A Study of Tirzepatide (LY3298176) In Participants After A Lifestyle Weight Loss Program: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Tirzepatide (LY3298176) In Participants After A Lifestyle Weight Loss Program (NCT04657016), a randomized phase 3 study. The primary outcome is Percent Change From Baseline in Body Weight.

2. Clinical Question

SURMOUNT-3 asks whether body weight changed differently over 72 weeks among participants assigned to tirzepatide or placebo after an intensive lifestyle program. Its primary endpoint is percent change in body weight from baseline. Secondary outcomes include weight-loss thresholds, maintenance of weight lost during the lifestyle program, and changes in waist circumference, body mass index, and systolic blood pressure.

3. Trial Design

This was a phase 3, randomized, double-blind, parallel-group trial sponsored by Eli Lilly and Company. The registry record lists enrollment of 579 and describes participants without type 2 diabetes who had obesity or were overweight with weight-related comorbidities. The study evaluated tirzepatide after an intensive lifestyle modification program, with placebo as the comparator. The primary completion date was 2023-04-20, and the registry lists the study status as completed.

Participants assigned to tirzepatide received weekly subcutaneous doses for 72 weeks. Treatment started at 2.5 milligrams for 4 weeks; the dose then increased by 2.5 mg every 4 weeks up to a maximum tolerated dose of either 10 mg or 15 mg. The placebo group received matching subcutaneous placebo once weekly. The registry's brief summary describes a study lasting about 2 years and involving 29 visits.

The primary endpoint was percent change from baseline in body weight, assessed at baseline and 72 weeks. The primary analysis used a mixed-model repeated-measures approach and was limited to randomly assigned participants who took at least 1 dose of study drug, had a baseline value and at least 1 post-baseline value for the endpoint, and contributed no data after study-drug discontinuation. The registry lists 291 for placebo and 284 for tirzepatide in the primary endpoint results. These are the arm-specific endpoint counts, not a replacement for the separately reported overall enrollment.

The registry record does not report the original sample-size justification or the assumptions used to plan the trial. It also does not report the allocation ratio, details of the randomization procedure, or further operational details of the intensive lifestyle program. Those omissions limit how fully the design can be reconstructed from the record alone.

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: There was a 12-week lead-in period wherein the participants underwent intensive lifestyle modification program to reduce their daily caloric energy intake and increase their physical activity, followed by which they were randomized to receive Tirzepatide or Placebo.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All randomly assigned participants who took at least 1 dose of study drug, had a baseline and at least 1 post- baseline value for this outcome, excluding data after discontinuation of study drug.

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 randomized participants.

6. Primary Endpoint

The registry lists the primary outcome as Percent Change From Baseline in Body Weight, measured over Baseline, 72 Weeks.

The primary endpoint is percent change in body weight from baseline, evaluated at 72 weeks. It expresses weight change relative to each participant's baseline weight, rather than as an absolute amount of weight. A negative value indicates a decrease from baseline, while a positive value indicates an increase. The reported group estimates are least-squares means from a model, not simply the raw arithmetic average of individual changes.

Weight is a direct measure of the outcome the study was designed to examine after the lifestyle program. Using percent change allows changes to be interpreted in relation to starting body weight. Comparing groups at the same follow-up point also addresses whether their average trajectories differ during the study period. The primary result is a between-group contrast, so it is more informative for the randomized comparison than considering either arm's estimate alone.

The endpoint is not a measure of every aspect of health or of long-term outcomes beyond the reported follow-up. The registry lists additional measures, including maintenance of weight lost during the lifestyle program and thresholds for weight reduction, as secondary outcomes. The primary endpoint analysis population excluded data after study-drug discontinuation, as specified in the registry record; the consequences of that choice depend on post-discontinuation outcomes and missing-data assumptions, which are not fully reported.

7. Results Posted in the Registry

• Tirzepatide: mean -21.10 versus 3.30 on control; difference -24.40 (95% CI -26.05 to -22.75), standardized effect d = -2.42; 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.

Percentage of Participants Who Maintain ≥80% of the Body Weight Lost During Intensive Lifestyle Program

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Tirzepatide: 98.6% (280/284) versus 37.8% (110/291) on control; risk difference 60.8 percentage points (95% CI 55.1 to 66.5), risk ratio 2.61 (2.25–3.02); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Tirzepatide than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants Who Achieve ≥10%Body Weight Reduction

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Tirzepatide: 88.0% (250/284) versus 4.8% (14/291) on control; risk difference 83.2 percentage points (95% CI 78.7 to 87.7), risk ratio 18.30 (10.96–30.56); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Tirzepatide than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants Who Achieve ≥15% Body Weight Reduction

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Tirzepatide: 73.9% (210/284) versus 2.1% (6/291) on control; risk difference 71.9 percentage points (95% CI 66.5 to 77.2), risk ratio 35.86 (16.20–79.40); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Tirzepatide than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary outcomes show a consistent descriptive pattern: the tirzepatide group had higher percentages than the placebo group for maintaining at least 80% of the body weight lost during the intensive lifestyle program and for meeting the listed weight-reduction thresholds. The registry also lists changes in waist circumference, body weight, body mass index, and systolic blood pressure among the secondary outcomes, but the selected endpoint results do not include all of those measures.

This pattern is directionally consistent with the primary result, which reports a larger average decrease in body weight for the tirzepatide group than for placebo. Threshold outcomes and the continuous primary endpoint answer different questions. A mean change describes the average group contrast, while a threshold percentage describes how many participants met a defined criterion. The threshold outcomes can therefore add context without duplicating the primary endpoint's information.

The registry reports sponsor logistic-regression analyses for the selected binary outcomes, with odds ratios and confidence intervals. The unadjusted recalculations express group differences in percentage points. These measures are on different scales and should not be directly compared as if they were the same effect measure.

Secondary findings generally carry less weight than the primary result because they are additional questions rather than the main endpoint specified for the trial. Testing multiple outcomes can increase the chance of finding at least one apparently notable result by chance. Whether multiplicity was controlled, and how the secondary analyses were prioritized, are not reported in the registry record. The individual secondary results are therefore informative as supporting descriptions, but their p-values should not be treated as if they were an isolated primary test with a fully documented multiplicity plan.

The registry lists 20 secondary outcomes as posted. That broad set includes measures with different scales and interpretations. The overall pattern adds coherence to the body-weight finding, but it does not remove the need to consider each outcome's definition, analysis method, and uncertainty separately.

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

• Percentage of Participants Who Maintain ≥80% of the Body Weight Lost During Intensive Lifestyle Program
• Percentage of Participants Who Achieve ≥10%Body Weight Reduction
• Percentage of Participants Who Achieve ≥15% Body Weight Reduction
• Percentage of Participants Who Achieve ≥20% Body Weight Reduction
• Change From Baseline in Waist Circumference
• Change From Baseline in Body Weight
• Change From Baseline in Body Mass Index (BMI)
• Change From Baseline in Systolic Blood Pressure (SBP)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary analysis uses a mixed-model repeated-measures model, abbreviated MMRM in the registry description. In general, this kind of model uses repeated measurements over time and accounts for the fact that measurements from the same participant are related. It can estimate average changes at follow-up while incorporating the treatment group and specified covariates. The registry says the model included baseline, analysis country, sex, treatment, time, and treatment-by-time terms, with Type III sums of squares. The record does not provide further details of the model implementation.

The reported arm estimates are least-squares means. In general, a least-squares mean is a model-based estimate adjusted for the variables included in the analysis, rather than a raw group average. Here, that distinction matters because the sponsor's result is based on the stated repeated-measures model. It is therefore appropriate to treat the sponsor-reported mixed-model contrast as the main inferential result, and to understand a simple recalculation from the arm summaries as an unadjusted comparison.

What does the standard error describe?

The arm-level results report standard errors of 0.6 for placebo and 0.59 for tirzepatide. In general, a standard error describes the estimated sampling variability of an estimate: it reflects how much the estimate would be expected to vary across repeated samples under the model and assumptions. It is not the spread of individual participants' body-weight changes. A standard error can be used in constructing a confidence interval, but the registry's primary confidence interval is reported for the between-group contrast from the sponsor's analysis.

The recalculated interval uses posted summary statistics and an approximate pooled standard deviation of 10.09. In general, a standard deviation describes variation among observations, while a standard error describes uncertainty in an estimated quantity. The standard deviation used for this recalculation is marked as approximate in the analysis details. Consequently, the recalculated interval is an illustrative unadjusted result, not the same calculation as the sponsor's model-based confidence interval.

How should the p-value be read?

The registry reports p < 0.001 for the sponsor's primary comparison, and the unadjusted recalculation also gives p < 0.001. In general, a p-value measures how unusual data at least as incompatible with a null comparison would be if that null and the analysis assumptions were true. It is not the probability that the observed result is due to chance, nor does it quantify the clinical importance of the estimate. The effect estimate and confidence interval provide information about magnitude and precision that the p-value alone does not.

The registry also describes the reported analysis as a superiority analysis. In general, a superiority analysis tests whether the compared groups differ in the direction of the prespecified comparison, rather than testing whether their outcomes are equivalent. The record does not report further details of the formal hypothesis, significance-level allocation, or analysis plan, so those details should not be inferred from the label alone.

What do the binary comparisons add?

Several secondary outcomes are reported as percentages meeting a threshold. The recalculation expresses their contrast as a risk difference, meaning the difference between the two observed proportions in the compared groups. In general, a risk difference is measured on the percentage-point scale and answers how much more or less frequent an outcome was in one group than the other. The registry also reports sponsor analyses for these outcomes using logistic regression and odds ratios. An odds ratio compares odds, not probabilities; it can look numerically very different from a risk difference, especially when an outcome is common.

The registry's sponsor analyses and the recalculations are not interchangeable. The former are the reported analyses, while the latter are simple unadjusted re-calculations from posted summary information. The sponsor's primary analysis models repeated outcome measurements and includes listed covariates; the basic recalculation does not reproduce that model. These methodological differences are reasons to distinguish results clearly rather than interpret small differences between estimates as a new finding.

General statistical interpretation also depends on assumptions about the data and analysis. The registry gives the primary model terms and the analysis population, but it does not report the full statistical analysis plan, the model's covariance structure, diagnostics, or detailed missing-data assumptions. The registry record does not report whether any interim analysis was planned or conducted. These missing details prevent a complete independent appraisal of every model choice.

11. Confidence Intervals and Precision

The sponsor-reported primary mean difference is -24.5 percentage points, with a 95% confidence interval from -26.1 to -22.8. In general, a confidence interval describes a range of values compatible with the observed data and the method's assumptions. A 95% confidence procedure would produce intervals containing the true parameter in 95% of repeated applications under those assumptions; it does not assign a 95% probability to this particular interval or to the true value.

The reported interval is entirely below zero. In the context of the defined contrast, negative values correspond to a larger average decrease in the tirzepatide group relative to placebo. Its endpoints indicate that the estimated contrast is not just a point estimate: the model's uncertainty is represented by a range. The interval does not capture every source of uncertainty, including limitations in the study design, analysis choices, or applicability to other populations and settings.

The unadjusted recalculation gives -24.4, with a 95% confidence interval from -26.05 to -22.75. Its interval is close to the sponsor-reported interval, but it comes from a different calculation using posted arm summaries and an approximate pooled standard deviation. It should not be mistaken for a second sponsor analysis or for a replacement for the reported MMRM result.

The reported arm-level standard errors, 0.6 for placebo and 0.59 for tirzepatide, indicate uncertainty in the estimated arm means, not the range of individual responses. The registry does not report arm-level confidence intervals for these estimates. Precision should therefore be assessed chiefly through the sponsor-reported contrast interval and its stated analysis, alongside the limitations of the analysis population and unreported modeling details.

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 illustrations express sample size as participants per arm for the continuous and binary outcomes. They do not give total events. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. The original trial's own design assumptions and sample-size justification are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm that would correspond to an 80% chance of rejecting a null comparison under the listed planning assumptions if the selected effect were the true effect. The 90% power column uses a higher target chance of detecting that effect, so it generally requires more participants per arm. These are planning calculations, not a description of the actual trial design.

The rows for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect illustrate how the required sample size changes as the target effect changes. Smaller effects are harder to distinguish from random variation and therefore require more participants for the same target power, under the same assumptions. The illustrations are based on the effects observed in this trial and should not be treated as universal sample-size recommendations.

This is not observed power. Observed power calculated from an estimated effect does not provide a useful substitute for the effect estimate and its confidence interval. The planning figures are best read as sensitivity illustrations: they show how assumptions about the effect to detect and the desired power affect a hypothetical planning calculation.

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 = -24.400; pooled standard deviation σ = 10.092.
2. n = 2 × (1.96 + 0.842)² × 10.092² / -24.400² = 2.7, rounded up to 3 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 <- -24.4                      # observed difference in means
sigma <- 10.0918896125123                     # 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(4657016)
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 = 3 (80% power), 4 (90% power)
 75% of observed effect: n per arm = 5 (80% power), 7 (90% power)
 50% of observed effect: n per arm = 11 (80% power), 15 (90% power)
power.t.test: 3.9 
Simulated power at n = 3 per arm: 0.606 (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 the sponsor's primary analysis as a mixed-model mean difference (net) of -24.5, with a 95% confidence interval from -26.1 to -22.8 and p < 0.001. It labels the analysis as superiority. This is the trial's reported inferential result and is based on the MMRM model described in the registry, including baseline, analysis country, sex, treatment, time, and treatment-by-time terms.

The simple recalculation from the posted arm summaries gives a mean difference of -24.4, with a 95% confidence interval from -26.05 to -22.75 and p < 0.001. It is an unadjusted check and does not reproduce the sponsor's model. The two estimates are close, but the sponsor result remains the appropriate result to cite as the registry's own analysis; the recalculation provides context about the posted summary data.

The registry record does not flag a discrepancy between the sponsor's result and the recalculation. Differences in the estimates and interval endpoints should not be interpreted as contradictory findings or as evidence of a particular model effect. The recalculation relies on summary-statistic assumptions, including an approximate pooled standard deviation, whereas the sponsor analysis uses repeated measurements and covariates. The record does not provide the full analysis details needed to replicate the sponsor model independently.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a trial tests several outcomes or contrasts. In general, the more statistical tests conducted, the greater the chance that at least one result will appear statistically notable even if no corresponding effect exists. Methods such as prespecifying a hierarchy, adjusting significance thresholds, or controlling a family-wise error rate can address this issue, but the registry record does not report whether such methods were used for the secondary outcomes. The primary p-value is reported, but the record does not provide enough detail to reconstruct the complete testing strategy.

The registry lists 20 posted secondary outcomes. Their results should therefore be read as secondary evidence rather than as a collection of independent confirmations. It is not reported in the registry record which outcomes were included in any multiplicity family, whether an adjustment was made, or whether the outcome order was prespecified. The statistical analysis plan and full protocol details are not reported in the registry record.

The primary population is defined as randomly assigned participants who took at least 1 dose of study drug, had a baseline and at least 1 post-baseline value for the outcome, and were analyzed without data after study-drug discontinuation. This is not the same as including all randomized participants regardless of treatment exposure and follow-up. The registry record does not report the number excluded from this analysis population or the reasons for their exclusion.

The registry describes an MMRM for post-baseline measures, but does not report detailed assumptions about missing outcomes, the method for handling missingness, or sensitivity analyses to assess alternative assumptions. In general, a repeated-measures model can use observed measurements without filling in every missing value, but its validity depends on assumptions about how missingness relates to observed and unobserved outcomes. Those assumptions cannot be verified from the registry description alone.

Data after discontinuation of study drug were excluded from the primary endpoint analysis. The record does not report how discontinuation was distributed between arms, whether follow-up continued after discontinuation, or how outcomes after discontinuation were handled in other analyses. It also does not report whether interim analyses were conducted. These omissions limit assessment of potential bias and of the broader analysis plan.

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: nephrolithiasis (1.0%), acute myocardial infarction (0.3%), colitis microscopic (0.3%), haematochezia (0.3%), pancreatitis (0.3%).

The registry reports serious adverse events in 14 participants at risk in the placebo group and 17 participants at risk in the tirzepatide group. The corresponding groups at risk are reported as 292 and 287. These are counts from the registry record; the figures alone do not show the type, timing, duration, or causal relationship of every event.

Other adverse events are reported for 172 participants at risk in the placebo group and 230 participants at risk in the tirzepatide group. The registry lists 1 death in each group. Serious events, other adverse events, and deaths are distinct summaries and should not be treated as interchangeable measures. The counts do not establish that an event was caused by the assigned treatment.

The registry lists nephrolithiasis at 1.0% among the serious-event terms and acute myocardial infarction, colitis microscopic, haematochezia, pancreatitis, and small intestinal obstruction at 0.3% each. The registry record does not provide enough context in this summary to attribute individual events or evaluate their timing and clinical course. These figures describe reported events in the trial; they are not, by themselves, a complete comparative safety assessment.

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

The primary result means that, in the specified analysis population and over the reported 72-week period, the sponsor's model estimated a lower average percent change in body weight in the tirzepatide group than in the placebo group. The reported between-group difference was -24.5 percentage points, with a 95% confidence interval from -26.1 to -22.8. The result is an average contrast and does not describe the experience of every participant.

It does not establish that every participant lost weight, that every participant maintained weight lost during the lifestyle program, or that the observed group difference would apply to every population or care setting. The trial enrolled participants without type 2 diabetes who had obesity or were overweight with weight-related comorbidities, after an intensive lifestyle program. Results should be interpreted in that study context rather than generalized without qualification.

The result also does not establish outcomes beyond the reported follow-up. The primary endpoint was evaluated at baseline and 72 weeks; longer-term effects and outcomes after that period are not reported in the registry record. Nor does the primary weight endpoint by itself establish changes in every health outcome that may matter to participants. Secondary outcomes provide additional measures, but they are subject to their own uncertainty and the unreported multiplicity approach.

A small p-value does not mean the estimated difference is large in a practical or clinical sense, and a confidence interval does not account for all sources of bias or uncertainty. Interpretation depends on the analysis population, the MMRM assumptions, and handling of missing and post-discontinuation data. The registry record does not provide sufficient detail to resolve all of those issues.

The posted results are evidence about a randomized comparison under the study's conditions. They do not amount to individualized medical guidance, and they do not determine what choice is appropriate for any particular person. They also cannot settle questions about treatment effects in groups not represented by the stated eligibility criteria or about outcomes not reported in the registry.

18. Why This Trial Matters Statistically

Statistically, SURMOUNT-3 is useful because it examines body-weight change after an intensive lifestyle program rather than framing the comparison only around weight change without that context. The primary endpoint directly compares average percent change from baseline between randomized groups at 72 weeks. This allows the estimate to address the added contrast between assigned study treatments within the trial setting.

The registry reports both model-based estimates and enough arm-level summary information for a simple unadjusted recalculation. Their close values offer a useful example of how a transparent summary-based check can be compared with, but should not replace, the trial's own repeated-measures analysis. The distinction illustrates why readers need to identify whether an estimate is adjusted, model-based, or calculated directly from group summaries.

The endpoint set also illustrates how continuous and binary outcomes complement each other. Average percent change captures the overall shift in the measure, while threshold outcomes describe the proportion meeting particular criteria. Their consistent direction contributes context, although multiple secondary comparisons make their interpretation less definitive when multiplicity methods are not reported.

The result's precision is expressed through a confidence interval rather than a p-value alone. Reading the estimate alongside its interval, the analysis population, and the model description gives a more complete account of what the data support. The record also makes visible important gaps: the original planning assumptions, full analysis plan, missing-data methods, and multiplicity approach are not reported.

These features make the trial a practical case for separating statistical evidence from broader claims. The primary comparison is clearly reported, but a careful reading still distinguishes the trial's defined population and follow-up from unanswered questions about durability, generalizability, and analysis assumptions.

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 full protocol or statistical analysis plan, the original sample-size justification, or the assumptions used to plan the trial. It also does not give detailed information about the randomization procedure or allocation ratio. Those omissions limit independent assessment of prespecification and trial conduct.

The primary analysis excludes data after study-drug discontinuation and requires participants to have a baseline and at least 1 post-baseline measurement. The record does not report the number of participants who did not meet these criteria, the reasons for discontinuation or missing measurements, or sensitivity analyses under alternative missing-data assumptions. This limits assessment of how the analysis population may differ from all randomized participants.

The MMRM description identifies the included terms, but does not report the covariance structure, model diagnostics, or detailed assumptions. The registry record does not report whether interim analyses occurred or how multiplicity was addressed across the 20 posted secondary outcomes. These details matter for interpreting the strength and robustness of the reported evidence.

The primary follow-up is 72 weeks. The registry's brief summary says the study lasted about 2 years, but longer-term outcomes beyond the reported endpoint are not included in the primary result. The record also does not provide full details of the intensive lifestyle program, limiting assessment of how the post-program setting might compare with other contexts.

The recalculated primary comparison relies on summary statistics and an approximate pooled standard deviation. It is not a replication of the sponsor's model-based analysis. The registry reports only selected safety summaries, which do not provide the timing, duration, clinical course, or causal attribution of individual events. These limitations do not invalidate the posted results, but they narrow what can be concluded from the registry record.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study of Tirzepatide (LY3298176) In Participants After A Lifestyle Weight Loss Program — NCT04657016.
• Registry record last updated 2024-05-16; status last verified 2024-05.
• Results first posted 2024-05-16.
• 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

SURMOUNT-3 is a completed phase 3 randomized, double-blind, parallel-group trial of tirzepatide versus matching placebo after an intensive lifestyle program. The registry lists enrollment of 579 and evaluates participants without type 2 diabetes who had obesity or were overweight with weight-related comorbidities.

The primary endpoint was percent change in body weight from baseline to 72 weeks. The sponsor-reported MMRM analysis found a mean difference of -24.5 percentage points for tirzepatide versus placebo, with a 95% confidence interval from -26.1 to -22.8 and p < 0.001. A simple unadjusted recalculation gives a similar estimate, but is not a substitute for the sponsor's model.

Selected secondary outcomes show a consistent direction, while the registry record does not report the multiplicity approach or detailed missing-data assumptions. Serious adverse events, other adverse events, and deaths are reported as counts and should be interpreted as trial summaries, not causal conclusions. The results are specific to the reported population, treatment context, analysis, and follow-up.