Improving Aphasia Using Electrical Brain Stimulation: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Improving Aphasia Using Electrical Brain Stimulation (NCT04963803), a randomized na study. The primary outcome is Language Specific Attention Treatment Probe.

2. Clinical Question

The trial hypothesis is that participants receiving active stimulation paired with speech and language therapy will improve more on a language comprehension task than participants receiving the same therapy paired with sham stimulation. The primary measure is the Language Specific Attention Treatment Probe. Participants listen to sentences and answer questions about what they heard; results are expressed as the percentage of correct responses. Higher scores indicate better performance and less auditory comprehension impairment.

This is a question about a between-group difference in change or performance, not simply whether scores improve within either group. An improvement within one group alone would not establish that active stimulation adds benefit relative to sham. A direct comparison, with its estimated size and uncertainty, is needed to address the stated question. The registry record does not report a single primary-outcome comparison, so the direction, magnitude, and precision of any group difference cannot be described from the record.

3. Trial Design

The registry record describes a randomized, double-masked, parallel-group trial. Participants were assigned to active tDCS paired with speech-language therapy or sham tDCS paired with speech-language therapy. The planned enrollment was 23. The trial is marked completed, and its phase is listed as NA. Syracuse University is named as the sponsor.

The record gives a start date of 2021-07-12 and a primary completion date of 2024-11-15. These are registry dates, not a description of when individual participants received treatment or completed follow-up. The primary outcome timeframe identifies a baseline measure before treatment, a post-treatment measure 5-6 weeks after baseline, and a follow-up measure 10-11 weeks after baseline.

Active stimulation is described as 2 milliamperes delivered for 20 minutes per session using a Soterix mini-CT device, with electrode locations specified in the arm description. The sham description states that 2 milliamperes are delivered for 1 minute to simulate tDCS, after which the current is ramped down to zero. The full descriptions of the treatment procedures are not available in the registry record, so details beyond those statements should not be inferred.

Randomization is intended to support a fair comparison by balancing participant characteristics across groups on average, while masking is intended to limit influences of treatment knowledge on care, participant responses, or assessment. The record identifies the design features but does not report how randomization was generated, who was masked, or whether masking was assessed. Further details of treatment scheduling, adherence, and allocation procedures are not reported in the registry record.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

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 Language Specific Attention Treatment Probe, measured over Pre-treatment: Baseline measure collected before treatment; Post-treatment: 5-6 weeks after baseline; Follow-up: 10-11 weeks after baseline.

The primary endpoint is the Language Specific Attention Treatment Probe, a continuous measure reported as a percentage. It assesses auditory comprehension for untrained sentence-level stimuli. Participants listen to a sentence and answer comprehension questions about it; the score is the percentage of correct responses among the items administered. Higher percentages represent better performance and less auditory comprehension impairment.

The endpoint addresses comprehension of sentences that were not used for training, making it a measure of performance on untrained stimuli rather than a direct tally of treatment exercises. It is relevant to the trial's stated question because the investigators hypothesize that pairing therapy with active stimulation will lead to greater improvement on a language comprehension task than pairing therapy with sham stimulation.

The registry specifies measurements at baseline before treatment, post-treatment 5-6 weeks after baseline, and follow-up 10-11 weeks after baseline. These time points can describe performance at different stages, but a group comparison requires a defined analysis that accounts for the repeated measurements and specifies which time point or contrast is primary. The record does not report a single primary-outcome comparison, estimate, confidence interval, or analysis method. It therefore does not establish from the posted information how the endpoint changed between groups or how precisely a difference was estimated.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a randomized, double-masked, parallel-group design. In general, random allocation is used to reduce systematic differences between treatment groups at the start of a trial. It does not guarantee that every characteristic will be balanced, especially in a small study. Masking is intended to reduce the influence of knowing treatment assignment on participant behavior, treatment delivery, or outcome assessment. The registry record does not report the randomization procedure, who was masked, or whether masking was successful.

What does a continuous outcome mean?

The primary outcome is a continuous measure reported as a percentage of correct responses. Continuous outcomes retain information across a range of scores rather than classifying each participant into only a small set of categories. The registry identifies the mean as the summary parameter and the standard deviation as the measure of dispersion. In general, the mean describes the group's average score, and the standard deviation describes how variable scores are around that average. These summaries do not replace a direct comparison between groups.

For a randomized trial, the central statistical question is usually whether the outcome differs between assigned groups, according to a prespecified contrast. The contrast might concern scores at a particular assessment or changes across assessments. The choice matters: a comparison of follow-up scores is not necessarily the same as a comparison of change from baseline. The registry lists multiple primary-outcome time points but does not report one analysis that resolves which comparison is primary.

Why does uncertainty matter?

A point estimate describes the observed size and direction of a difference. A confidence interval is a statistical range that represents uncertainty about the estimate under the assumptions of the analysis. A narrow interval indicates greater precision than a wide interval. The interval also helps readers judge which effect sizes remain compatible with the data. A p-value, when reported, addresses compatibility with a specified null hypothesis under the analysis assumptions; it is not a measure of the probability that a treatment works.

Neither a confidence interval nor a p-value can be interpreted without the estimate, its scale, and the analysis context. The registry record summarized does not report an estimate, interval, or primary-outcome p-value. Therefore, it is not possible to assess the observed effect's magnitude or precision using the registry information.

How should repeated measurements be understood?

The primary measure is scheduled at baseline, post-treatment, and follow-up. Repeated assessments can provide information about initial performance, performance after treatment, and later performance. Because the same participants may contribute observations at more than one time, an analysis may need to account for the relationship among measurements from the same person. This is a general statistical consideration, not a claim about the method used in this trial. The statistical model and the handling of repeated observations are not reported in the registry record.

Missing observations also matter. If scores are unavailable at later assessments, the number and characteristics of participants contributing to each comparison can change. Different methods for handling missing outcome data rely on different assumptions. The registry record does not report outcome-specific missingness, reasons for missing observations, or a method for handling them.

What do the planning assumptions say?

The record describes sample-size planning illustrations based on the observed effect. It states a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; the effect to detect is described as a stated fraction of the observed effect. These are planning assumptions, not observed trial results. The registry record does not report the resulting planning counts in the information summarized, and those illustrations must not be read as observed power or as evidence that the trial detected an effect.

In general, planning calculations estimate how much information a future study might require under chosen assumptions. Their outputs depend on the assumed effect and variability, the statistical test, and the desired power. A calculation based on an observed effect can be particularly sensitive to uncertainty in that effect. A planning illustration cannot replace the trial's actual estimate and confidence interval. The record does not report the protocol's original sample-size justification or establish that the described planning illustration was used to design this trial.

What methods are not documented?

The registry record does not report the primary statistical model, adjustment for baseline score, analysis population, missing-data procedure, multiplicity strategy, or any interim analysis. It also does not provide analysis results for the primary endpoint. These omissions prevent a detailed appraisal of how the primary comparison was conducted and how robust it might be to alternative analytical choices. General statistical explanations describe considerations that matter for this kind of outcome; they should not be mistaken for methods confirmed to have been used in this trial.

10. Multiplicity, Missing Data and Analysis Populations

The primary outcome is listed at baseline, post-treatment 5-6 weeks after baseline, and follow-up 10-11 weeks after baseline. Considering multiple time points can create more than one possible comparison. In general, testing multiple comparisons without a prespecified strategy can increase the chance of finding an apparently notable result by chance. A multiplicity plan can define which comparison is primary or adjust interpretation across tests. The registry record does not report such a strategy, and it indicates that the primary outcome has several categories or time points and was not reduced to one comparison.

The record lists secondary outcomes including the Continuous Performance Test, Attention Network Test, Scenario Test, and Revised Token Test. It does not report how those outcomes were analyzed, whether they were treated as confirmatory or exploratory, or how multiplicity across outcomes was addressed. The record reports zero analyses, so no analysis-level results are available in the information summarized.

Missing-data handling is also not reported in the registry record. There is no outcome-specific account of missing measurements, reasons for missingness, or the method used to address incomplete data. In general, conclusions from repeated-measures analyses can depend on assumptions about why observations are missing and on how available observations are used.

The analysis populations are not reported. The record does not state whether results were analyzed according to randomized assignment, restricted to participants who completed treatment, or based on another population. Those choices can affect interpretation, so the absence of this information limits assessment of the planned and completed analyses. It is not possible to infer the analysis population from the completed status or enrollment figure.

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 5% reporting threshold.

The registry lists zero serious adverse events in the active-stimulation group and zero in the sham-stimulation group. It also lists zero other adverse events in each group and zero deaths in each group. The at-risk count is 11 for each listed group. These are the recorded counts in the registry, not estimates of the probability of harm or results of a comparative safety analysis.

A count of zero means that no event in the listed category was recorded for the stated group in this record. It does not establish that the risk is absent, nor does it show that the groups have identical safety profiles. Rare or delayed events may be difficult to characterize from a limited trial, and a zero count does not provide a precise upper bound on risk without further analysis. This is a general statistical caution, not an additional finding about this trial.

The registry record does not report definitions for serious and other adverse events, event descriptions, severity, timing, relatedness to treatment, or the method of collecting events. It lists no top serious-event terms. Consequently, these figures support only a narrow description of the recorded counts; they do not support a broader conclusion about safety or a formal comparison between interventions.

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

The trial is designed to compare speech-language therapy paired with active tDCS against the same therapy paired with sham tDCS among people with aphasia. Its primary measure assesses auditory comprehension for untrained sentence-level stimuli, with higher percentages indicating better performance and less auditory comprehension impairment. The record also specifies a schedule extending from baseline to post-treatment and follow-up.

What the posted information does not establish is whether the active group had better, worse, or similar scores relative to the sham group. The registry record does not report a single primary-outcome comparison or an effect estimate with an uncertainty interval. Without those results, neither the size of a difference nor its precision can be described. The completed status and the stated hypothesis do not substitute for comparative outcome data.

The absence of a reported primary comparison should not be read as evidence that there was no difference. It also should not be read as evidence that there was a difference. Both interpretations require results that are not reported in the registry record. A within-group change, if one were reported, would likewise not by itself answer whether active stimulation adds to therapy relative to sham; that question depends on the between-group comparison.

The safety counts are limited to the adverse-event categories and group figures recorded. Zero recorded events does not establish zero risk. The record does not include the detail necessary to assess event definitions, collection, timing, or comparative uncertainty.

More broadly, a trial result applies to the population, treatment procedures, and outcome it studied. The registry information does not supply enough detail to judge how closely those conditions match other clinical settings, nor does it report broader outcomes beyond the listed measures. The most defensible reading is therefore descriptive: the record identifies the design and intended question, but does not report the primary comparative result needed to answer that question statistically.

13. Why This Trial Matters Statistically

This trial poses a clear statistical comparison: whether adding active electrical brain stimulation to speech-language therapy produces a different result on an auditory-comprehension measure than pairing the therapy with sham stimulation. The sham condition is important to the comparison because the stated procedure is intended to simulate the experience of stimulation, while the randomized, double-masked design aims to limit biases associated with treatment assignment.

The primary measure focuses on comprehension of untrained sentence-level material rather than only on trained items. That makes the choice of endpoint relevant to the question of performance beyond the treatment exercises. The repeated schedule also raises an important design and analysis issue: the trial needs a clearly defined primary comparison across baseline, post-treatment, and follow-up measurements so that the result is interpretable and not selected from multiple possible time points.

Statistically, an estimate and its uncertainty are needed to distinguish the observed difference from the range of effects compatible with the data. Reporting only whether a result is statistically significant would be incomplete; the size and precision of the comparison matter. The registry record does not report a single primary-outcome comparison, so the key evidence needed to evaluate that question is unavailable in the record summarized.

The safety counts illustrate another general point: descriptive event counts need context, definitions, and uncertainty before they can support broad conclusions. The recorded zeros are worth reporting accurately, but do not establish an absence of risk. Taken together, the trial's design and outcome provide a framework for a useful comparison, while the lack of reported comparative results limits what can be learned statistically from the registry record.

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 central limitation for interpreting this record is that it does not report a single primary-outcome comparison. There is no reported group-level endpoint estimate, between-group effect, confidence interval, or primary-outcome p-value in the information summarized. As a result, the primary research question cannot be answered from the registry record.

The primary outcome includes multiple assessment times: baseline, post-treatment, and follow-up. The record states that the outcome has several categories or time points and was not reduced to one comparison. Without a clearly reported primary contrast, readers cannot determine which time point or pattern of change was intended to carry the main inferential weight. The statistical model for repeated measurements is also not reported.

The listed enrollment is 23, but enrollment does not identify the number of observations available at each assessment or the precision of an outcome estimate. Outcome-specific denominators, missingness, reasons for missing observations, and analysis populations are not reported. The record also does not report the method used to handle missing data, the statistical model, or a multiplicity strategy for the primary time points and secondary outcomes.

The treatment descriptions in the registry record are incomplete in the information summarized. Details beyond the stated active and sham stimulation procedures should not be inferred. The record does not report how randomization was implemented, who was masked, or whether masking was maintained. Such details would help readers assess potential sources of bias.

Finally, the adverse-event record gives counts but not event definitions, descriptions, timing, severity, relatedness, or collection procedures. The zero counts therefore have limited interpretive scope. These limitations concern what the registry record reports; they do not establish that unreported procedures or results do not exist elsewhere.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Improving Aphasia Using Electrical Brain Stimulation — NCT04963803.
• Registry record last updated 2026-01-30; status last verified 2026-01.
• Results first posted 2026-01-30.
• 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

NCT04963803 is a completed, randomized, double-masked, parallel-group trial sponsored by Syracuse University. The registry lists planned enrollment of 23 and conditions of stroke and aphasia. The primary outcome is the Language Specific Attention Treatment Probe, a percentage-correct measure of auditory comprehension for untrained sentence-level stimuli, assessed at baseline, post-treatment, and follow-up.

The comparison is speech-language therapy paired with active tDCS versus speech-language therapy paired with sham tDCS. The registry record does not report a single primary-outcome comparison or its estimate and uncertainty, so the comparative result cannot be interpreted from the record summarized. Safety listings show zero serious adverse events, zero other adverse events, and zero deaths in each group, with 11 at risk in each group. These counts are descriptive and do not establish absence of risk.