Vagus Nerve Stimulation and Stress Reduction Training for Migraine: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Vagus Nerve Stimulation and Stress Reduction Training for Migraine (NCT03592329), a randomized na study. The primary outcome is Brain Activity Changes in Migraine Patients in Response to Treatment.

2. Clinical Question

3. Trial Design

The registry describes a randomized, triple-masked, parallel study with two components. One is a cross-sectional assessment of brain activity and inflammation in migraine patients compared with healthy controls. The other assesses an 8-week combination therapy approach for migraine. The reported primary treatment outcome is an fMRI measure taken at the post-treatment time point, with change assessed against baseline.

In the treatment component, the four listed combinations are active tVNS plus MBSR, active tVNS plus NEC, sham tVNS plus MBSR, and sham tVNS plus NEC. The registry describes the training interventions as Stress Reduction Training A and Stress Reduction Training B, while the posted arm labels use MBSR and NEC. The primary comparison summaries use sham tVNS plus MBSR as the reference. They do not report a single overall comparison of active versus sham stimulation or a formal comparison of the two training programs.

The registry reports enrollment of 193 and a COMPLETED status. The posted primary outcome analysis is narrower than overall study enrollment: it includes only participants who successfully completed both pre- and post-treatment 7T fMRI scans. The record notes that some study completers were excluded from scanning because of MRI contraindications. The precise number of people assigned to each treatment arm is not stated in the primary outcome summaries as the overall randomized group size; those summaries give the analyzed fMRI counts.

The study is listed as phase NA. The registry gives a start date of 2019-08-20 and a primary completion date of 2025-01-14. The protocol's original sample-size assumptions and the identities of people covered by triple masking are not reported in the registry record.

4. Participant Flow and Accounting

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

Before assignment: Participants completed a daily headache log run-in to assess eligibility prior to randomization. Participants were excluded if they reported fewer than 4 or greater than 20 headache days within a 28-day period. Final eligibility was assessed by doctoral-level review, and eligible participants were randomized to one of 4 study arms. In total, 193 participants began the run-in period; 14 healthy controls and 43 migraine participants were found ineligible and excluded prior to randomization.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Only participants who successfully completed a pre- and post-treatment 7T fMRI scan are included in this analysis. Some participants completed the overall study but were excluded from fMRI scanning due to MRI contraindications, e.g. metal implants, claustrophobia.

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 Brain Activity Changes in Migraine Patients in Response to Treatment, measured over 8 weeks (i.e. post-treatment).

The primary endpoint is the mean post-treatment versus baseline change in an intervention-evoked cortical amplification ratio. It is based on fMRI blood-oxygen-level-dependent (BOLD) percent signal change in the left posterior insula relative to the spinal trigeminal nucleus during trigeminal afferent stimulation, described as forehead stimulation. The reported unit is a log ratio of percent signal change (Insula/Sp5), and the endpoint is assessed at 8 weeks, after treatment.

This measure is used to quantify a brain response to a specified sensory stimulation rather than a self-reported symptom or a count of migraine episodes. A ratio compares the signal in two named brain regions, and a log transformation expresses that ratio on a transformed scale. In general, this type of fMRI measure can help characterize neural responses relevant to the study's brain-mechanism question. It is not itself a direct measure of how participants feel or function.

The analysis population matters to interpretation. Only people with usable pre- and post-treatment 7T fMRI scans are included, so the endpoint does not represent every participant who completed the overall study. The registry does not report the original protocol rationale for selecting this measure or its clinical threshold for meaningful change.

7. Results Posted in the Registry

• Active tVNS + MBSR: mean -0.25 versus 0.68 on control; difference -0.93 (95% CI -1.92 to 0.06), standardized effect d = -0.73; z-test p = 0.067.
• Active tVNS + NEC: mean 0.68 versus 0.68 on control; difference -0.00 (95% CI -0.88 to 0.87), standardized effect d = -0.00; z-test p = 0.998.
• Sham tVNS + NEC: mean 1.06 versus 0.68 on control; difference 0.38 (95% CI -0.63 to 1.39), standardized effect d = 0.28; z-test p = 0.460.

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.

Brain Activity Differences Between Migraine Patients and Healthy Controls

• 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:
  • Migraine Patients: mean 0.01 versus -0.11 on control; difference 0.13 (95% CI 0.07 to 0.19), standardized effect d = 0.86; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Migraine Patients than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry identifies secondary questions about brain activity and brain inflammation differences between migraine patients and healthy controls. The posted numeric secondary result is a baseline-window comparison of percent signal change in brain activity. The registry does not post a numeric result for the brain-inflammation outcome, and it lists no reported analysis entries. The secondary information therefore provides only a partial view of the planned or registered secondary questions.

The cross-sectional brain-activity comparison addresses a different scientific question from the primary treatment endpoint. It compares migraine patients with healthy controls during the baseline window, while the primary result compares changes after treatment across randomized treatment combinations. A baseline group difference cannot establish that a treatment changed brain activity, and the randomized treatment contrasts do not answer whether migraine patients differ from healthy controls. The endpoints should therefore be interpreted in their separate contexts rather than combined into one conclusion.

The secondary comparison has a posted unadjusted p-value, and the registry record does not report a plan for controlling multiplicity across secondary outcomes or analyses. When many outcomes or comparisons are examined, some may appear more striking by chance. This general concern means an unadjusted secondary result carries less confirmatory weight than a result from a prespecified, multiplicity-controlled analysis. The record does not provide enough information to determine whether this comparison was prespecified as a key hypothesis or how it fits into a broader analysis plan.

The secondary result can be viewed as evidence of an observed baseline association in the analyzed groups, with its own confidence interval and uncertainty. It does not confirm the primary treatment finding, establish a mechanism, or substitute for an unreported inflammation result. The relationship between the cross-sectional outcomes and the treatment outcome is conceptual: both concern brain measures in migraine, but they use different populations, time points, and comparisons.

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

• Brain Activity Differences Between Migraine Patients and Healthy Controls
• Brain Inflammation Differences Between Migraine Patients and Healthy Controls

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it is a mean change on the log-ratio fMRI scale. For each arm, the registry posts an analyzed sample size, a mean, and a standard deviation. The mean describes the center of the observed values; the standard deviation describes how dispersed individual values are around that mean. A standard deviation is not the same as a confidence interval for the mean, and it does not indicate how certain the treatment contrast is on its own.

What does a mean difference represent?

A mean difference compares the average outcome in one group with the average in a reference group, on the outcome's original reported scale. The posted comparisons use sham tVNS plus MBSR as the reference. A negative estimate means the mean in the named comparison group is lower than the reference mean under the direction used for the contrast; a positive estimate means it is higher. These estimates describe differences in group averages, not the range of individual responses and not the probability that an individual would have a particular outcome.

The comparison summaries are unadjusted re-calculations from the posted group summaries. They are not described as the sponsor's prespecified model. The registry record does not report the original statistical analysis plan, any adjustment variables, or the model used for the sponsor's own analysis. Since the discrepancy flag is false, there is no flagged disagreement between the posted summaries and these calculations. The registry record does not report a sponsor analysis result that should be substituted for them.

Why are standard deviations relevant?

Variation within groups affects the precision of a mean difference. When outcomes vary widely among participants, a group mean is less informative about the mean that might be expected in a broader population, especially when the analyzed group is small. General statistical principle: if outcomes are more dispersed, a study usually needs more observations to estimate a group contrast with a similar degree of precision, all else being equal. The posted standard deviations are relatively large compared with some of the estimated contrasts, which is one reason the intervals are broad.

The posted comparisons use pooled standard deviations and report standardized effect estimates as well as mean differences. A standardized effect expresses the mean difference relative to a measure of within-group variation. It can help compare effects across scales, but it does not replace the original-scale estimate or its confidence interval. The reported standardized estimates are -0.725 for active tVNS plus MBSR, -0.001 for active tVNS plus NEC, and 0.279 for sham tVNS plus NEC, each relative to sham tVNS plus MBSR.

How should the p-values be read?

A p-value describes how incompatible the observed data are with a specified null model, under the assumptions of the statistical test. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of a difference. The posted values are unadjusted: they do not account for the fact that multiple comparisons or outcomes may be examined. The registry record does not report a multiplicity adjustment or the full testing plan, so the p-values should be treated as descriptive evidence for the individual contrasts, not as a complete confirmatory testing framework.

A p-value above a conventional threshold does not show that groups are the same. To support equivalence or non-inferiority, researchers need a suitable design, a prespecified margin, and an analysis targeted to that question. Those details are not reported in the registry record. Conversely, a p-value alone does not convey whether a difference is large enough to matter; the estimate, its uncertainty, the endpoint's meaning, and the study context all matter.

What does randomization contribute?

Random allocation is intended to make treatment groups comparable on average before treatment, reducing the risk that observed differences are explained by systematic allocation choices. It does not guarantee balance in every characteristic, particularly in modest analyzed groups, and it cannot eliminate uncertainty due to sampling variation. The registry identifies the design as randomized, triple-masked, and parallel, but does not report the allocation sequence method or who was masked. The posted summaries alone do not show whether randomization was preserved in the fMRI analysis population after exclusions for scan completion or MRI contraindications.

Why can over-dispersion matter?

General explanation: over-dispersion means that observed variation is greater than a statistical model expects. For continuous outcomes, substantial variation can make estimates less precise; for count outcomes, over-dispersion can make a simple model understate uncertainty if it assumes too little variability. The primary endpoint is continuous, and the registry does not report a count model or an over-dispersion assessment. The relevant general point is that the observed standard deviations and confidence intervals deserve attention rather than relying on a single test statistic.

The registry also posts a cross-sectional comparison of brain activity between migraine patients and healthy controls. That is a different question from the randomized treatment comparison: it compares groups at baseline rather than estimating a treatment contrast. The posted analysis reports a mean difference of 0.129, a 95% confidence interval from 0.065 to 0.192, and an unadjusted p-value of <0.001. This comparison is observational with respect to migraine status and does not show that migraine caused the difference. The registry lists secondary outcomes involving brain activity and brain inflammation, but no reported analysis entries for them.

11. Confidence Intervals and Precision

Confidence intervals describe a range of values reasonably compatible with the observed data under the calculation's assumptions. They are useful for judging both the direction and precision of an estimate. A narrow interval indicates greater precision than a wide interval on the same scale; an interval that includes zero leaves open the possibility of no mean difference as well as differences in either direction. The intervals posted for all three primary treatment contrasts include zero.

For active tVNS plus MBSR versus sham tVNS plus MBSR, the estimated mean difference is -0.929 and the 95% confidence interval runs from -1.922 to 0.064. The interval permits a range of plausible contrasts, extending from a sizable negative difference to a small positive one. For active tVNS plus NEC versus the reference, the estimate is -0.001 and the interval runs from -0.876 to 0.874. The estimate is near zero, but the interval is broad enough to include meaningful differences in either direction on this scale. For sham tVNS plus NEC versus the reference, the estimate is 0.38 and the interval runs from -0.627 to 1.387, again spanning zero and values on both sides.

These intervals are not probability statements about a fixed effect, nor do they indicate the range containing most individual outcomes. They quantify uncertainty in the estimated mean contrast under the calculation's assumptions. They also do not incorporate any unreported multiplicity adjustment, missing-data model, or other analysis choices. The registry record does not report the original protocol's target precision or a threshold for a meaningful change in the fMRI measure, so the practical importance of values within these intervals cannot be determined from the intervals alone.

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 planning figures are participants per arm for these comparisons, not total events. The calculations are illustrations based on the observed effect rather than the trial's original design assumptions. Their stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning scenario targets the stated fraction of the observed effect. The original trial's sample-size assumptions are not reported in the registry record.

The 80% power column gives the estimated participants per arm for a hypothetical study to have an 80% chance of detecting the specified effect under the planning assumptions, if that effect is the true one and the assumptions hold. The 90% power column describes the corresponding planning target for a higher chance of detection. These are design calculations, not guarantees: actual power and precision depend on the true effect, variability, analysis, and study conduct.

Rows planning for 100% of the observed effect use the full observed estimate as the target; rows planning for 75% or 50% use smaller target effects. Smaller effects need more participants to distinguish them from random variation at the same planned power. This is especially consequential when an observed estimate is close to zero, since planning around that estimate can produce very large requirements. Such figures should not be read as a recommendation to conduct a study at that size, nor as evidence about what the original study planned. The sample-size figures are planning illustrations, not observed power.

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 = -0.929; pooled standard deviation σ = 1.281.
2. n = 2 × (1.96 + 0.842)² × 1.281² / -0.929² = 29.8, rounded up to 30 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 <- -0.929                      # observed difference in means
sigma <- 1.28085908670704                     # 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(3592329)
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 = 30 (80% power), 40 (90% power)
 75% of observed effect: n per arm = 54 (80% power), 72 (90% power)
 50% of observed effect: n per arm = 120 (80% power), 160 (90% power)
power.t.test: 30.8 
Simulated power at n = 30 per arm: 0.790 (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. Multiplicity, Missing Data and Analysis Populations

The posted treatment results include three contrasts against the same reference group, and the registry identifies multiple outcomes, including primary and secondary brain measures. Testing more than one contrast or outcome increases the chance that at least one result may appear notable by chance when each is considered separately. The posted p-values are unadjusted. The registry record does not report a multiplicity-control procedure, a hierarchy of hypotheses, or a prespecified plan that explains how the several comparisons were to be interpreted together.

The record lists six outcome measures and two secondary outcome topics, but does not report analysis entries for the secondary outcomes. It also does not report a detailed analysis plan, the treatment of any interim analyses, or whether the posted comparisons were prespecified. These missing details prevent a full assessment of whether the reported analyses represent all planned analyses or how selective reporting was addressed.

The primary fMRI analysis includes only participants who successfully completed pre- and post-treatment scans. The registry notes that some participants completed the overall study but did not undergo scanning because of MRI contraindications. It does not report the reasons for all missing scans, the number of missing endpoint observations by treatment arm, or an analysis of whether the fMRI-completer groups remained comparable after these exclusions. It also does not report an imputation method, a missing-data sensitivity analysis, or an intention-to-treat analysis for this endpoint.

These omissions do not show that the analysis is biased, but they limit assessment of robustness. If scan completion is related to treatment assignment or to outcome-related characteristics, the analyzed groups may differ from the randomized groups in ways that affect the comparison. The registry record does not report enough information to determine whether that occurred. The interpretation should therefore remain tied to the fMRI-completer population and the unadjusted comparisons that are posted.

15. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: surgery (3.6%), kidney infection (3.6%).

The registry posts serious adverse event figures separately for the four treatment combinations and healthy controls. The number at risk for serious adverse events is 25 for active tVNS plus MBSR, 28 for active tVNS plus NEC, 25 for sham tVNS plus MBSR, 29 for sham tVNS plus NEC, and 29 for healthy controls. The recorded serious adverse event counts are 0, 2, 0, 0, and 0, respectively. These are registry counts for the listed groups; the record does not provide further detail about timing, relatedness to an intervention, or event-level circumstances.

For other adverse events, the numbers at risk and affected are 25 and 14 for active tVNS plus MBSR, 28 and 12 for active tVNS plus NEC, 25 and 18 for sham tVNS plus MBSR, 29 and 13 for sham tVNS plus NEC, and 29 and 6 for healthy controls. These figures should be read as counts of affected participants in the registry's category, not as an assessment of severity or causal attribution.

The registry lists Surgery and Kidney Infection as top serious adverse event terms, each reported as 3.6%. No deaths are recorded in the listed groups. The posted information alone does not establish whether an event was caused by treatment or whether the groups differ in safety; such conclusions require context about definitions, follow-up, and attribution that is not reported in the registry record.

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

The primary result describes how group-average changes in a particular fMRI ratio differed across treatment combinations at the post-treatment assessment. The three posted comparisons use sham tVNS plus MBSR as the reference, and each confidence interval includes zero. This means the posted data do not provide a precise basis for distinguishing those group means from the reference on this measure. It does not prove that the combinations are identical or that there is no effect.

The endpoint is a brain-activity measure during a specified stimulation task. It is not a direct measure of migraine frequency, pain, disability, or quality of life. The result therefore cannot by itself establish a clinical benefit or lack of benefit. The registry record does not report a clinical threshold for a meaningful change on this fMRI scale, and the primary analysis applies only to participants who completed both required scans.

The estimates also do not isolate the separate contributions of tVNS and stress reduction training. The posted contrasts compare particular combinations, rather than reporting a complete factorial analysis of stimulation, training, and their interaction. A difference between two combinations, if observed, would not automatically identify which component produced it. The registry record does not report a formal interaction analysis.

The cross-sectional brain-activity comparison addresses migraine patients versus healthy controls at baseline. It is not randomized by migraine status and cannot determine whether migraine caused any observed difference. The brain-inflammation outcome has no numeric analysis posted in the record. More broadly, unadjusted comparisons across several outcomes and contrasts require caution, particularly without a reported multiplicity plan.

Overall, the posted evidence is best understood as a set of uncertain estimates on a mechanistic imaging endpoint, not as a definitive test of clinical effectiveness, equivalence, or biological mechanism. The data can inform future study planning and help describe the observed scale and variability, but they do not settle the broader clinical question.

17. Why This Trial Matters Statistically

This study illustrates why clinical trial interpretation needs more than a single p-value. The primary outcome is a specialized continuous imaging measure, and its interpretation depends on the population scanned, the change scale, the group variability, and the precision of each contrast. The reported confidence intervals make uncertainty visible: all three treatment comparisons include zero and extend across a range of possible differences.

The study also combines two related but distinct research aims: a cross-sectional comparison of migraine patients with healthy controls and a randomized assessment of treatment combinations. Keeping those aims separate is statistically important. A baseline association between clinical groups is not evidence of a treatment effect, while a randomized treatment contrast is not a test of why those groups might differ.

The treatment combinations make the choice of comparison especially important. A comparison of one combined intervention with another does not automatically identify the separate contribution of stimulation or training. Without a reported factorial or interaction analysis, readers should avoid attributing an observed contrast to one component alone.

Finally, the primary analysis depends on completing specialized imaging before and after treatment. That makes the analyzed population narrower than the overall enrolled population and raises questions about missing scans and generalizability. The registry record does not provide the detail needed to evaluate those issues fully. These features make the trial a useful example of how design, outcome selection, missingness, and multiplicity shape the claims that can be supported by posted results.

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.

The primary endpoint analysis is restricted to participants with both pre-treatment and post-treatment 7T fMRI scans. The registry notes MRI contraindications as one reason some overall study completers were not scanned, but it does not report the complete reasons for missing scans or the arm-specific number of participants excluded from the primary analysis. This limits assessment of selection into the analyzed population and how representative it is of all enrolled participants.

The posted primary group summaries are based on relatively small analyzed groups, and the standard deviations indicate appreciable variability. The confidence intervals are broad and include zero for all posted treatment contrasts. As a result, the estimates do not sharply distinguish among possible directions and magnitudes of difference. The record does not report a clinically meaningful threshold for this transformed fMRI outcome.

The comparisons are unadjusted and use one treatment combination as the reference. No formal factorial effects or interaction results are posted, so the separate roles of active versus sham stimulation and the two training programs cannot be inferred from these contrasts. The registry record also does not report a multiplicity adjustment, full statistical analysis plan, or details of any sponsor model.

The cross-sectional comparison is not a randomized contrast, and the posted brain-inflammation secondary outcome has no numeric result. The record lists six outcome measures but no reported analysis entries. These gaps constrain interpretation of the breadth and completeness of the results. No inference should be made from the absence of a posted result alone about what the unreported analysis would have shown.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Vagus Nerve Stimulation and Stress Reduction Training for Migraine — NCT03592329.
• Registry record last updated 2026-02-25; status last verified 2026-02.
• Results first posted 2026-02-25.
• 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

NCT03592329 is a COMPLETED randomized, triple-masked, parallel study of vagus nerve stimulation and stress reduction training for migraine, with a cross-sectional brain assessment alongside an 8-week combination-treatment component. The primary outcome is mean change from baseline in an fMRI cortical amplification ratio at post-treatment, analyzed among participants who completed pre- and post-treatment scans.

The registry posts four treatment-arm means and three unadjusted comparisons against sham tVNS plus MBSR. Each primary comparison has a confidence interval that includes zero. A separate baseline-window brain-activity comparison is posted for migraine patients and healthy controls; no numeric brain-inflammation result is reported. Safety counts are listed by group, with no deaths recorded. The registry record does not report the original sample-size assumptions, a detailed analysis plan, multiplicity procedures, or a complete account of missing scans.