External Trigeminal Nerve Stimulation for Epilepsy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for External Trigeminal Nerve Stimulation for Epilepsy (NCT01159431), a randomized phase 2 study. The primary outcome is 50% Responder Rate.

2. Clinical Question

The central statistical question is about a binary outcome: whether each participant did or did not reach a defined seizure-reduction threshold. It is not a question about the average change in seizures for every participant. That distinction matters because a responder rate describes how many participants crossed a threshold, while a mean change describes the average numerical change in a measure across participants.

The registry also lists response ratio, expressed as mean percent change in seizures, and mood as secondary outcomes. Those outcomes address different aspects of the study and do not replace the primary responder-rate question. The trial’s results can describe the observed group patterns, but the registry does not report enough information to establish all details of the analysis plan, the handling of missing observations, or the precise composition of the control condition.

3. Trial Design

This was a randomized, quadruple-masked, parallel-group phase 2 study. The registry describes it as a double-blind study and records quadruple masking; it does not specify which roles were masked or describe how masking was maintained. The study status is completed. The recorded start was 2008-01, and primary completion was 2011-06.

The enrollment listed in the registry is 50. The primary outcome results are posted for 23 participants in the treatment arm and 19 in the control arm. The registry does not state why these outcome denominators differ from the enrollment figure, whether participants were excluded from a particular analysis, or how missing outcome information was handled. Those details should not be inferred from the arm counts alone.

The treatment is named Trigeminal Nerve Stimulation and is classified as a device intervention. The study summary describes external electrical stimulation over the forehead and above the eyes. The arm descriptions do not provide further details for either arm. In particular, the control group’s procedure and the degree to which it resembled the active intervention are not reported in the registry record.

The design allows a comparison between randomized groups, but the registry record does not report the sequence-generation method, allocation concealment, planned sample-size justification, analysis population, or statistical analysis plan. It also does not report whether interim analyses were planned or conducted. The record identifies five outcome measures, while results are posted for the primary outcome and two secondary outcomes. The other outcome results are not reported in the registry record.

Randomization and masking are important design features, but their presence alone does not resolve all questions about the reliability or scope of a result. Interpretation also depends on how the outcomes were defined and measured, how missing data were handled, which analyses were planned, and whether the reported groups and denominators correspond to those plans. Those details are not available in this registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: 6-week pretreatment baseline

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as 50% Responder Rate, measured over Treatment period, 18 weeks (end of double blind period) compared with first 6 weeks.

The primary endpoint is the 50% responder rate. It is the percentage of participants whose seizures were reduced by at least 50%, as defined in the registry description. The timeframe is the 18-week treatment period, at the end of the double-blind period, compared with the first 6 weeks. The registry emphasizes that this outcome is a fixed percentage of participants, not a mean or median value.

A responder endpoint converts each participant’s seizure experience into a binary classification: meeting or not meeting the specified threshold. The resulting rate answers a practical group-level question about the proportion crossing that threshold. It differs from an average percentage change, which can reflect both large changes in some participants and smaller or opposite changes in others.

Using a stated threshold can make the outcome easier to interpret, but it also means that participants with changes close to the threshold fall into different categories even when their observed changes are similar. The responder rate therefore gives only one view of the seizure data. The registry separately lists mean percent change in seizures as a secondary outcome, which captures a different summary of change.

The posted responder rates are 40.5% in treatment and 15.6% in control. The registry record does not report additional details about the seizure-counting method, the precise baseline assessment procedures, or how incomplete seizure data were treated. Those omissions limit how fully the endpoint’s measurement and analysis can be evaluated.

7. Results Posted in the Registry

• Treatment: 39.1% (9/23) versus 15.8% (3/19) on control; risk difference 23.3 percentage points (95% CI -2.5 to 49.2), risk ratio 2.48 (0.78–7.88); unadjusted two-proportion z-test p = 0.096.

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.

Response Ratio: Mean Percent Change in Seizures

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Treatment: mean -13.90 versus -9.00 on control; difference -4.90 (95% CI -9.00 to -0.80), standardized effect d = -0.73; z-test p = 0.019.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Treatment than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Mood

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Active: mean -8.13 versus -3.95 on control; difference -4.18 (95% CI -4.96 to -3.40), standardized effect d = -3.23; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Active than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry posts two secondary outcomes: mean percent change in seizures and mood. Together, they broaden the description beyond the primary binary responder threshold. The seizure measure concerns average change rather than the proportion reaching a fixed cutoff, while mood is a separate scale-based outcome. These measures therefore address different questions and are not interchangeable with the primary endpoint.

The posted secondary comparisons point in directions that differ in their substantive interpretation: the seizure response-ratio measure has a mean change in both groups, while the mood measure has a larger negative mean change in the active group than in control. The registry does not define the favorable direction or clinical interpretation of the mood scale in the posted outcome description, so the direction of that numerical change should not be assigned a clinical meaning from the registry alone.

Secondary findings can provide context, but their evidential weight depends on the number of outcomes and analyses considered, which outcomes were specified in advance, and whether the analysis accounted for multiple comparisons. The registry lists five outcome measures but posts results for only two secondary outcomes. It does not report a multiplicity plan, a hierarchy among secondary endpoints, or the full set of planned analyses. Without those details, the unadjusted secondary results should be treated as less conclusive than a clearly prespecified primary analysis.

The secondary summaries also do not resolve uncertainty in the primary result. A mean change in seizures may complement a threshold-based responder rate, but a different summary can produce a different pattern without being contradictory. Mood is a distinct outcome and cannot confirm the seizure endpoint. The registry provides no reported analyses linking the secondary measures to the primary result or establishing that any observed pattern was planned as confirmatory.

In general, multiple unadjusted comparisons increase the chance that at least one result will look notable by chance, even if none represents a reliable underlying difference. This is why secondary outcomes are useful for describing the range of measured outcomes while requiring caution in interpretation. The absence of a reported multiplicity strategy means that the displayed secondary p-values, where present, should not be treated as if they were adjusted confirmatory tests.

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

• Response Ratio: Mean Percent Change in Seizures
• Mood

9. Statistical Interpretation

10. Statistical Methods Explained

The statistical summaries for this trial involve both binary and continuous outcomes. The primary endpoint is binary because each participant either meets the 50% responder definition or does not. The secondary outcomes are continuous summaries: mean percent change in seizures and a mood measure reported in scale units. These outcome types require different summaries and comparisons.

What does a responder rate describe?

A responder rate is the proportion of participants who meet a specified criterion. Here, the criterion is a 50% reduction in seizures within the registry’s stated comparison of the treatment period with the first 6 weeks. A rate describes the frequency of threshold attainment within a group; it does not show the full distribution of individual changes. A participant just below the threshold and a participant with no reduction both count as nonresponders under a binary definition, even though their individual experiences may differ.

The posted rates are 40.5% for treatment and 15.6% for control. The reported event counts are 9 of 23 and 3 of 19. Because percentages are rounded in the registry, they need not reproduce a count exactly if recalculated from the displayed values. The analysis should be read using the reported results rather than by attempting to reconstruct unrounded data.

What does a risk difference mean?

A risk difference compares the two group probabilities by subtracting one group’s rate from the other. It is expressed on the probability scale, so zero represents equal rates. The posted primary comparison includes a risk-difference estimate and an interval. The interval crosses zero, meaning the displayed range includes both a possible advantage for treatment and a possible advantage for control, as well as no difference. The range’s width indicates that the estimate is not highly precise.

The registry also includes relative measures for the primary binary outcome, including a risk ratio and an odds ratio. These express the group contrast on relative rather than absolute scales. Relative measures can look different from an absolute difference because their interpretation depends on the control-group rate. They do not replace the responder rates themselves, and the registry does not report an adjusted model for these measures.

How should the p-value be read?

The posted unadjusted p-value for the primary comparison is 0.096. General explanation: a p-value measures how unusual data at least as discrepant as those observed would be under a specified statistical model in which the null hypothesis is true. It is not the probability that the null hypothesis is true, nor is it a measure of the probability that the intervention works. A p-value also does not convey the magnitude or precision of an effect; those require the estimate and its interval.

The statistical test and significance threshold used for the trial’s original analysis are not reported in the registry record. The displayed p-value is an unadjusted re-calculation from the posted summary statistics. The registry lists no reported primary analyses, so it does not establish that this calculation was the trial’s prespecified analysis.

Why do confidence intervals matter?

General explanation: a confidence interval gives a range of values compatible with the observed data under a particular statistical method and its assumptions. It is useful to consider both the point estimate and the interval. A point estimate alone can appear definitive even when the data leave substantial uncertainty. An interval crossing zero for a difference does not establish equality; it indicates that the data and method do not rule out zero within that interval.

The confidence level and interval-construction method are not reported in the registry record. The posted interval for the primary risk difference extends across zero. This is consistent with the unadjusted p-value not providing strong evidence against a zero difference under the calculation used, but neither statistic alone answers whether a difference would be important in a clinical setting.

What do the continuous-outcome summaries add?

For continuous outcomes, the registry reports group means and standard deviations. A mean summarizes the group average, while a standard deviation describes variation among observations around that average. The mean difference compares the group averages, and its interval conveys uncertainty in that contrast. These summaries can retain more information about the magnitude of change than a responder threshold, but they depend on how the measure was collected and analyzed.

The registry posts unadjusted re-calculations for the secondary comparisons, with no reported analyses in the record. The statistical tests, any covariate adjustments, and the handling of repeated measurements are not reported. The posted summaries should therefore be interpreted as group-level descriptive comparisons rather than evidence that every analysis detail is known.

What does unadjusted mean in this context?

An unadjusted comparison contrasts the observed groups without accounting for other variables in a statistical model. General explanation: adjustment may be used to account for prespecified baseline factors or design features, but it can also introduce additional modeling assumptions. Whether adjustment is suitable depends on the protocol and analysis plan. Those plans are not reported in the registry record, so there is no basis for attributing a particular adjusted method to this trial.

The posted calculations are simple, unadjusted re-calculations from the summary statistics. The registry does not report the original trial’s statistical analysis plan, the exact tests used by the study team, or whether an analysis adjusted for baseline seizure frequency or other characteristics. The re-calculated results should not be described as a complete reconstruction of the trial’s original analysis.

General explanation: statistical methods rely on assumptions about the data, the sampling process, and the form of the comparison. For binary outcomes, the amount of information depends in part on the number of events and participants. For continuous outcomes, the distribution of measurements and the variability within groups can affect precision. The registry’s summary data do not allow all assumptions to be checked, and they do not include participant-level observations.

11. Confidence Intervals and Precision

Precision concerns how narrowly the data constrain the estimated group difference. The primary comparison’s point estimate is accompanied by a confidence interval that extends across zero. That interval therefore permits values indicating a difference in either direction, as well as no difference. Its breadth signals considerable uncertainty around the estimate. The confidence level and method used to construct the interval are not reported in the registry record.

The primary outcome denominators are 23 for treatment and 19 for control. These are modest group sizes for estimating a percentage, particularly when the outcome is based on relatively few responder events. The posted event counts are 9 in treatment and 3 in control. Estimates based on limited numbers of events can be sensitive to the outcomes of individual participants, so the observed rates should not be treated as exact underlying probabilities.

The p-value of 0.096 is one part of the statistical summary, but it does not express precision. It does not indicate how large the group difference is or how wide the range of plausible values may be. The interval is more informative about uncertainty in the estimated difference, while the arm-specific rates give the observed group proportions. These summaries answer related but not interchangeable questions.

There are also reporting limits. The registry does not explain the relationship between the enrollment of 50 and the primary outcome denominators, report missing-data handling, or state the analysis population. The primary endpoint’s confidence level and the trial’s planned analysis details are not reported. These gaps prevent a complete assessment of whether the displayed precision reflects the intended analysis and all randomized participants.

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 own design justification. For these outcomes, the figures refer to participants per arm; they are not total event counts. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect equal to the stated fraction of the observed effect. The trial’s original sample-size assumptions and justification are not reported in the registry record.

The 80% power column gives the number of participants per arm in the illustration for a design with an 80% chance of detecting the specified effect under the assumptions. The 90% power column gives the corresponding per-arm number for a 90% chance. In general, higher planned power requires more participants when other assumptions are held constant. The figures are not observed power.

Each row considers a different target effect: the full observed effect, 75% of the observed effect, or 50% of the observed effect. Smaller target effects are harder to distinguish from random variation, so the planning illustrations call for more participants per arm as the target effect becomes smaller. The rows are not alternative descriptions of what happened in the completed trial; they show how planning requirements change when the assumed effect changes.

These calculations use the observed effect as a planning reference and depend on assumptions that may not match the trial’s original design. They should not be read as evidence that the completed study had a particular probability of finding a result, nor as a claim about the sample size that should be used in another study without a separate design assessment.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.158; treatment proportion p₁ = 0.391; average p̄ = 0.275.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6312 = 1.2371.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6092 = 0.5127.
4. n = (1.2371 + 0.5127)² / (0.2334)² = 56.2, rounded up to 57 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
p0 <- 0.157894736842105                          # control proportion
p1 <- 0.391304347826087                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(1159431)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 57 (80% power), 75 (90% power)
 75% of observed effect: n per arm = 94 (80% power), 125 (90% power)
 50% of observed effect: n per arm = 195 (80% power), 260 (90% power)
power.prop.test: 56.2 
Simulated power at n = 57 per arm: 0.819 (20000 replicates)

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

14. Multiplicity, Missing Data and Analysis Populations

The registry record does not report how the investigators addressed multiplicity. It lists five outcome measures, with results posted for the primary responder-rate endpoint and two secondary outcomes. The record does not state how many hypotheses were prespecified, whether the primary endpoint was the only confirmatory test, whether secondary outcomes followed a testing hierarchy, or whether any adjustment to account for multiple comparisons was planned.

General explanation: when several outcomes or comparisons are tested, the chance of at least one apparently unusual result can increase. Procedures such as a prespecified hierarchy or adjustment can control how results are interpreted across a family of tests. Whether such procedures are needed and which approach is appropriate depend on the design and analysis plan. No particular approach can be attributed to this trial based on the registry record.

Missing-data procedures are also not reported. The enrollment is 50, while the posted primary outcome denominators are 23 and 19. The record does not explain the difference, describe whether all randomized participants were included in the analysis, or state how participants without usable outcome data were handled. It does not provide reasons for missing observations or report sensitivity analyses. The arm denominators should therefore not be taken as proof of a particular exclusion or analysis rule.

The analysis population is not reported in the registry record. It is not possible to determine from the posted information whether the primary comparison used all randomized participants, participants who completed treatment, or another defined group. The record also does not provide participant-level data to assess whether the groups contributing to the outcome summaries remained comparable after randomization.

The statistical analysis plan, prespecified model, and interim-analysis approach are not reported. The registry lists no reported analyses. Consequently, the unadjusted comparisons should be understood as calculations based on the posted summary statistics, not as confirmation that every element of the trial’s original analysis was prespecified or performed in that form.

15. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 4% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: seizure cluster (4.2%).

The registry reports serious adverse events for 25 participants at risk in each group. There were 0 serious adverse events in the active group and 1 in the control group. These counts describe the posted safety results; they do not establish whether the difference between groups is meaningful, particularly with these denominators.

Other adverse events are reported for 5 of 25 participants in the active group and 4 of 25 in control. The registry lists seizure cluster as a serious adverse-event term at 4.2%. The record does not provide further detail in the posted summary about event timing, severity beyond the serious-event classification, relatedness to the intervention, or whether participants could experience more than one event.

Deaths are not reported in the registry record. The safety figures should be interpreted as counts from the record, not as a complete comparative safety analysis. The registry does not report an analysis plan for safety outcomes or provide enough detail to assess how adverse events were collected and classified.

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

The registry records a higher observed 50% responder rate in the treatment group than in the control group. It also reports a primary unadjusted p-value of 0.096 and a confidence interval for the group difference that crosses zero. Taken together, these statistics describe a result with an observed difference but substantial uncertainty about the size and direction of the underlying group contrast.

This result does not establish that the intervention works, nor does it establish that it does not work. A group difference in a completed trial is not the same as a definitive causal conclusion. Interpretation depends on the study design, the analysis population, measurement quality, missing data, and prespecified statistical methods. Several of these details are not reported in the registry record.

The primary endpoint captures whether participants crossed a 50% seizure-reduction threshold. It does not describe the full distribution of seizure changes, the duration of any change, or the experience of each participant. The secondary measures add different kinds of information but do not remove the uncertainty around the primary comparison. The registry does not establish a clinical threshold for interpreting the secondary changes or a multiplicity plan for their analyses.

The record describes a randomized, quadruple-masked, parallel-group phase 2 study, but does not provide all operational details needed to assess those design features. It also does not explain why the primary outcome denominators are smaller than the enrollment count. These limitations narrow what can be concluded from the posted summaries.

The most defensible reading is descriptive: the posted responder rates favor the treatment group numerically, while the interval and p-value indicate that the comparison remains uncertain. The registry results alone do not support a broad conclusion about clinical effectiveness, effects in other populations, long-term outcomes, or the balance of benefits and harms.

17. Why This Trial Matters Statistically

This trial illustrates why statistical interpretation requires more than comparing two percentages. The primary outcome is a threshold-based responder rate, so its estimate depends on both the number of participants and the number who meet the defined criterion. Reporting arm-specific rates, event counts, an effect estimate, and an interval helps distinguish the observed pattern from the uncertainty around it.

The result also shows why a p-value should not be treated as a verdict. The primary p-value is 0.096, while the confidence interval for the risk difference extends across zero. The p-value does not describe the likely range of effect sizes, and the interval does not establish whether any possible difference would be important. Each statistic contributes a different piece of information.

The secondary outcomes demonstrate that a trial can measure more than one aspect of an intervention, while raising questions about how multiple outcomes should be interpreted together. A responder rate, a mean seizure-change measure, and a mood score do not answer the same question. Without a reported hierarchy or multiplicity strategy, isolated secondary findings warrant caution.

Finally, the registry’s missing details matter statistically. The difference between enrollment and the primary outcome denominators, the absence of a stated analysis population, and the lack of reported missing-data and analysis plans limit the evaluation of the estimates. These are not merely administrative omissions: they affect how confidently readers can connect the posted summaries to the randomized trial as a whole.

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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The registry does not explain why the enrollment of 50 differs from the primary outcome denominators of 23 and 19. It does not report participant flow, reasons for missing outcome data, or the method used to handle missingness. Without this information, it is not possible to determine whether the posted comparison reflects all randomized participants or a subset defined by completion or data availability.

Although the study is described as randomized and quadruple-masked, the registry does not report the randomization procedure, allocation concealment, which groups or roles were masked, or how masking was maintained. The control arm description is blank, and the exact control intervention is not reported. These details are relevant to understanding the comparison and the potential for bias.

The primary endpoint description gives a 50% responder threshold and a timeframe, but the record does not provide detailed seizure-measurement procedures or explain how the first 6 weeks and end-of-treatment assessment were operationalized. The registry does not report the confidence level or method for the primary interval, the test used for the primary p-value, or the original statistical analysis plan. Interim analyses and the trial’s original sample-size justification are also not reported.

There are additional limits to the secondary results. The registry lists five outcome measures but posts only two secondary outcomes, and it does not report the full set of planned analyses, multiplicity handling, or a prespecified hierarchy. The mood scale’s favorable direction and interpretation are not described in the outcome information. The safety summary gives event counts but limited detail on event timing, relatedness, and collection methods; deaths are not reported in the registry record.

These limitations mean that the posted results can be described, but several aspects of their design and analysis cannot be independently evaluated from the registry record alone.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: External Trigeminal Nerve Stimulation for Epilepsy — NCT01159431.
• Registry record last updated 2013-06-03; status last verified 2013-05.
• Results first posted 2013-06-03.
• 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 record describes a completed, randomized, quadruple-masked, parallel-group phase 2 study of external trigeminal nerve stimulation for epilepsy. Enrollment is listed as 50. The primary outcome is the 50% responder rate at the end of an 18-week treatment period compared with the first 6 weeks. Posted rates are 40.5% for treatment and 15.6% for control, with a primary unadjusted p-value of 0.096.

The record also posts secondary results for mean percent change in seizures and mood. It reports serious adverse events in 0 of 25 active participants and 1 of 25 control participants, and other adverse events in 5 and 4 participants, respectively. Several details needed for fuller interpretation—including missing-data handling, analysis populations, the statistical analysis plan, multiplicity procedures, and the specific control intervention—are not reported in the registry record.