Evaluation of Cognitive State in Senior Subjects Using Neurosteer EEG System: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Evaluation of Cognitive State in Senior Subjects Using Neurosteer EEG System (NCT04683835), a registered phase not stated study. The primary outcome is Correlation Between Cognition Level Changes as Evaluated by Current Clinical Tools (i.e. MMSE) and Brain Activity Features Extracted Using the Neurosteer Technology..

2. Clinical Question

3. Trial Design

This was a completed observational study sponsored by Neurosteer Ltd. The registry lists an enrollment of 80 and Mild Cognitive Impairment as the condition. The study start date was 2020-10-15 and the primary completion date was 2021-08-01. The record describes neurologic evaluation and EEG recordings during an auditory cognitive assessment, with the EEG processed using proprietary computational analyses.

The device intervention is named as the Neurosteer Aurora system. The registry’s arm description identifies patients assessed by clinical staff using the Mini-Mental State Exam (MMSE). It does not describe randomized allocation, a parallel treatment arm, or a like-for-like control comparison. Allocation is not stated, and the record explicitly characterizes the study as observational. Consequently, the question is about a relationship between measurements, not a contrast between assigned treatments.

The primary outcome timeframe states that the MMSE score was taken from a previous evaluation at the institute. Brain activity features were assessed twice within 7 days using the Neurosteer EEG system. A secondary outcome concerns inter-patient variability between two consecutive measurement sessions. The registry lists two outcome measures and zero analyses.

The registry record does not report the inclusion or exclusion criteria, the number of participants with usable paired or complete measurements, or the detailed analysis population. It also does not report a protocol-based sample-size justification. The described sample-size planning illustration is separate from the study design and should not be read as an account of randomized arms or as evidence of observed power.

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.

6. Primary Endpoint

The registry lists the primary outcome as Correlation Between Cognition Level Changes as Evaluated by Current Clinical Tools (i.e. MMSE) and Brain Activity Features Extracted Using the Neurosteer Technology., measured over MMSE score will be taken from previous evaluation performed in the institute. Through study completion, brain activity features will be assessed twice within 7 days using the Neurosteer EEG system..

The primary endpoint is stated as a Pearson correlation coefficient relating cognition assessed with the MMSE to EEG-derived brain activity features. The outcome description says the correlation is calculated between mean activity of the EEG features and an individual’s MMSE score. The MMSE is described as a commonly used assessment of cognitive functioning, with a total possible score of 30 points; the record notes that scores below 24 are typically suspected of cognitive decline. It does not say that this threshold defined study eligibility or the primary analysis.

The EEG recordings were made during auditory cognitive tasks. The record says the signal is transformed into brain activity features, with ST4 and A0 given as examples, using a variant of wavelet packet analysis and the best basis algorithm. These features are intended to represent signal characteristics that can be compared with MMSE scores. A correlation coefficient can describe the direction and strength of a statistical association between measurements; it does not by itself establish that one measurement causes a change in the other.

There is an important distinction between the endpoint title and its operational description. The title refers to cognition level changes, while the description specifies MMSE scores from a previous evaluation and EEG features assessed twice within 7 days. The registry record does not explain how a change in MMSE would be calculated or how the repeated EEG measurements would be combined beyond referring to mean activity. No endpoint estimate or uncertainty measure is reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is framed around Pearson correlation. In general, correlation describes the direction and degree to which two measured variables vary together. In this study, the record identifies the variables as mean activity of EEG-derived brain activity features and an individual’s MMSE score. A positive association would mean that higher values of one tend to occur with higher values of the other; a negative association would indicate that higher values of one tend to occur with lower values of the other. These are general interpretations, not results reported for this study.

What does a Pearson correlation describe?

In general, Pearson correlation summarizes linear association between quantitative measurements. It does not show that one variable causes the other, and it may not adequately describe a relationship that is strongly curved or otherwise not linear. It can also be affected by unusual observations. A scatterplot and checks of the observed data can help assess whether a linear summary is appropriate. The registry record does not report such checks, the observed distribution of MMSE scores or EEG features, or whether individual feature values were analyzed separately or combined. ST4 and A0 are examples of features in the outcome description, not reported results.

Correlation also depends on how the measurements are defined. The registry says EEG features are assessed twice within 7 days and refers to mean activity, but does not state in detail how observations from the two sessions are combined for the primary analysis. It identifies the MMSE score as coming from a previous institute evaluation. The outcome title uses the language of changes, yet the description does not specify repeated MMSE measurement or a change-score calculation. That difference limits the ability to reconstruct the intended estimand—the precise quantity the analysis is meant to estimate.

Why does repeated measurement matter?

Repeated EEG measurements can help characterize whether measurements vary across sessions, which connects to the named secondary outcome of inter-patient variability between consecutive sessions. In general, repeated observations from the same participant are related rather than independent. An analysis should account for the structure of those observations if it uses session-level data. The record does not report whether the primary analysis used session averages, one session at a time, or another approach. It also does not report a reliability statistic or the result of the secondary outcome.

In general, measurement variability can affect an observed association. If EEG features fluctuate between sessions, that variability may make their relationship with MMSE less precise or harder to interpret. The registry’s mention of a secondary variability outcome indicates that repeat-session variation was of interest, but does not establish how large it was or how it was incorporated into the primary analysis.

What would uncertainty measures add?

A confidence interval is a statistical summary of uncertainty around an estimate under the assumptions of the analysis. For a correlation, it would help show which association values are compatible with the observed data and method. A p value addresses a different question: how compatible the data are with a specified null model, under the assumptions used. Neither measure alone establishes practical importance or causality. The registry record reports no correlation estimate, confidence interval, or p value, so neither the magnitude nor the statistical uncertainty of the primary result can be assessed.

The sample-size planning illustration is described as using a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. These are assumptions for a planning illustration, not a description of this observational study’s allocation or a reported protocol-based design calculation. The numerical planning results and effect fractions are not reported in the registry record. Sample-size illustrations are planning scenarios, not observed power and not evidence that an analysis had a particular probability of detecting an effect.

Why can over-dispersion matter?

In general, over-dispersion means that data vary more than a chosen statistical model expects. When present, a model that assumes less variability than the data contain can understate uncertainty and make results appear more precise than they are. The primary outcome is described as a correlation, and the record does not say that a count model was used or report an over-dispersion assessment. This general concept should not be mistaken for a finding about this study.

More broadly, interpretation depends on clear definitions of the analysis population, measurement processing, missing-data handling, and statistical model. These details are not reported in the registry record. With no posted analysis estimate, the statistical methods can be discussed only at the level of the planned outcome description and general principles, not as a verified account of the analysis actually performed.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists two outcome measures: the primary correlation outcome and a secondary outcome concerning inter-patient variability between two consecutive measurement sessions. It reports zero analyses. The registry record does not describe a multiplicity strategy, such as whether one outcome was designated as the principal basis for inference or whether any adjustment was planned for examining multiple outcomes or EEG features. It also does not report how many feature-level comparisons were ultimately made. Without those details, the extent of any multiple-testing issue cannot be assessed.

The registry record does not report missing-data rates, reasons for missing measurements, or a method for handling incomplete MMSE or EEG data. It also does not identify an analysis population, such as whether the primary analysis required complete measurements from both EEG sessions. Enrollment of 80 does not establish the number with usable data for each outcome. The number analyzed and the characteristics of any participants omitted from analysis are not reported in the registry record.

In general, missing measurements can affect an association analysis when the availability of data is related to participant characteristics or measured values. Complete-case analysis may then describe a different subset from the enrolled population. Alternative methods make their own assumptions and should be described explicitly. These are general considerations; the record does not say which method, if any, was used.

The registry record also does not report a statistical analysis plan, rules for handling outliers, or procedures for selecting EEG features. Those details matter because analytic decisions can influence estimates and uncertainty, particularly when several candidate features are available. No conclusions about such decisions can be drawn from the registry information.

11. 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.

The registry lists 80 participants at risk in the Patients group. It reports zero serious adverse events among those 80 participants and zero deaths. It also reports zero other adverse events among 80 participants. These are the counts recorded in the registry for the listed group.

These figures describe what was entered for this study; they do not show that no adverse events of any kind occurred outside the categories or reporting procedures represented in the registry record. The record does not provide event terms, definitions, collection procedures, follow-up detail, or an explanation of how adverse events were assessed. No serious adverse-event terms are listed.

The study is observational and does not report a comparator arm for safety contrasts. The figures therefore cannot establish a difference in safety between groups. They should be read as registry counts, not as an estimate that rules out risk or as evidence about a treatment effect. The record’s adverse-event entries do not change the absence of a reported primary correlation analysis.

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

The registry record describes an observational evaluation of whether EEG-derived brain activity features are associated with cognition measured by the MMSE. It gives an enrollment of 80, names the Neurosteer Aurora system, and states that EEG features were assessed twice within 7 days. It also identifies a secondary question about inter-patient variability between consecutive measurement sessions. These details establish the intended measurement context, but not the result of the analysis.

No primary correlation estimate, confidence interval, p value, or secondary-outcome result is reported in the registry record. As a result, the posted information does not establish whether the EEG features were associated with MMSE scores, how strong any association was, or how precisely it was estimated. The record does not report the number of participants included in the analyses, so the enrollment figure should not be treated as the analyzed sample size.

Even if an association estimate were available, association would not by itself demonstrate that EEG features cause a change in cognition or that the device improves cognitive function. This is an observational design without a reported randomized comparator. The record also does not specify a paired change analysis despite the primary outcome title’s reference to changes in cognition. The operational description instead identifies a previous MMSE score and repeated EEG assessments.

The registry information can therefore support a description of the question and measurement plan, and a cautious account of the adverse-event counts entered for the study. It cannot support a conclusion about the magnitude, direction, precision, or practical significance of the primary relationship. It also cannot establish the repeatability of the EEG measurements, because no secondary-outcome result is reported.

13. Why This Trial Matters Statistically

This study raises a useful statistical question: whether features extracted from EEG recordings during an auditory cognitive task vary in association with a clinical cognitive assessment. The outcome description makes the measurement link explicit by naming mean EEG feature activity and MMSE score. It also identifies repeat-session variability as a secondary question. Together, these aims highlight that an association result and the consistency of repeated measurements are distinct statistical issues.

The design is observational, so interpretation depends on the clarity of measurement timing and analysis definitions. The MMSE score comes from a previous institute evaluation, while EEG features were assessed twice within 7 days. The outcome title refers to changes in cognition, but the description does not explain how change would be calculated. Distinguishing a comparison of levels from a comparison of changes is important because they represent different statistical questions.

The registry record also illustrates why a planned outcome is not a result. It names the primary measure and study population but lists zero analyses and gives no correlation estimate or uncertainty interval. Without those results, readers cannot assess the strength or precision of the relationship. Enrollment alone cannot answer that question, and a planning illustration does not substitute for observed analysis.

Statistically, the record points to issues that require transparent reporting: how repeated EEG sessions are combined, which EEG features are analyzed, who is included in each analysis, and how incomplete measurements are handled. These details are not reported in the registry record. Their absence limits reproducibility and makes it difficult to determine exactly what quantity the primary analysis was intended to estimate.

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.

The principal limitation of the registry information is that it reports no primary or secondary analysis results. Although the outcome measures are named, zero analyses are listed. No correlation estimate, confidence interval, p value, or repeat-session variability result is reported in the registry record. The direction, magnitude, and precision of any association therefore remain unknown.

The primary endpoint is not fully aligned in wording and operational detail. Its title refers to cognition level changes, while its description identifies an MMSE score from a previous evaluation and EEG features assessed twice within 7 days. The registry record does not explain how MMSE change is determined, how session-level EEG data are combined for the primary analysis, or how the repeated measurements relate to the stated correlation.

The record does not report the inclusion and exclusion criteria, participant characteristics beyond the stated condition and senior population, or the number of participants with complete measurements. It also does not describe the analysis population, missing-data method, treatment of unusual values, or selection of EEG features. These omissions prevent assessment of selection into analysis and limit reproducibility.

There is no reported randomized comparator, and the allocation is not stated. The study therefore cannot support a randomized treatment contrast. The registry also does not report a protocol-based sample-size justification or analysis plan. The planning illustration’s assumptions should not be interpreted as the study’s own design or as observed power. Finally, the adverse-event counts are registry entries without further information about collection or follow-up, which limits their interpretation.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Evaluation of Cognitive State in Senior Subjects Using Neurosteer EEG System — NCT04683835.
• Registry record last updated 2023-11-13; status last verified 2023-11.
• Results first posted 2023-11-13.
• 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

NCT04683835 is a completed observational study sponsored by Neurosteer Ltd. It lists Mild Cognitive Impairment as the condition and enrollment of 80. The study start date was 2020-10-15, and the primary completion date was 2021-08-01.

The study assessed EEG-derived brain activity features using the Neurosteer Aurora system during an auditory cognitive assessment and related them to MMSE scores. The MMSE score was taken from a previous evaluation, while EEG features were assessed twice within 7 days. A secondary outcome concerned inter-patient variability between consecutive sessions.

The registry record lists two outcome measures and zero analyses. It reports no primary correlation estimate or uncertainty measure. It records zero serious adverse events, zero other adverse events, and zero deaths among 80 participants at risk. The record does not report the analysis population, missing-data approach, or a detailed statistical analysis plan.