Study Comparing Discontinuation Symptoms Of DVS SR In Subjects With Major Depressive Disorder (MDD): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study Comparing Discontinuation Symptoms Of DVS SR In Subjects With Major Depressive Disorder (MDD) (NCT01056289), a randomized phase 4 study. The primary outcome is Total Discontinuation - Emergent Signs and Symptoms (DESS) Score Over the First 2 Weeks of the Double-blind Phase.

2. Clinical Question

3. Trial Design

This completed Phase 4 study enrolled 480 subjects with major depressive disorder. The registry lists a randomized, quadruple-masked, parallel-group design. The start is recorded as 2010-03 and primary completion as 2011-02. The study title and official title describe a comparison of discontinuation symptoms following 24 weeks of open-label treatment with 50 mg DVS SR, including abrupt discontinuation and a 1-week tapering regimen.

The interventions listed in the record are DVS SR 50 mg, DVS SR 25 mg, and placebo. The double-blind arm descriptions are blank, so the precise sequence of treatments and the correspondence between each listed arm and the abrupt or taper strategy are not reported in the registry record. The intended comparison can therefore be described from the official title, but the detailed arm-level treatment schedule cannot be reconstructed from the posted descriptions.

The primary outcome was total DESS score over the first 2 weeks of the double-blind phase, with Week 1 identified as Study Day 175 and Week 2 as Study Day 182. The primary analysis population included randomized participants who had at least one postrandomization DESS record. The posted primary outcome group sizes were 72 for DVS SR 50 mg, 139 for DVS SR 25 mg, and 146 for placebo. The safety denominators differ, and the reasons for those differences are not reported in the registry record.

The record also lists two secondary outcomes: the percentage of participants with taper adverse events and the percentage unable to complete tapering because of the number and/or severity of discontinuation symptoms. Their results are not included in the registry record for this analysis, so their findings 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: Eligible participants entered a 24 Week Open-label treatment phase; participants who completed the Open-label phase were randomized to a 4 Week Double-blind treatment phase.

Open-label Phase

Reasons for not completing this period:

Double-blind Phase

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set (FAS): all randomized participants who had at least 1 postrandomization DESS record. Mean was adjusted for baseline DESS score and study center.

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 Total Discontinuation - Emergent Signs and Symptoms (DESS) Score Over the First 2 Weeks of the Double-blind Phase, measured over Double-blind phase: Week 1 (Study Day 175), Week 2 (Study Day 182).

The primary endpoint is total Discontinuation-Emergent Signs and Symptoms (DESS) score over the first 2 weeks of the double-blind phase. A clinician administered the 43-item assessment. The total score counts new symptoms and old symptoms that had worsened as 1, while absent, unchanged, or improved old symptoms score 0. The possible score range is 0 to 43, and higher values represent more symptoms.

The registry describes the reported endpoint as 2 multiplied by the mean of the Week 1 and Week 2 DESS assessments. It also specifies that a new symptom appeared within 7 days before assessment, while an old symptom appeared earlier and continued into the assessment period. The endpoint is intended to quantify the symptom burden associated with discontinuation during the early double-blind period. Comparing group means offers a summary of average symptom burden, although it does not show the distribution of scores or the experience of each participant.

The primary outcome mean was adjusted for baseline DESS score and study center. This adjustment accounts for those specified factors in the mean comparison; other details of the model are not reported in the registry record. The reported outcome is continuous, so the results are expressed as means and mean differences rather than as counts of participants experiencing symptoms.

7. Results Posted in the Registry

• DVS SR 50 mg: mean 4.10 versus 5.30 on control; difference -1.20 (95% CI -2.94 to 0.54), standardized effect d = -0.19; z-test p = 0.177.
• DVS SR 25 mg: mean 4.80 versus 5.30 on control; difference -0.50 (95% CI -1.97 to 0.97), standardized effect d = -0.08; z-test p = 0.505.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: each participant has a DESS score, and the results summarize those scores with group means. A mean difference expresses the estimated separation between the average scores of two groups. Its direction depends on the order in which the groups are subtracted. Because higher DESS scores indicate more symptoms, a negative difference when the active group is stated first indicates a lower observed mean in that group than in placebo. Contrast orientation should be checked before comparing signs across analyses.

What does the adjusted analysis do?

The registry says the primary mean was adjusted for baseline DESS score and study center. In general, an analysis of covariance (ANCOVA) estimates a group comparison while accounting for specified covariates. Baseline adjustment can improve comparability in the analysis by incorporating participants’ baseline measure; adjustment for study center accounts for the centers named in the model. The exact ANCOVA specification, including any additional terms or assumptions, is not reported in the registry record.

The sponsor’s posted pairwise analyses are identified as ANCOVA analyses of mean differences in final values. They include comparisons of DVS SR 50 mg with placebo, DVS SR 50 mg with DVS SR 25 mg, and DVS SR 25 mg with placebo. Each has a 95% confidence interval. The sponsor’s p-values are not reported in the registry record, and the contrast direction is not explicitly defined there. The registry labels these analyses as non-inferiority or equivalence analyses, but does not report the associated margin or decision rule.

What are the recalculated comparisons?

The recalculated estimates use posted group summary statistics to form simple, unadjusted comparisons. They are not the sponsor’s adjusted ANCOVA estimates and do not replace them. For DVS SR 50 mg versus placebo, the recalculated mean difference is -1.2, with p = 0.177. For DVS SR 25 mg versus placebo, it is -0.5, with p = 0.505. These recalculated p-values are unadjusted; they should not be read as the sponsor’s inferential results or as a substitute for the registered analysis.

A p-value describes how compatible the observed data are with a specified null hypothesis under the analysis assumptions. It is not the probability that the null hypothesis is true, nor does it measure clinical importance. The unadjusted values do not account for the set of outcomes and comparisons as a whole. The registry record does not report a multiplicity adjustment or a prespecified testing hierarchy.

Why does a confidence interval matter?

A confidence interval gives a range of estimates compatible with the data and the model at its stated confidence level. It communicates uncertainty and precision as well as the estimated direction. When a mean-difference interval includes zero, the data are compatible with no mean difference under that analysis, as well as with differences in either direction within the interval. This does not prove that the groups are identical. For equivalence or non-inferiority, interpretation also depends on a prespecified margin and a suitable analysis plan; those details are not reported in the registry record.

The posted primary group means are described as adjusted for baseline DESS and study center, while the posted standard errors are associated with those means. The recalculation uses the summary statistics and is explicitly unadjusted. A simple recalculation cannot reproduce an adjusted ANCOVA unless all relevant model details and data are available. Differences between estimates, intervals, or p-values can therefore arise from differences in adjustment and analytical approach.

What is the analysis population?

The primary analysis used the Full Analysis Set, defined as all randomized participants with at least one postrandomization DESS record. This definition excludes randomized participants without such a record from the primary analysis, but the number and reasons for missing postrandomization DESS records are not reported in the registry record. The study’s method for handling missing outcome values, if any, is likewise not reported. The available outcome group sizes should be understood as analysis counts, not assumed to be the full enrollment in each arm.

Finally, this endpoint averages symptom scores across the stated Week 1 and Week 2 assessments using the registry’s stated calculation. A mean can make group-level summaries easy to compare, but it may conceal variation between participants or across individual symptoms. The registry record does not report distributions, item-level results, or additional diagnostics needed to assess those features.

10. Confidence Intervals and Precision

The sponsor-reported 95% confidence intervals are broad enough to include zero for each listed pairwise comparison. For DVS SR 50 mg versus placebo, the estimate is 1.16 and the interval runs from -0.51 to 2.83. For DVS SR 50 mg versus DVS SR 25 mg, the estimate is 0.66, with an interval from -1.03 to 2.35. For DVS SR 25 mg versus placebo, the estimate is 0.50, with an interval from -0.88 to 1.89. These intervals indicate uncertainty about the size and direction of the mean differences under the reported ANCOVA analyses.

The unadjusted recalculations also have intervals that include zero. The recalculated mean differences are -1.2 for DVS SR 50 mg versus placebo and -0.5 for DVS SR 25 mg versus placebo. The posted recalculated upper confidence limits contain more than 3 decimal places, so they are not reproduced. The unadjusted p-values are 0.177 and 0.505, respectively. These are distinct from the sponsor analyses, for which p-values are not reported in the registry record.

Precision depends in part on the number of participants contributing outcome data and on variability in the endpoint. The posted primary analysis counts differ by arm, and the mean standard errors are 0.72 for DVS SR 50 mg, 0.54 for DVS SR 25 mg, and 0.52 for placebo. Standard errors describe uncertainty around the estimated means, not the spread of individual participant scores. The registry record does not provide individual-level results or enough information to assess whether the intervals capture all relevant sources of uncertainty.

Intervals that include zero are not proof of no effect or equivalence. A clinically meaningful interpretation would require context about the acceptable difference in DESS score and, for the sponsor’s stated non-inferiority or equivalence analyses, the prespecified margin. That margin is not reported in the registry record.

11. Planning a Similar Trial: Sample Size

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

The planning figures are participants per arm for the pairwise comparisons shown, not total events. They are illustrations based on the observed effect, rather than the trial’s original design assumptions. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning scenario uses the stated fraction of the observed effect as the effect to detect. The trial’s own original sample-size justification and design assumptions are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm for a design with an 80% chance of detecting the specified effect under the stated planning assumptions, if that effect and those assumptions hold. The 90% power column expresses the corresponding planning target for a 90% chance. These are design calculations, not statements about the chance that the completed trial detected an effect or about the probability that a result is true.

For the DVS SR 50 mg versus placebo comparison, planning for 100% of the observed effect gives 423 participants per arm for 80% power and 566 for 90% power. Planning for 75% of the observed effect gives 752 and 1,006 participants per arm, respectively. Planning for 50% of the observed effect gives 1,691 and 2,264 participants per arm, respectively. The corresponding figures for DVS SR 25 mg versus placebo are 2,512 and 3,362 for 100% of the observed effect; 4,465 and 5,977 for 75%; and 10,045 and 13,447 for 50%.

Smaller effects require more participants to reach the same power target under the same assumptions. These are planning illustrations, not observed power. They do not establish what sample size the original investigators intended or what effect would be clinically important.

12. 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 = -1.200; pooled standard deviation σ = 6.227.
2. n = 2 × (1.96 + 0.842)² × 6.227² / -1.200² = 422.6, rounded up to 423 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 <- -1.2                      # observed difference in means
sigma <- 6.2265964783101                     # 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(1056289)
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 = 423 (80% power), 566 (90% power)
 75% of observed effect: n per arm = 752 (80% power), 1,006 (90% power)
 50% of observed effect: n per arm = 1,691 (80% power), 2,264 (90% power)
power.t.test: 423.6 
Simulated power at n = 423 per arm: 0.804 (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.

13. Analyses Reported by the Sponsor

The sponsor’s posted comparisons use ANCOVA and report mean differences with 95% confidence intervals. They are distinct from the unadjusted recalculations based on the posted summary statistics. The sponsor estimates are 1.16 for DVS SR 50 mg versus placebo, 0.66 for DVS SR 50 mg versus DVS SR 25 mg, and 0.50 for DVS SR 25 mg versus placebo. Their confidence intervals all include zero. Sponsor p-values are not reported in the registry record.

The recalculated estimates are -1.2 for DVS SR 50 mg versus placebo and -0.5 for DVS SR 25 mg versus placebo, with unadjusted p-values of 0.177 and 0.505. They are simple unadjusted comparisons, while the sponsor analyses are adjusted ANCOVA results. Also, the registry does not define the order used to calculate the sponsor contrasts, so the signs should not be compared without caution. Adjustment and contrast orientation can affect how the estimates are expressed.

The discrepancy indicator is false, so the registry entries do not flag a discrepancy requiring the sponsor’s analysis to take precedence over a recalculation. Even so, the sponsor’s reported analysis remains the registry’s stated adjusted analysis, while recalculations are supplementary. The non-inferiority or equivalence label does not by itself show that either criterion was met: the margin and decision rule are not reported in the registry record.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary DESS outcome and two secondary outcomes concerning taper adverse events and inability to complete tapering because of discontinuation symptoms. It also posts three ANCOVA pairwise comparisons for the primary outcome. The registry record does not report a multiplicity adjustment, a testing hierarchy, or whether any comparisons were designated as confirmatory versus exploratory. The unadjusted recalculated p-values therefore should not be interpreted as controlling the overall false-positive risk across outcomes or comparisons.

The Full Analysis Set for the primary outcome includes randomized participants with at least one postrandomization DESS record. The registry record does not state how many randomized participants lacked such a record or why. It also does not describe imputation, a missing-data model, sensitivity analyses, or another method for handling missing DESS values. The outcome group counts are 72, 139, and 146, but the reasons these analysis counts differ from enrollment and from safety denominators are not reported in the registry record.

The primary mean is stated to be adjusted for baseline DESS score and study center. The complete model specification, including the method for estimating standard errors and any assessment of model assumptions, is not reported in the registry record. The sponsor’s ANCOVA contrasts are listed as non-inferiority or equivalence analyses, but the non-inferiority or equivalence margin and the formal decision criteria are not reported. Without those details, the estimates and confidence intervals can be described, but the registry record does not support an independent assessment of whether a formal non-inferiority or equivalence conclusion was reached.

These omissions limit assessment of how missingness, multiple comparisons, and model choices affect interpretation. They do not establish that the analyses were inappropriate; they mean the relevant methods are not described in the registry record.

15. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: coronary artery disease (0.2%), dehydration (0.2%), arthritis (0.2%), muscle twitching (0.2%), intentional self-injury (0.2%).

The safety record lists 5 serious adverse events among 480 participants in the open-label DVS SR 50 mg phase. In the double-blind phase, the serious adverse event count is 0 among 73 participants in the DVS SR 50 mg group, 0 among 140 in the DVS SR 25 mg group, and 0 among 148 in the placebo group. These are the posted counts and denominators; no rate has been calculated. Death counts are not reported in the registry record.

The registry also lists other adverse events affecting 197 of 480 participants in the open-label phase, and 9 of 73, 18 of 140, and 33 of 148 participants in the respective double-blind groups. Those are not serious adverse event counts. The serious-event terms listed include coronary artery disease, dehydration, arthritis, muscle twitching, intentional self-injury, and suicidal ideation, each at 0.2%.

These figures describe events recorded in the registry and should be read with their phase-specific denominators. The double-blind denominators differ from the primary DESS analysis counts. The record does not provide further detail about event timing, attribution, or the reasons for differences in participants at risk. The counts alone do not establish whether an event was caused by a study treatment.

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

The posted results show average DESS scores of 4.1 for DVS SR 50 mg, 4.8 for DVS SR 25 mg, and 5.3 for placebo during the stated primary outcome period. Because higher DESS scores indicate more discontinuation-emergent symptoms, these means are descriptively lower in the DVS SR groups than in placebo. The unadjusted recalculations estimate mean differences of -1.2 and -0.5 for the respective active groups compared with placebo, but both recalculated confidence intervals include zero and their unadjusted p-values are 0.177 and 0.505.

The sponsor’s adjusted ANCOVA comparisons also have confidence intervals that include zero. This means the posted intervals remain compatible with no mean difference under those analyses, as well as with differences in either direction within their limits. It does not prove that the groups were equivalent, that no difference exists, or that an observed difference is clinically important. The registry’s non-inferiority or equivalence labels cannot be interpreted as a successful formal conclusion without the margin and decision rule, which are not reported in the registry record.

The outcome is a clinician-administered symptom score averaged across the first 2 weeks of the double-blind phase. It does not describe every symptom, the course after that period, or each participant’s experience. The arm descriptions do not clarify which listed group corresponds to abrupt discontinuation or to the 1-week taper. Thus, the results can be reported for the named DVS SR and placebo groups, but the detailed treatment-regimen mapping is not reported in the registry record.

These results are statistical summaries of the posted study data, not medical advice. They do not establish an individual outcome, provide a treatment recommendation, or show that one discontinuation approach is preferable.

17. Why This Trial Matters Statistically

This study raises several statistical issues that are common in clinical research on discontinuation: the choice of a symptom endpoint, the distinction between adjusted and unadjusted estimates, and the interpretation of non-inferiority or equivalence analyses. The DESS endpoint is a continuous score with a defined direction, so the group means and mean differences provide a direct summary of average symptom burden. At the same time, a mean can conceal variation in participant-level scores and symptom patterns.

The posted primary means are adjusted for baseline DESS score and study center, while the recalculated contrasts are unadjusted. Keeping those analyses distinct matters because they answer related but not identical statistical questions. Comparing their numerical estimates also requires attention to the direction of each contrast. The registry does not specify the sponsor contrast orientation, so sign comparisons are uncertain without further analysis details.

The sponsor labels its analyses as non-inferiority or equivalence analyses. Such claims depend on a prespecified margin and a decision rule; neither is reported in the registry record. Confidence intervals that include zero alone do not establish equivalence or non-inferiority. The posted results therefore illustrate why an analysis label and a confidence interval are not enough to reconstruct a formal conclusion.

Finally, the record leaves important methods unreported, including missing-data handling, multiplicity procedures, and the original sample-size justification. The planning illustrations show how required sample size changes as the assumed effect becomes smaller, but they are not the original trial plan and are not observed power. These distinctions help readers separate what the posted estimates show from what cannot be concluded from the available statistical details.

18. 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 leaves the mapping between the listed double-blind arms and the official title’s abrupt-discontinuation and 1-week tapering approaches unclear. The arm descriptions are blank. This prevents a precise interpretation of the named arms as specific discontinuation regimens from the posted descriptions alone.

The primary analysis population is defined as randomized participants with at least one postrandomization DESS record, but the number of participants excluded for lacking such a record and the handling of missing outcomes are not reported in the registry record. The outcome group sizes also differ from the phase-specific safety denominators. The record does not explain these differences.

The primary results are adjusted for baseline DESS and study center, but the full model specification is not reported. The sponsor analyses are labeled non-inferiority or equivalence analyses, but the margin, decision criteria, and p-values are not reported. Multiplicity methods and a hierarchy for the primary and secondary outcomes are also not reported. These details are important for evaluating the formal interpretation of the posted analyses.

The registry gives group means and standard errors but not individual-level score distributions, symptom-level findings, or additional diagnostics. A group mean may not represent the range of experiences within an arm. The posted endpoint covers the first 2 weeks of the double-blind phase; longer-term results are not reported in the registry record. The original sample-size justification and any interim analyses are also not reported in the registry record.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Study Comparing Discontinuation Symptoms Of DVS SR In Subjects With Major Depressive Disorder (MDD) — NCT01056289.
• Registry record last updated 2012-02-28; status last verified 2011-10.
• Results first posted 2012-02-17.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

This completed Phase 4 randomized, quadruple-masked, parallel-group study enrolled 480 subjects with major depressive disorder. The official title describes 24 weeks of open-label 50 mg DVS SR followed by a comparison of abrupt discontinuation and a 1-week tapering regimen.

The primary outcome was total DESS score over the first 2 weeks of the double-blind phase. Posted means were 4.1 for DVS SR 50 mg, 4.8 for DVS SR 25 mg, and 5.3 for placebo. The unadjusted recalculated mean differences versus placebo were -1.2 and -0.5; the sponsor’s adjusted ANCOVA confidence intervals for all posted comparisons include zero.

The registry does not report the arm-to-regimen mapping, missing-data methods, multiplicity procedures, non-inferiority or equivalence margin, or original sample-size justification. Those omissions limit how far the posted statistical findings can be interpreted.