Safety and Efficacy of Levomilnacipran ER (F2695 SR) in Adults With Fatigue Associated With Major Depressive Disorder: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy of Levomilnacipran ER (F2695 SR) in Adults With Fatigue Associated With Major Depressive Disorder (NCT01254305), a randomized phase 2 study. The primary outcome is Change in Clinical Global Impression of Severity (CGI-S) for Fatigue Score.

2. Clinical Question

The clinical question concerns a specific population: adults with major depressive disorder whose study entry included fatigue. It is not a question about fatigue in all populations or about major depressive disorder without the fatigue context. The registry describes the trial’s purpose as evaluating efficacy, safety, and tolerability of levomilnacipran ER for fatigue associated with major depressive disorder.

The three groups support comparisons with placebo and with an active treatment category. The active comparator was not a single drug: participants in that group were randomized to one of four listed SSRIs. The registry record does not report separate outcome results for each SSRI, so the active-group result represents the group as reported rather than a drug-by-drug comparison.

The outcome captures investigators’ clinical ratings of current fatigue severity. The results describe average score changes over the stated interval; they do not, by themselves, establish how individual participants changed or whether any group difference is important to patients. Those interpretations depend on the scale, uncertainty, study conduct, and analysis choices as well as the observed averages.

3. Trial Design

This was a phase 2, randomized, triple-masked, parallel-group study with placebo and active controls. The registry record lists Forest Laboratories as the sponsor and describes the study as completed. It gives the start as 2011-04 and primary completion as 2012-07. The enrollment was 262.

Participants were assigned to oral levomilnacipran ER, matching placebo, or an active-comparator group receiving one of four SSRIs: paroxetine, sertraline, citalopram, or fluoxetine. The levomilnacipran ER description gives a dose range of 40 -120 mg/day. The SSRI group is described as once-daily oral dosing. The masking level is recorded as triple, but the identities of the groups covered by masking are not reported in the registry record.

The primary outcome was change in the Clinical Global Impression of Severity for Fatigue score from baseline to Week 8. The registry describes a randomized population of 262 patients and a safety population of 251 patients who took at least one dose of double-blind treatment. It also reports an Intent-to-Treat population of 248 patients from the safety population who had a baseline and at least one postbaseline assessment. Arm-level outcome summaries use different reported group sizes, reflecting the population contributing to the outcome results rather than necessarily the randomized population.

The registry record does not report the original sample-size justification, the planned allocation ratio, detailed randomization procedures, or an analysis plan. The reported results include group means and standard deviations and recalculated unadjusted comparisons with placebo. No sponsor-reported analyses are listed. It is therefore important to distinguish the randomized design and the observed summaries from any assumptions that might have been used in the original protocol.

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: Patients went through a 1-week single-blind placebo run-in period, followed by an 8-week double-blind treatment period.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The Randomized Population consisted of 262 patients, with 251 patients who took at least 1 dose of double-blind treatment to comprise the Safety population. The Intent-to-Treat (ITT) Population consisted of 248 patients in the Safety Population who had a baseline and at least 1 postbaseline assessment of either the CGI-S or PGI-I fatigue score.

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: A total of 262 patients were randomized to receive double-blind treatment. The Baseline Participant population is based on the 251 randomized patients received at least 1 dose of double-blind treatment (Safety Population).

6. Primary Endpoint

The registry lists the primary outcome as Change in Clinical Global Impression of Severity (CGI-S) for Fatigue Score, measured over From Baseline to Week 8.

The primary endpoint was change in the Clinical Global Impression of Severity for Fatigue score from baseline to Week 8. The registry describes the CGI-S for Fatigue as a clinician-rated assessment of the patient’s current fatigue severity, based on the investigator’s clinical opinion with regard to the patient population with major depressive disorder. The scale ranges from 1 to 7, with 1 indicating a normal state and 7 indicating that the patient is among the most extremely fatigued.

A change score expresses how the group’s average recorded score shifted over the assessment interval. In the posted results, each group’s mean change is negative. Since the scale describes severity, a negative change represents a decline in the recorded score; the size and uncertainty of the difference between treatment groups and placebo are needed to interpret the comparative result. A group mean does not indicate that every participant had the same change, and it does not describe the distribution of individual responses in full.

Clinician ratings are used to capture an overall judgment of severity, rather than only a participant’s self-report. That perspective can be useful for evaluating a clinical state, although ratings may still reflect judgment and measurement variability. The registry record does not report a threshold for a clinically meaningful change, so the observed mean differences should not be translated into a claim about clinical importance. The endpoint is one measure of fatigue in this study; it does not represent every aspect of functioning, symptoms, safety, or tolerability.

7. Results Posted in the Registry

• Levomilnacipran ER: mean -1.80 versus -1.50 on control; difference -0.30 (95% CI -0.70 to 0.10), standardized effect d = -0.22; z-test p = 0.144.
• SSRI: mean -1.90 versus -1.50 on control; difference -0.40 (95% CI -0.82 to 0.02), standardized effect d = -0.30; z-test p = 0.060.

8. Secondary Efficacy Endpoints

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

Change in Cognitive and Physical Functioning Questionnaire (CPFQ), Last Observation Carried Forward

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Levomilnacipran ER: mean -7.00 versus -5.90 on control; difference -1.10 (95% CI -3.37 to 1.17), standardized effect d = -0.14; z-test p = 0.341.
  • SSRI: mean -6.40 versus -5.90 on control; difference -0.50 (95% CI -2.79 to 1.79), standardized effect d = -0.07; z-test p = 0.668.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Levomilnacipran ER 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 SSRI and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcome is change in the Cognitive and Physical Functioning Questionnaire from baseline to Week 8, analyzed using last observation carried forward. Its group means all show negative changes, and the recalculated unadjusted comparisons with placebo have intervals that include zero. This overall pattern does not provide clear evidence of separation between either active group and placebo on the posted secondary measure.

The secondary result is broadly consistent with the primary result in the limited sense that neither set of posted group comparisons clearly distinguishes the active groups from placebo. The outcomes assess different aspects of the study: the primary measure is clinician-rated fatigue severity, while the CPFQ concerns cognitive and physical functioning. Agreement in the broad pattern does not make the two outcomes interchangeable, nor does it establish that the study has ruled out effects on either construct.

Secondary results generally carry less weight than a designated primary outcome because they may be among several outcomes or comparisons examined. If each is interpreted separately without an adjustment strategy, the chance of at least one chance-positive finding can increase. The registry record lists one secondary outcome but does not report a multiplicity adjustment, a hierarchy among outcomes, or a sponsor-reported analysis for it. Its result is best treated as supportive descriptive information, not independent confirmation.

The last-observation-carried-forward designation also matters. It indicates a particular approach to missing measurements, but the registry record does not give the extent or reasons for missingness or document sensitivity analyses using alternative assumptions. The secondary result therefore reflects both the observed data and the stated imputation approach. That limitation reinforces the need to interpret it alongside, rather than in place of, the primary outcome and its uncertainty.

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

• Change in Cognitive and Physical Functioning Questionnaire (CPFQ), Last Observation Carried Forward

9. Statistical Interpretation

10. Statistical Methods Explained

What does a mean change represent?

The outcome is continuous and is summarized by a mean and standard deviation in each group. The mean is the group’s average change in score over the stated period. It provides a compact description of the center of the observed changes, but it does not imply that all participants had changes near that average. The standard deviation describes the spread of individual values around the mean. A larger standard deviation indicates more variability among the recorded changes; it is not the uncertainty in the estimated mean.

The primary outcome summaries are -1.5 with a standard deviation of 1.3 for placebo, -1.8 with a standard deviation of 1.4 for levomilnacipran ER, and -1.9 with a standard deviation of 1.4 for SSRI. The comparisons report differences in average change relative to placebo. A negative estimated difference means the treatment-group mean change was numerically more negative than the placebo mean change. The estimates do not alone establish whether that difference is distinguishable from random variation.

How should the comparisons be read?

The reported treatment contrasts are simple, unadjusted comparisons of group means. The primary estimated mean differences are -0.3 for levomilnacipran ER versus placebo and -0.4 for SSRI versus placebo. Their uncertainty intervals describe a range of values compatible with the estimate under the statistical model used for the recalculation. Because each interval includes zero, the data are compatible with no difference in average change as well as with nonzero differences in either direction within the interval.

The listed p-values are unadjusted: 0.144 for the levomilnacipran ER comparison and 0.06 for the SSRI comparison. In general, a p-value measures how unusual data at least as incompatible with a null hypothesis would be under that hypothesis and the assumptions of the test. It is not the probability that the null hypothesis is true, and it does not measure the probability that a finding will replicate. These p-values should not be treated as a ranking of treatments or as a direct measure of clinical relevance.

The reported comparisons are recalculated from posted summary statistics and are unadjusted. The registry record lists no sponsor-reported analyses. They should therefore be understood as descriptive statistical comparisons rather than as a substitute for a prespecified sponsor analysis. The registry record does not report a full analysis plan, such as covariate adjustment, model specification, or a multiplicity strategy for the primary and secondary outcomes.

Why does variability matter?

Differences between group means must be read alongside the variability of measurements and the number of observations contributing to each group summary. When individual scores vary widely, an estimated mean difference can be imprecise even if the group averages differ. The standard deviations for the primary outcome are similar in scale across the three groups, while the reported outcome group sizes are 88 for placebo, 85 for levomilnacipran ER, and 75 for SSRI. The standard error values in the comparison record quantify sampling uncertainty in the group means; they are not a measure of participant-level variation.

For a continuous outcome, the mean difference is often a useful summary when the mean is an appropriate description of the data. It can conceal skewed distributions, outliers, or subgroups with different responses. The registry record does not report distributional diagnostics or individual-level results, so those features cannot be assessed from the posted summaries alone.

What does over-dispersion mean in general?

As a general statistical concept, over-dispersion occurs when observed variation is greater than a model assumes. It matters because a model that understates variability can produce overly narrow uncertainty intervals and overly optimistic tests. The posted endpoints are continuous means summarized with standard deviations; the registry record does not report a model-based over-dispersion assessment. This concept should not be mistaken for a finding that over-dispersion occurred in this trial.

How should the secondary outcome be handled?

The CPFQ change is also reported as a continuous mean with standard deviation. Its posted analysis is labeled last observation carried forward. This is a missing-data procedure in which an earlier recorded value is carried forward to represent a later missing measurement. In general, that procedure makes an assumption about how an unobserved later value relates to the last observed one. Its suitability depends on the reason data are missing and the clinical setting; the registry record does not report enough detail to assess those assumptions.

The secondary comparisons are unadjusted and are not listed as sponsor-reported analyses. In general, examining multiple outcomes or comparisons without adjustment raises the chance that at least one result appears notable by chance. The record does not report whether the outcomes were prespecified in a hierarchy or whether any multiplicity control was planned. The secondary comparisons should therefore be treated as exploratory or descriptive rather than as independent confirmation of the primary result.

More broadly, statistical significance, precision, and clinical importance are distinct. A confidence interval describes uncertainty in an estimate; a p-value addresses compatibility with a specified null model; clinical importance requires a meaningful interpretation of the scale and an appropriate threshold. The registry record does not report a minimally important difference for the fatigue scale or CPFQ. Statistical summaries alone cannot fill that gap.

11. Confidence Intervals and Precision

A confidence interval expresses uncertainty around an estimated group difference. For the primary comparison of levomilnacipran ER with placebo, the estimated mean difference was -0.3, and the reported interval extends from -0.703 to 0.103. For SSRI versus placebo, the estimated mean difference was -0.4, with an interval from -0.817 to 0.017. Both intervals include zero, the value corresponding to no average difference between groups.

These intervals are wide enough to include values on either side of zero. They therefore do not establish a precise direction or size of the underlying average difference. The point estimates are the single best estimates in the recalculated comparisons, but the interval limits indicate that other values remain compatible with the observed data under the calculation’s assumptions. The intervals are not a range containing a fixed percentage of individual participants’ changes; they concern uncertainty in the estimated mean difference.

The interval for the SSRI comparison comes close to zero at its upper end, while still including it. That detail should not be turned into evidence of a near-significant or clinically important effect. An interval should be interpreted as a whole, with attention to both the direction and the range of plausible effects. The interval for levomilnacipran ER likewise includes no difference and possible effects in either direction.

Precision depends on both the amount of information and the variability in the measurements. The reported standard deviations indicate variation in participant changes, while the confidence intervals reflect uncertainty in the estimated comparisons. The registry record does not report a threshold for meaningful change, so the intervals cannot be judged against a registry-defined clinical benchmark. They show uncertainty in the statistical contrast, not certainty about whether a patient would benefit or experience a particular change.

12. Planning a Similar Trial: Sample Size

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

The planning figures are illustrations based on the observed effect, not a reconstruction of the trial’s original design assumptions. For these continuous outcomes, the figures are participants per arm, not total events. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect.

The 80% power column gives the illustrative number of participants per arm that would be needed for an 80% chance of detecting the assumed effect under those planning assumptions, if the effect and model assumptions held. The 90% power column uses a higher target chance of detection and therefore generally requires more participants per arm. These are design-planning quantities, not descriptions of the number enrolled in this study or evidence about the probability that the completed study detected an effect.

The rows represent planning for 100%, 75%, or 50% of the observed effect. A smaller target effect is harder to distinguish from random variation, so the planning illustration requires more participants per arm. The figures for a given endpoint and comparison can differ because the observed effects and variability differ. They should not be treated as recommended sample sizes without considering the research question, analysis plan, missingness, multiplicity, and other design features.

The registry record does not report the trial’s original sample-size justification or its original design assumptions. These figures are planning illustrations computed from the observed effect and are not observed power. Observed power is not a useful substitute for reading the estimated effect and its uncertainty; the confidence intervals give a more direct account of precision in the posted comparisons.

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 = -0.300; pooled standard deviation σ = 1.350.
2. n = 2 × (1.96 + 0.842)² × 1.350² / -0.300² = 317.9, rounded up to 318 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 <- -0.3                      # observed difference in means
sigma <- 1.35004873206391                     # 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(1254305)
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 = 318 (80% power), 426 (90% power)
 75% of observed effect: n per arm = 566 (80% power), 757 (90% power)
 50% of observed effect: n per arm = 1,272 (80% power), 1,703 (90% power)
power.t.test: 318.9 
Simulated power at n = 318 per arm: 0.793 (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. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several hypotheses, outcomes, time points, or treatment comparisons are tested. If each result is assessed separately without an adjustment or prespecified hierarchy, the chance of a chance-positive result across the set can increase. This is a general statistical concern, not evidence that any particular reported result is false. The registry record includes a primary endpoint and a posted secondary outcome, but does not report a multiplicity adjustment, a testing hierarchy, or a detailed statistical analysis plan. The reported comparison p-values are unadjusted.

The registry record describes the randomized population as 262 patients and the safety population as 251 patients who took at least one dose of double-blind treatment. It describes the Intent-to-Treat population as 248 patients in the safety population with a baseline and at least one postbaseline assessment. That definition indicates that the reported ITT group was not simply everyone randomized. The record’s description is cut off after “assessm”; further details of the population definition are not reported in the registry record.

For the CPFQ secondary outcome, the analysis is explicitly labeled last observation carried forward. The registry record does not report the extent or pattern of missing data, the reasons assessments were missing, or sensitivity analyses under alternative missing-data assumptions. The primary endpoint’s missing-data method is not reported in the registry record. Missingness can affect estimates if the participants with recorded outcomes differ systematically from those whose measurements are absent; whether that occurred cannot be determined from these summaries.

The registry record lists no sponsor-reported analyses, although it posts outcome summaries from which unadjusted comparisons have been recalculated. It does not report whether an interim analysis was planned or conducted, or the original sample-size assumptions. These omissions do not imply that a procedure was or was not used; they mean the relevant information is not reported in the registry record. Consequently, interpretation should stay close to the posted estimates, their uncertainty, and the stated analysis population, without assuming unreported design or analysis choices.

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: pneumonia (1.2%).

The serious adverse event figures are counts of participants affected among those at risk in each reported group. The record lists 0 affected among 89 at risk for placebo, 1 among 85 at risk for levomilnacipran ER, and 0 among 77 at risk for SSRI. These figures describe serious adverse events as reported; they do not establish that the treatments caused or prevented the events.

Pneumonia is listed as a top serious adverse event term, at 1.2%. The record does not provide further event-level detail in the information, such as timing, outcome, or an assessment of relationship to study treatment. Death counts are not reported in the registry record.

The serious-event counts should be considered separately from the counts for other adverse events, which are recorded as 39 among 89 at risk for placebo, 41 among 85 at risk for levomilnacipran ER, and 37 among 77 at risk for SSRI. Neither kind of count by itself summarizes severity, duration, recurrence, or causal attribution. Group sizes and follow-up can affect how event counts are interpreted, and the registry record does not report all the detail needed for a full safety comparison.

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

The posted primary results show decreases in mean fatigue-severity scores in all three groups from baseline to Week 8. The active-group averages were numerically more negative than the placebo average, but the recalculated unadjusted mean differences have confidence intervals that include zero. The posted comparisons therefore do not establish a clear average difference from placebo. They also do not demonstrate equivalence or prove that no difference exists.

The findings are limited to the population and outcome studied: adults with major depressive disorder and associated fatigue, assessed using the CGI-S for Fatigue. The registry record reports a phase 2 randomized trial with placebo and active controls, but the summaries do not explain how individual participants responded, whether particular subgroups differed, or whether the observed changes were meaningful to patients. No threshold for clinically meaningful change is reported in the registry record.

The active comparator is an SSRI group that combined participants assigned to one of four named drugs. The posted result cannot be interpreted as a separate estimate for each drug. The results also should not be generalized automatically to other populations, settings, doses, or durations. The record reports the levomilnacipran ER dose range and Week 8 outcome timeframe, but does not supply enough information to establish how results would apply under other conditions.

The secondary CPFQ summaries likewise do not establish a clear difference from placebo in the unadjusted comparisons. They provide additional descriptive information about functioning, but they do not override the primary outcome or independently settle the clinical question. Safety event counts are also descriptive and do not establish causation. Overall, the registry record supports a careful account of observed averages and statistical uncertainty, not a broad conclusion about treatment effectiveness or safety.

17. Why This Trial Matters Statistically

This trial illustrates why a randomized comparison should be read through both its estimated effects and their uncertainty. The group means alone show that average fatigue scores changed in every arm. The comparisons add context: the estimated differences from placebo are relatively small on the scale, and the confidence intervals include zero. Looking only at the direction of the averages could overstate what the data establish; looking only at the p-values would also miss the range of effects compatible with the estimates.

The design includes both placebo and an active-comparator group, which places the levomilnacipran ER comparison in a broader treatment context. At the same time, the SSRI group combines four drugs, so it cannot answer drug-specific questions. This is a useful reminder that the meaning of a comparator depends on how it is defined and how its results are reported.

The registry record also highlights the importance of analysis populations and missing-data methods. It distinguishes the randomized and safety populations from an ITT population defined using postbaseline assessment availability. The CPFQ outcome is labeled last observation carried forward, while the primary endpoint’s missing-data approach is not reported. Those distinctions affect what the summary estimates represent and how robust their interpretation may be.

Finally, the planning illustrations emphasize the connection between effect size, variability, and precision. Smaller effects require more participants per arm to detect under the stated planning assumptions. These figures do not replace a trial’s original design justification, which is not reported in the registry record. Taken together, the study provides a practical example of separating descriptive change, comparative evidence, uncertainty, and design considerations.

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.

Several trial-specific details needed for a fuller evaluation are not reported in the registry record. These include the original sample-size justification, the allocation ratio, the randomization procedures, the detailed statistical analysis plan, and any adjustment for multiple outcomes or comparisons. The record also does not state whether interim analyses were planned or performed. Their absence from the registry record should not be taken as evidence that any particular procedure was or was not used.

The reported ITT population is described as members of the safety population with baseline and at least one postbaseline assessment, which means the label does not describe all randomized patients in the usual broad sense. The record does not report the reasons that participants lacked assessments, the extent of missing primary outcome data, or sensitivity analyses. Last observation carried forward is specified for the secondary CPFQ measure, but its assumptions and robustness are not assessed in the registry summaries.

The active-comparator group combines four SSRIs and does not provide drug-specific estimates. The outcome summaries are group-level, so participant-level distributions, response patterns, and subgroups cannot be assessed. The registry record does not give a threshold for a clinically meaningful change on the fatigue scale or CPFQ, limiting interpretation of clinical relevance.

The primary comparisons are recalculated and unadjusted, with no sponsor-reported analyses listed. The posted estimates and intervals are informative for describing the observed results, but the lack of a reported analysis plan makes it difficult to evaluate whether these comparisons reproduce the intended primary analysis. As with any registry summary, details not reported cannot be reconstructed from the mean, standard deviation, and group size alone.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Safety and Efficacy of Levomilnacipran ER (F2695 SR) in Adults With Fatigue Associated With Major Depressive Disorder — NCT01254305.
• Registry record last updated 2014-08-06; status last verified 2014-08.
• Results first posted 2014-08-06.
• 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

NCT01254305 is a completed, randomized, triple-masked, parallel-group phase 2 study sponsored by Forest Laboratories. It enrolled 262 adults with major depressive disorder and associated fatigue and compared oral levomilnacipran ER, an SSRI active-comparator group, and placebo.

The primary outcome was change in clinician-rated CGI-S for Fatigue from baseline to Week 8. The mean changes were -1.5 for placebo, -1.8 for levomilnacipran ER, and -1.9 for SSRI. The recalculated unadjusted comparisons versus placebo had confidence intervals including zero. The secondary CPFQ outcome also had unadjusted comparisons whose intervals included zero.

The registry describes randomized, safety, and ITT populations, but does not report the original design assumptions, detailed statistical plan, multiplicity strategy, or primary missing-data method. The displayed planning figures are illustrations based on observed effects, not the original trial justification or observed power.