A Study of Orforglipron in Adult Participants With Obesity or Overweight and Type 2 Diabetes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Orforglipron in Adult Participants With Obesity or Overweight and Type 2 Diabetes (NCT05872620), a randomized phase 3 study. The primary outcome is Percent Change From Baseline in Body Weight.

2. Clinical Question

3. Trial Design

This was a phase 3, randomized, double-blind, parallel-group study comparing once-daily oral orforglipron with matching placebo. The registry record reports enrollment of 1613 and lists the study status as completed. The study began on 2023-06-05, and primary completion was recorded as 2025-08-08. The brief study description says the study would last about 77 weeks and could include up to 22 visits.

Participants were assigned to placebo or one of three orforglipron target doses: 6 mg, 12 mg, or 36 mg once daily. Each active regimen began at 1 mg, followed by dose escalation every 4 weeks. Placebo used corresponding escalation to preserve masking. Target treatment was described through Week 72, with dose modification possible. The registry record does not report the allocation ratio, the method used to generate the random sequence, or the procedures used to conceal allocation.

The primary endpoint was percent change in body weight from baseline to Week 72. Its analysis included all randomized participants, but the included measurements were restricted to the treatment period: at or after baseline and before the earliest of study-drug discontinuation or initiation of prohibited weight-management treatments. This defines which observed measurements contributed to the primary analysis; it does not, by itself, describe how missing measurements were handled. Specific details of the missing-data strategy and the trial’s original sample-size justification are not reported in the registry record.

The study’s placebo comparison can characterize the posted differences among these assigned regimens under the trial’s analysis definitions. It does not directly compare orforglipron with another active treatment. The registry record does not report additional design details such as a planned interim analysis or the intended allocation ratio.

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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All randomized participants. All data points obtained during the treatment period, defined as at or after baseline up to the earliest date of discontinuation of study drug or initiation of prohibited weight management treatments were included in this analysis.

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, Week 72.

The primary endpoint is percent change in body weight from baseline to Week 72. It expresses change relative to each participant’s starting weight, so the measure is not in kilograms or pounds. A negative value indicates a reduction from baseline; a more negative group estimate represents a larger mean reduction on this scale. The reported group values are least-squares means, which are model-based estimates rather than simple arithmetic averages of the recorded measurements.

Body weight is a direct measure of the outcome the study set out to investigate. Measuring change from baseline makes the analysis focus on change over the treatment period rather than on the final weight alone. Comparing the modeled changes between each dose group and placebo addresses whether the estimated change differed between assigned groups within the trial’s analysis framework. The endpoint does not, on its own, describe changes in body composition, health events, or other outcomes.

The registry record specifies baseline and Week 72 as the primary endpoint timeframe, while its analysis description includes post-baseline measurements during the defined treatment period. That distinction matters: the model uses repeated measurements, but the reported endpoint is an estimated treatment contrast for percent change in body weight, not a count of measurements or an individual-level prediction. The record does not report the clinical threshold, if any, that was used to interpret a particular percentage change as important.

7. Results Posted in the Registry

• 6 mg Orforglipron QD: mean -5.50 versus -2.21 on control; difference -3.29 (95% CI -4.11 to -2.47), standardized effect d = -0.57; z-test p = <0.001.
• 12 mg Orforglipron QD: mean -7.78 versus -2.21 on control; difference -5.57 (95% CI -6.48 to -4.66), standardized effect d = -0.90; z-test p = <0.001.
• 36 mg Orforglipron QD: mean -10.54 versus -2.21 on control; difference -8.33 (95% CI -9.34 to -7.32), standardized effect d = -1.27; 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 Achieved With Greater Than or Equal to (≥) 5% Body Weight Reduction From Baseline

• 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:
  • 6 mg Orforglipron QD: 49.8% (164/329) versus 24.4% (154/630) on control; risk difference 25.4 percentage points (95% CI 19.0 to 31.8), risk ratio 2.04 (1.71–2.43); unadjusted two-proportion z-test p = <0.001.
  • 12 mg Orforglipron QD: 60.2% (200/332) versus 24.4% (154/630) on control; risk difference 35.8 percentage points (95% CI 29.6 to 42.0), risk ratio 2.46 (2.09–2.90); 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 6 mg Orforglipron QD 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 proportion with the event is statistically higher for 12 mg Orforglipron QD than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants Who Achieved ≥10% Body Weight Reduction From Baseline

• 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:
  • 6 mg Orforglipron QD: 24.0% (79/329) versus 7.0% (44/630) on control; risk difference 17.0 percentage points (95% CI 12.0 to 22.1), risk ratio 3.44 (2.44–4.85); unadjusted two-proportion z-test p = <0.001.
  • 12 mg Orforglipron QD: 35.5% (118/332) versus 7.0% (44/630) on control; risk difference 28.6 percentage points (95% CI 23.0 to 34.1), risk ratio 5.09 (3.70–7.01); 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 6 mg Orforglipron QD 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 proportion with the event is statistically higher for 12 mg Orforglipron QD than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants Who Achieved ≥15% Body Weight Reduction From Baseline

• 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:
  • 6 mg Orforglipron QD: 7.3% (24/329) versus 1.9% (12/630) on control; risk difference 5.4 percentage points (95% CI 2.4 to 8.4), risk ratio 3.83 (1.94–7.56); unadjusted two-proportion z-test p = <0.001.
  • 12 mg Orforglipron QD: 17.8% (59/332) versus 1.9% (12/630) on control; risk difference 15.9 percentage points (95% CI 11.6 to 20.1), risk ratio 9.33 (5.09–17.11); 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 6 mg Orforglipron QD 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 proportion with the event is statistically higher for 12 mg Orforglipron QD than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary weight-threshold results show a consistent descriptive ordering across the three listed thresholds: the reported percentages are higher in the orforglipron dose groups than in the placebo group, with larger percentages at the higher thresholds and a gradient across the listed doses. This provides a complementary view of the weight outcome. The primary endpoint summarizes average percent change, whereas threshold outcomes describe the share of participants meeting specified levels of reduction.

These endpoint types answer different questions. A group can have a change in its mean without every participant experiencing that change, and a threshold percentage does not show the size of change among participants who fall above or below the threshold. The secondary results therefore complement, rather than replace, the continuous primary outcome. The registry also lists waist circumference, hemoglobin A1c, fasting serum glucose, and hemoglobin A1c target outcomes, but their numerical results are not reported in the registry record.

Unadjusted secondary comparisons carry less confirmatory weight when several outcomes and dose comparisons are examined. In general, conducting many tests increases the chance that some results will appear statistically notable by chance. A prespecified multiplicity strategy can address this issue, but the strategy is not reported in the registry record. The posted secondary outcomes can be described as supportive patterns, but their statistical significance should not be treated as equivalent to a single prespecified primary test without the analysis plan.

The registry includes sponsor-reported logistic regression risk differences for some threshold comparisons and also permits simple unadjusted recalculations from group summaries. Those are different analytic approaches and may produce slightly different estimates. The sponsor’s own reported analysis remains the reference for its comparisons. The information available does not establish how the full set of secondary outcomes was prioritized or whether testing was controlled across the endpoint family.

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

• Percentage of Participants Who Achieved With Greater Than or Equal to (≥) 5% Body Weight Reduction From Baseline
• Percentage of Participants Who Achieved ≥10% Body Weight Reduction From Baseline
• Percentage of Participants Who Achieved ≥15% Body Weight Reduction From Baseline
• Change From Baseline in Waist Circumference
• Change From Baseline in Hemoglobin A1c (HbA1c) %
• Change From Baseline in Fasting Serum Glucose (FSG)
• Percentage of Participants Who Achieved Hemoglobin A1c (HbA1c) Target Value (<6.5%)
• Percentage of Participants Who Achieved Hemoglobin A1c (HbA1c) Target Value (<7.0%)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint analysis used a mixed model for repeated measures, commonly abbreviated MMRM. In general, this type of model analyzes repeated outcome measurements while accounting for their relationship within a participant. It can use measurements from different post-baseline visits together, rather than reducing every participant’s record to a single final observation. The registry reports least-squares means, which are model-based estimates of the group outcomes under the fitted model.

What does the model adjust for?

The registry describes region and interactions among baseline value, time, treatment, strata, and time and treatment in the model. The strata were defined using the joint levels of sex and background oral antihyperglycemic medication categories, classified by their potential effect on body weight. This means that the estimate is not simply the raw average change in each arm: it comes from a model that includes the listed covariates and interactions. The registry does not give further detail about the number or identity of regions, nor does it report a full model specification beyond this description.

The record states that the variance-covariance structure for change from baseline was unstructured. In general, a variance-covariance structure describes how the model represents variation at each measurement time and the correlations among measurements over time. An unstructured form allows those variances and correlations to be estimated without imposing a simpler pattern. This flexibility can be useful when the relationship among repeated measurements is not assumed to follow a particular shape, although it also relies on adequate information to estimate the model parameters. No additional information about model diagnostics or convergence is reported in the registry record.

What is a least-squares mean difference?

A least-squares mean is an adjusted, model-based group estimate. The reported least-squares mean difference is the modeled contrast between an orforglipron group and placebo. It is expressed on the primary endpoint’s percent-change scale. The group-specific least-squares means and their differences should not be interpreted as raw percentages of participants or as estimates for every individual. They describe average group outcomes under the statistical model and the analysis population definition.

The sponsor’s reported comparisons use mixed models and include confidence intervals and p-values. The separately recalculated comparisons are simple, unadjusted calculations based on posted summary statistics. Their standard deviations are approximate, derived from reported standard errors, and their intervals are not the same as intervals from the trial’s full repeated-measures model. General statistical principle: an unadjusted summary-based calculation cannot reproduce adjustments, covariance handling, or other features of a fitted model unless those details are available and incorporated. The sponsor’s own analysis therefore remains the relevant result for the registry’s planned comparisons.

Why does the treatment-period definition matter?

The primary analysis population is described as all randomized participants, but the record also defines which measurements were included: observations at or after baseline and before the earliest of study-drug discontinuation or starting prohibited weight-management treatment. This is an analysis rule for the measurements, not a statement that every randomized participant contributed a complete Week 72 record. The registry’s description does not establish how participants with missing measurements were handled, whether measurements after treatment discontinuation were modeled in another analysis, or how departures from assigned treatment were addressed beyond the stated inclusion rule.

In general, missing-data assumptions can affect model-based estimates because measurements that are not observed cannot directly contribute their recorded outcomes. An MMRM can use available repeated observations, but the validity of its estimates depends on assumptions about the process leading to missingness and on the model specification. The particular assumptions and sensitivity analyses are not reported in the registry record, so their adequacy cannot be assessed from the listed endpoint results alone.

What does a p-value contribute?

The registry reports p-values below .001 for the three primary comparisons. In general, a p-value measures how compatible the data are with a specified null hypothesis under the analysis model; it is not a measure of effect size or clinical relevance. A small p-value does not eliminate uncertainty about the estimated difference, and it does not establish that the analysis assumptions are correct. The estimated contrasts and their confidence intervals provide complementary information about magnitude and precision.

The registry labels the sponsor’s comparisons as superiority analyses. It does not report the complete statistical analysis plan, the prespecified hierarchy among outcomes and doses, or how multiplicity was controlled. Those details matter when interpreting several comparisons as a family of confirmatory tests. The reported primary results can be described as posted, while the strength of claims across multiple tests cannot be fully assessed without the missing plan details.

11. Confidence Intervals and Precision

The sponsor reports 95% confidence intervals for the three primary contrasts. For 6 mg versus placebo, the interval is -4.10 to -2.47 around an estimated difference of -3.28. For 12 mg versus placebo, it is -6.48 to -4.65 around -5.56. For 36 mg versus placebo, it is -9.34 to -7.32 around -8.33. Each interval lies below zero, so the range of values compatible with the model-based interval procedure is on the side corresponding to a greater estimated mean reduction in the orforglipron group.

In general, a confidence interval is a range calculated by a method that, over repeated studies conducted under the same assumptions, would contain the true parameter at the stated rate. It is not a range containing a stated proportion of individual participant outcomes, and it is not a probability distribution for the true effect. The intervals communicate uncertainty about the estimated group contrast. Their interpretation depends on the fitted model and the assumptions used to calculate them.

The recalculated intervals are -4.105 to -2.475 for 6 mg, -6.479 to -4.661 for 12 mg, and -9.336 to -7.324 for 36 mg. These are simple unadjusted recalculations based on the posted summary statistics and approximate pooled standard deviations. They should not be treated as replacements for the sponsor’s model-based intervals. The close numerical agreement offers a check that the posted group summaries and reported contrasts are broadly aligned, but it does not verify the full model or its assumptions.

Each group estimate also has a reported standard error: 0.215 for placebo, 0.356 for 6 mg, 0.411 for 12 mg, and 0.466 for 36 mg. A standard error describes the estimated sampling variability of a group estimate under its analysis framework; it is not the spread of individual body-weight changes. The registry does not provide confidence limits for these group-specific estimates. The sponsor’s comparison intervals are the most direct registry-based description of precision for the primary treatment contrasts.

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 primary-endpoint planning estimates are participants per arm, not total events. The calculation uses a two-sided significance level of 0.05, equal allocation between the two compared arms, and a normal approximation. It considers effects set at the stated fraction of the observed effect. These are planning illustrations based on the observed contrast, not the original trial design assumptions; the trial’s own sample-size justification and design assumptions are not reported in the registry record.

The 80% power column gives the estimated participants per arm for a future two-arm study to have 80% power under the stated assumptions and assumed effect. In plain language, power is the chance of detecting an effect of the assumed size at the chosen significance level if that effect is present and the assumptions hold. The 90% power column uses a higher target chance of detection, so its required participant count is larger for the same assumed effect. For example, planning for 75% of the observed effect gives estimates of 87 participants per arm for 80% power and 116 participants per arm for 90% power for the 6 mg comparison.

Smaller assumed effects require more participants to distinguish them from random variation under the same planning framework. For that reason, the estimates for 50% of the observed effect are larger than those for the full observed effect. The plotted planning relationship represents the same link between assumed effect and participants per arm. These estimates are not observed power: they do not describe the probability that this completed trial detected its actual effect, and they should not be read as evidence about its design adequacy.

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 = -3.290; pooled standard deviation σ = 5.782.
2. n = 2 × (1.96 + 0.842)² × 5.782² / -3.290² = 48.5, rounded up to 49 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 <- -3.29                      # observed difference in means
sigma <- 5.78199525732207                     # 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(5872620)
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 = 49 (80% power), 65 (90% power)
 75% of observed effect: n per arm = 87 (80% power), 116 (90% power)
 50% of observed effect: n per arm = 194 (80% power), 260 (90% power)
power.t.test: 49.5 
Simulated power at n = 49 per arm: 0.795 (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’s own primary analyses use mixed models and report least-squares mean differences. Their estimates are -3.28 for 6 mg, -5.56 for 12 mg, and -8.33 for 36 mg versus placebo. The reported 95% intervals are -4.10 to -2.47, -6.48 to -4.65, and -9.34 to -7.32, respectively; each comparison has p <.001.

The recalculated contrasts from the posted arm summaries are -3.29, -5.57, and -8.33. They use an approximate pooled standard deviation derived from the reported standard errors and are simple unadjusted calculations. The close values are consistent with the sponsor’s reported contrasts at the displayed precision. The recalculated intervals are also close to, but not identical with, the sponsor’s intervals because they do not reproduce the full mixed-model analysis.

The sponsor’s analysis takes precedence for interpreting the trial’s planned comparisons. The recalculation is a transparent numerical cross-check, not an alternative analysis plan or a way to recover all features of the model. It does not independently validate the model assumptions, missing-data handling, or any adjustments described in the registry. No discrepancy between the sponsor results and recalculation is flagged in the registry record.

15. Multiplicity, Missing Data and Analysis Populations

The primary endpoint includes three comparisons with placebo, one for each orforglipron target dose. The registry reports superiority analyses for these comparisons, as well as multiple secondary outcomes. When a study tests several hypotheses, the probability of at least one apparently positive result can increase if each test is interpreted in isolation. In general, multiplicity procedures may control this familywise error, allocate significance levels, or use a hierarchy that specifies which comparisons are tested first.

The multiplicity strategy, if any, is not reported in the registry record. The record also does not provide the complete statistical analysis plan or enough information to determine whether the three primary dose comparisons were treated as a single confirmatory family, whether secondary testing followed a hierarchy, or whether p-values were adjusted. Therefore the reported p-values can be quoted as posted, but the record does not support a complete assessment of error control across all analyses.

The primary analysis population is described as all randomized participants. However, the primary measurement rule includes data only from baseline through the earliest of study-drug discontinuation or initiation of prohibited weight-management treatments. The registry does not report the number of participants with missing Week 72 measurements, reasons for missingness, or a detailed method for handling missing data. It also does not report sensitivity analyses assessing the effect of alternative missing-data assumptions. General statistical principle: conclusions from repeated-measures models can depend on assumptions about how the observed and missing outcomes are related.

The description does not establish whether the primary endpoint is an intention-to-treat estimand that includes outcomes after treatment discontinuation or prohibited treatment initiation. The stated data window ends at the specified event, so the estimand and interpretation are tied to that rule. Further estimand details, including how intercurrent events were handled in analyses beyond the stated measurement inclusion rule, are not reported in the registry record.

The record identifies the primary analysis as model-based and lists the included factors and unstructured covariance specification, but it does not report model diagnostics, sensitivity models, or the full set of analysis decisions. These omissions limit independent evaluation of robustness. They do not negate the posted estimates; they define what cannot be determined from the registry record.

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 fibrillation (0.9%), urinary tract infection (0.9%), coronary artery disease (0.6%), myocardial infarction (0.6%), nephrolithiasis (0.6%).

The safety summary reports serious adverse events among participants at risk in each group. The registry lists 55 serious adverse events among 628 participants at risk in the placebo group, 24 among 328 at risk in the 6 mg group, 34 among 331 at risk in the 12 mg group, and 35 among 321 at risk in the 36 mg group. These are reported counts and denominators; they should be read as safety summaries, not as estimates of treatment effect.

The record also lists other adverse events for the same at-risk group sizes: 452 in placebo, 231 in the 6 mg group, 262 in the 12 mg group, and 255 in the 36 mg group. Deaths are reported as 4, 0, 4, and 2, respectively. The record does not establish from these counts alone whether an event was caused by the study treatment, nor does it give the clinical details needed to assess individual events.

Serious adverse event terms listed among the top terms include atrial fibrillation and urinary tract infection at 0.9% each, and coronary artery disease, myocardial infarction, nephrolithiasis, and osteoarthritis at 0.6% each. The registry summary does not provide enough detail to compare event timing, duration, severity beyond the serious classification, or relatedness by treatment group. Counts and percentages should be interpreted within the record’s safety definitions and not as a standalone causal comparison.

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

The primary results describe model-based differences in mean percent change in body weight between assigned groups from baseline to Week 72, using the treatment-period measurements defined in the registry record. The posted estimates are more negative for each orforglipron group than for placebo, and the sponsor’s reported confidence intervals for the three contrasts lie below zero. This describes the observed trial results and their statistical uncertainty under the sponsor’s model.

The estimates do not predict the result for any particular person. Group averages can conceal variation in individual responses, and the endpoint does not indicate how many participants experienced a specific change unless a separate threshold outcome is considered. Nor does the primary endpoint establish changes in other health outcomes. Although the registry lists several glycemic and waist-related secondary outcomes, their numerical results are not reported in the registry record.

The data also do not resolve how results would generalize to populations outside the adult participants with obesity or overweight and type 2 diabetes studied in this trial. The record compares orforglipron with placebo, not with another active treatment. The primary analysis uses a defined treatment-period window, so it does not by itself describe outcomes after study-drug discontinuation or after initiation of prohibited weight-management treatments.

Statistical evidence of a difference is not the same as evidence of clinical importance. The record does not report a threshold for a clinically meaningful difference, and the primary outcome alone cannot establish the overall balance of benefits and risks. The safety summaries are descriptive counts and do not prove that an event was caused or not caused by treatment. Taken together, these results are evidence about the specified outcome, population, comparison, and analysis period—not a complete assessment of every outcome that could matter in clinical care.

18. Why This Trial Matters Statistically

This trial illustrates why endpoint definitions and analysis populations matter as much as the headline group estimates. The primary outcome is a model-based estimate of percent change, and the measurement window ends at the earliest of study-drug discontinuation or initiation of prohibited weight-management treatments. Understanding that rule is necessary to interpret what the reported contrast represents.

The registry also provides a useful example of the distinction between an analysis performed by the study sponsor and an unadjusted recalculation based on group summaries. The recalculated contrasts are close to the sponsor’s mixed-model estimates, but the two approaches are not interchangeable. Model-based results incorporate repeated measurements and the listed covariates; simple summary-statistic calculations do not recreate those features. Reporting both can help readers check numerical coherence while keeping the planned analysis in view.

Precision and multiplicity are separate questions. The primary contrasts have reported confidence intervals, which describe uncertainty around each estimate under the sponsor’s model. Several doses and secondary endpoints are also examined, but the registry does not report the multiplicity strategy. That makes it important to distinguish the posted primary comparisons from exploratory or supportive secondary patterns rather than treating every result as having the same confirmatory status.

Finally, the safety counts show why outcome definitions and denominators should be read carefully. The listed figures describe events and participants at risk; they are not causal estimates. Statistical interpretation is strongest when it stays tied to the population, comparison, endpoint, analysis window, and uncertainty that the record actually specifies.

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, allocation ratio, random-sequence method, allocation concealment procedure, or a complete statistical analysis plan. It also does not report the multiplicity strategy, so the intended control of error across the dose comparisons and secondary outcomes cannot be assessed. These missing design and analysis details limit independent evaluation of the confirmatory status of the full set of findings.

The primary analysis uses measurements collected during a defined treatment period rather than describing outcomes after the earliest of study-drug discontinuation or initiation of prohibited weight-management treatments. The record does not report how missing Week 72 measurements were handled in detail, the amount or reasons for missing data, or sensitivity analyses for alternative assumptions. It is therefore not possible to assess from the registry summary how robust the estimates are to missingness or to the treatment-period rule.

The study compares three orforglipron target doses with placebo; it does not provide an active-treatment comparison. The primary endpoint is body-weight change and does not by itself establish effects on broader health outcomes. The registry lists other outcomes, but numerical results for waist circumference and glycemic measures are not reported in the registry record. The available safety summary gives aggregate counts and selected serious-event terms, without enough detail to establish causality, timing, or event-level context.

The recalculated estimates are approximate and unadjusted, using pooled standard deviations derived from standard errors. They are useful as a numerical cross-check but do not reproduce the sponsor’s repeated-measures model. Finally, the record’s population is adults with obesity or overweight and type 2 diabetes; generalization beyond that study population is not established by these results.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study of Orforglipron in Adult Participants With Obesity or Overweight and Type 2 Diabetes — NCT05872620.
• Registry record last updated 2026-09-04; status last verified 2026-09.
• Results first posted 2026-09-04.
• 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

NCT05872620 is a completed phase 3 randomized, double-blind, parallel-group study sponsored by Eli Lilly and Company. It evaluated once-daily oral orforglipron at target doses of 6 mg, 12 mg, and 36 mg against matching placebo in adults with obesity or overweight and type 2 diabetes. Enrollment was 1613. The study began on 2023-06-05, and primary completion was recorded on 2025-08-08.

The primary endpoint was percent change in body weight from baseline to Week 72, analyzed using a mixed model for repeated measures and reported as least-squares means. The sponsor’s reported mean differences versus placebo were -3.28, -5.56, and -8.33 for the 6 mg, 12 mg, and 36 mg groups, respectively; each reported comparison had p <.001. The registry also reports secondary weight-threshold outcomes and aggregate safety counts. Multiplicity procedures and detailed missing-data methods are not reported in the registry record.