Efficacy Study Measuring the Impact of Pregabalin on Peripheral Neuropathic Pain.: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy Study Measuring the Impact of Pregabalin on Peripheral Neuropathic Pain. (NCT00219544), a randomized phase 3 study. The primary outcome is Neuropathic Pain in Subjects With Peripheral Neuropathic Pain Conditions During the Double-blind Phase.

2. Clinical Question

3. Trial Design

This was a phase 3 randomized, double-blind, parallel-group, placebo-controlled trial. The registry lists an enrollment of 158 and marks the study completed. Its recorded start was 2005-11 and its primary completion was 2008-02. Those dates describe the registry record; they do not establish when results were analyzed or reported.

The study addressed peripheral neuropathic pain. The brief summary describes adult men and women with this diagnosis, including examples such as diabetic peripheral neuropathy and postherpetic neuralgia. The registry lists pregabalin and placebo interventions. Although multiple pregabalin entries appear, the arm descriptions are blank, and doses and detailed treatment assignments are not reported in the registry record.

The primary analysis population was defined more narrowly than the overall enrollment: it included participants randomized into the double-blind treatment phase who received at least one dose of double-blind study medication and completed at least one postrandomization efficacy assessment. For the primary endpoint, the posted arm summaries list 78 participants in the pregabalin group and 75 in the placebo group. These are endpoint analysis counts, not a replacement for the registry's overall enrollment figure.

The record also describes single-blind and double-blind treatment phases in its outcomes. Specific details of the phase sequence, allocation procedures, concealment, treatment duration outside the stated primary timeframe, and dose regimens are not reported in the registry record. The record does not provide a sample-size justification or the original design assumptions.

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: 405 subjects enrolled - 234 completed the single-blind phase. Subjects who responded (≥30% decrease in weekly mean pain score) during single-blind entered double-blind. 158 subjects assigned to study drug in double-blind phase. One subject in placebo group was randomized in error, not treated, not included as discontinuation;157 received treatment.

Single Blind Treatment Period

Reasons for not completing this period:

Double Blind Treatment Period

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set (FAS) = Intent-to-Treat (ITT) population, defined as all subjects who are randomized into the Double-Blind treatment phase, who receive at least one dose of Double-Blind study medication and complete at least one postrandomization efficacy assessment.

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 Neuropathic Pain in Subjects With Peripheral Neuropathic Pain Conditions During the Double-blind Phase, measured over 9 weeks.

The primary endpoint was the pain score at the end of the double-blind treatment phase, with a stated timeframe of 9 weeks. It was calculated as the mean of the last 7 available daily pain scores recorded while participants were receiving double-blind treatment. The rating used an 11-point numerical scale ranging from 0, no pain, to 10, worst possible pain. The posted summaries report least-squares means and standard errors for the two groups.

A daily diary can capture a participant's pain ratings across time rather than relying on a single retrospective assessment. Averaging the last 7 available scores is the endpoint definition in the registry; it should not be interpreted as a complete description of how missing diary entries were handled. That handling is not reported in the registry record.

In general, a continuous pain score preserves gradations across the scale, allowing the comparison to represent a difference in average scores rather than only a count of participants meeting a threshold. Lower values on this scale correspond to ratings nearer to no pain, while higher values are nearer to worst possible pain. The endpoint alone does not describe every aspect of pain, functioning, or an individual's experience. The registry does not report a threshold for judging whether a difference is clinically important.

7. Results Posted in the Registry

• Pregabalin: mean 2.84 versus 3.62 on control; difference -0.78 (95% CI -1.28 to -0.28), standardized effect d = -0.50; z-test p = 0.002.

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.

Time to Meaningful Increase in Pain During Double-blind Treatment Phase (Number of Participants)

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Pregabalin: 35.0% (28/80) versus 36.4% (28/77) on control; risk difference -1.4 percentage points (95% CI -16.4 to 13.6), risk ratio 0.96 (0.63–1.47); unadjusted two-proportion z-test p = 0.858.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Pregabalin and the control. The interval gives the range of effects that remain compatible with the data.

Mean Sleep Interference Score

• 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:
  • Pregabalin: mean 1.81 versus 2.52 on control; difference -0.71 (95% CI -1.20 to -0.22), standardized effect d = -0.46; z-test p = 0.004.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Pregabalin than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Intensity of Neuropathic Pain -Visual Analog Scale (NeP - VAS)

• 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:
  • Pregabalin: mean 6.90 versus 15.90 on control; difference -9.00 (95% CI -15.79 to -2.21), standardized effect d = -0.42; z-test p = 0.009.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Pregabalin than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes cover different aspects of pain and its effects, including pain scores across treatment phases, change from baseline, responder definitions, categorized daily scores, sleep interference, visual analog scale intensity, and time to a meaningful increase in pain. The registry lists 18 posted secondary outcomes, but the full set of details is not represented by the selected numerical summaries. The measures should be understood as complementary rather than interchangeable.

Across the posted numerical comparisons, the continuous secondary score estimates generally point in the same direction as the primary comparison. Their reported intervals and p-values indicate varying degrees of precision. The binary comparison of participants with a meaningful increase in pain is less conclusive: its risk-difference interval includes zero, and its unadjusted p-value is 0.858. This does not establish equivalence; it indicates that the posted estimate is uncertain and compatible with a range of differences.

The secondary findings can add context to the primary pain-score result, but they do not independently confirm it simply by pointing in a similar direction. These outcomes differ in definition, timing, and statistical scale. Some are based on repeated or categorized diary information, while others use a single summary or participant count. A secondary estimate may also be less precise or may reflect a different subset of participants, as the posted analysis counts vary by outcome.

In general, testing many outcomes raises the chance of obtaining small p-values even when there is no true effect on any particular outcome. Adjustment procedures or a prespecified hierarchy can address that issue, but whether either was used is not reported in the registry record. Accordingly, unadjusted secondary results carry less weight than a clearly designated primary analysis and should be read as supportive or hypothesis-generating rather than as a collection of independent confirmatory findings.

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

• Weekly Mean Pain Scores During the Single-blind Treatment Phase
• Weekly Mean Pain Scores During the Double Blind Treatment Phase
• Change in Pain Scores During Double Blind Treatment Phase
• Number of Subjects With >= 30% Reduction in Mean Pain Score During Single-blind Treatment
• Mean Pain Score for Responders at End of Single-blind Treatment Phase. Change From Baseline of Mean of Last 7 Available Pain Scores From Daily Pain Diary While on Single-blind Treatment.
• Mean Pain Score for Non-responders at End of Single-blind Treatment Phase
• Categorized Daily Pain Score
• Time to Meaningful Increase in Pain During Double-blind Treatment Phase (Number of Participants)

9. Statistical Interpretation

10. Statistical Methods Explained

The registry's primary result and the simple recalculation answer related but not identical statistical questions. The reported primary analysis used ANCOVA and described its estimate as a net mean difference. The recalculation uses the posted group summaries to estimate an unadjusted difference between group means. General statistical principles help explain why the estimates can be similar without being numerically identical.

What does the mean difference represent?

A mean difference is the mean in one group minus the mean in the comparator group, using the stated order of groups. Here the contrast is pregabalin minus placebo, and the reported estimate is -0.78. The negative sign indicates that the pregabalin group's estimated mean score was lower. It does not mean that every participant's score changed by that amount. A difference in group averages describes the contrast between group-level summaries, not the distribution of individual responses.

The endpoint summaries are least-squares means, which are model-based adjusted means in the sponsor's analysis. ANCOVA, or analysis of covariance, generally compares outcomes while accounting for specified covariates. Which covariates were included, how they were coded, and the exact model specification are not reported in the registry record. Therefore, the posted ANCOVA result cannot be independently reconstructed from the summary fields alone.

Why is the recalculation unadjusted?

The recalculated estimate is derived from the posted means and standard errors rather than individual participant data or a full ANCOVA model. It is labeled unadjusted because it does not reproduce covariate adjustment. The recalculation also uses an approximate pooled standard deviation inferred from the standard errors, as indicated by the posted approximation flag. In general, converting standard errors to an estimate of spread relies on assumptions about the summaries and sample sizes. That approximation is useful for descriptive comparison and planning illustrations, but it is not a substitute for the sponsor's model-based result.

When trial-level analysis details are available, the prespecified and sponsor-reported analysis is ordinarily the primary basis for interpreting the trial's own result. The registry does not report the full statistical analysis plan, so it is not possible to assess whether the posted model follows a prespecified plan or which adjustments were intended.

What do the p-value and interval say?

In general, a p-value evaluates how compatible the observed data are with a null hypothesis under a statistical model. The sponsor's primary result reports p = 0.002. This is evidence against a zero mean difference under that analysis, not a measure of the probability that the finding occurred by chance, and not a measure of clinical importance.

A confidence interval expresses uncertainty around an estimate under the method's assumptions. The sponsor's 95% interval is -1.27 to -0.30. Because the interval does not include zero, it is consistent with a nonzero difference in the direction of the point estimate under the reported analysis. It still allows a range of effect sizes, and it does not capture every source of uncertainty, including uncertainty from unreported analysis choices or missing-data handling.

Why can over-dispersion matter for count outcomes?

In general, binary or count outcomes may show more variability than a simple statistical model assumes. Over-dispersion means that observed variability exceeds the model's expected variability; if ignored, uncertainty can be understated. The primary endpoint is continuous, and the registry does not state that over-dispersion was assessed. The secondary time-to-meaningful-increase outcome is reported as participant counts at a stated time, not as a detailed time-to-event model in the posted comparison. The record does not describe model diagnostics or variance adjustments for that outcome.

What can be inferred from the secondary analyses?

For secondary outcomes, the posted summaries include continuous score comparisons and a binary participant-count comparison. The continuous outcomes use reported least-squares means and standard errors, while the binary comparison is expressed as a risk difference with an interval. These effect measures answer different questions and should not be combined as if they were one endpoint. The registry lists 19 outcome measures and 22 analyses, but the analysis plan, hierarchy among outcomes, and multiplicity adjustment are not reported. Consequently, secondary p-values are best treated as supportive or exploratory rather than as a set of independent confirmatory tests.

11. Confidence Intervals and Precision

The sponsor-reported primary estimate is a mean difference of -0.78, with a 95% confidence interval from -1.27 to -0.30. This interval gives a range of values compatible with the posted ANCOVA analysis under its assumptions. Its endpoints are both negative, indicating that the range does not include a zero mean difference. The interval is more informative than the point estimate alone because it communicates uncertainty about the magnitude of the group contrast.

The recalculated unadjusted interval is -1.279 to -0.281. It is based on posted summary statistics and an approximate pooled standard deviation, whereas the sponsor's interval comes from ANCOVA. The similar direction and scale of these estimates provide a useful cross-check, but their different methods mean that their endpoints should not be expected to match exactly. The recalculation is not a replacement for the sponsor's analysis.

In general, a confidence interval is not a range containing a known fixed effect with a probability assigned after observing the data. It is produced by a procedure that, under its assumptions, would cover the fixed effect at the stated rate over repeated applications. Its width reflects sampling uncertainty under the model, but not necessarily all uncertainty about study conduct, outcome measurement, missingness, or analysis choices.

The registry does not state a minimum difference considered clinically important. Therefore, the interval alone cannot establish whether all plausible values are meaningful to participants. The secondary results also vary in precision: some reported intervals exclude zero, while the binary secondary comparison has an interval spanning both negative and positive risk differences. That pattern supports caution in interpreting individual secondary estimates, particularly without a reported multiplicity plan.

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 effects, not the trial's original design calculations. For continuous outcomes, the entries are participants per arm. For the time-to-event planning figures, the registry's planning note says they represent the total number of events. The illustrations assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. They use the stated fraction of the observed effect as the effect to detect.

The column for 80% power describes the number of participants per arm, or total events for the time-to-event calculation, that the planning method estimates would be needed to have an 80% chance of detecting the specified effect under its assumptions if that effect were the true one. The 90% power column describes the corresponding planning target for a 90% chance. These are design concepts: power is calculated before observing a study's results for a specified effect and model, not inferred from whether a result was statistically significant.

Rows that plan for 75% or 50% of the observed effect represent smaller target effects than the row using 100% of the observed effect. Smaller effects are generally harder to detect, so the illustrations require more participants per arm as the target effect becomes smaller. The figures are not observed power and do not describe the actual power of this trial. The registry does not report the trial's original sample-size justification or design assumptions.

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 = -0.780; pooled standard deviation σ = 1.575.
2. n = 2 × (1.96 + 0.842)² × 1.575² / -0.780² = 64.0, rounded up to 64 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 <- -0.78                      # observed difference in means
sigma <- 1.57466363484838                     # 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(219544)
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 = 64 (80% power), 86 (90% power)
 75% of observed effect: n per arm = 114 (80% power), 153 (90% power)
 50% of observed effect: n per arm = 256 (80% power), 343 (90% power)
power.t.test: 65 
Simulated power at n = 64 per arm: 0.792 (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 primary analysis is reported as ANCOVA, with a mean difference of -0.78, a 95% confidence interval from -1.27 to -0.30, and p = 0.002. This is the registry's own model-based result for the primary comparison. The recalculated result is an unadjusted estimate from the posted summary statistics, with a mean difference of -0.78, a 95% confidence interval from -1.279 to -0.281, and p = 0.002.

The estimates are close in magnitude and direction, but the recalculation does not reproduce ANCOVA. It uses summary-level information, including an approximate pooled standard deviation inferred from standard errors, and lacks the covariate details and participant-level data needed to reconstruct the sponsor's model. Differences in intervals and p-values are therefore expected possibilities rather than evidence of an error.

The registry does not mark a discrepancy between the posted sponsor analysis and the recalculation. The sponsor's result remains the analysis reported for the trial; the recalculation is a transparent unadjusted cross-check, not an alternative trial analysis. No claim about prespecification can be made because the statistical analysis plan is not reported in the registry record.

15. Multiplicity, Missing Data and Analysis Populations

The registry identifies a primary endpoint and lists 19 outcome measures and 22 analyses. It also presents multiple secondary outcomes, including several related pain scores, responder measures, and phase-specific assessments. When many outcomes are tested, the chance of at least one apparently notable result can increase even if no underlying differences exist. In general, multiplicity can be managed through prespecified outcome hierarchies, adjusted significance thresholds, or other planned procedures.

Whether a multiplicity adjustment was used, which outcomes were considered confirmatory, and whether any testing hierarchy was prespecified are not reported in the registry record. The posted secondary p-values should therefore be interpreted as unadjusted unless the record explicitly reports otherwise. A small unadjusted p-value does not by itself establish a confirmatory finding across a family of outcomes.

The primary population is defined as the Full Analysis Set: participants randomized into the double-blind phase, receiving at least one dose of double-blind study medication, and completing at least one postrandomization efficacy assessment. This definition is reported, but details about exclusions, participant flow, and the handling of participants who did not meet the efficacy-assessment criterion are not reported in the registry record.

The primary endpoint uses the last 7 available daily pain scores during double-blind treatment. The word available indicates that the endpoint can draw on diary observations that were recorded, but it does not reveal how missing diary days were addressed, whether scores were imputed, or what assumptions governed incomplete records. The statistical model's approach to missing data and sensitivity analyses are not reported in the registry record.

In general, missing-data methods matter because participants with incomplete outcomes may differ systematically from those with complete outcomes. An analysis limited to available observations can be affected if that possibility is not appropriately addressed. The registry's summary does not provide enough information to evaluate this issue, to reproduce the primary model fully, or to determine the robustness of the result to alternative assumptions.

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

The posted primary result describes a lower average pain score in the pregabalin group than in the placebo group during the double-blind phase. The sponsor's ANCOVA estimate is -0.78, with a 95% confidence interval from -1.27 to -0.30. This is a group-level comparison on the registry's 0-to-10 pain scale, where 0 is no pain and 10 is worst possible pain. It summarizes the specified endpoint, the mean of the last 7 available daily scores, rather than every experience during the study.

The result does not mean that every participant's pain score was lower, that all participants responded in the same way, or that the estimated difference applies to every person with peripheral neuropathic pain. A mean can conceal variation across individuals. The registry does not provide participant-level distributions or a threshold for determining whether a score difference is clinically important.

The result also does not establish what would happen outside the trial's defined population, under treatment regimens not described in the record, or over periods beyond the reported outcome timeframes. The dose details and the sponsor's full model specification are not reported in the registry record. The primary outcome is a pain rating; it does not on its own establish effects on every aspect of function, quality of life, or safety.

Statistical significance and clinical importance are distinct concepts. A p-value evaluates compatibility with a null model under assumptions, while the confidence interval describes uncertainty in the estimated difference. Neither supplies a clinical-importance threshold. Interpretation should therefore retain the scale, endpoint definition, model, population, and uncertainty rather than reducing the result to a yes-or-no statement.

These findings are statistical descriptions of the registry record, not medical advice. The registry's results do not determine the right choice for any individual. Clinical decisions require considerations beyond this single endpoint and are outside the scope of these statistical comparisons.

17. Why This Trial Matters Statistically

This trial illustrates why statistical interpretation needs to distinguish a sponsor-reported model from a summary-level recalculation. The primary analysis used ANCOVA, while the recalculated unadjusted contrast is based on group summaries and an approximate estimate of spread. Their close estimates provide a useful consistency check, but the two analyses do not answer precisely the same modeling question. A reported trial analysis and a transparent recalculation have complementary roles.

It also shows the value of keeping effect size, uncertainty, and statistical evidence together. The estimated difference alone does not reveal its precision; the interval helps communicate the range of values compatible with the analysis. The p-value adds information about compatibility with a null model, but cannot replace the interval or determine clinical importance. Interpreting all three prevents overreliance on a significance label.

The record's multiple secondary outcomes highlight another central issue in clinical research: repeated testing requires attention to multiplicity and to the prespecified status of each outcome. Directionally similar secondary estimates can support context, but they are not necessarily independent confirmations. Without a reported hierarchy or adjustment strategy, the primary endpoint has a clearer interpretive role.

Finally, the registry makes explicit an analysis population and endpoint definition, yet leaves important details such as missing-data handling and model covariates unreported. This is a reminder that numerical results are inseparable from how outcomes and populations are defined. Statistical confidence in a displayed estimate should not be confused with complete knowledge of every design and analysis decision.

18. Limitations

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

Several trial-specific details needed for a fuller assessment are not reported in the registry record. The arm descriptions are blank, and the listed pregabalin entries do not specify doses or explain the detailed allocation among intervention groups. The statistical analysis plan, ANCOVA covariates, model specification, and original sample-size justification are also not reported. These gaps prevent a complete reconstruction of the sponsor's analysis or an assessment of whether the posted methods match prespecified choices.

The primary endpoint uses the last 7 available daily pain scores, but the registry does not explain how missing diary observations were handled. It does not report imputation methods, sensitivity analyses, or participant flow through the double-blind phase. The stated Full Analysis Set definition is useful, but the record does not provide enough information to evaluate the consequences of that inclusion criterion or incomplete outcome data.

Analysis counts differ across posted outcomes. For example, the primary endpoint summaries list 78 participants for pregabalin and 75 for placebo, while the secondary outcome summaries have their own analysis counts. Those differences should be recognized as outcome-specific; the registry record does not explain all reasons for them. The overall enrollment figure does not resolve those analysis-population details.

The secondary results are also limited by the absence of a reported multiplicity strategy and by the fact that the registry summary does not include all outcome-level details in the numerical comparisons. No threshold for clinical importance is reported for the primary scale. These limitations constrain what can be concluded from the record, without establishing that the trial's conduct or analysis was flawed.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Efficacy Study Measuring the Impact of Pregabalin on Peripheral Neuropathic Pain. — NCT00219544.
• Registry record last updated 2021-01-22; status last verified 2018-11.
• Results first posted 2009-06-02.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

The registry describes a completed phase 3 randomized, double-blind, parallel, placebo-controlled study of pregabalin for peripheral neuropathic pain, with an enrollment of 158. The primary endpoint was the mean of the last 7 available daily pain scores during the 9-week double-blind phase, on a scale from 0, no pain, to 10, worst possible pain.

The sponsor-reported ANCOVA estimated a mean difference of -0.78 between pregabalin and placebo, with a 95% confidence interval from -1.27 to -0.30 and p = 0.002. A summary-statistic recalculation gives a similar unadjusted estimate, but does not reproduce the sponsor's adjusted model. Secondary outcomes cover several related measures; multiplicity and missing-data procedures are not reported in the registry record.