Vagus Nerve Stimulation (VNS) Paired With Tones for Tinnitus: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Vagus Nerve Stimulation (VNS) Paired With Tones for Tinnitus (NCT01962558), a randomized na study. The primary outcome is Number of Participants With Serious Adverse Events.

2. Clinical Question

3. Trial Design

The registry identifies NCT01962558 as a randomized, triple-masked, parallel-group pilot study. Its official title describes a comparison of VNS paired with tones for tinnitus against VNS with unpaired tones. The listed phase is NA. The sponsor is MicroTransponder Inc., and the registry status is completed. The recorded study start is 2014-02, and primary completion is 2015-05.

Enrollment is recorded as 30. The posted arm summaries list 16 participants for VNS Treatment and 14 for VNS Control. Treatment is described as VNS and tones used 2.5 hours per day for a 6-week period. The control description says VNS and tones were given daily over a 2.5-hour period, but not in the same way as the experimental group, with the arrangement intended to be ineffective while preserving blinding.

The registry’s brief summary states that the study assessed safety and efficacy using a randomized, controlled, parallel design. Its primary outcome, however, is the number of participants with serious adverse events during 6-weeks. Other listed outcomes concern adverse events, minimum masking level, and tinnitus handicap measures. The posted record gives four secondary outcomes, while recording six outcome measures in total; the reason for that difference is not reported in the registry record.

Details of the randomization procedure, concealment, masking assessments, site setting, eligibility rules, and treatment adherence are not reported in the registry record. The record also does not describe a protocol-defined primary analysis population or the original statistical design assumptions. Those omissions limit how fully the design and analysis can be reconstructed from the posted results.

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.

Overall Study

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.

Population: Patients with tinnitus

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With Serious Adverse Events, measured over 6-weeks.

The primary endpoint is the number of participants with serious adverse events during the 6-weeks timeframe. It is a count of participants who had at least one event, rather than a measure of event severity, duration, or total event episodes. The registry describes the purpose as assessing serious adverse events during the study and comparing groups to indicate whether there were more in the treatment group than in the control group.

As a safety-oriented endpoint, this count records whether a serious adverse event was observed in each arm. In the posted results, 1 participant in VNS Treatment and 2 participants in VNS Control are recorded with serious adverse events. The arm summaries list 16 and 14 participants, respectively. These counts describe the posted outcome; they do not, by themselves, explain why an event occurred or establish whether it was caused by the device or the tone procedure.

The endpoint is not a direct measure of tinnitus severity or improvement. The registry separately lists change in minimum masking level, percent change in Tinnitus Handicap Inventory, and change in Tinnitus Handicap Questionnaire among its secondary outcomes. Those symptom-related measures should not be conflated with the primary serious-adverse-event count. Results for those secondary outcomes are not included in the registry record for this analysis, so their findings are not reported in the registry record.

General: a participant-level event count is useful for comparing how often participants experience at least one event, but it does not capture repeated events in the same participant as separate participants. Interpretation also depends on consistent event definitions, follow-up, and ascertainment; further details on those procedures are not reported in the registry record.

7. Results Posted in the Registry

• VNS Treatment: 6.2% (1/16) versus 14.3% (2/14) on control; risk difference -8.0 percentage points (95% CI -29.9 to 13.8), risk ratio 0.44 (0.04–4.32); unadjusted two-proportion z-test p = 0.464.

8. Statistical Interpretation

9. Statistical Methods Explained

General statistical concepts help distinguish the event counts from the comparison estimate and explain why the result remains uncertain. The registry record identifies the primary outcome as a participant count and posts an unadjusted comparison, but it does not report a complete statistical analysis plan. The explanations below describe the general meaning of these methods without assuming that unreported procedures were used.

What does a participant count measure?

A participant-level event count records how many people experienced at least one serious adverse event in the outcome timeframe. In this trial, the posted counts are 1 in VNS Treatment and 2 in VNS Control. The arm sizes are 16 and 14, respectively. This structure differs from an event-episode count: if one participant experienced multiple events, a count of participants would still represent that person once for this endpoint. Whether repeated events were handled separately in other safety summaries is not reported in the registry record.

Comparing the proportions rather than only the raw counts accounts for the different arm sizes. The posted comparison is expressed as a risk difference, which is the treatment-arm event proportion minus the control-arm event proportion. Its negative direction corresponds to fewer observed events in the treatment arm. The registry also lists a risk ratio and an odds ratio with confidence intervals, but the primary interpretation is the risk-difference comparison. The registry entries do not describe which measure was designated as the main analysis beyond the primary outcome itself.

What does the p-value describe?

The posted unadjusted p-value is 0.464. In general, a p-value measures how unusual data at least as incompatible with a specified null model would be if that model and the test assumptions were correct. It is not the probability that a treatment effect is real, and it does not measure the size or clinical importance of a difference. A value that does not indicate strong evidence against the null does not establish equivalence; demonstrating equivalence requires a defined margin and an analysis designed to assess it. Such a margin and equivalence analysis are not reported in the registry record.

“Unadjusted” indicates that the posted p-value is not described as adjusted for other variables or for multiple comparisons. It does not reveal all details of the test itself. The specific test, test assumptions, and any prespecified analysis choices are not reported in the registry record. The registry records zero analyses, so additional analysis records cannot clarify these details.

Why does the confidence interval matter?

A confidence interval gives a range of effect values compatible with the data and the method used, subject to that method’s assumptions. It expresses uncertainty around an estimate rather than showing every possible value with equal plausibility. The interval for the risk difference includes zero and values on both sides of zero. Zero represents no difference in the arm-level event proportions. Thus, the interval does not isolate a single direction of effect, and the result is not precise enough to rule out either direction.

General: a wide interval commonly indicates limited information about the effect estimate. Sparse event data can make an interval wide because the observed counts provide little information about the underlying rates. Here the posted serious-event counts are small, and the interval is broad. The registry does not state whether an exact, asymptotic, or other interval method was used, so the interval’s construction cannot be independently characterized from the record.

What does randomization contribute?

Random assignment is intended to make treatment groups comparable in expectation, reducing systematic differences in measured and unmeasured baseline factors. It does not guarantee balance in every realized small trial. The record labels the study randomized, but it does not report the sequence-generation procedure, allocation concealment, or baseline balance information. The fact that a study is randomized therefore supports the design description but does not remove uncertainty arising from few participants and few events.

The record also labels the study triple masked. Masking can reduce the influence of expectations on care, reporting, or outcome assessment, depending on who is masked and how. The identities of the masked parties and any assessment of masking success are not reported in the registry record. The control description says that VNS and tones were delivered in a manner intended to maintain blinding, but further operational details are not available.

Why might rare outcomes need particular care?

General: binary outcomes with uncommon events can be sensitive to individual observations. Standard approximations may be less reliable when event counts are sparse, and exact or alternative methods may be considered in some analyses. The registry’s sample-size illustration specifies a normal approximation, but that assumption pertains to the planning illustration and does not establish which method was used for the trial’s reported comparison. The trial’s primary analysis method is not reported in the registry record.

These considerations do not invalidate the posted result. They clarify why the small observed counts and broad interval warrant cautious interpretation, and why a p-value alone cannot settle the question. The available record supports a descriptive statement about the observed counts and an imprecise unadjusted comparison, not a claim of a definitive safety difference.

10. Confidence Intervals and Precision

The confidence interval around the risk-difference estimate is broad and crosses zero. In practical terms, the data are compatible with no difference in the proportion of participants experiencing a serious adverse event, as well as with differences in either direction. The point estimate favors fewer observed events in VNS Treatment, but the interval shows that this direction is uncertain.

Precision concerns how narrowly the data locate an effect estimate; it is distinct from whether a test passes a significance threshold. Here, the interval’s breadth means the estimate does not tightly determine the size or direction of a difference. The unadjusted p-value of 0.464 is consistent with that uncertainty, but the interval conveys more directly that materially different effect values remain compatible with the result.

The sparse serious-event counts contribute to this limited precision. When only a few participants have an event, the estimate can be sensitive to individual observations. The recorded arm sizes are 16 for treatment and 14 for control, and the primary outcome timeframe is 6-weeks. The registry does not report a precision target or a clinically important risk-difference threshold, so the interval cannot be assessed against a stated decision threshold.

General: confidence intervals depend on the statistical method and its assumptions. The method used to construct the posted interval is not reported in the registry record. The interval should therefore be interpreted as the posted uncertainty summary, without assuming a specific calculation method or treating its endpoints as absolute bounds on the true effect.

11. 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 planning figures are participants per arm, not total event counts. They are illustrations based on the observed effect rather than the trial’s original design. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the arms, a normal approximation, and an effect to detect set to the stated fraction of the observed effect. The trial’s own original design assumptions are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm for an analysis to have an 80% chance of detecting the specified effect under those assumptions, if that effect is the true one. The 90% power column gives the corresponding planning number for a 90% chance. These are design-planning concepts, not guarantees that a future study will reach a particular conclusion; actual results also depend on the data and whether assumptions hold.

For planning at 100% of the observed effect, the figures are 223 participants per arm for 80% power and 298 participants per arm for 90% power. At 75% of the observed effect, they are 432 and 577 participants per arm, respectively. At 50% of the observed effect, they are 1,047 and 1,400 participants per arm, respectively. The pattern illustrates that targeting a smaller effect requires more participants under the stated assumptions.

These are planning illustrations, not observed power. They are not the enrollment target documented for this pilot and do not show what power the completed study actually had. The registry record does not report the assumptions used to plan the trial itself.

12. 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 = 0.143; treatment proportion p₁ = 0.062; average p̄ = 0.103.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4293 = 0.8414.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4255 = 0.3581.
4. n = (0.8414 + 0.3581)² / (-0.0804)² = 222.8, rounded up to 223 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 <- 0.142857142857143                          # control proportion
p1 <- 0.0625                          # 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(1962558)
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 = 223 (80% power), 298 (90% power)
 75% of observed effect: n per arm = 432 (80% power), 577 (90% power)
 50% of observed effect: n per arm = 1,047 (80% power), 1,400 (90% power)
power.prop.test: 222.8 
Simulated power at n = 223 per arm: 0.816 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists four secondary outcomes: number of participants with adverse events, change in Minimum Masking Level in units of dB, percent change in Tinnitus Handicap Inventory, and change in Tinnitus Handicap Questionnaire. It also records six outcome measures in total, but results for the listed secondary outcomes are not reported in the registry record. The reason the total outcome-measure count differs from the number of listed secondary outcomes is not reported.

When a study examines multiple outcomes or multiple comparisons, the chance of at least one apparently noteworthy result can increase. General methods for addressing this include prespecifying a primary comparison, adjusting significance thresholds, or treating secondary analyses as exploratory. The registry record does not report a multiplicity strategy, so it is not possible to determine whether or how the secondary outcomes were accounted for in statistical inference. The posted primary outcome remains the serious-adverse-event count.

The record identifies the study as randomized and triple masked, but does not specify the analysis population. It is not reported whether the primary comparison included all randomized participants, only participants who completed treatment, or another defined group. The arm summaries report 16 for VNS Treatment and 14 for VNS Control, but the population rules behind those summaries are not described.

Missing-data handling, participant withdrawals, outcome ascertainment completeness, and any imputation are not reported in the registry record. These details matter because missing outcome information can affect the observed comparison, particularly when event counts are small. The registry also records zero analyses and does not report sensitivity analyses, covariate adjustment, or interim analyses. None of those procedures should be assumed to have occurred or not occurred based solely on the posted primary comparison.

14. Safety Results

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

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hoarseness (7.1%), surgery (7.1%).

The serious-adverse-event figures count participants with at least one serious adverse event during the study timeframe. The posted results record 1 participant with a serious adverse event among 16 in VNS Treatment and 2 among 14 in VNS Control. These are observed counts, not estimates of causation. The registry’s primary outcome description says the counts are compared to assess whether there were more serious events in the treatment arm than in control.

The adverse-event record also lists Hoarseness and Surgery among the top serious terms, each at 7.1%. These terms describe reported serious-event categories; the record does not provide enough detail to connect either term to an individual participant, explain the circumstances, or establish attribution to an intervention. The number of deaths is not reported in the registry record.

Other adverse events are recorded for 2 participants in VNS Treatment and 1 in VNS Control. These figures are separate from the serious-adverse-event counts. The registry record does not provide additional details about severity categories, event timing, recurrence, or assessment of relatedness in the registry record. The small counts and short 6-week timeframe limit what can be learned about less common or later events from this comparison.

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

The primary result describes a small numerical difference in serious-adverse-event counts: 1 participant in VNS Treatment and 2 in VNS Control. The point estimate is in the direction of fewer observed serious adverse events in treatment, but its confidence interval includes no difference and values favoring either arm. The unadjusted p-value is 0.464. Taken together, the posted data do not provide a precise estimate of the difference between groups.

This result does not establish that the interventions are equivalent, nor does it establish that one caused fewer serious adverse events. A non-significant comparison is not proof of no effect. The registry does not report an equivalence margin or an analysis designed to demonstrate equivalence. The counts also do not explain whether an event was caused by VNS, tones, the procedure, or another factor.

The study is described as a pilot, with enrollment of 30 and a 6-week primary outcome timeframe. The planning illustrations based on the observed effect are substantially larger than the arm sizes in the posted results, but they are not the trial’s original design assumptions and do not determine the interpretation of its observed outcome. The record does not report the rationale for the pilot’s enrollment or its intended precision.

Although the trial concerns tinnitus and its title emphasizes VNS paired with tones, the primary result addresses serious adverse events rather than symptom response. The registry lists tinnitus-related secondary measures, but their results are not reported in the registry record. Consequently, the primary result cannot establish whether tinnitus symptoms changed or whether the paired-tone approach differed from the control procedure on those measures.

The most supportable reading is limited: the recorded serious-event counts were low, the estimated between-arm difference is uncertain, and the posted data do not resolve whether the groups differed. These observations should not be extended to other patient populations, longer follow-up, or outcomes not reported in the registry record.

16. Why This Trial Matters Statistically

This study is statistically informative because it illustrates the difference between an observed direction and a precise conclusion. The treatment arm had fewer recorded serious-adverse-event participants than the control arm, but the confidence interval includes no difference and allows either direction. A point estimate alone can be compelling at a glance; uncertainty around it is essential to judging what the data establish.

The outcome is based on sparse event counts in a pilot study. With few observed events, a small change in the count can shift the estimated comparison, while the confidence interval remains broad. This makes it especially important not to interpret a p-value as proof of no difference or as a measure of the probability that the intervention is safe. The posted p-value and interval instead indicate that the comparison is uncertain.

The registry’s planning illustration also makes the relationship between effect size and sample size visible. Under the listed assumptions, planning for a smaller fraction of the observed effect requires more participants per arm. Such illustrations can help readers understand the scale of a future study needed to examine a target effect, but they are not evidence about the design rationale for this completed pilot.

Finally, the mismatch between the trial’s tinnitus focus and its posted primary safety endpoint matters for interpretation. Safety counts and symptom outcomes answer different questions. The registry lists symptom measures, but their results are not reported in the registry record. Keeping the endpoint distinction clear prevents a safety comparison from being mistaken for evidence about tinnitus response.

17. 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 comparison is based on few serious adverse events, and the resulting confidence interval is broad. The trial is a pilot with enrollment recorded as 30, so the posted comparison has limited precision for detecting or excluding differences in this outcome. The study’s original sample-size assumptions and rationale are not reported in the registry record.

The primary outcome timeframe is 6-weeks. This record therefore does not describe longer-term serious adverse events. The registry does not report details about event adjudication, how seriousness was defined, event timing, or how relatedness to the intervention was assessed. These gaps limit interpretation of the event categories and their causes.

The record describes the control procedure as VNS and tones delivered differently from the experimental group, in a way intended to be ineffective while maintaining blinding. Additional details about the procedure, implementation, treatment adherence, and success of masking are not reported in the registry record. Randomization and triple masking are identified, but the allocation method and the identities of masked parties are not specified.

Although several secondary outcomes are listed, their results are not reported in the registry record. The record does not report the analysis population, missing-data handling, withdrawals, multiplicity strategy, or sensitivity analyses. It also records zero analyses, limiting the available detail on the statistical procedures behind the posted comparison. These limitations mean the primary result should be treated as a preliminary, imprecise safety comparison rather than a complete account of efficacy or longer-term safety.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Vagus Nerve Stimulation (VNS) Paired With Tones for Tinnitus — NCT01962558.
• Registry record last updated 2018-04-06; status last verified 2018-03.
• Results first posted 2018-04-06.
• 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.

21. Record Summary

NCT01962558 is a completed randomized, triple-masked, parallel pilot study sponsored by MicroTransponder Inc. It compared VNS Treatment, described as VNS paired with tones, with VNS Control, described as VNS and tones delivered differently to maintain blinding. Enrollment is recorded as 30; the arm summaries list 16 in treatment and 14 in control. The primary outcome is the number of participants with serious adverse events during 6-weeks.

The posted counts are 1 in VNS Treatment and 2 in VNS Control. The unadjusted p-value is 0.464, and the confidence interval for the risk difference includes no difference and values in either direction. The result is therefore uncertain and does not establish equivalence or causation. The registry lists tinnitus-related secondary outcomes, but their results are not reported in the registry record.