Circuitry Assessment and Reinforcement Training Effects on Recovery: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Circuitry Assessment and Reinforcement Training Effects on Recovery (NCT04290988), a randomized na study. The primary outcome is Change in Number of Items Correctly Named on the Philadelphia Naming Test.

2. Clinical Question

The study is framed around whether EEG neurofeedback is more beneficial than sham feedback for communication, anxiety, and sleep quality in individuals with aphasia. Its designated primary outcome focuses more narrowly on picture naming: the number of items correctly named on the Philadelphia Naming Test. Other listed outcomes include a word association test, sleep quality, anxiety, sleep medication use, and EEG measures.

This is a question about a comparison within a randomized crossover design, rather than a simple comparison between separate groups receiving only one intervention. The summary says half of participants receive active feedback first and then sham feedback, while the other half receive sham first and then active feedback. The registry record does not report the observed changes, a treatment contrast, or an uncertainty estimate. It therefore does not establish what the comparison showed.

3. Trial Design

The registry describes Circuitry Assessment and Reinforcement Training Effects on Recovery, or CARTER, as a randomized, quadruple-masked crossover study. Johns Hopkins University is listed as the sponsor. The conditions are aphasia, primary progressive aphasia, and stroke. The enrollment field lists 7 participants. The study is marked terminated; the reason for termination is not reported in the registry record.

The two interventions are EEG Neurofeedback and Sham Feedback, both classified as devices. Each intervention is described as 15 sessions at a frequency of 3-5 sessions per week over 3-5 weeks. The brief summary describes two sequences: active EEG neurofeedback followed by sham feedback, or sham feedback followed by active neurofeedback. The registry record does not report the exact sequence allocation, whether every participant completed both periods, or whether there was a washout period.

Randomization can help balance factors that affect outcomes, while masking is intended to limit expectations or other knowledge from influencing trial conduct or assessment. The registry labels masking as quadruple but does not identify which roles were masked or describe how masking was maintained. It also does not report details of allocation concealment or the randomization procedure.

The primary outcome is change in the number of items correctly named on the Philadelphia Naming Test, assessed at baseline and 1 week following each intervention period. The record gives a mean and standard deviation as the outcome measure and lists a score range of 0 - 175, with higher scores reflecting more accurate naming. The record also lists 11 outcome measures, including secondary measures of word association, sleep, anxiety, medication use, and EEG. It does not report a statistical analysis for these measures.

The study start is listed as 2020-09-23 and primary completion as 2025-09-01. These dates and the terminated status describe the registry record; they do not, by themselves, show how many participants completed treatment or when study activities ended.

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: 7 participants were enrolled. 1 participant withdrew before randomization.

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 Change in Number of Items Correctly Named on the Philadelphia Naming Test, measured over Baseline, 1 week following each intervention period.

The primary endpoint is change in the number of items correctly named on the Philadelphia Naming Test. The test is described as a behavioral picture-naming assessment, with a score range of 0 - 175. Higher scores reflect more accurate naming and better naming ability according to the registry description. The stated time points are baseline and 1 week following each intervention period.

Picture naming is a direct way to measure performance on a naming task. In this study, it provides a focused behavioral measure relevant to communication in people with aphasia. It is not a comprehensive measure of communication, and it does not directly measure the other areas named in the study summary, such as anxiety or sleep quality. Those topics are represented by separate listed outcomes.

Change from baseline is intended to describe how the score differs after an intervention period relative to the starting assessment. In a crossover study, the comparison of interest may also depend on how outcomes change across the active and sham periods and on the order in which participants receive them. The registry record does not report the exact endpoint contrast or how period and sequence were handled in analysis.

The outcome is specified as a mean with standard deviation, which describes an average score change and the spread of individual changes. However, no numerical primary outcome results, treatment comparison, or uncertainty estimate are posted in the registry record. The endpoint description explains what was intended to be assessed; it is not itself evidence of an observed effect.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary outcome as a continuous measure summarized by a mean and standard deviation. The outcome is change in the number of items correctly named on the Philadelphia Naming Test, measured at baseline and 1 week following each intervention period. The registry does not give an analysis model, a treatment-effect estimate, or a test result. The explanation below describes general statistical ideas relevant to this design; it does not describe an analysis confirmed to have been used in this study.

What a mean and standard deviation describe

General: a mean summarizes the average value in a group or set of measurements. A standard deviation describes how spread out individual values are around that average. Both can help characterize outcome data, but neither alone gives the comparison between interventions. For a treatment comparison, an analysis would need to define which changes are compared and account for the study design. No numerical mean or standard deviation for the primary endpoint is reported in the registry record.

Why the crossover design matters

General: in a crossover study, participants receive more than one intervention in sequence, so outcomes after the different interventions can be compared within the same participant. This can account for stable differences between people, but measurements within a person are related rather than independent. The analysis therefore needs to reflect the repeated observations. In addition, outcome values can vary over time, and sequence or period may matter. The registry describes randomized sequences of active then sham feedback or sham then active feedback, but the statistical handling of within-person correlation, period, sequence, or carryover is not reported in the registry record.

A carryover effect is a general concern when an intervention might influence measurements in a later period. Whether that is relevant here, and whether any steps were planned to address it, is not reported in the registry record. The record also does not state whether a washout period was used. These omissions prevent readers from evaluating how the crossover analysis was intended to separate intervention effects from other changes over time.

What an effect estimate and confidence interval would add

General: an effect estimate expresses the size and direction of a comparison using a defined scale. A confidence interval adds information about uncertainty around that estimate. An interval that is narrow indicates greater statistical precision than a wide interval, subject to the analysis assumptions. The registry record has no posted estimate or interval for naming change, so neither the direction nor precision of a treatment contrast can be evaluated.

A p-value, when reported, describes how compatible the observed data are with a specified null hypothesis under the model. It is not a measure of clinical importance, and a p-value alone does not show the size or precision of an effect. No primary-outcome p-value is reported. The sample-size assumptions listed for planning illustrations refer to a two-sided significance level of 0.05 and a normal approximation, but no planning result is needed to interpret the absent primary analysis. Such illustrations are planning calculations, not observed power and not a substitute for observed results.

Multiplicity and multiple measurements

General: multiplicity arises when several outcomes, time points, or comparisons are examined. Looking across many results can increase the chance that some appear notable by chance, so statistical plans may specify a primary comparison or adjustment approach. This record lists 11 outcome measures and identifies multiple time points for the primary outcome. It also indicates that the primary outcome has several categories or time points and was not reduced to one comparison. The multiplicity strategy, if any, is not reported in the registry record.

Why missing information limits interpretation

General: results depend on which participants have usable measurements and how missing observations are handled. In a crossover study, completion of both periods may be important to a within-participant comparison, but the registry record does not report completion by period, missingness, or the analysis population. It does not specify whether all randomized participants, participants who received treatment, or only participants with complete measurements were included in an analysis.

The record also does not report a sample-size justification, planned interim analyses, or a detailed statistical plan. These items should not be inferred from the enrollment field or from the study status. In particular, the enrollment of 7 is not a reported sample-size calculation and does not establish the precision of any result. Because no primary endpoint analysis is posted, readers cannot assess the model assumptions or determine whether the planned analysis was followed.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists 11 outcome measures and specifies baseline measurement and measurement 1 week following each intervention period for the primary naming outcome. It states that the primary outcome has several categories or time points and was not reduced to one comparison. This leaves the reader without a single posted primary contrast to interpret. The record does not report how the study planned to address multiplicity across outcomes, time points, or comparisons.

General: testing or estimating effects across multiple outcomes and time points can complicate interpretation. A statistical plan may identify a primary comparison and describe whether other analyses are confirmatory or exploratory, or whether adjustment is used. No such approach is reported in the registry record. The listed secondary outcomes cover word association, sleep quality, anxiety, medication dosage and frequency, and EEG measures, but no analyses for these outcomes are posted.

The record does not report missing-data counts, reasons for missingness, completion by intervention period, or the analysis population. In a crossover design, it matters whether participants contributed measurements after both periods and how incomplete sequences were handled. Whether any imputation or other missing-data method was planned or used is not reported in the registry record.

The registry lists 0 analyses. Therefore, there is no posted account of exclusions, adherence, protocol deviations, sensitivity analyses, or which participants contributed to an outcome analysis. A sample-size justification, interim analysis plan, and detailed statistical plan are also not reported in the registry record. These gaps limit assessment of both the intended analysis and the extent to which the posted outcome data represent the enrolled participants.

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 adverse-event section lists 6 participants at risk for each intervention. For active EEG neurofeedback, it reports 0 participants with serious adverse events, 0 with other adverse events, and 0 deaths. For sham feedback, it reports the same figures: 0 serious adverse events, 0 other adverse events, and 0 deaths. The registry lists no top serious-event terms.

These are the figures recorded in the adverse-event section, not a complete description of all safety monitoring or participant experience. The enrollment field lists 7 participants, while the adverse-event groups list 6 at risk in each group; the registry does not explain this difference. The group figures also do not establish how many participants completed each intervention or contributed to outcome analyses.

General: an observed count of zero does not demonstrate that an event is impossible or that risk is absent. The amount of information available depends on the number of people observed, the duration of observation, and how events were collected and reported. The registry record does not provide further details on event definitions, observation periods, collection procedures, or reasons for the listed denominators. These counts should therefore be read as the posted adverse-event figures, not as evidence that the interventions are risk-free.

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

The registry record supports a description of the study design and intended assessment: a randomized, quadruple-masked crossover comparison of active EEG neurofeedback and sham feedback, with change in picture naming designated as the primary outcome. It also records the intervention schedules, the listed outcome domains, and the study's terminated status. These design details explain what the study set out to examine.

The record does not post a numerical analysis of naming change. It therefore does not show whether scores changed after either intervention, whether the changes differed, or how precise any comparison was. No effect size, confidence interval, or p-value for the primary outcome is reported. The absence of these results should not be interpreted as a finding of no difference, nor as evidence that one intervention is better.

The record's enrollment field lists 7, but the number of participants with complete data for the primary outcome is not reported in the registry record. The status is terminated, but no reason is given. These details limit what can be concluded about the progress and analysis of the study; they do not, on their own, explain why the record contains no outcome analysis.

The listed safety figures show 0 participants with serious adverse events, 0 with other adverse events, and 0 deaths in each intervention group, with 6 at risk listed in each group. Those are descriptive registry entries, not proof of absence of risk. Further information about event monitoring and study follow-up is not reported in the registry record.

In short, this record describes a clinically relevant research question and the planned crossover comparison, but it does not provide results with which to judge the primary endpoint. Any conclusion about the relative effects of active and sham feedback would go beyond the posted data.

13. Why This Trial Matters Statistically

This study is statistically notable because its crossover design makes the comparison more than a simple contrast between two separate groups. Participants are described as receiving both active EEG neurofeedback and sham feedback in different orders. General: a within-participant comparison can account for stable differences between people, but it also requires analysis that recognizes repeated measurements and considers period and sequence. The registry record does not report how those features were handled.

The primary outcome is a behavioral naming score with a defined range and direction, measured at baseline and after each intervention period. That gives the research question a concrete outcome, while the listed secondary measures extend into sleep, anxiety, medication use, and EEG. The breadth of outcomes also makes transparent reporting of the primary comparison and the approach to multiplicity important. The record lists 11 outcome measures but does not report a multiplicity strategy or analyses.

The lack of posted endpoint results is statistically important in its own right. Without an estimate and a measure of uncertainty, readers cannot distinguish a small and precise comparison from a large but imprecise one, or determine whether the observed data support a meaningful direction. The enrollment field alone cannot answer these questions, and the terminated status does not explain why results are absent.

For clinical researchers and biostatisticians, the record illustrates why registration details and analyzed results serve different purposes. The design description identifies the intended question; the results are needed to assess the estimated effect and its precision. Here, the registry documents the former but does not provide the latter. That distinction prevents the design or the safety counts from being mistaken for evidence about efficacy.

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 for interpretation is that the registry record contains no posted primary-outcome analysis. Although the endpoint, score range, direction, and assessment time points are described, no observed change, intervention contrast, confidence interval, or p-value is available. The record lists 0 analyses and notes that the primary outcome has several categories or time points and was not reduced to one comparison.

The enrollment field lists 7 participants, while the adverse-event section lists 6 at risk in each intervention group. The record does not explain the denominator difference or report how many participants completed each intervention period. It also does not report the number included in any primary-outcome analysis, participant characteristics, missing observations, or reasons for missingness.

Important details for understanding a crossover analysis are absent. The registry does not report the exact analysis model, how within-participant dependence was handled, or how period, sequence, and possible carryover were considered. The summary describes both intervention orders but does not state whether participants completed both periods or whether there was a washout period. The masking label is quadruple, but the roles masked and the procedures used to maintain masking are not described.

The multiplicity approach, sample-size justification, interim analysis plan, and detailed statistical plan are not reported in the registry record. The study is marked terminated, but the reason for that status is also not reported. These gaps limit interpretation of the conduct and analysis; they do not support an assumption about why the trial stopped or what its results would have been.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Circuitry Assessment and Reinforcement Training Effects on Recovery — NCT04290988.
• Registry record last updated 2026-02-05; status last verified 2026-01.
• Results first posted 2026-02-05.
• 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

CARTER is listed as a randomized, quadruple-masked crossover study sponsored by Johns Hopkins University. It compares active EEG neurofeedback with sham feedback in people with aphasia-related conditions. Each intervention is described as 15 sessions delivered at a frequency of 3-5 sessions per week over 3-5 weeks, with the two intervention orders described in the study summary.

The primary outcome is change in the number of items correctly named on the Philadelphia Naming Test, assessed at baseline and 1 week following each intervention period. The registry lists 7 participants enrolled, 11 outcome measures, and 0 analyses. The study status is terminated; the reason is not reported in the registry record.

No numerical primary-outcome comparison or uncertainty estimate is posted. The adverse-event section lists 0 serious adverse events, 0 other adverse events, and 0 deaths for each intervention group, with 6 at risk in each. These entries describe the posted record but do not replace an analysis of naming outcomes or establish absence of risk.