95% confidence interval from -11.97 to -6.80. A confidence interval describes uncertainty in the estimated treatment difference under the analysis and its assumptions. It is not a range of individual participant outcomes, and it does not guarantee that the true value lies inside the interval. The interval is separated from zero, consistent with the reported p < 0.001.

The unadjusted recalculation gives a mean difference of 10.0, with a 95% confidence interval from 7.641 to 12.36. This estimate uses the opposite contrast order from the negative sponsor estimate: liraglutide minus semaglutide rather than a contrast ordered semaglutide minus liraglutide. The recalculated interval is likewise separated from zero. Its endpoints and estimate should be read as an unadjusted summary-statistic comparison, not as an alternative presentation of the ANCOVA model.

The width of a confidence interval reflects precision as well as the scale of the outcome. In general, intervals tend to be narrower when estimates are more precise and wider when uncertainty is greater. Here, the two intervals come from different methods, so their width should not be used as a direct comparison of method quality without knowing the model details. The covariates and other implementation details of the ANCOVA are not reported in the registry record.

Precision of a group mean and precision of a between-group contrast are also distinct from the standard deviation among participants. The primary group standard deviations, 10.5 and 7.7, describe participant-level spread; the treatment-difference interval describes uncertainty about the estimated average contrast. Neither statistic conveys the full range of individual responses.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are illustrations based on the observed effect and the posted variability, not the original sample-size design for the trial. For these continuous outcomes, the entries are participants per arm, not total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Each scenario asks how many participants per arm would be needed to detect a stated fraction of the observed effect.

The 80% power column gives the illustrative per-arm number for a design with an 80% chance of detecting the assumed effect under the planning model, if that effect and the model assumptions were correct. The 90% power column gives the corresponding number for a 90% chance. These are design-planning probabilities, not guarantees about a future study or about the chance that a result is true. The values are not observed power.

Rows for 100% of the observed effect use the full recalculated effect as the planning target. Rows for 75% or 50% use smaller target effects. Smaller effects are harder to detect against the same variability, so they require more participants per arm in the illustration. The figures do not account for features of the original trial design that are not recorded, and they should not be read as evidence that the study was planned using these particular assumptions.

The trial’s own original sample-size justification and assumptions are not reported in the registry record. The posted planning calculations are therefore useful for illustrating how assumed effect size, variability, and desired power relate, but they do not explain the enrolled sample or establish what power the completed trial had for its actual analysis.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 10.000; pooled standard deviation σ = 9.207.
2. n = 2 × (1.96 + 0.842)² × 9.207² / 10.000² = 13.3, rounded up to 14 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- 10                      # observed difference in means
sigma <- 9.20706250657613                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(4074161)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 14 (80% power), 18 (90% power)
 75% of observed effect: n per arm = 24 (80% power), 32 (90% power)
 50% of observed effect: n per arm = 54 (80% power), 72 (90% power)
power.t.test: 14.3 
Simulated power at n = 14 per arm: 0.792 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The registry reports an ANCOVA treatment difference of -9.38, with a 95% confidence interval from -11.97 to -6.80 and p < 0.001. This is the sponsor-reported primary analysis and is the result to prioritize for interpretation of the planned primary endpoint. It is adjusted, whereas the recalculation is an unadjusted comparison derived from posted means and standard deviations.

The recalculated mean difference is 10.0, with a 95% confidence interval from 7.641 to 12.36 and p < 0.001. Its contrast is liraglutide minus semaglutide. The sponsor estimate has the opposite sign, consistent with expressing the contrast in the reverse order. The two estimates therefore point in a consistent direction when their contrast orders are taken into account, but their values are not identical because the methods differ.

The registry does not report the ANCOVA covariates or all model details. It is not possible to reconstruct the sponsor analysis from the posted group summaries alone. The recalculation is a transparent descriptive check of the group-level results, not a replacement for the reported adjusted analysis. There is no recorded discrepancy between the reported and recalculated results.

15. Multiplicity, Missing Data and Analysis Populations

The registry identifies one primary endpoint and reports one primary analysis, an ANCOVA comparison of semaglutide 2.4 mg and liraglutide 3.0 mg. It also lists many secondary outcomes, including threshold weight reductions, waist circumference, body weight in kilograms and percentage, and blood pressure measures. Considering multiple endpoints and comparisons raises a multiplicity question: without a prespecified approach to control the overall chance of false-positive findings, individual secondary p-values can overstate the strength of evidence across the collection of tests.

The registry record does not report whether the trial used a multiplicity adjustment, a hierarchical testing strategy, or another procedure to manage multiple comparisons. The unadjusted recalculated p-values should therefore be treated as exploratory context rather than as proof that every secondary contrast is independently confirmed. The primary result should remain distinct from the secondary comparisons, both because it addresses the planned active-treatment question and because its reported analysis is the sponsor’s ANCOVA.

The registry describes the full analysis set as all randomized participants, while defining the analyzed number for the primary endpoint as those with available data. The endpoint summaries include 117 participants in each active-treatment group. This wording indicates that the analysis depends on available measurements, but does not specify how missing observations were handled in the ANCOVA, whether any values were imputed, or whether sensitivity analyses were conducted. Those procedures are not reported in the registry record.

Randomization supports comparison of the assigned groups, but missing outcome data can complicate interpretation if availability differs in ways related to outcomes or treatment. The available record does not provide enough detail to assess those mechanisms or the robustness of the result to alternative missing-data assumptions. No interim analyses or stopping rules are reported in the registry record. These unknowns should be kept separate from the observed estimate and its confidence interval.

16. 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: atrial flutter (1.2%), bradyarrhythmia (1.2%), cardiac failure congestive (1.2%), cholelithiasis (1.2%), invasive ductal breast carcinoma (1.2%).

The registry reports serious adverse events for 10 of 126 participants at risk in the semaglutide 2.4 mg group, 14 of 127 in the liraglutide 3.0 mg group, and 6 of 85 in the pooled-placebo group. It reports zero deaths in each of those groups. These counts describe the recorded groups and their stated at-risk denominators; they are not the primary efficacy analysis and do not by themselves establish why an event occurred.

The registry lists atrial flutter, bradyarrhythmia, congestive cardiac failure, cholelithiasis, invasive ductal breast carcinoma, and lumbar radiculopathy among the top serious-event terms, each recorded as 1.2%. These terms are descriptive listings. They should not be interpreted as evidence of a causal relationship with a study treatment from these data alone.

Adverse-event counts are not a complete account of safety. The figures do not show event timing, duration, severity beyond the serious classification, or individual clinical context. Group sizes and follow-up can also affect observed counts. The registry record does not provide enough information to make a detailed comparative safety assessment.

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

The primary results describe an average difference in percentage change in body weight between randomized groups at week 68. The mean change was -16.4 in the semaglutide group and -6.4 in the liraglutide group. The reported ANCOVA estimate and its confidence interval indicate that the registry analysis estimated a difference between the groups, with the interval separated from zero. The unadjusted recalculation points in the same direction when its contrast order is taken into account.

This does not mean that every participant had a change equal to the group mean. The standard deviations indicate participant-level variation, and the averages do not reveal the distribution of individual outcomes. Nor does the primary endpoint state the proportion of participants reaching a specific threshold. Those questions require participant-level or separately reported categorical results.

The findings apply to the enrolled trial population, the assigned interventions, the stated lifestyle measures, and the in-trial period described in the registry. They do not establish what would happen in different populations, under different conditions, or over a different duration. The registry record does not provide detailed information about the ANCOVA covariates, missing-data procedures, or multiplicity approach, which limits assessment of those analytic choices.

Statistical evidence about an average contrast is not the same as a judgment about clinical importance. The result does not, by itself, answer questions about an individual’s likely outcome, preferences, suitability, or care. It also does not establish that one intervention is superior on every outcome: the primary endpoint is specifically percentage change in body weight, while secondary measures address other outcomes and comparisons.

18. Why This Trial Matters Statistically

This trial illustrates why the treatment contrast and the group changes need to be read separately. Both active-treatment group means show decreases from baseline, while the primary question concerns the difference between those mean changes. A within-group change alone cannot answer a randomized between-group question; the comparison does.

It also demonstrates the distinction between an adjusted reported analysis and a simple unadjusted recalculation. The ANCOVA result is the registry’s primary inferential analysis. Recalculation from posted summaries offers a useful check on direction and approximate magnitude, but it cannot reproduce unreported model details. Giving the sponsor analysis precedence for the primary endpoint avoids presenting the summary-statistic calculation as if it were the trial’s fitted model.

The confidence interval adds information beyond a significance test by expressing uncertainty around the estimated contrast. At the same time, its interpretation depends on the analysis and available data. The registry’s definition of the full analysis set and its available-data count highlights why the analysis population and missing-data procedures matter to a complete appraisal.

Finally, the trial has a primary active-treatment comparison alongside multiple secondary outcomes and placebo comparisons. Keeping those questions distinct helps prevent exploratory or unadjusted results from being treated as if they had the same status as the primary analysis. The planning illustrations also show how assumptions about effect size and variability inform prospective sample-size planning, without revealing the original design rationale.

19. 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 does not report the original sample-size justification or the assumptions used to plan enrollment. The sample-size illustrations based on observed effects are not a substitute for that missing design rationale. The record also does not state the covariates used in ANCOVA, the detailed model specification, or the missing-data methods and sensitivity analyses. These omissions limit independent assessment of how the adjusted estimate was obtained and how robust it is to analytic choices.

The primary endpoint is available for 117 participants in each active-treatment group, while the study enrollment is 338. These figures refer to different things: the former are endpoint analysis counts, and the latter is enrollment. The registry defines the analysis count as participants with available outcome data but does not give a complete account of missing observations or reasons for their absence.

The reported primary period is the in-trial period, ending at last contact with the trial site, and the endpoint is at week 68. The registry does not provide results beyond that period. Generalization to other populations or circumstances is not established by this comparison alone. The outcome is a group-average percentage change and does not convey the full participant-level distribution or the proportion crossing each possible threshold.

The registry record does not report multiplicity procedures for the secondary outcomes, nor interim analyses. Secondary comparisons should consequently be read with caution, especially when several measures and contrasts are considered. Finally, serious adverse-event counts are descriptive; the information recorded does not provide the event-level detail needed to assess causality or fully characterize comparative safety.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Research Study to Investigate How Well Semaglutide Works Compared to Liraglutide in People Living With Overweight or Obesity — NCT04074161.
• Registry record last updated 2023-05-19; status last verified 2023-05.
• Results first posted 2022-04-27.
• 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.

23. Record Summary

NCT04074161 is a completed randomized, quadruple-masked, parallel-group phase 3 study of semaglutide 2.4 mg and liraglutide 3.0 mg in people with overweight or obesity. The primary endpoint was percentage change in body weight from week 0 to week 68. The registry reports an ANCOVA treatment difference of -9.38, with a 95% confidence interval from -11.97 to -6.80 and p < 0.001.

The primary group means were -16.4 for semaglutide and -6.4 for liraglutide, with 117 analyzed participants in each group. An unadjusted recalculation gives a mean difference of 10.0 when expressed as liraglutide minus semaglutide, consistent in direction with the sponsor estimate under the opposite contrast order. Secondary summaries include weight and waist measures compared with pooled placebo. Interpret the adjusted primary result separately from unadjusted secondary comparisons, and note that model, missing-data, multiplicity, and original sample-size planning details are not reported in the registry record.