Simulated Driving Performance, Daytime Sedation and Cognition in Healthy Volunteers Taking Gralise, Neurontin or Lyrica: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Simulated Driving Performance, Daytime Sedation and Cognition in Healthy Volunteers Taking Gralise, Neurontin or Lyrica (NCT03179345), a randomized phase 4 study. The primary outcome is Change From Baseline in the "Standard Deviation of the Lateral Position" (SDLP) Measured on the Driving Simulator Between Gralise® and Neurontin®.

2. Clinical Question

3. Trial Design

This was a completed Phase 4, randomized, double-blind, placebo-controlled crossover study in healthy volunteers. Its stated aim was to compare simulated driving performance, daytime sedation, and cognition during therapeutic-dose treatment with Gralise, Neurontin, Lyrica, and placebo. The registry describes four treatments and four sequences. In a crossover design, participants receive more than one treatment condition, so comparisons can reflect repeated observations from the same people rather than entirely separate groups.

The registry records an enrollment of 32. The primary outcome was change from baseline in SDLP on a driving simulator at baseline and Hour 3 on Day 3. The primary comparison named in the record is Gralise versus Neurontin. Secondary outcomes include another SDLP comparison, cognitive tasks from the Cogstate battery, and sedation assessments using the Karolinska Sleepiness Scale and Portland Neurotoxicity Scale.

The treatment descriptions distinguish the dosing schedules and forms. Gralise was described as 1800 mg once daily at 7:00 pm on Day 1 and Day 2. Neurontin was described as a 600 mg tablet given at several stated times across Day 1, Day 2, and Day 3. Lyrica was described as a 150 mg capsule on a different schedule across those days. The summary says dosing occurred under fed conditions. These are descriptions of the trial regimen, not evidence about how participants responded.

The registry status is completed. The original sample-size justification, planned assumptions for the trial’s own design, details of any interim analyses, and a full protocol-level statistical plan are not reported in the registry record. The planning illustrations accompanying the posted results are separate from the original design rationale.

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: A total of 32 healthy subjects enrolled in this crossover study. Each subject was randomly assigned to 1 out of 4 treatment sequences. Abbreviations: Gral = Gralise®; Neur = Neurontin®; Lyr = Lyrica®; Plbo = Placebo.

Overall Study

Analysis population for the primary outcome, as defined in the registry: The above is the only Primary Outcome Measure (comparing the SDLP between Gralise® and Neurontin®).

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in the "Standard Deviation of the Lateral Position" (SDLP) Measured on the Driving Simulator Between Gralise® and Neurontin®, measured over Baseline and Hour 3 on Day 3.

The primary endpoint was change from baseline in the Standard Deviation of the Lateral Position (SDLP), measured in feet on a driving simulator. The comparison of interest was Gralise versus Neurontin, assessed at baseline and Hour 3 on Day 3. The record describes SDLP as a measure of change from maintaining the normal driving position in the lane over time or distance.

As a continuous outcome, SDLP can represent the amount of variation in lane position rather than assigning each participant to a simple category. A comparison of changes from baseline focuses on how the measure shifts over the stated assessment period, rather than only comparing a single later measurement. In general, a driving-simulator outcome provides a standardized task for measuring performance under study conditions; it does not by itself establish how a person would drive in everyday settings.

The endpoint is relevant to the stated study question because the trial examined simulated driving performance alongside daytime sedation and cognition. Its interpretation depends on the estimated treatment contrast, its uncertainty, and the analysis used. The registry identifies least-squares means and standard errors for the outcome summaries, while the sponsor’s reported direct comparison is an ANOVA estimate. The registry record does not report a minimally important difference or a threshold defining a meaningful change in SDLP. Without such a threshold, statistical estimates describe the observed scale and precision, but do not by themselves establish the practical importance of a difference.

7. Results Posted in the Registry

• Gralise® (Gabapentin): mean 0.26 versus 0.14 on control; difference 0.12 (95% CI -0.09 to 0.33), standardized effect d = 0.29; z-test p = 0.274.
• Neurontin® (Gabapentin): mean 0.40 versus 0.14 on control; difference 0.26 (95% CI 0.05 to 0.47), standardized effect d = 0.61; z-test p = 0.016.

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 From Baseline in the "Standard Deviation of the Lateral Position" (SDLP) Measured on the Driving Simulator Between Gralise® and Lyrica®

• 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:
  • Gralise® (Gabapentin): mean 0.26 versus 0.14 on control; difference 0.12 (95% CI -0.09 to 0.33), standardized effect d = 0.29; z-test p = 0.274.
  • Lyrica® (Pregabalin): mean 0.36 versus 0.14 on control; difference 0.22 (95% CI 0.01 to 0.43), standardized effect d = 0.52; z-test p = 0.042.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Gralise® (Gabapentin) and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Lyrica® (Pregabalin) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline Between, Gralise® and Neurontin®, Gralise® and Lyrica® for Cognitive Evaluation of Cogstate - Detection Task (DET)

• 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:
  • Gralise® (Gabapentin): mean 0.00 versus -0.00 on control; difference 0.00 (95% CI -0.03 to 0.03), standardized effect d = 0.00; z-test p = 1.000.
  • Neurontin® (Gabapentin): mean 0.00 versus -0.00 on control; difference 0.00 (95% CI -0.03 to 0.03), standardized effect d = 0.00; z-test p = 1.000.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Gralise® (Gabapentin) 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 Neurontin® (Gabapentin) and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline Between, Gralise® and Neurontin®, Gralise® and Lyrica® for Cognitive Evaluation of Cogstate - Groton Maze Learning Test (GMLT).

• 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:
  • Gralise® (Gabapentin): mean -0.40 versus 3.32 on control; difference -3.72 (95% CI -10.52 to 3.08), standardized effect d = -0.28; z-test p = 0.283.
  • Neurontin® (Gabapentin): mean -0.01 versus 3.32 on control; difference -3.33 (95% CI -9.99 to 3.33), standardized effect d = -0.25; z-test p = 0.327.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Gralise® (Gabapentin) 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 Neurontin® (Gabapentin) and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover simulated driving, several cognitive tasks, and sedation scales. Across the posted results, the evidence is not a single uniform pattern across domains: the reported estimates and intervals differ by measure and comparison. The cognitive comparisons reported for detection and maze-learning tasks have confidence intervals that include zero. The posted driving comparison between Gralise and Lyrica has a sponsor interval that includes zero, while the sponsor’s comparison of Gralise with placebo also has an interval that includes zero. These secondary findings should be viewed as distinct from the primary Gralise-versus-Neurontin question.

Secondary outcomes can help describe whether the broader set of measured domains appears consistent with the primary finding, but they do not independently confirm it. Measures may differ in scale, variability, assessment properties, and analysis model. A result from one task should not be treated as interchangeable with a result from another, even when both are intended to capture aspects of cognition or performance.

The registry lists 12 outcome measures and 71 analyses. With multiple outcomes and treatment contrasts, the chance of at least one nominally notable result can rise if many tests are considered without adjustment. The record does not report a multiplicity procedure or clarify which secondary analyses were prespecified as confirmatory. Unadjusted secondary results therefore carry less weight than the named primary analysis and should be treated as exploratory unless additional protocol information establishes their role.

The overall secondary pattern does not establish a broad, consistent treatment effect across driving, cognition, and sedation. Nor does it establish that no differences exist: confidence intervals and analysis details matter, and some posted comparisons may be imprecise. The secondary outcomes add context, but the registry record does not justify elevating an isolated result over the primary comparison.

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

• Change From Baseline in the "Standard Deviation of the Lateral Position" (SDLP) Measured on the Driving Simulator Between Gralise® and Lyrica®
• Change From Baseline Between, Gralise® and Neurontin®, Gralise® and Lyrica® for Cognitive Evaluation of Cogstate - Detection Task (DET)
• Change From Baseline Between, Gralise® and Neurontin®, Gralise® and Lyrica® for Cognitive Evaluation of Cogstate - Groton Maze Learning Test (GMLT).
• Change From Baseline Between, Gralise® and Neurontin®, Gralise® and Lyrica® for Cognitive Evaluation of Cogstate - Identification Task (IDN).
• Change From Baseline Between, Gralise® and Neurontin®, Gralise® and Lyrica® for Cognitive Evaluation of Cogstate - International Shopping List Test (ISL)
• Change From Baseline Between, Gralise® and Neurontin®, Gralise® and Lyrica® for Cognitive Evaluation of Cogstate - One Card Learning (OCLT).
• Change From Baseline Between, Gralise® and Neurontin®, Gralise® and Lyrica® for Sedation Evaluation - Karolinska Sleepiness Scale (KSS).
• Change From Baseline Between, Gralise® and Neurontin®, Gralise® and Lyrica® for Sedation Evaluation - Portland Neurotoxicity Scale (PNS).

9. Statistical Interpretation

10. Statistical Methods Explained

The registry identifies the primary outcome parameter as a least-squares mean and gives standard errors for the treatment summaries. A mean describes the average value of a continuous measure, while a standard error describes the estimated uncertainty in that mean. In general, a least-squares mean is a model-based mean adjusted for terms included in the statistical model. The exact model specification for the sponsor’s primary analysis is not reported in the registry record.

The sponsor reports an ANOVA comparison for the primary Gralise-versus-Neurontin outcome. ANOVA, or analysis of variance, is a general framework for comparing means. In a crossover study, a suitable analysis may need to account for repeated measurements on the same participants and other design features. The registry labels the sponsor’s analysis as ANOVA but does not report enough detail to determine the full model structure, adjustment terms, or treatment of sequence and period effects. The registry also does not report the analysis population definition or the approach to missing observations.

The recalculated summaries are different in purpose. They use posted arm means and standard errors to form simple, unadjusted comparisons of each active treatment with placebo. The calculation uses an approximate standard deviation inferred from the reported standard errors and participant counts. This is a reconstruction from summary statistics, not the sponsor’s crossover model. In general, such a calculation can be useful as a descriptive check, but it cannot recover participant-level relationships or model adjustments that are absent from the summaries. These estimates therefore should not be read as an independent reproduction of the sponsor’s direct primary contrast.

What does a confidence interval convey?

A confidence interval gives a range of values compatible with the data and the method used to calculate it. In general, a narrower interval indicates greater precision, while a wider interval indicates more uncertainty. The interval is not a guarantee that the true value lies within its endpoints, and it is not a range of individual participant responses. For the primary result, the sponsor reports a 95% interval. Its placement relative to zero gives information about whether the data are compatible with no mean difference under the analysis, but it does not establish practical importance.

What does a p-value convey?

In general, a p-value measures how surprising data at least as incompatible with a specified null hypothesis would be if that hypothesis and the analysis assumptions held. It is not the probability that the null hypothesis is true, nor a measure of the size or importance of an effect. The registry reports sponsor p-values for specified contrasts and recalculated unadjusted p-values for summary-statistic comparisons. These have different analytical bases, so they should not be treated as interchangeable.

Why does the crossover design matter?

In a crossover study, a participant can contribute observations under multiple treatments. A comparison can therefore use information about how the same participant’s outcome differs across conditions. In general, this can make within-participant variation important to the estimate and its standard error. A calculation that treats arm summaries as independent does not account for that paired structure. The registry says the study had four treatments and four sequences, but the detailed sequence allocation, period structure, washout information, and model handling of repeated observations are not reported in the registry record.

What is meant by over-dispersion?

For continuous outcomes, variability may be greater than a chosen model expects. In general, over-dispersion matters because underestimated variability can make intervals too narrow and p-values too small. The registry record does not state whether over-dispersion was assessed or whether any special correction was used. The endpoint summaries do report standard errors, but those summaries alone do not show the diagnostics or assumptions behind the sponsor’s analysis.

The reported analyses also include ANOVA and ANCOVA labels for secondary outcomes. ANCOVA, in general, compares outcomes while adjusting for one or more covariates. The registry does not give the complete model terms or explain the rationale for using different methods across the posted secondary comparisons. This limits how fully an outside reader can reproduce or evaluate the adjustment strategy.

The record lists 12 outcome measures and 71 analyses. The number of analyses indicates that results span multiple outcomes and comparisons, but does not specify which analyses were prespecified, how they were prioritized, or whether p-values were adjusted for multiple testing. Those details are not reported in the registry record. Consequently, statistical significance in an isolated secondary comparison should be interpreted cautiously, particularly when many opportunities for comparison exist.

11. Confidence Intervals and Precision

Precision is conveyed most directly by the confidence intervals around estimates. For the sponsor’s primary comparison, the 95% interval runs from -0.266 to -0.016 around the reported mean difference (net) of -0.141. It does not include zero under the sponsor’s reported analysis. The interval nevertheless spans a range of possible contrasts, so the point estimate should not be treated as exact. The registry does not give a threshold for a meaningful SDLP difference, which prevents judging the interval against a stated practical benchmark.

The recalculated Gralise-versus-placebo interval extends from -0.095 to 0.335, while the recalculated Neurontin-versus-placebo interval extends from 0.048 to 0.472. These are unadjusted comparisons based on arm summaries, not the direct crossover contrast reported by the sponsor. They therefore illustrate uncertainty for different comparisons and should not be used to infer the precision of the sponsor’s primary estimate.

In general, a confidence interval’s width reflects both the amount of information and the variability assumed by the analysis. A result can have an estimate near zero but still be imprecise if the interval spans materially different values. Conversely, an interval that excludes zero does not establish that the difference is large enough to matter in practice. The registry’s standard errors and posted intervals describe statistical uncertainty under the reported methods; they do not capture every uncertainty about the design, missing data, or model assumptions.

Because the treatment comparisons occur in a crossover design, within-participant correlation may affect precision. The sponsor’s model would need to reflect the study structure for its interval to represent that design appropriately. The registry record labels the primary analysis as ANOVA, but the detailed model and variance structure are not reported. The sponsor’s own analysis takes precedence over the simple recalculation when interpreting the stated primary contrast.

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 sample-size justification. The original design assumptions and sample-size rationale are not reported in the registry record. The listed assumptions for the illustrations are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect.

For these continuous-outcome illustrations, the entries are participants per arm, not total participants across both arms. The 80% power column indicates the approximate number of participants per arm needed for an 80% chance of detecting the specified effect under those planning assumptions. The 90% column gives the corresponding per-arm planning figure for a higher chance of detection. These are design calculations, not estimates of the trial’s observed ability to detect an effect.

The rows consider planning for 100%, 75%, and 50% of the observed effect. Smaller target effects require more participants in the illustration, because smaller differences are harder to distinguish from variability. The figures are not a reconstruction of the crossover trial’s own design: the stated assumptions describe equal allocation between two arms and a normal approximation, while the registry describes four treatments and four sequences. They should be read as general planning references for the specified pairwise effect, not as the number of participants the trial should have enrolled.

This is not observed power. The figures depend on the observed effect used as the planning basis and on the stated assumptions; they do not change the interpretation of the trial’s confidence intervals or p-values.

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 = 0.120; pooled standard deviation σ = 0.418.
2. n = 2 × (1.96 + 0.842)² × 0.418² / 0.120² = 190.2, rounded up to 191 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 <- 0.12                      # observed difference in means
sigma <- 0.41767928485861                     # 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(3179345)
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 = 191 (80% power), 255 (90% power)
 75% of observed effect: n per arm = 339 (80% power), 453 (90% power)
 50% of observed effect: n per arm = 761 (80% power), 1,019 (90% power)
power.t.test: 191.1 
Simulated power at n = 191 per arm: 0.796 (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 sponsor’s reported primary analysis directly compares Gralise with Neurontin. It gives a mean difference (net) of -0.141, a 95% confidence interval from -0.266 to -0.016, and p = 0.028. This is the analysis aligned with the primary question named in the registry.

The simple recalculated comparisons do not reproduce that direct contrast. They compare Gralise with placebo and Neurontin with placebo using posted arm summaries, with an approximate standard deviation and an unadjusted calculation. The design is crossover, so these summary-based calculations also do not reconstruct any within-participant model used by the sponsor. A difference between the recalculated figures and the sponsor’s analysis is therefore not a like-for-like disagreement about the same estimate.

The registry flags a discrepancy between the posted sponsor analysis and the recalculation. In that circumstance, the sponsor’s own analysis takes precedence for the trial’s stated primary comparison. The recalculated results may offer a rough descriptive check of the arm summaries, but they should not be substituted for the sponsor’s direct Gralise-versus-Neurontin result or used to explain its direction or magnitude. The record does not report enough model detail to reproduce the sponsor’s analysis independently.

15. Multiplicity, Missing Data and Analysis Populations

The record lists 12 outcome measures and 71 analyses, spanning the primary driving outcome and secondary driving, cognition, and sedation measures. When many outcomes or contrasts are tested, some results may appear statistically notable by chance even if there is no underlying difference for those tests. In general, multiplicity procedures adjust interpretation or testing to account for the number of comparisons, but the registry record does not report whether any such procedure was planned or used.

The primary outcome is explicitly identified, which helps distinguish the principal question from the wider set of analyses. However, the record does not provide a complete statistical analysis plan showing which comparisons were confirmatory, how the secondary outcomes were prioritized, or whether the reported p-values are adjusted. The unadjusted recalculations should not be interpreted as multiplicity-controlled evidence. Results from secondary comparisons consequently need caution, especially if they are considered one at a time.

Missing-data handling is not reported in the registry record. The registry gives outcome-specific participant counts that vary by treatment, but does not explain why the counts differ, whether observations were missing, or how incomplete crossover periods were handled. It also does not report whether participants were analyzed according to a specified intention-to-treat or per-protocol population, nor does it define the analysis population for each endpoint. These omissions make it difficult to assess how missing observations or exclusions might affect estimates.

The record identifies the trial as randomized and crossover but does not report the detailed analysis model, the handling of sequence and period, or the assumptions used for incomplete data. Those details can matter because the same participants may contribute observations under multiple treatments. Without them, outside readers cannot fully reproduce the sponsor’s estimates or evaluate the robustness of the results to alternative missing-data assumptions.

These reporting gaps do not prove that the trial lacked a multiplicity or missing-data plan. They mean that the relevant methods are not reported in the registry record, limiting what can be concluded from the posted results alone.

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 non-serious adverse event terms

The registry’s serious adverse event figures show zero affected participants in each listed treatment group: Gralise, Neurontin, Lyrica, and placebo. The numbers at risk are 28 for Gralise, 31 for Neurontin, 31 for Lyrica, and 30 for placebo. These figures describe the serious adverse event reporting in the registry, not all possible harms or events outside the recorded study period.

The record lists no serious adverse event terms. Death counts are not reported in the registry record. The separate other-adverse-event counts are not serious-event counts and should not be combined with the serious-event figures when describing serious events.

In general, a zero count in a small trial does not prove that an event cannot occur or that treatments have equivalent safety. The study involved healthy volunteers, and the registry’s adverse-event summaries do not provide enough information to characterize every potential risk or to compare safety precisely. These figures are descriptive registry results, not medical advice.

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

The primary result describes a difference in change in simulated-driving SDLP between Gralise and Neurontin under the sponsor’s reported analysis. Its estimate is negative, and its 95% confidence interval excludes zero. This is evidence of a statistical difference under that analysis, but statistical evidence is not the same as evidence of practical importance. The registry does not report a threshold for a meaningful change in SDLP.

The endpoint is a simulator measure in healthy volunteers, assessed at a specified time in the study. It does not directly establish how participants would perform in real-world driving, how people with a medical condition would respond, or what would happen with use beyond the study conditions. The study also included cognitive and sedation measures, but those outcomes do not turn the primary result into a comprehensive assessment of every aspect of functioning.

The trial compared multiple treatment conditions in a crossover design. The sponsor’s direct primary analysis should be distinguished from the simple recalculated comparisons against placebo. Those recalculations use different contrasts and do not reproduce the crossover model. The sponsor’s analysis takes precedence for the primary question, while missing model details limit independent reproduction.

The posted secondary results cover several outcomes and comparisons. Without a reported multiplicity strategy and a complete analysis plan, isolated secondary findings warrant caution. The registry also does not report the sample-size rationale, detailed missing-data methods, or full analysis-population definitions. These limitations affect how confidently the posted results can be generalized or independently assessed.

Overall, the record supports describing the reported statistical estimates and their uncertainty. It does not, by itself, establish clinical benefit, lack of benefit, or a treatment recommendation. The findings should be understood within the healthy-volunteer, simulated-task, and registry-reporting context.

18. Why This Trial Matters Statistically

This trial illustrates why the named primary comparison and the analysis design need to be read together. The central question is a direct Gralise-versus-Neurontin comparison in a crossover study. Simple comparisons of each treatment with placebo do not answer that same question, even if they are calculated from the same posted outcome summaries.

It also shows how a confidence interval adds information beyond a p-value. The sponsor’s estimate and interval describe both the direction and uncertainty of the primary contrast. Yet statistical evidence against zero does not settle whether the size of the difference matters in practice. That judgment would require an interpretable benchmark, which is not reported for SDLP in the registry.

The record contains a broad set of outcomes and analyses. This makes the distinction between a prespecified primary endpoint and secondary comparisons especially important. Without reported multiplicity handling, secondary results should not be treated as equivalent confirmatory evidence. Reporting the analysis plan, missing-data approach, and model details would help readers assess how the results were obtained and how robust they may be.

Finally, the sample-size illustrations demonstrate how planning changes when a smaller effect is targeted, but they are not the trial’s original design rationale. Keeping planning calculations separate from observed findings prevents sample-size figures from being mistaken for evidence about the trial’s actual power. These are general statistical lessons arising from the specific reporting and design features of this record.

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 describes a crossover design with four treatments and four sequences, but does not report the detailed sequence allocation, period structure, or model terms. It is therefore not possible to reconstruct from the record how the primary ANOVA handled repeated observations, sequence, or period. The sponsor’s reported analysis remains the relevant result for the primary question, but its full specification is unavailable.

Participant counts in the posted outcome summaries differ by treatment, while the record does not explain missing observations, incomplete treatment periods, or exclusions. The analysis-population definitions and missing-data methods are not reported in the registry record. These gaps limit assessment of whether the available summaries represent all randomized participants consistently.

The primary endpoint uses simulated driving in healthy volunteers, at a specified assessment time. The registry does not report a threshold for a meaningful SDLP change, and the result cannot by itself establish real-world driving performance or outcomes in a clinical population. The record also provides no basis for extending the findings beyond the studied treatment schedules and conditions.

The statistical reporting includes many outcomes and analyses, but multiplicity handling, the complete prespecified analysis plan, and interim-analysis details are not reported. As a result, secondary findings require particular caution. The sample-size illustrations are based on observed effects and separate planning assumptions; they do not document the original trial’s justification. The registry’s summary-level data also do not permit an independent reproduction of the crossover model.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Simulated Driving Performance, Daytime Sedation and Cognition in Healthy Volunteers Taking Gralise, Neurontin or Lyrica — NCT03179345.
• Registry record last updated 2020-06-01; status last verified 2020-04.
• Results first posted 2020-05-13.
• 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

This completed Phase 4 trial randomized 32 healthy volunteers in a double-blind, placebo-controlled crossover study of Gralise, Neurontin, Lyrica, and placebo. The primary outcome was change from baseline in simulated-driving SDLP, comparing Gralise with Neurontin at Hour 3 on Day 3. The sponsor reported an ANOVA mean difference (net) of -0.141, with a 95% confidence interval from -0.266 to -0.016 and p = 0.028.

The registry also reports cognitive, sedation, and additional driving outcomes. The simple recalculations compare treatment summaries with placebo and do not reproduce the sponsor’s direct crossover comparison; the sponsor’s own analysis takes precedence. Full model details, the original sample-size rationale, missing-data methods, analysis populations, and multiplicity handling are not reported in the registry record.