Efficacy and Safety of Eslicarbazepine Acetate as Therapy in Subjects With Fibromyalgia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of Eslicarbazepine Acetate as Therapy in Subjects With Fibromyalgia (NCT01820585), a randomized phase 2 study. The primary outcome is Absolute Change From Baseline to Endpoint in Mean Pain.

2. Clinical Question

3. Trial Design

This was a randomized, double-blind, placebo-controlled, parallel-group, multicentre, multinational Phase II study in people with pain due to fibromyalgia syndrome. The registry records an enrollment of 528. Participants were allocated in a 1:1:1:1 ratio to placebo, eslicarbazepine acetate 400 mg once daily, 800 mg once daily or 1200 mg once daily. The study began in 2009-04 and primary completion was recorded in 2010-09; its status is completed.

The parallel-group design assigned each participant to one group rather than exposing each person to multiple groups in sequence. Random allocation is intended to make the groups comparable on average at the start of a trial, while masking is intended to reduce the influence of treatment knowledge on participants’ reports and study conduct. The registry identifies the study as double-blind, but does not specify which parties were masked or how masking was maintained.

The primary outcome was a continuous measure: absolute change in mean pain from baseline to endpoint. Pain intensity was rated daily on an 11-point Numeric Pain Rating Scale and recorded in a diary upon awakening. The stated endpoint timeframe was baseline and 13 weeks. The registry describes the main analysis population as the full analysis set, which included randomized subjects who received at least one dose of study medication and had at least one post-randomization rating of 24-hour average pain.

The registry does not report the original sample-size justification, the planned number of participants in each arm, or details of the allocation procedure beyond the ratio. Details such as site count, inclusion and exclusion criteria, treatment duration beyond the endpoint timeframe, and whether any interim analyses were planned 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: After completing procedures at V1, subjects returned to the study centre for V2. At V2, subjects, who had completed at least 4 subject diary pain assessments satisfactorily within the past 7 days, had an average pain score that was ≥4 and ≤9 and continued to meet all study entry criteria, were randomly assigned to 1 of the 4 treatment groups.

Overall Study

Reasons for not completing this period:

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 Absolute Change From Baseline to Endpoint in Mean Pain, measured over Baseline and 13 Weeks.

The primary endpoint was absolute change from baseline to endpoint in mean pain intensity, measured on an 11-point Numeric Pain Rating Scale. The scale ran from 0, meaning no pain, to 10, meaning worst possible pain. Participants recorded pain in a diary each morning upon awakening. The outcome description specifies a baseline and 13-week timeframe.

A change score expresses how a participant’s recorded pain level at endpoint differs from baseline. The reported group means are negative, indicating a decrease in the scale value from baseline on average; the size and direction of a group mean alone do not establish whether groups differed statistically. The comparison of interest is the difference in change between each eslicarbazepine acetate dose group and placebo.

Using a continuous outcome retains information about the amount of change on the rating scale rather than reducing the outcome to a yes-or-no category. General statistical context: an analysis of change can account for baseline values when estimating treatment-group differences, but the precise model specification, including any covariates, is not reported in the registry record. The registry identifies ANCOVA as the method for the primary analysis and states that it found no statistically significant difference between any of the three dose groups and placebo after 13 weeks. The registry does not report a clinically important difference threshold for this scale, so the clinical importance of a particular numerical change cannot be assessed from the record alone.

7. Results Posted in the Registry

• ESL 400 mg: mean -0.80 versus -1.00 on control; difference 0.20 (95% CI -0.28 to 0.68), standardized effect d = 0.10; z-test p = 0.414.
• ESL 800 mg: mean -1.20 versus -1.00 on control; difference -0.20 (95% CI -0.68 to 0.28), standardized effect d = -0.10; z-test p = 0.411.
• ESL 1200 mg: mean -0.80 versus -1.00 on control; difference 0.20 (95% CI -0.28 to 0.68), standardized effect d = 0.10; z-test p = 0.415.

8. Statistical Interpretation

9. Statistical Methods Explained

General statistical context: the aim of a randomized comparison is to estimate how outcomes differ between assigned groups while reducing the influence of baseline differences. Randomization supports that aim, but it does not guarantee that every measured or unmeasured characteristic is exactly balanced. The reported result therefore depends on both the trial design and the analysis population and model.

What does the primary analysis test?

The registry says the primary efficacy analysis used ANCOVA on the full analysis set. ANCOVA is a model-based method that compares a continuous outcome between groups while accounting for one or more covariates. General statistical context: adjusting for baseline outcome values can improve precision when baseline and follow-up measurements are related. However, the registry record does not specify the covariates in the model, how baseline pain was represented, whether any additional adjustment was used, or the exact model contrasts. Those details are not reported in the registry record.

The stated primary outcome is absolute change from baseline to endpoint in mean pain. The registry’s description says the ANCOVA found no statistically significant differences between the three eslicarbazepine acetate groups and placebo after 13 weeks. The numerical ANCOVA estimates, their confidence intervals and p-values are not reported in the registry record. The sponsor’s stated analysis is therefore the central inferential result available for the primary endpoint.

What do the recalculated comparisons represent?

The arm means and standard errors in the registry permit simple, unadjusted comparisons of each dose group with placebo. These recalculated comparisons use the posted summary statistics rather than the sponsor’s adjusted ANCOVA model. Their estimated mean differences are 0.2 for 400 mg, -0.2 for 800 mg and 0.2 for 1200 mg, with unadjusted p-values of 0.414, 0.411 and 0.415, respectively. The corresponding unadjusted confidence intervals all include zero. They are useful as a descriptive check on the magnitude and uncertainty implied by the summaries, but they cannot reproduce an adjusted analysis without the model specification and participant-level information.

The pooled standard deviations accompanying these comparisons are 1.976, 1.979 and 1.974 for the respective comparisons. These were approximated from the posted standard errors. General statistical context: a standard deviation describes variability among individual measurements, while a standard error describes uncertainty in an estimated mean. They answer different questions. An estimated standard deviation derived from standard errors is approximate, so it should not be confused with a directly reported participant-level standard deviation.

How should p-values be understood?

A p-value describes how incompatible the observed result is with a specified null model, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, nor is it a measure of the size or clinical importance of a difference. The unadjusted p-values are all above conventional thresholds often used for statistical significance, but they are not the sponsor’s ANCOVA p-values. The registry’s explicit statement about the primary analysis is that no statistically significant differences were found.

General statistical context: a confidence interval combines an estimate with a range of values that are compatible with the data under the model and its assumptions. When a confidence interval for a difference includes zero, the data do not rule out no difference at the corresponding confidence level. Such an interval may also include differences in either direction. The intervals for the unadjusted calculations include zero, and their width indicates that the estimates do not pinpoint a single value. The exact ANCOVA confidence intervals are not reported in the registry record.

Why can over-dispersion matter?

General statistical context: over-dispersion occurs when observed variability is greater than a model assumes. It is especially relevant to count outcomes when a simple model assumes variability tied closely to the mean. This primary endpoint is a continuous pain rating, and the registry does not describe an over-dispersion assessment or a count-outcome model. It would therefore be inappropriate to infer that over-dispersion was assessed or affected this analysis. More generally, model assumptions matter because incorrect assumptions can affect standard errors, confidence intervals and tests.

What is known about the outcome distribution?

The endpoint is based on daily ratings on a bounded 0-to-10 scale, and the reported statistic is a least-squares mean. General statistical context: a least-squares mean is a model-based mean, often adjusted for terms included in the model; its interpretation depends on the specification. The registry lists the primary parameter as a least-squares mean and dispersion as standard error, but does not report the specific ANCOVA model or adjusted least-squares means by group. The arm values and standard errors posted alongside the result should not be assumed to be the full adjusted output of that ANCOVA.

Repeated diary observations create a further analysis question: the registry identifies the endpoint as mean pain change but does not state how daily ratings were combined into the baseline and endpoint means, how many diary days were required, or how incomplete diaries were handled. These decisions can influence both the estimate and its precision. The available description establishes that pain was recorded daily upon awakening and that participants needed at least one post-randomization rating to enter the full analysis set; further diary-processing rules are not reported in the registry record.

10. Confidence Intervals and Precision

Precision concerns how narrowly the data locate a treatment-group difference. The unadjusted estimates are 0.2 units for 400 mg versus placebo, -0.2 units for 800 mg versus placebo and 0.2 units for 1200 mg versus placebo. Each recalculated confidence interval crosses zero, so each is compatible with no difference and also allows values on either side of zero. The intervals do not establish a precise direction of the unadjusted group differences.

The interval endpoints are not repeated; their interpretation is more important than treating a point estimate as exact. A confidence interval is not a guarantee that the true value lies within the particular interval, and it does not describe the probability that a treatment effect has a particular value. General statistical context: under the model and repeated use of the same method, a confidence-interval procedure has a stated long-run coverage property. Its practical meaning depends on assumptions, the analysis population and the method used.

The registry does not post confidence intervals for the sponsor’s ANCOVA estimates. Thus, the precision of the primary adjusted comparisons cannot be evaluated directly from the record. The recalculated intervals are unadjusted and based on summary statistics; they should not be treated as the ANCOVA intervals. No threshold for a clinically meaningful difference is reported, so the intervals cannot be classified against such a threshold from the registry information.

The standard errors associated with the arm means are close in size, but a standard error is not a measure of individual variation. The pooled standard deviations used for the simple comparisons are approximations derived from those standard errors. These calculations offer one view of uncertainty, not a replacement for a model-based precision estimate from the primary analysis.

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 illustrations based on the observed unadjusted effect for each dose-versus-placebo comparison, not the trial’s original sample-size plan. They give participants per arm, rather than a total across both arms. The assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect equal to the stated fraction of the observed effect. The trial’s own original sample-size assumptions and justification are not reported in the registry record.

The 80% power column describes the planned number of participants per arm under those assumptions for a design with an 80% chance of detecting the specified effect if that effect were the true value. The 90% power column gives the corresponding planning number per arm for a design with a 90% chance under the same assumptions. These are design probabilities conditional on the assumed effect and model, not guarantees about what a future trial will find.

The rows for planning at 100%, 75% and 50% of the observed effect show how the illustrative target changes when the effect to detect is made smaller. Smaller effects need more participants for the same nominal power and significance level. The displayed figures therefore grow as the assumed effect fraction falls. Since the observed differences in this trial are small and uncertain, these scenarios should be read as sensitivity illustrations rather than as a recommended sample size or a prediction of a future result.

This is not observed power. It is a planning exercise calculated from the observed effect, with the stated assumptions, and does not measure the probability that the completed trial detected an effect.

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 = 0.200; pooled standard deviation σ = 1.976.
2. n = 2 × (1.96 + 0.842)² × 1.976² / 0.200² = 1,532.2, rounded up to 1,533 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 <- 0.2                      # observed difference in means
sigma <- 1.97592532753644                     # 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(1820585)
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 = 1,533 (80% power), 2,052 (90% power)
 75% of observed effect: n per arm = 2,724 (80% power), 3,647 (90% power)
 50% of observed effect: n per arm = 6,129 (80% power), 8,205 (90% power)
power.t.test: 1533.2 
Simulated power at n = 1,533 per arm: 0.800 (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. Multiplicity, Missing Data and Analysis Populations

There are three dose-versus-placebo comparisons for the primary pain outcome: 400 mg, 800 mg and 1200 mg. When several hypotheses are tested, the chance of at least one false-positive finding can increase if each test is interpreted independently. General statistical context: procedures such as family-wise error control can address this issue, but the choice of procedure depends on the planned testing strategy.

The registry does not report whether the primary comparisons were adjusted for multiplicity, whether a testing hierarchy was prespecified, or whether a particular comparison was designated as the main contrast. The sponsor’s summary states that ANCOVA found no statistically significant differences between the three dose groups and placebo. The separate recalculated p-values are unadjusted and should not be interpreted as multiplicity-adjusted results. A statistical analysis plan or further detail on the testing strategy is not reported in the registry record.

The registry records one outcome measure and no posted analyses. Its primary outcome description identifies the full analysis set as all randomized subjects who received at least one dose of study medication and had at least one post-randomization rating of 24-hour average pain. This definition is narrower than every randomized participant because receipt of medication and a post-randomization rating are required. The record does not provide the number excluded from this analysis set or reasons for exclusion.

Missing-data handling is not reported in the registry record. The description does not state how missing daily diary entries, missing endpoint measurements or discontinuations were handled, whether any imputation was used, or whether sensitivity analyses were performed. It also does not report an interim-analysis plan. These omissions limit assessment of how incomplete observations and repeated testing may affect the interpretation of the primary result.

General statistical context: missing data can affect both the estimate and its uncertainty, especially when the chance of an observation being missing is related to outcome or treatment. The full analysis set definition establishes a minimum post-randomization rating requirement but does not, by itself, explain the handling of additional missing observations. No conclusion about the effect of missingness on this trial’s result can be drawn from the record.

14. 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 3% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: abdominal pain (0.8%), abortion induced (0.8%), blood creatine phosphokinase increased (0.8%), gastroenteritis (0.8%), inguinal hernia (0.8%).

The registry lists serious adverse events affecting 3 participants in the placebo group, 5 in the 400 mg group, 6 in the 800 mg group and 1 in the 1200 mg group. The corresponding numbers at risk are 131, 130, 135 and 132. These are group-level counts recorded in the safety results, not estimates of comparative risk adjusted for exposure or other factors. The registry does not provide exposure-adjusted rates or a formal statistical comparison of serious adverse events.

Other adverse events are listed as affecting 75 participants in the placebo group, 86 in the 400 mg group, 92 in the 800 mg group and 93 in the 1200 mg group. These counts should be read with their recorded denominators and do not establish whether an event was caused by study medication. Serious adverse-event terms listed include Abdominal pain, Abortion induced, Blood creatine phosphokinase increased, Gastroenteritis, Inguinal hernia and Incisional hernia; each is reported as 0.8%.

The number of deaths is not reported in the registry record. Safety counts and the primary efficacy analysis have different recorded group denominators; the registry does not explain the population differences. These figures describe what the record reports and do not, by themselves, establish a comparative safety conclusion.

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

The primary result is a comparison of mean pain-score change from baseline to 13 weeks among placebo and three eslicarbazepine acetate dose groups. The registry states that the sponsor’s ANCOVA found no statistically significant differences between the dose groups and placebo. The posted group means range from -0.8 to -1.2 units, while the simple unadjusted recalculations produce mean differences close to zero and confidence intervals that include zero.

These findings do not establish that the groups were identical. A non-significant result means that the primary analysis did not identify a statistically significant difference under its method and assumptions; it does not prove an absence of any difference. The precision of the sponsor’s adjusted estimates cannot be judged because those estimates and their confidence intervals are not reported. The unadjusted recalculations are not a substitute for the primary ANCOVA.

The record also does not establish whether any difference would be clinically meaningful. No threshold for a meaningful change in the pain scale is reported, and the sponsor’s adjusted effect estimates are unavailable. The results concern the stated pain endpoint and timeframe. They do not establish outcomes beyond the 13-week endpoint, outcomes not reported in the registry, or what an individual participant would experience.

Safety counts offer descriptive information about reported adverse events in the listed groups, but they do not provide a formal comparative analysis or establish that an event was caused by a study treatment. The registry does not report deaths, and the safety denominators differ from the primary outcome group sizes without an explanation in the record.

General statistical context: interpretation of a randomized trial depends on the design, outcome definition, analysis population, missing-data handling, multiplicity strategy and precision of estimates. Several of those details are not reported in this registry record. The appropriate conclusion is therefore limited to the recorded primary analysis statement and the descriptive and unadjusted summary results, without extending them to claims about treatment benefit, lack of benefit or clinical recommendations.

16. Why This Trial Matters Statistically

This trial is statistically informative because it places three dose groups alongside a placebo group and uses a randomized, double-blind, parallel-group design. The primary outcome is a continuous pain rating recorded repeatedly in a diary, and the sponsor analyzed change in mean pain using ANCOVA. Those features make the analysis question concrete: whether the mean change differed between each dose group and placebo over the stated 13-week timeframe.

The results also illustrate why a non-significant test is not a finding of equivalence. The reported ANCOVA did not find statistically significant differences, while the unadjusted estimates are small and their confidence intervals include zero. The intervals leave uncertainty about the direction and size of any difference. Without adjusted estimates, intervals and a clinically meaningful threshold, the registry record cannot establish that the groups are equivalent or determine the practical importance of the estimates.

The study highlights the distinction between planned analysis and post hoc calculation. The ANCOVA is the sponsor’s stated primary analysis, whereas the summary-statistic comparisons are unadjusted recalculations. Comparing their roles helps readers avoid treating a convenient recalculation as if it were the prespecified model. The registry does not report the model details or multiplicity approach, so the primary analysis cannot be independently reconstructed from the posted summary alone.

Finally, the registry shows how interpretation depends on more than a point estimate. The analysis-set definition, diary completeness, missing-data methods and the distinction between efficacy and safety populations all affect how results should be understood. Several relevant details are not reported. Recognizing these reporting limits is part of sound statistical interpretation, not a reason to infer unrecorded methods or results.

17. 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 gives a high-level description of the primary ANCOVA but does not report its adjusted estimates, confidence intervals, p-values or full model specification. The sponsor’s statement that no statistically significant differences were found is available, but the numerical output needed to assess the magnitude and precision of those adjusted comparisons is not. Recalculated comparisons from summary statistics are unadjusted and cannot fill that gap.

Important operational and analysis details are not reported in the registry record. These include the original sample-size justification, covariates in the ANCOVA, the multiplicity strategy, missing diary-data methods, endpoint diary-completion rules, and any sensitivity or interim analyses. The record defines the full analysis set but does not report how many randomized participants were excluded from it or why.

The outcome is a diary-based pain rating on a bounded scale, but the registry does not report a threshold for clinically meaningful change. The sponsor’s adjusted estimates therefore cannot be compared with such a threshold, and the practical importance of the unadjusted estimates cannot be judged from the record alone. Only the primary outcome is described; secondary outcome results are not posted.

The safety results are counts, with different denominators from the primary efficacy group sizes. The registry does not explain the differences, provide exposure-adjusted comparisons, or report deaths. The listed adverse-event counts do not establish causation. The record also does not provide enough detail to evaluate treatment adherence or the effects of discontinuation on outcomes; those details are not reported in the registry record.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Efficacy and Safety of Eslicarbazepine Acetate as Therapy in Subjects With Fibromyalgia — NCT01820585.
• Registry record last updated 2013-07-19; status last verified 2013-06.
• Results first posted 2013-07-19.
• 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.

21. Record Summary

NCT01820585 is a completed Phase II randomized, double-blind, placebo-controlled, parallel-group study of eslicarbazepine acetate in people with fibromyalgia. The registry records 528 enrolled subjects and allocation in a 1:1:1:1 ratio to placebo or 400 mg, 800 mg or 1200 mg once daily.

The primary endpoint was absolute change from baseline to endpoint in mean pain, assessed with a daily 0-to-10 Numeric Pain Rating Scale and a stated timeframe of baseline and 13 weeks. The registry says the sponsor’s ANCOVA found no statistically significant differences between the three dose groups and placebo.

Unadjusted recalculations from the posted summaries yield small mean differences, with confidence intervals that include zero. They do not replace the sponsor’s primary analysis. Several details needed to assess the analysis fully, including adjusted estimates, missing-data methods and multiplicity handling, are not reported in the registry record.