Study of Efficacy and Safety of Vortioxetine (Lu AA21004) in Treating Generalized Anxiety Disorder: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Efficacy and Safety of Vortioxetine (Lu AA21004) in Treating Generalized Anxiety Disorder (NCT00731120), a randomized phase 3 study. The primary outcome is Change From Baseline in the Hamilton Anxiety Scale (HAM-A) Total Score.

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, double-blind, parallel-group, placebo-controlled, fixed-dose study of vortioxetine in adults with generalized anxiety disorder. Participants were assigned to placebo, vortioxetine 2.5 mg, or vortioxetine 10 mg. The registry describes oral dosing once daily for up to 8 weeks in each arm. The reported enrollment was 457; the registry record does not explain how that enrollment target was selected.

The primary outcome was change from baseline in HAM-A total score at Week 8. The primary analysis population was the full analysis set, defined in the record as randomized patients who received at least 1 dose of study drug and had at least 1 post-baseline value for assessment of primary efficacy. The record states that last observation carried forward (LOCF) was used. It reports primary arm summaries for 148 participants in the placebo group, 144 in the vortioxetine 2.5 mg group, and 146 in the vortioxetine 10 mg group. These are the reported analysis counts, not the registry enrollment figure.

The study started in 2008-06, had primary completion in 2009-01, and is listed as completed. The registry record does not report details of the randomization sequence, allocation concealment, treatment adherence, protocol deviations, or follow-up after the treatment period. The primary outcome analysis is described as an analysis of covariance (ANCOVA) with treatment and center as fixed factors and baseline value as a covariate. Other statistical-plan details, including the original sample-size justification, 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: Participants with a diagnosis of generalized anxiety disorder were enrolled equally in one of three treatment groups, once a day placebo or 2.5 or 10 mg vortioxetine.

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 patients who received at least 1 dose of study drug, and had at least 1 post-baseline value for assessment of primary efficacy. Last observation carried forward (LOCF) was used.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in the Hamilton Anxiety Scale (HAM-A) Total Score, measured over Baseline and Week 8.

The primary endpoint was change from baseline to Week 8 in the total score on the Hamilton Anxiety Scale (HAM-A). The scale contains 14 items covering anxious mood, tension, fear, insomnia, cognitive and depressed-mood symptoms, behavior at interview, and several somatic symptom domains. Each item is rated from 0, absent, to 4, maximum severity; the total ranges from 0 to 56. In general, because higher item ratings indicate greater symptom severity, a more negative change reflects a larger reduction in the measured score.

A total score gives a continuous summary of symptom ratings across the included domains. Continuous outcomes can capture graded change rather than classifying participants only as responders or nonresponders. The registry also reports response and remission as secondary outcomes, but those classifications answer different questions from average score change. A group mean can describe the overall shift in scores while concealing variation among individuals; it does not show how every participant changed.

The posted primary values are least-squares means, not simple raw averages. The registry says the ANCOVA model included treatment and center as fixed factors and baseline HAM-A value as a covariate. This adjustment estimates group means while accounting for the stated model terms. The record describes LOCF for the primary analysis population; the implications and limitations of that approach are discussed in the statistical interpretation.

7. Results Posted in the Registry

• Vortioxetine 2.5 mg: mean -10.75 versus -9.87 on control; difference -0.88 (95% CI -2.48 to 0.72), standardized effect d = -0.13; z-test p = 0.280.
• Vortioxetine 10 mg: mean -10.68 versus -9.87 on control; difference -0.81 (95% CI -2.42 to 0.80), standardized effect d = -0.11; z-test p = 0.325.

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 Responders in HAM-A Total Score at Week 8

• 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:
  • Vortioxetine 2.5 mg: 46.5% (67/144) versus 41.9% (62/148) on control; risk difference 4.6 percentage points (95% CI -6.7 to 16.0), risk ratio 1.11 (0.86–1.44); unadjusted two-proportion z-test p = 0.425.
  • Vortioxetine 10 mg: 41.8% (61/146) versus 41.9% (62/148) on control; risk difference -0.1 percentage points (95% CI -11.4 to 11.2), risk ratio 1.00 (0.76–1.31); unadjusted two-proportion z-test p = 0.985.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Vortioxetine 2.5 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Vortioxetine 10 mg and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Participants in HAM-A Remission at Week 8

• 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:
  • Vortioxetine 2.5 mg: 21.5% (31/144) versus 21.6% (32/148) on control; risk difference -0.1 percentage points (95% CI -9.5 to 9.3), risk ratio 1.00 (0.64–1.54); unadjusted two-proportion z-test p = 0.984.
  • Vortioxetine 10 mg: 19.2% (28/146) versus 21.6% (32/148) on control; risk difference -2.4 percentage points (95% CI -11.7 to 6.8), risk ratio 0.89 (0.56–1.39); unadjusted two-proportion z-test p = 0.603.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Vortioxetine 2.5 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Vortioxetine 10 mg and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in Sheehan Disability Scale (SDS) at Week 8

• 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 errors.
• Result:
  • Vortioxetine 2.5 mg: mean -5.73 versus -4.26 on control; difference -1.47 (95% CI -3.19 to 0.25), standardized effect d = -0.22; z-test p = 0.094.
  • Vortioxetine 10 mg: mean -5.24 versus -4.26 on control; difference -0.98 (95% CI -2.75 to 0.79), standardized effect d = -0.14; z-test p = 0.277.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Vortioxetine 2.5 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Vortioxetine 10 mg and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcome summaries cover several types of measures, including symptom response and remission, global clinical ratings, anxiety and depression scales, disability, and health status. The detailed posted results in the registry record include responder status, remission, and Sheehan Disability Scale change. Their pattern is not uniform across outcomes or doses: the relative group patterns vary rather than pointing in one consistent direction. That mixed pattern does not provide a clear, independent confirmation of the primary result.

Secondary measures can complement the primary HAM-A analysis because they describe different aspects of symptoms, functioning, or health status. They do not answer exactly the same question. A responder threshold reduces a continuous score to a category, remission applies a separate outcome definition, and disability scores measure a different domain. Agreement across related outcomes can sometimes add context, while disagreement can reflect differences in what is measured, statistical uncertainty, or other factors.

The registry record lists additional secondary outcomes, but detailed numerical results for those outcomes are not reported in the registry record. No sponsor-reported analyses are included for the secondary results with detailed summaries. The displayed recalculated comparisons are unadjusted, and their p-values do not account for the fact that multiple outcomes and dose comparisons were considered.

When many outcomes are examined, the chance of seeing apparently notable results by chance can increase. This is a general multiplicity concern, not a claim that any particular secondary result is false. The registry record does not report a multiplicity strategy. Accordingly, secondary findings carry less inferential weight than a prespecified primary analysis, and should be treated as supporting context rather than as definitive evidence on their own. They do not change the need to interpret the primary estimate and its uncertainty directly.

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

• Change From Baseline in Hamilton Anxiety Scale (HAM-A) Total Score at Other Weeks Assessed
• Percentage of Responders in HAM-A Total Score at Week 8
• Percentage of Participants in HAM-A Remission at Week 8
• Clinical Global Impression Scale-Global Improvement (CGI-I) at Each Week Assessed
• Change From Baseline in Clinical Global Impression Scale-Severity of Illness (CGI-S)
• Change From Baseline in Hospital Anxiety and Depression (HAD) Scales
• Change From Baseline in Sheehan Disability Scale (SDS) at Week 8
• Change From Baseline in 36-item Short-form Health Survey (SF-36)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is continuous: it is a score, rather than a yes-or-no event. The analysis therefore compares average score changes between groups. The reported arm values are least-squares means, with standard errors, from an ANCOVA model. A standard error describes the estimated sampling variability of a mean; it is not the same as the spread of individual participant scores.

What does ANCOVA contribute?

The registry describes an analysis of covariance model with treatment and center as fixed factors and baseline HAM-A value as a covariate. In general, ANCOVA estimates group outcomes while accounting for the included covariates. Baseline adjustment can account for differences in starting scores and may improve precision when baseline score is related to the follow-up outcome. Including center allows the model to account for the centers represented in the trial. The record does not provide model diagnostics, details of the center structure, or other model specifications, so those cannot be assessed from the registry record.

The least-squares mean difference is the model-based comparison between a vortioxetine group and placebo. The reported contrast is negative for both doses, which indicates a numerically more negative change in the active-treatment group under the subtraction order of treatment minus placebo. The contrast is not the same as a percentage change, and it should be interpreted in the original HAM-A score units. No new standardized effect or clinical-importance threshold is needed to read the reported contrast.

How should the confidence interval and p-value be read?

In general, a confidence interval represents a range of estimates reasonably compatible with the observed data and the statistical model at the stated confidence level. The sponsor-reported 95% intervals for both primary comparisons cross zero. Zero represents no difference in mean change under the contrast used. Because the intervals extend to both sides of zero, the results do not pin down the direction of the between-group difference with precision sufficient to exclude no difference.

A p-value measures how unusual data at least as discordant with a specified null hypothesis would be under that hypothesis and the model assumptions. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of an effect. The sponsor-reported primary p-values are 0.279 and 0.306. The registry record does not state a decision threshold or describe an adjustment for the two primary dose comparisons, so the p-values should be read as reported rather than as a complete account of the trial's testing strategy.

What is the role of the recalculation?

The recalculated mean differences use the posted group means and standard errors to form simple, unadjusted comparisons. Such a calculation does not incorporate the sponsor model's adjustment for center and baseline HAM-A value. The posted standard errors also do not contain the underlying participant-level data needed to reproduce that adjusted analysis. The recalculated estimates are close to the sponsor estimates in this trial, but the sponsor's own ANCOVA takes precedence for interpretation.

For secondary binary outcomes, recalculations express group differences as risk differences. In general, a risk difference is the difference between the proportions meeting an outcome in two groups. The direction depends on which group is subtracted from which; an interval spanning zero does not rule out no difference. These secondary calculations are unadjusted, and the registry record does not report sponsor analyses for the three secondary outcomes with detailed results in the record.

Why can variability affect inference?

General statistical principle: estimates become less precise when there is substantial variation among participant outcomes relative to the size of the comparison. The standard error reflects uncertainty in an estimated mean, while a confidence interval combines the estimate and its uncertainty. A relatively small estimate with substantial uncertainty can yield an interval that includes both a modest difference and no difference. This helps explain why the point estimate alone is not enough: its precision matters.

The sample-size illustrations use planning assumptions stated in the registry-derived material: a two-sided significance level of 0.05, equal allocation between the two compared arms, a normal approximation, and an effect to detect set as a stated fraction of the observed effect. In general, smaller target effects require more participants to detect with the same planned power, because they are harder to distinguish from random variation. The original trial design assumptions are not reported in the registry record. These planning calculations are not observed power.

What limitations follow from the analysis population and missing-data approach?

The primary analysis set excluded randomized participants who did not receive at least 1 dose or had no post-baseline primary efficacy value, according to the record's definition. That differs from an analysis of every randomized participant. The record says LOCF was used, meaning the last observed value was carried forward for the primary analysis. In general, this approach can be problematic if a participant's outcome would have changed after the last observed visit; it relies on an assumption that the carried-forward value is an acceptable substitute. The record does not report sensitivity analyses or enough detail to evaluate the impact of this assumption.

11. Confidence Intervals and Precision

Precision is best judged from the confidence interval alongside the point estimate. For vortioxetine 2.5 mg versus placebo, the sponsor-reported least-squares mean difference was -0.87, with a 95% confidence interval from -2.45 to 0.71. For vortioxetine 10 mg versus placebo, it was -0.81, with a 95% confidence interval from -2.36 to 0.74. Both intervals include zero and extend to values on either side of it.

In general, a confidence interval shows the range of values compatible with the data and the assumptions of the analysis at its stated confidence level. These intervals leave uncertainty about whether either dose had a smaller or larger mean reduction than placebo. They also show that the point estimates should not be treated as exact effects. The record does not report a clinically important difference threshold, so the clinical importance of values within the intervals cannot be classified from these results alone.

The sponsor-reported p-values, 0.279 and 0.306, are consistent with the intervals including zero, but p-values do not describe the range or magnitude of uncertainty. The posted arm summaries include standard errors, which quantify uncertainty in the estimated least-squares means rather than the range of individual participants' changes. The registry record does not provide individual-level distributions or additional interval estimates for the primary arm values.

Precision also depends on the analysis model and the handling of missing data. The reported primary estimates come from the stated ANCOVA and use LOCF. The registry record does not report sensitivity analyses to show how estimates might change under other missing-data assumptions. Thus, the intervals quantify uncertainty under the reported analysis, not every source of uncertainty about the result.

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 sample-size display is a planning illustration based on the observed effect, not a report of the trial's original design calculation. For these continuous and binary outcomes, the displayed figures are participants per arm, not total participants or total events. The assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and a target effect equal to the stated fraction of the observed effect. The registry record does not report the assumptions used when the trial was originally designed.

The 80% power column gives the planned number of participants per arm for a hypothetical study to have an 80% chance of detecting the specified effect under those assumptions, if that effect and the model assumptions were correct. The 90% power column describes the corresponding planning target for a 90% chance. Power is a design property under assumed conditions; it is not a measure of the probability that a result is true.

Rows for smaller fractions of the observed effect call for more participants per arm. This is the expected general pattern: a smaller difference is more difficult to distinguish from random variation at the same significance level and power target. Planning values based on an effect estimated in a single trial can be unstable, particularly when the observed effect is small. They should be treated as illustrations for assumptions, not as a recommended design or a guarantee of future results. These figures are not observed power.

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 = -0.880; pooled standard deviation σ = 6.970.
2. n = 2 × (1.96 + 0.842)² × 6.970² / -0.880² = 984.7, rounded up to 985 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 <- -0.880000000000001                      # observed difference in means
sigma <- 6.96961406144682                     # 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(731120)
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 = 985 (80% power), 1,319 (90% power)
 75% of observed effect: n per arm = 1,751 (80% power), 2,344 (90% power)
 50% of observed effect: n per arm = 3,939 (80% power), 5,273 (90% power)
power.t.test: 985.6 
Simulated power at n = 985 per arm: 0.802 (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 analyses as ANCOVA comparisons of least-squares means. For vortioxetine 2.5 mg versus placebo, the reported difference is -0.87, with a 95% confidence interval from -2.45 to 0.71 and p = 0.279. For vortioxetine 10 mg versus placebo, the reported difference is -0.81, with a 95% confidence interval from -2.36 to 0.74 and p = 0.306.

The simple recalculations from the posted summary statistics give mean differences of -0.88 and -0.81, respectively. They are unadjusted comparisons and do not reproduce the ANCOVA adjustment for treatment, center, and baseline value. Small differences between an unadjusted calculation and the sponsor's model-based result can arise because they use different analysis methods. The sponsor's own analysis takes precedence.

The registry record does not report sponsor analyses for the secondary outcomes with detailed summaries in the record. The recalculated secondary comparisons should therefore be read as descriptive, unadjusted checks rather than trial-reported inferential analyses. They do not replace an analysis specified and conducted by the sponsor.

15. Multiplicity, Missing Data and Analysis Populations

The registry record reports one primary outcome and two dose-versus-placebo comparisons. It does not report whether the trial's statistical plan adjusted for the number of primary comparisons, or what testing sequence or decision rule was used. Because more than one dose is compared with placebo, multiplicity is relevant: in general, conducting several tests can increase the chance of at least one apparently unusual result under a setting where none of the tested effects exists.

Several secondary outcomes are also listed, and the record includes detailed summaries for some of them. The record does not report a multiplicity adjustment for secondary analyses or a hierarchy that would define which outcomes were confirmatory and which were exploratory. The secondary recalculations are unadjusted. Their p-values should therefore not be treated as if they arose from a single isolated test.

The primary analysis population is defined as randomized participants who received at least 1 dose and had at least 1 post-baseline primary efficacy value. The record says LOCF was used. In general, carrying the last observed value forward can simplify an analysis but may not reflect an unobserved outcome after a participant's last assessment. Its validity depends on assumptions about how missing outcomes relate to earlier observations and subsequent change. The registry record does not report the reasons for missing data, the amount of missingness by visit, or sensitivity analyses using other assumptions.

The record does not report an interim-analysis plan, stopping boundaries, a detailed missing-data analysis plan, or the original sample-size justification. Their absence from the record does not establish that they were absent from the full study documents; it means they cannot be evaluated from the registry record. These gaps limit assessment of the full testing strategy and the robustness of the reported estimates.

16. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: inguinal hernia (0.7%), pyrexia (0.7%), abortion spontaneous (0.7%), vertigo positional (0.7%).

The registry reports serious adverse events for the safety groups. One serious adverse event was reported among 153 participants at risk in the placebo group, 2 among 151 at risk in the vortioxetine 2.5 mg group, and 1 among 152 at risk in the vortioxetine 10 mg group. The registry also lists four serious-event terms at 0.7% each: inguinal hernia, pyrexia, spontaneous abortion, and positional vertigo.

These counts describe events recorded in the trial; they do not establish that study treatment caused an event. Small counts are imprecise and cannot characterize uncommon risks reliably. The record also reports other adverse events, but these serious-event figures alone do not describe the overall tolerability profile. Death counts are not reported in the registry record. Treatment was planned for up to 8 weeks, so the safety information should be read in the context of the reported study period rather than as evidence about longer-term experience.

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

The central statistical result is a comparison of average HAM-A change at Week 8, not a test of whether participants improved at all. Each group had a negative reported mean change from baseline. The sponsor's adjusted comparisons estimated somewhat more negative changes in both vortioxetine groups than in placebo, but each 95% confidence interval included zero. The results therefore do not establish a clear between-group difference on the primary outcome.

That conclusion is narrower than saying the treatments are equivalent or that no effect exists. A nonsignificant result does not prove equality, and the confidence intervals remain compatible with differences in either direction. The registry record does not report an equivalence or noninferiority framework, a clinical-importance threshold, or a prespecified margin that would support those kinds of conclusions.

The primary measure captures rated anxiety symptoms across 14 HAM-A items. It does not by itself describe every aspect of a participant's health, functioning, or experience. Secondary outcomes provide other perspectives, but their detailed results are limited in the registry record and the available recalculations are unadjusted. They should not be combined into a single claim about overall benefit.

The findings are limited to the reported Week 8 comparison in the defined analysis population. The registry record does not provide a basis for conclusions about effects after treatment, longer-term safety, or other populations. The use of LOCF and the absence of reported sensitivity analyses also leave uncertainty about how missing data might affect the estimates. These results are statistical evidence about the posted trial outcomes, not medical advice or an individual treatment recommendation.

18. Why This Trial Matters Statistically

This trial illustrates why within-group change and between-group comparison must be kept separate. The HAM-A score fell on average in all three groups, but the primary question concerns whether change differed between each vortioxetine dose and placebo. Improvement from baseline in one arm, considered alone, does not identify how much change is attributable to treatment rather than other influences over the trial period.

It also demonstrates why a point estimate cannot stand alone. The sponsor-reported differences are modest in the original scale units, and their confidence intervals include zero. The interval communicates uncertainty that a single estimate and p-value cannot convey: data compatible with the analysis include no difference and differences in either direction. This is not evidence of equivalence; it is a reason to avoid overstating precision.

The registry's comparison of two doses, multiple secondary outcomes, and reported recalculations highlights the importance of a prespecified statistical plan. Adjustment for baseline and center affects the primary estimate, while unadjusted secondary comparisons answer a less controlled set of questions. Reporting the analysis population and LOCF approach also matters because the handling of missing outcomes can shape estimates.

Finally, planning calculations based on observed effects should not be confused with the trial's original design or with observed power. They show how assumptions about target effect, significance level, allocation, and desired power influence hypothetical sample sizes. The original design assumptions are not reported in the registry record, so the planning illustrations cannot explain why this trial enrolled its reported number of participants.

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 primary results concern a short treatment period, with dosing described as lasting up to 8 weeks. The registry record does not report longer-term outcomes, so the primary comparison cannot address persistence of effects after Week 8. The listed primary analysis counts are smaller than the enrollment figure, and the record defines the analysis population using receipt of study drug and availability of a post-baseline primary efficacy assessment. Details of participant flow and reasons for exclusion or missing assessment are not reported in the registry record.

The primary analysis used LOCF, but the record does not provide missing-data patterns, reasons for missingness, or sensitivity analyses under alternative assumptions. This limits assessment of how dependent the results are on the imputation method. The ANCOVA model is described at a high level, but model diagnostics and further specifications are not reported in the registry record.

The sponsor's primary analysis is available, while the detailed secondary comparisons do not include sponsor-reported analyses. Recalculated results are unadjusted and do not reproduce the primary ANCOVA. The record also does not report the multiplicity plan for the dose comparisons and secondary outcomes, the original sample-size justification, or an interim-analysis plan. These omissions constrain interpretation of the complete statistical strategy.

Finally, the registry provides group-level summaries rather than individual outcome distributions. Group means and standard errors do not reveal the full range of participant responses. Serious adverse event counts are small and provide limited precision for characterizing uncommon events. These are limits of what can be assessed from the registry record, not evidence that unreported procedures or analyses were not conducted.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study of Efficacy and Safety of Vortioxetine (Lu AA21004) in Treating Generalized Anxiety Disorder — NCT00731120.
• Registry record last updated 2013-12-18; status last verified 2013-10.
• Results first posted 2013-12-18.
• 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

The registry describes a completed Phase 3 randomized, double-blind, parallel-group, placebo-controlled study of vortioxetine 2.5 mg and 10 mg in adults with generalized anxiety disorder. Treatment was oral and once daily for up to 8 weeks. The primary endpoint was change from baseline to Week 8 in HAM-A total score, analyzed in a full analysis set with LOCF and an ANCOVA model adjusting for treatment, center, and baseline value.

All three groups had negative mean score changes. Sponsor-reported comparisons of each vortioxetine dose with placebo had 95% confidence intervals that included zero. The registry record lists secondary outcomes and serious adverse event counts, but does not report several details needed to assess the full statistical plan, missing-data robustness, multiplicity strategy, or original design assumptions.