Polysomnography Study of GSK1838262 Extended Release Tablets Versus Placebo in RLS and Associated Sleep Disturbance: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Polysomnography Study of GSK1838262 Extended Release Tablets Versus Placebo in RLS and Associated Sleep Disturbance (NCT00748098), a randomized phase 3 study. The primary outcome is Adjusted Mean Change From Baseline in Wake Time During Sleep (WTDS) at Week 4/10 Measured by Polysomnography (PSG) (Sleep Study) Using Last Observation Carried Forward (LOCF).

2. Clinical Question

3. Trial Design

This was a Phase 3 randomized, triple-masked, crossover study sponsored by XenoPort, Inc. The registry lists enrollment of 136 and a completed study. The recorded start was 2008-10, and primary completion was 2009-07. The study concerned restless legs syndrome and associated sleep disturbance.

The two treatment sequences were GEn 1200 mg followed by placebo, and placebo followed by GEn 1200 mg. Each sequence therefore included both treatments, given in different intervention periods. The primary endpoint timeframe identifies Week 4 and Week 10 as the last week of each intervention period. The registry says adjusted means accounted for treatment, pooled center, and period effects. Details such as the duration between periods or whether a washout was used are not reported in the registry record.

The primary analysis is described for the ITT population of 131 participants, with the analysis of the primary endpoint excluding some participants from the second-period data. Specifically, the record notes participants who did not take the assigned treatment in the second period and participants without second-period polysomnography data. The analysis-specific participant counts are reported alongside the endpoint results. The exact allocation ratio, randomization procedure, masking procedures, eligibility criteria, and the trial’s original sample-size assumptions are not reported in the registry record.

The primary endpoint was a continuous measure: adjusted mean change from baseline in wake time during sleep. Secondary outcomes included polysomnographic measures, periodic limb movements associated with arousal, and scores from the International Restless Legs Rating Scale. The registry lists 28 outcome measures and 27 posted secondary outcomes. The presence of several outcomes matters when interpreting secondary findings, because a collection of unadjusted tests can produce findings that are less conclusive than a prespecified primary comparison.

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.

First Treatment Intervention Period

Reasons for not completing this period:

First Taper Period

Reasons for not completing this period:

1-Week Washout Period

Reasons for not completing this period:

Second Treatment Intervention Period

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: ITT Population: Of the 131 ITT participants, 2 and 6 participants did not take Placebo and GEn 1200 mg, respectively, in the second period. In addition, 6 placebo and 4 GEn 1200 participants did not have PSG data during the second intervention period and were therefore not included in this analysis.

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Adjusted Mean Change From Baseline in Wake Time During Sleep (WTDS) at Week 4/10 Measured by Polysomnography (PSG) (Sleep Study) Using Last Observation Carried Forward (LOCF), measured over Baseline, Week 4/10 (representing the last week of each intervention period, i.e. Weeks 4 and 10).

Wake time during sleep (WTDS) measures how much time a participant is awake after falling asleep and before the last awakening. It was measured with polysomnography, a comprehensive recording of physiological changes during sleep. The measure is reported in minutes, and the endpoint is the change from baseline at the end of each intervention period: Week 4 and Week 10.

Change was calculated by subtracting the baseline value from the value at the relevant endpoint week. The reported mean change was adjusted for treatment, pooled center, and period effects. Adjustment for period is relevant in a crossover design because observations occur in different treatment periods; the posted analysis accounts for period effects, but further details of the model are not reported in the registry record.

WTDS provides a quantified sleep-related outcome rather than relying only on a symptom rating. Polysomnography records sleep-related physiology under a sleep-study setting, while patient-reported scales capture participants’ reported symptoms and experiences. These measures address related but not identical aspects of restless legs syndrome and sleep disturbance. The endpoint is therefore useful for assessing a particular aspect of sleep continuity, but it is not a comprehensive measure of all sleep quality or all effects of restless legs syndrome. A change in WTDS should be interpreted as a change in this defined measure, not as a direct measure of every participant’s overall experience.

7. Results Posted in the Registry

• GEn 1200 mg: mean -32.02 versus -6.02 on control; difference -26.00 (95% CI -39.87 to -12.13), standardized effect d = -0.47; z-test p = <0.001.

8. Secondary Efficacy Endpoints

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

Adjusted Mean Change From Baseline in Periodic Limb Movements Associated With Arousal (PLMAI) at Week 4/10 as Measured by Polysomnography Using LOCF

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • GEn 1200 mg: mean -4.59 versus -1.52 on control; difference -3.07 (95% CI -5.55 to -0.59), standardized effect d = -0.31; z-test p = 0.015.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for GEn 1200 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Adjusted Mean Change From Baseline in the International Restless Legs Rating Scale (IRLS) Total Score at Week 4/10 Using LOCF

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • GEn 1200 mg: mean -14.99 versus -8.42 on control; difference -6.57 (95% CI -8.56 to -4.58), standardized effect d = -0.82; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for GEn 1200 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Adjusted Mean Change From Baseline in the Item 4 (Sleep Disturbance) Scores of the IRLS Rating Scale at Week 4/10 Using LOCF

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • GEn 1200 mg: mean -2.27 versus -1.47 on control; difference -0.80 (95% CI -1.06 to -0.54), standardized effect d = -0.77; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for GEn 1200 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes cover several related domains: polysomnographic sleep measures, periodic limb movements associated with arousal, and restless legs symptom ratings. The posted results generally show changes from baseline in a direction more negative for GEn 1200 mg than placebo across the outcomes with numerical summaries. This broad pattern is directionally consistent with the primary WTDS comparison, but the measures are not interchangeable: a symptom score, sleep-stage measure, and limb-movement measure each describe a different aspect of the condition or sleep.

The registry reports an ANCOVA comparison for the periodic limb movement measure, with an adjusted mean difference of -3.07 limb movements per hour, a 95% confidence interval from -5.04 to -1.10, and p = 0.002. The secondary symptom outcomes include the IRLS total score and its sleep-disturbance item. Their posted group changes are more negative in the GEn 1200 mg group, but the registry does not include sponsor-reported analyses for those outcomes in the record summarized by these results.

Secondary results carry less weight than a designated primary result when many outcomes are examined. Unless a multiplicity strategy is reported, repeated statistical testing can increase the chance that some comparisons appear persuasive by chance. The registry does not report a multiplicity adjustment or a hierarchy for interpreting the secondary analyses. Thus, secondary findings can add context and indicate whether other measures move in a compatible direction, but they should not be treated as independent confirmation of the primary endpoint.

The secondary pattern also does not establish that all sleep domains changed to the same extent or that each change is clinically important. The registry record lists multiple secondary outcomes and does not report a threshold for meaningful change for these measures. Their interpretation should account for the outcome-specific measurement, available data, analysis method, and the uncertainty from multiple comparisons.

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

• Adjusted Mean Change From Baseline in Periodic Limb Movements Associated With Arousal (PLMAI) at Week 4/10 as Measured by Polysomnography Using LOCF
• Adjusted Mean Change From Baseline in the International Restless Legs Rating Scale (IRLS) Total Score at Week 4/10 Using LOCF
• Adjusted Mean Change From Baseline in the Item 4 (Sleep Disturbance) Scores of the IRLS Rating Scale at Week 4/10 Using LOCF
• Adjusted Mean Change From Baseline in Time Spent in N1 Sleep as Measured by Polysomnography at Week 4/10 Using LOCF
• Adjusted Mean Change From Baseline in the Percentage of Total Sleep Time Spent in the N1 Sleep Stage at Week 4/10 Using LOCF
• Adjusted Mean Change From Baseline in Time Spent in the N2 Sleep Stage as Measured by Polysomnography at Week 4/10 Using LOCF
• Adjusted Mean Change From Baseline in the Percentage of Total Sleep Time Spent in the N2 Sleep Stage at Week 4/10
• Mean Change From Baseline in the Total Time Spent in Stage N3 Sleep Time at Week 4/10 Measured by PSG Using LOCF

9. Statistical Interpretation

10. Statistical Methods Explained

What does an adjusted mean change describe?

A mean is the average value in a group. Here, the endpoint is change from baseline in WTDS, measured in minutes. A negative mean change represents a reduction relative to baseline under the stated calculation. The adjusted means account for treatment, pooled center, and period effects, according to the registry. Adjustment means the model estimates treatment results while accounting for the listed factors; it is not the same calculation as simply subtracting two unadjusted group means.

The reported treatment contrast is the adjusted mean difference versus placebo. Its negative direction indicates that the estimated change was more negative with GEn 1200 mg than with placebo. This is a comparison on the trial’s defined endpoint and does not establish that the same direction or size applies to other outcomes.

Why use ANCOVA?

ANCOVA is a regression-based method that estimates group differences while accounting for specified covariates. The sponsor’s primary analysis is reported as ANCOVA, with treatment, pooled center, and period effects described for the adjusted mean change. In general, covariate adjustment can improve comparability and precision when the chosen model and covariates are appropriate. The precise model specification, covariate coding, and diagnostic checks are not reported in the registry record.

What does the confidence interval convey?

A confidence interval expresses the uncertainty around an estimated treatment contrast under the statistical model. A narrower interval indicates greater precision on that model’s scale; a wider interval indicates more uncertainty. The sponsor’s 95% confidence interval for the primary adjusted mean difference ranges from -35.64 to -16.36 minutes. It does not include zero, which is consistent with the reported p-value of < 0.001 for a test of no difference under the analysis. The interval does not give the probability that the true effect lies within its limits, nor does it address biases that may arise from missing observations, model assumptions, or trial conduct.

What is the role of the unadjusted recalculation?

The recalculation compares posted group summary statistics without the sponsor’s ANCOVA adjustment. It reports a mean difference of -26.0 minutes and a 95% confidence interval from -39.87 to -12.13. This calculation is a useful check on the direction and approximate scale of the group contrast, but it is not the same as reproducing the participant-level model. It cannot account for covariates or the crossover structure beyond what is contained in the posted summaries. Differences between its interval and the sponsor’s interval can arise because the methods and information used differ; the registry does not flag a discrepancy.

Why does the crossover design matter?

In a crossover study, participants receive more than one treatment across separate periods. General statistical reasoning is that within-participant comparisons can account for stable differences between participants, while period effects may arise because outcomes can differ over time. The posted primary analysis explicitly adjusted for period effects. A crossover interpretation also depends on appropriate conduct and analysis across periods. Whether a washout period was used, how sequence effects were evaluated, and whether carryover was assessed are not reported in the registry record, so those features cannot be evaluated from the posted record.

What does LOCF mean?

Last observation carried forward is a missing-data method that substitutes a participant’s last observed value for a later value that is missing. In general, this can preserve an analysis set but relies on assumptions about the unobserved outcome. If outcomes would have continued to change after the last measurement, carrying that value forward may not represent the missing value accurately. The registry identifies LOCF for the primary endpoint but does not report the reasons for missing observations or sensitivity analyses using other methods. The ITT population description also notes missing second-period treatment exposure and missing polysomnography data, making missing-data handling relevant to interpretation.

How should p-values be read?

A p-value measures how incompatible the observed data are with a specified null model, assuming that model and the analysis conditions hold. A small p-value is not the probability that the null hypothesis is true, and it does not measure clinical importance. The sponsor reports p < 0.001 for the primary ANCOVA comparison. The recalculation reports p < 0.001. These results concern the primary WTDS comparison; they do not resolve uncertainty about unreported analyses, missing-data assumptions, or the value of the measured difference to participants.

Why can over-dispersion matter generally?

Over-dispersion occurs when outcome variability is greater than a model assumes. In general, a model that understates variability may give standard errors that are too small and intervals that are too narrow. The primary endpoint is continuous, and the record reports standard errors for its group estimates. It does not report an over-dispersion assessment or diagnostic findings. This general issue should not be treated as evidence that over-dispersion occurred in this trial.

11. Confidence Intervals and Precision

The sponsor’s primary estimate is an adjusted mean difference of -26.00 minutes, with a 95% confidence interval from -35.64 to -16.36. The interval stays below zero, indicating that the model-based estimate is consistent with a greater reduction in WTDS during GEn 1200 mg treatment than during placebo treatment. Its width also shows that the exact size of the difference is uncertain: the data are compatible with a range of negative differences rather than one exact value.

The unadjusted recalculation gives a mean difference of -26.0 minutes and a 95% confidence interval from -39.87 to -12.13. Its interval also stays below zero, but it is wider than the sponsor’s interval. These are not two estimates from the same analysis: one is a recalculation from posted summary statistics, and the other is the sponsor’s ANCOVA result. Their difference in precision should therefore not be interpreted as a direct test of which analysis is correct.

Confidence intervals quantify sampling uncertainty under the assumptions of the analysis. They do not account automatically for possible bias from missing data, deviations from the model, unmeasured factors, or limitations of the trial design. Nor do they indicate whether a difference is large enough to matter to participants; the registry does not report a threshold for a clinically important WTDS change.

The sponsor’s confidence interval and p-value provide evidence against a zero difference under the reported model. Precision is not the same as certainty about generalizability: the result concerns the analyzed participants and the measured endpoint in this crossover study. Interpretation outside that setting requires information that is not reported in the registry record.

12. Planning a Similar Trial: Sample Size

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

The sample-size figures are planning illustrations based on the observed effect, not the trial’s original design calculations. For these continuous outcomes, the figures refer to participants per arm. The listed assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; each planning scenario uses the stated fraction of the observed effect as the effect to detect. The original trial’s sample-size assumptions and justification are not reported in the registry record.

The 80% power column gives the planned number of participants per arm under the stated assumptions for a design with an 80% chance of detecting an effect of the specified size, if that effect and the assumptions hold. The 90% power column gives the corresponding planning target for a design with a 90% chance. These are design probabilities, not guarantees that a future study will produce a statistically significant result.

The rows that plan for 100% of the observed effect use that observed estimate as the target effect. Rows planning for 75% or 50% of the observed effect use smaller target effects. Smaller effects are harder to distinguish from random variation, so they require more participants per arm under the same assumptions. The figures are illustrations, not recommendations: their usefulness depends on whether the assumed effect is realistic and whether the planned design and analysis match the eventual study.

This is not observed power. It does not describe the probability that the completed trial detected its effect, and it should not be used to reinterpret the observed p-value. Planning calculations based on an observed effect also inherit uncertainty in that estimate, which may not match the effect a future study should target.

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 = -26.000; pooled standard deviation σ = 55.259.
2. n = 2 × (1.96 + 0.842)² × 55.259² / -26.000² = 70.9, rounded up to 71 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 <- -26                      # observed difference in means
sigma <- 55.2590409631337                     # 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(748098)
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 = 71 (80% power), 95 (90% power)
 75% of observed effect: n per arm = 127 (80% power), 169 (90% power)
 50% of observed effect: n per arm = 284 (80% power), 380 (90% power)
power.t.test: 71.9 
Simulated power at n = 71 per arm: 0.794 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary analysis as an ANCOVA adjusted mean difference versus placebo of -26.00 minutes, with a 95% confidence interval from -35.64 to -16.36 and p < 0.001. This is the model-based result reported for the primary endpoint. The registry also reports adjusted means of -6.02 minutes for placebo and -32.02 minutes for GEn 1200 mg.

The separate recalculation from posted summary statistics gives a mean difference of -26.0 minutes, with a 95% confidence interval from -39.87 to -12.13 and p < 0.001. It is an unadjusted calculation and does not reproduce the sponsor’s ANCOVA. The point estimates are similar in direction and scale, while the intervals differ. That difference is expected to be possible when the calculations use different information and methods; the record does not flag a discrepancy.

The sponsor analysis should be read as the trial’s reported primary comparison, while the recalculation is a supplementary check based on summaries. Neither estimate by itself resolves the limitations associated with missing second-period data, LOCF, or unreported details of the model and trial conduct. The sponsor’s result addresses the specified WTDS endpoint, not every outcome or the overall experience of participants.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists 28 outcome measures, including 27 posted secondary outcomes. Testing many outcomes creates a multiplicity issue: even if none of the underlying treatment effects differs, repeated tests can produce apparently notable results by chance. Statistical approaches can control this risk, for example through a prespecified hierarchy or adjustment to significance thresholds. Whether such an approach was used for this trial is not reported in the registry record. The secondary p-values should therefore be interpreted cautiously and are not equivalent to a single prespecified primary test.

The primary endpoint is identified in the record, and its sponsor analysis is reported as ANCOVA. However, the full statistical analysis plan, any interim analysis plan, and the methods used to handle multiplicity are not reported in the registry record. The original sample-size justification and analysis plan are also not reported. The record therefore does not establish what other analyses were prespecified or how choices among outcomes were made.

The primary endpoint description specifies LOCF. In general, LOCF substitutes the last available observation when a later value is missing; it can produce biased estimates if the unobserved outcome would have changed after the last measurement. The registry does not report sensitivity analyses under alternative missing-data assumptions or the reasons observations were missing.

The ITT description states that the primary analysis involved 131 ITT participants. It also notes that 2 participants did not take placebo and 6 did not take GEn 1200 mg in the second period. In addition, 6 placebo and 4 GEn 1200 mg participants did not have second-period polysomnography data and were not included in that analysis. The record does not explain whether those categories overlap, so they should not be combined into a new total. The ITT label describes the population, but the endpoint analysis data availability differs by period. Further details about the treatment of incomplete crossover observations are not reported.

These omissions do not invalidate the reported comparison, but they limit independent assessment of analytic choices. The reported estimate should be considered alongside the stated analysis population, LOCF method, crossover structure, and the absence of reported multiplicity and sensitivity-analysis details.

16. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: cellulitis (0.8%), transient ischaemic attack (0.8%).

The serious adverse event figures are reported for the safety groups, not as a direct comparison of event rates adjusted for exposure or time. The registry lists 0 serious adverse events among 132 participants at risk in the placebo group and 2 among 127 at risk in the GEn 1200 mg group. The listed serious event terms are cellulitis and transient ischaemic attack, each reported as 0.8%.

These counts describe the events recorded in this trial and do not establish whether treatment caused an event. The registry does not report event timing, participant-level details, exposure-adjusted rates, or a causal assessment in the registry record. The death fields contain no numerical results, and deaths are not reported in the registry record. The record also lists other adverse events affecting 27 of 132 participants in the placebo group and 54 of 127 in the GEn 1200 mg group; these are not serious adverse event counts and should not be conflated with them.

Safety counts should be interpreted in light of the study’s duration, crossover structure, group-specific risk sets, and available reporting. The registry record does not provide enough detail to infer comparative safety beyond these recorded figures, and these trial data alone do not establish a general safety profile.

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

The primary result shows that the adjusted mean change in WTDS was more negative during GEn 1200 mg treatment than during placebo treatment in the analyzed trial data. The sponsor’s ANCOVA estimate was -26.00 minutes, with a 95% confidence interval from -35.64 to -16.36 and p < 0.001. This is evidence of a difference on the specified polysomnographic measure under the reported analysis.

The result does not establish the size of a benefit that participants would consider meaningful. The registry does not report a threshold for clinical importance for WTDS. It also does not show that every participant experienced a reduction or that changes in a group mean describe individual responses. Statistical evidence for an average difference is not the same as evidence about the experience of each participant.

The endpoint captures time awake after falling asleep until the last awakening. It does not directly measure every part of sleep quality, daytime functioning, symptom burden, or long-term outcomes. Secondary outcomes provide additional perspectives, but they involve distinct measures and multiple comparisons. The record does not report a multiplicity adjustment for those outcomes.

The finding is specific to a randomized, triple-masked crossover study comparing GEn 1200 mg and placebo across two periods. Some second-period observations were unavailable or participants did not take the treatment, and the primary outcome used LOCF. The registry does not report sensitivity analyses that would show how the result changes under alternative missing-data assumptions. These details limit certainty about the robustness and broader applicability of the estimate.

Accordingly, the statistical result should be read as a treatment contrast on WTDS in this trial, not as proof of an effect on all restless legs symptoms, all sleep outcomes, or every population. The trial result alone does not provide medical advice or establish what treatment choice is appropriate for an individual.

18. Why This Trial Matters Statistically

This trial illustrates how outcome definition shapes statistical interpretation. WTDS is a specific polysomnographic measure, with an explicit definition and a stated time window. The result can therefore be interpreted on a clear scale in minutes, while remaining distinct from patient-reported symptom ratings and other sleep-stage measures.

It also highlights the importance of analyzing a crossover study according to its structure. Treatment occurs across periods, and the reported primary analysis accounts for period effects. A simple comparison of group summaries can provide a check on the direction of a result, but it is not a substitute for the sponsor’s adjusted ANCOVA analysis or the participant-level crossover model.

The posted primary analysis reports a sizeable negative mean contrast with a confidence interval that excludes zero. At the same time, the registry documents missing or unavailable second-period information and use of LOCF. Statistical significance and data completeness are separate considerations: a small p-value does not eliminate uncertainty related to missing outcomes or analytical assumptions.

The record also illustrates why primary and secondary outcomes should not be treated alike. A designated primary endpoint has a central role in the evidence, while numerous secondary comparisons are useful for context but raise multiplicity concerns when no adjustment strategy is reported. Finally, planning illustrations based on an observed effect should not be confused with the trial’s original design or with observed power. Together, these considerations support reading the numerical result alongside its design, endpoint, analysis, and reporting limits.

19. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record does not report the full eligibility criteria, participant characteristics, randomization procedure, masking procedures, or detailed trial protocol. These omissions limit assessment of who was enrolled, how treatment sequences were assigned, and how closely the study procedures support the stated design.

The study is a crossover trial, but the registry does not report whether a washout period was used or how potential carryover and sequence effects were assessed. The primary description states that the analysis adjusted for treatment, pooled center, and period effects, but the full model specification and diagnostics are not reported. The crossover results therefore cannot be independently evaluated for all design-related assumptions from the registry record.

The primary outcome uses LOCF, and the record identifies participants without second-period data or treatment exposure. Reasons for missing data, the timing of missingness, and sensitivity analyses are not reported. This limits assessment of whether the primary estimate is robust to alternative assumptions about missing observations.

The registry does not report the original sample-size justification, any interim analyses, a multiplicity strategy, or a complete statistical analysis plan. Nor does it report a threshold for clinically important change in WTDS or the secondary measures. These gaps limit interpretation of the study’s planning, the weight of secondary findings, and the practical meaning of the estimated difference.

Safety reporting is also limited in the registry summary. The serious event counts and listed terms are available, but event timing, exposure-adjusted rates, participant-level descriptions, and causal assessments are not reported. The posted record alone cannot settle these questions.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Polysomnography Study of GSK1838262 Extended Release Tablets Versus Placebo in RLS and Associated Sleep Disturbance — NCT00748098.
• Registry record last updated 2013-07-22; status last verified 2013-01.
• Results first posted 2011-05-18.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

NCT00748098 is a completed Phase 3 randomized, triple-masked crossover study of GSK1838262 extended release tablets, listed in the results as GEn 1200 mg, versus placebo for restless legs syndrome and associated sleep disturbance. Enrollment was 136. The primary endpoint was adjusted mean change from baseline in wake time during sleep measured by polysomnography at Week 4 and Week 10.

The sponsor reported an ANCOVA adjusted mean difference versus placebo of -26.00 minutes, with a 95% confidence interval from -35.64 to -16.36 and p < 0.001. The analysis used LOCF and described adjustment for treatment, pooled center, and period effects. The registry also reports secondary symptom and polysomnographic outcomes, along with serious adverse event counts.

Interpretation should account for the crossover design, unavailable second-period data, LOCF, and unreported details of the analysis plan and multiplicity handling. The primary result concerns the specified sleep measure and does not by itself establish effects on every symptom or outcome.