Concussion Testing: A Novel Non-invasive Concussion Detection Device: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Concussion Testing: A Novel Non-invasive Concussion Detection Device (NCT02370914), a registered phase not stated study. The primary outcome is A Measure Assessing the Sensitivity and Specificity Rate Based on the Nautilus Neurowave Data.

2. Clinical Question

This is a diagnostic-classification question rather than a treatment-effect question. Sensitivity describes the proportion of concussed participants correctly classified as concussed; specificity describes the proportion of non-concussed participants correctly classified as non-concussed. These measures address different kinds of classification error, so one cannot substitute for the other.

The registry record identifies a threshold-based algorithm: a recording is classified as concussed when R1 is at least 1.0 and R2 is at least 0.66, and as non-concussed otherwise. It also states that individual assessments were made within 2 days of recording. However, it does not report the sensitivity or specificity estimates, their confidence intervals, or the underlying classification counts. The registry record therefore describes the question and method but does not provide the numerical results needed to assess observed diagnostic performance.

3. Trial Design

The registry record describes a completed study titled “Concussion Testing: A Novel Non-invasive Concussion Detection Device.” It lists Concussion as the condition, Jan Medical, Inc. as the sponsor, and an enrollment of 84. The recorded study start is 2011-08 and primary completion is 2011-12. The record lists the Nautilus NeuroWave System and SCAT as interventions.

The arm descriptions indicate that a baseline NeuroWave recording was obtained for participants at the start of the study and treated as a control recording from a non-concussed participant. The suspected-concussive-event description says a recording was obtained when concussion was suspected, evaluated by a Jan Medical algorithm, and classified as concussed or not. It also describes a comparison with SCAT, but the text is truncated. Some participants in the cohort were recorded again at the end of the study, according to the partial control-arm description.

The registry record does not describe a randomized comparison. It states that allocation was not stated and that the arms are not a like-for-like control comparison. This matters because baseline control recordings and recordings prompted by a suspected event may come from different assessment circumstances. Their comparison should not be interpreted as if participants had been randomly assigned to equivalent groups.

The record reports one outcome measure and no analysis entries. It does not give further design details such as the complete recruitment procedure, eligibility criteria, or the full reference-standard process. Those details are not reported in the registry record, so the design description must remain limited to the recorded information.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Analysis population includes all participants. Some of these participants were were declared concussed by the physician in the team during the course of the season. All baseline recordings were treated as from non-concussed participants.

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 A Measure Assessing the Sensitivity and Specificity Rate Based on the Nautilus Neurowave Data, measured over Individual assessments were made within 2 days of recording.

The primary endpoint is a measure of sensitivity and specificity based on the Nautilus NeuroWave data, expressed as a percentage of participants. Sensitivity is the proportion of participants considered concussed who are correctly identified as concussed by the algorithm. Specificity is the proportion considered non-concussed who are correctly identified as non-concussed. Sensitivity concerns false-negative classifications; specificity concerns false-positive classifications.

The algorithm’s classification rule uses two parameters, R1 and R2. The registry description says the frequency spectrum is computed using an FFT and that each parameter is a ratio of signal intensities at specified higher and lower frequencies. A recording is considered to come from a concussed participant if R1 is at least 1.0 and R2 is at least 0.66; otherwise it is considered non-concussed. Individual assessments were made within 2 days of recording.

These measures are used to describe diagnostic classification against a reference status. Their interpretation depends on how that status is established and on which observations enter each denominator. The registry record describes some physician declarations and states that baseline recordings were treated as non-concussed, but the complete reference-standard procedure is not reported. It also does not report the endpoint estimates or confidence intervals, so the registry record does not quantify the observed sensitivity or specificity.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What do sensitivity and specificity describe?

In general, sensitivity and specificity are conditional measures of diagnostic classification. Sensitivity asks, among observations that meet the chosen definition of the target condition, what proportion the test classifies as positive. Specificity asks, among observations that do not meet that definition, what proportion the test classifies as negative. In this study’s terminology, a positive classification is “concussed” and a negative classification is “non-concussed.”

These measures are distinct from overall accuracy. Overall accuracy combines classifications across both reference-status groups, while sensitivity and specificity keep the groups separate. The measures also do not by themselves give the probability of the condition after a positive or negative test result. That interpretation depends on the frequency of the condition in the population being assessed, as well as test performance.

How does the recorded algorithm make a classification?

The registry description says that R1 and R2 are calculated from the frequency spectrum using an FFT, with each parameter formed as a ratio of signal intensities at specified higher and lower frequencies. The stated rule classifies a recording as concussed when R1 is at least 1.0 and R2 is at least 0.66. A recording that does not meet both conditions is considered non-concussed. This is a deterministic rule as described: it maps the calculated values to a category.

Having a prespecified or clearly stated cutoff can make a classification rule easier to reproduce and evaluate. But a threshold alone does not demonstrate that classifications agree with the target condition. Evaluation requires applying the rule to recordings and comparing each resulting classification with the reference status. The registry record does not report those classification counts or the endpoint estimates.

What does a confidence interval add?

In general, a confidence interval describes statistical uncertainty around an estimated quantity. For sensitivity or specificity, the estimate is based on the proportion correctly classified within the relevant reference-status group. The interval reflects that a proportion estimated from observed data can vary from the underlying performance that would be seen across repeated samples under the model’s assumptions. A narrower interval indicates greater precision than a wider interval, although precision does not by itself establish that the reference status or study design is appropriate.

The endpoint description lists a 95% confidence interval as the dispersion measure. However, the registry record does not report the interval limits or the associated point estimates. Consequently, neither the magnitude nor the precision of sensitivity or specificity can be evaluated from the posted results.

Why does the reference status matter?

In general, sensitivity and specificity depend on how the target condition is defined and established. If reference classifications are uncertain or applied differently across observations, estimated test performance can be difficult to interpret. The record says some participants were declared concussed by the team physician during the season and that all baseline recordings were treated as non-concussed. It does not report the complete reference-standard process, including how physician assessments and baseline status were used for every individual classification.

The record also mentions SCAT as a comparison in a truncated arm description. It does not report enough detail to characterize the full comparison procedure or establish whether SCAT served as the reference standard for all assessments. This makes it important not to assume that the diagnostic outcome was validated against a single, fully described standard.

What is the relevance of the assessment timing?

The endpoint timeframe states that individual assessments were made within 2 days of recording. In general, the timing of a test relative to the recording and to determination of reference status can matter when interpreting diagnostic comparisons. Here, the recorded timeframe supplies a timing window for assessments, but the registry record does not provide further detail about the sequence or timing of all assessments. No additional timing assumptions can be made from the description.

How should study design inform interpretation?

General statistical interpretation depends not only on the endpoint but also on which observations are compared and how they were selected. The record states that allocation was not stated and that the arms are not a like-for-like control comparison. It describes baseline recordings as controls and suspected-event recordings in the other arm. These circumstances are not described as randomized, equivalent groups, so a comparison between arms should not be read as a randomized treatment contrast. The endpoint is diagnostic performance, but the study description does not resolve every question about selection, repeated recordings, or comparability.

Finally, the registry record reports one outcome measure and no analysis entries. A stated endpoint and a stated analysis are not the same thing: the endpoint defines what was intended to be measured, while analysis results would show the estimates and their uncertainty. Since the numerical results are not reported in the registry record, the methods description cannot substitute for evidence about observed performance.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists one outcome measure and no analysis entries. It does not report a multiplicity adjustment, a hierarchy of hypotheses, or a plan for handling multiple analyses. The absence of analysis entries does not establish that no other analyses were conducted; it means their details are not reported in the registry record.

The record describes the analysis population as including all participants. It notes that some were declared concussed by the team physician during the season and that all baseline recordings were treated as non-concussed. It does not report how many observations had missing or unusable recordings, how missing classifications were handled, or whether every participant contributed to each part of the sensitivity and specificity assessment. Missing-data handling is not reported in the registry record.

For diagnostic measures, the relevant denominators differ: sensitivity is calculated among observations with the reference status of concussed, while specificity is calculated among observations with the reference status of non-concussed. In general, the way missing reference statuses or test results are handled can affect these denominators and the resulting estimates. Since the record gives neither the classification counts nor the missing-data procedures, that aspect of the analysis cannot be assessed.

The registry record does not report a statistical analysis plan or details about prespecification of the endpoint, thresholds, or analysis population. Those details are not reported in the registry record, and no conclusion about their existence or implementation should be inferred.

11. 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 0% reporting threshold.

The adverse-event entry lists the Diagnostic Device - Nautilus NeuroWave group. It reports 82 at risk for serious adverse events and 0 affected. It also reports 82 at risk for other adverse events and 0 affected. The death field is null, so a death count is not reported in the registry record.

These entries describe the counts recorded for the listed group; they do not establish that no events occurred outside the categories or reporting procedures represented in the record. The registry record does not give further details about event definitions, collection methods, follow-up duration, severity grading, or whether events were judged related to the device. Those details are not reported in the registry record.

A zero count in a recorded category means that no affected participants were entered for that category in this registry entry. It should not be read as proof that the event risk is absent. In general, observed counts depend on the size and characteristics of the group, the duration and completeness of monitoring, and the rules used to collect and classify events. The posted information does not provide enough detail to assess those factors or to compare safety with another group.

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

The registry record establishes that the study was intended to evaluate whether data from the Nautilus NeuroWave system could distinguish recordings considered concussed from those considered non-concussed. It gives the algorithm’s R1 and R2 thresholds and names sensitivity and specificity as the primary outcome. It also states that individual assessments were made within 2 days of recording.

What the record does not establish is the observed diagnostic performance. It reports no sensitivity or specificity estimates, no confidence-interval limits, and no counts of true-positive, false-negative, true-negative, or false-positive classifications. Without those results, readers cannot assess either the apparent magnitude or the precision of classification performance.

The record also does not fully describe the reference-standard process. It notes physician declarations for some participants and says all baseline recordings were treated as non-concussed, while a truncated description refers to SCAT. The way these elements were combined for each assessment is not reported in the registry record. This matters because sensitivity and specificity are defined relative to the status used as the reference.

The study is not described as a randomized comparison, allocation is not stated, and the record says the arms are not a like-for-like control comparison. Thus, the available descriptions should not be interpreted as evidence from a randomized comparison between equivalent groups. Nor should the recorded zero adverse-event counts be taken as proof that there is no risk; they are registry entries for specified categories, with further monitoring details not reported.

In short, the record communicates the study’s aim, endpoint, and classification rule, but not the numerical endpoint findings needed to judge accuracy. The appropriate reading is descriptive and limited to what is recorded, without an efficacy or clinical-use conclusion.

13. Why This Trial Matters Statistically

This study is statistically notable because it frames concussion detection as a diagnostic-classification problem. That calls for measures such as sensitivity and specificity, which separate performance among observations classified by reference status. Reporting both measures is important because correct identification of concussed participants and correct identification of non-concussed participants are different properties of a test.

The record also illustrates why an algorithmic threshold is only one component of a diagnostic evaluation. The R1 and R2 rule makes the classification procedure explicit, but the threshold does not show how the system performed against a reference status. That assessment depends on the observed classifications, a clearly described reference process, and estimates with uncertainty. The registry record describes the rule but does not report those numerical results.

Another statistical issue is the structure of the comparison. Baseline recordings were treated as non-concussed controls, while the other arm describes recordings after suspected concussive events. The registry record explicitly cautions that the arms are not a like-for-like control comparison and does not state allocation. Recognizing this prevents a reader from treating the arms as randomized, equivalent groups.

Finally, the record makes clear that endpoint definitions and endpoint results are not interchangeable. A named primary measure, a timeframe, and a decision rule convey what was intended to be assessed. They do not supply the estimates needed to judge precision or performance. This distinction is important in reading registry records: methods can be described even when numerical results are not reported.

14. Limitations

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

Several trial-specific limitations follow from the information recorded. The most consequential is that the registry record does not report sensitivity or specificity estimates, confidence-interval limits, or classification counts. Diagnostic performance and its precision therefore cannot be assessed from the posted results.

The reference-status process is not fully described. The record notes that some participants were declared concussed by a team physician and that all baseline recordings were treated as non-concussed. A truncated arm description refers to comparison with SCAT, but the full role of that tool and the complete adjudication process are not reported in the registry record.

The arm descriptions are incomplete, and the record says allocation was not stated and the arms are not a like-for-like control comparison. It also does not provide the full details needed to understand participant selection, the handling of repeated recordings, or whether recording circumstances were comparable. These design details are not reported in the registry record.

The analysis population is described as including all participants, but the record does not report missing-data procedures, numbers of usable recordings, or the denominators used for sensitivity and specificity. It reports one outcome measure and no analysis entries, while the statistical analysis plan and any multiplicity procedures are not reported in the registry record. The safety entry gives zero affected for the recorded serious and other adverse-event categories, but does not detail event collection, follow-up, or relatedness assessments.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Concussion Testing: A Novel Non-invasive Concussion Detection Device — NCT02370914.
• Registry record last updated 2022-08-02; status last verified 2022-07.
• Results first posted 2017-04-20.
• 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.

18. Record Summary

The registry record describes a completed study of the Nautilus NeuroWave system for concussion detection, sponsored by Jan Medical, Inc. It lists an enrollment of 84, with a study start of 2011-08 and primary completion of 2011-12. The study’s stated purpose was to establish and evaluate a distinctive signal for concussion.

The primary outcome is sensitivity and specificity based on NeuroWave data. The described algorithm classifies a recording as concussed when R1 is at least 1.0 and R2 is at least 0.66. Assessments were made within 2 days of recording. The record describes baseline recordings as non-concussed controls and suspected-event recordings assessed by the algorithm, while noting that the arms are not a like-for-like control comparison.

Numerical diagnostic results and their confidence intervals are not reported in the registry record. The adverse-event entry lists 0 affected for both serious and other adverse events among 82 at risk in each category; the death count is not reported in the registry record.