TMS in Aphasia Recovery: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for TMS in Aphasia Recovery (NCT04777214), a randomized na study. The primary outcome is Boston Naming Test.

2. Clinical Question

The clinical question concerns whether language performance, measured by confrontation naming, changes after active stimulation and how those observed changes compare with sham stimulation. The registry record describes the trial as testing the hypothesis that stimulation to the side of the brain opposite the stroke may improve language function. That description is the study rationale, not a result.

The distinction between an outcome within one arm and a contrast between randomized arms is central. The posted primary-outcome values are means for the active arm at two follow-up times. Without corresponding sham-arm values or a reported between-arm estimate, they cannot establish whether the observed changes differ from what occurred with sham treatment. The registry record does not report a comparative effect estimate or its uncertainty.

3. Trial Design

The registry lists the study as randomized, double-masked, and parallel, with an enrollment of 24 and a completed status. The official title describes it as an incomplete crossover trial, while the registry’s design model is listed as parallel. These descriptions do not align, and the registry record does not explain the discrepancy. The record lists the sponsor as the University of Pennsylvania.

The intervention descriptions specify active TMS and sham TMS. Active treatment consists of 10 sessions over 2 consecutive weeks, with 20 minutes and 1200 pulses of 1 Hz stimulation directed to a previously determined optimal response site in the right frontal lobe. In the sham condition, the same stated session schedule, duration, pulse count, and frequency are described, but the coil is rotated 90 degrees during stimulation. The brief summary describes daily sessions to the right, unaffected side of the brain and language testing before and after treatment, as well as functional MRI scans before and after treatment.

The listed primary outcome is the Boston Naming Test, expressed as percent change from baseline. Its timeframe is baseline, 2 months and 6-months after the last rTMS treatment session. The primary-outcome population description specifies 9 participants who completed the Active TMS arm and were available for follow-up at both stated follow-up times. The posted outcome values are for the active arm only.

Further details about randomization procedures, allocation concealment, how masking was maintained or assessed, and the rationale for the design are not reported in the registry record. The record also does not report a trial-specific sample-size justification or a statistical analysis plan. The enrollment figure alone does not establish the planned power or precision for the primary comparison.

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: Patients were assessed against the study inclusion and exclusion criteria. Eligible patients then underwent language tests before and after 6 ten minute sessions of rTMS to determine if there was a right hemisphere cortical region which, when suppressed with 1 Hz rTMS, resulted in significantly improved naming. If no benefit was observed from stimulation, the patient did not continue on to the treatment phase.

Initial Randomization

Sham Crossover to Active TMS Arm

Analysis population for the primary outcome, as defined in the registry: 9 participants completed the Active TMS arm and were available for follow-up at 2- and 6- months post treatment.

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 Boston Naming Test, measured over Baseline, 2 months and 6-months after the last rTMS treatment session.

The primary endpoint is the Boston Naming Test, reported as percent change from baseline. The test assesses confrontation naming: a participant is shown an item and asked to name it. The registry describes a score range from 0, for no items named correctly, to 36, for all items named correctly. Naming is one aspect of language performance, and a standardized task gives investigators a defined outcome to assess over time.

The registry lists the measurement timeframe as baseline, 2 months and 6-months after the last rTMS treatment session. The posted active-arm means are 10.3 with a standard deviation of 20.0 at the 2-month follow-up, and 37.7 with a standard deviation of 31.2 at the 6-month follow-up. Each value is expressed as percentage change from baseline, not as a raw Boston Naming Test score. The number of participants is 9 at each time point.

The registry record does not describe the formula used to calculate percentage change, nor does it give baseline scores or individual participant results. Consequently, the reported percentages should be read as the registry’s stated endpoint values, not converted into raw test-score points. The record also does not report a sham-arm value for this endpoint, so these summaries alone do not quantify a treatment contrast.

In general, repeated outcome assessment can describe how a measure varies over follow-up, but an observed change within one group is not the same as a randomized comparison. The clinical interpretation of a difference on this test would also depend on details such as measurement variability and the meaning of a change for participants; those details are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean describe? The primary endpoint is continuous: it is a numeric measure reported as percentage change from baseline. A mean represents the arithmetic average of the values in the group being summarized. Here, the registry reports an active-arm mean of 10.3 at 2 months and 37.7 at 6-months, with 9 participants at each time. These are group summaries, not values that should be attributed to every participant.

What does the standard deviation describe? The standard deviation describes the spread of observed values around the mean. The reported standard deviations are 20.0 at the 2-month follow-up and 31.2 at the 6-month follow-up. In general, a larger standard deviation indicates greater dispersion among the observations on the scale being summarized. It is not a measure of the uncertainty of the mean, and it is not a range containing all participants’ values.

How is uncertainty usually expressed? In general, a confidence interval gives a range of values compatible with the data and the assumptions of the analysis. Its width reflects statistical precision: wider intervals indicate less precise estimates. The registry record does not report confidence intervals for these means or a confidence interval for an active-versus-sham contrast. The standard deviations cannot substitute for those missing quantities. No uncertainty measure should be inferred from them without the relevant analysis details.

Why is the randomized contrast important? In a randomized trial, the main comparative estimate ordinarily relates outcomes in the assigned groups. The causal question is not answered by showing that one group’s average outcome differs from its baseline. General influences on outcomes over time can affect both groups, and a comparator helps distinguish the observed group-level difference from such influences. In this record, the primary outcome has active-arm summaries but no sham-arm summary or reported contrast. A treatment effect estimate, its direction, and its precision therefore cannot be described from the posted primary results.

What does a p-value test? In general, a p-value summarizes how incompatible the observed data are with a specified statistical null hypothesis under the assumptions of a test. It does not measure the size or importance of an effect, and it is not the probability that a hypothesis is true. No p-value or test result for the primary outcome is reported in the registry record. Thus, the posted means do not establish statistical significance or lack of statistical significance.

What can be said about the follow-up times? The two values refer to separate stated follow-up times after the last rTMS treatment session. Although both summaries have 9 participants, the registry record does not provide participant-level values or a repeated-measures analysis. In general, measurements on the same participants are related rather than independent; a formal analysis of change across time should account for that relationship. The record does not say whether the same individuals contributed to both displayed summaries beyond describing the 9 participants as available at both follow-ups, and it does not report a model for their repeated measurements.

How should percentage change be read? The registry labels the endpoint percent change from baseline, while describing the underlying Boston Naming Test as ranging from 0 to 36. Those are different scales: a percentage summary is not a raw score. The record does not state the calculation formula, baseline values, or how any special cases were handled. General cautions about interpreting percentages therefore apply, but no alternative calculation can be reconstructed from the posted summaries alone.

What is over-dispersion? In general, over-dispersion describes data that vary more than a statistical model assumes. It can make model-based uncertainty estimates too small if the extra variability is ignored. The posted primary outcome is summarized with means and standard deviations, and the registry record does not describe a model or a dispersion assumption. There is no basis to conclude that over-dispersion was assessed or that a particular model addressed it.

What can planning calculations show? In general, sample-size calculations depend on a target effect, variability assumptions, a significance level, allocation, and an analysis method. The registry’s planning note describes an illustration based on the observed effect, a two-sided significance level of 0.05, equal allocation, and a normal approximation; the target effect is defined as a fraction of the observed effect. It also distinguishes participants per arm for ordinary outcomes from total events for time-to-event outcomes. These are planning assumptions, not results from participants and not observed power. A trial-specific sample-size justification is not reported in the registry record. No calculated planning totals are stated in the record details available for this analysis, so no such totals can be interpreted.

The registry marks the continuous endpoint comparison as unavailable because values are missing or not estimable for a two-arm comparison. It lists 0 analyses. The distinction matters: descriptive summaries remain informative about the reported active-arm observations, but they do not replace an analysis that estimates a randomized between-arm difference.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Active TMS 2-month Follow-Up", "Active TMS 6-month Follow-Up")
n   <- c(9, 9)
m   <- c(10.3, 37.7)
sp  <- c(20, 31.2)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Active TMS 2-month Follow-Up  9  10.300  20.000  -5.073  25.673
CARM  Active TMS 6-month Follow-Up  9  37.700  31.200  13.718  61.682

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 4 outcome measures and 0 posted analyses. The primary outcome described in detail is the Boston Naming Test, with follow-up assessments at 2 months and 6-months after the last rTMS treatment session. The registry record does not describe the other outcome measures in the details, identify which analyses were planned, or report a strategy for handling multiple outcomes or time points.

In general, examining several outcomes or time points can increase the chance of finding an apparently notable result by chance if the analyses are treated as independent opportunities for discovery. A prespecified hierarchy or multiplicity adjustment can address that issue, but neither is reported in the registry record. No claim can be made about whether an adjustment was used.

The primary-outcome population is described as 9 participants who completed the Active TMS arm and were available for follow-up at both stated times. The record does not explain the handling of missing outcome values, give a participant flow for the primary analysis, or state whether all randomized participants were included in an analysis population. The enrollment figure of 24 does not resolve those questions. The record also does not report imputation, a missing-data sensitivity analysis, or reasons for unavailable follow-up.

In general, analyses restricted to participants with observed follow-up can be affected if availability is related to outcomes. That possibility is not evidence that bias occurred in this trial; the registry record does not provide the information needed to assess it. The active-arm summaries should therefore be interpreted as applying to the described follow-up population, not automatically to all enrolled participants.

12. 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 figures list 10 participants at risk in the Active TMS group and 5 at risk in the Sham TMS group. The posted counts are zero serious adverse events and zero other adverse events in each group, with zero deaths in each group. The record also gives no serious adverse-event terms in its listed top terms.

These are counts for the groups and denominators stated in the registry record. They describe what was posted; they do not establish that no event could occur, or that the groups have identical safety profiles. In general, a zero count in a small group is compatible with uncertainty about the frequency of an event, particularly when the number at risk is limited.

The registry record does not provide additional detail about event definitions, collection procedures, severity grading beyond the serious and other categories, timing, or statistical comparison between groups. It also does not explain why the adverse-event denominators differ from the enrollment figure. The counts should not be used to infer a comparative risk estimate or to make a broader safety conclusion.

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

The posted primary results show mean percentage change from baseline on the Boston Naming Test in the active arm at two follow-up times: 10.3 at 2 months and 37.7 at 6-months. Each summary is based on 9 participants, and the corresponding standard deviations are 20.0 and 31.2. The values are descriptive summaries of the reported active-arm observations.

They do not show how active TMS compares with sham TMS on this outcome. Sham-arm primary-outcome values, an active-versus-sham estimate, and its uncertainty are not reported in the registry record. There is also no posted primary-outcome p-value or confidence interval. The results therefore do not establish a statistically significant difference between the randomized groups, nor do they establish that no difference exists.

The results also do not provide a formal estimate of change between the two follow-up times. The registry reports separate means rather than a stated repeated-measures analysis. A difference between those means should not be treated as a calculated time effect, and neither mean should be interpreted as the outcome for every participant.

In general, a randomized comparison can help distinguish treatment-group differences from changes that occur over time or under the comparator condition. That comparison is unavailable in the posted primary results. Likewise, a measure of naming performance is informative about the specific task assessed, but the registry record does not report how these values relate to broader language function or to outcomes that matter to participants.

The appropriate conclusion is limited to the numerical summaries and their stated population. The registry record does not report enough comparative or uncertainty information to determine the size or precision of a treatment effect. This is a limitation of the posted evidence, not proof that the trial did or did not detect a difference in an analysis that is absent from the record.

14. Why This Trial Matters Statistically

This record illustrates why reporting a group mean is not the same as reporting a randomized treatment effect. The primary endpoint is summarized for the active arm at two follow-up times, but the corresponding sham-arm values and an arm-to-arm estimate are not reported. A reader can describe the active-arm observations, yet cannot use those summaries to quantify the contrast that randomization is intended to support.

It also demonstrates why variability and precision should be distinguished. The standard deviations describe spread among the observed percentage-change values. They do not show the precision of the group means or the uncertainty around a treatment contrast. Without confidence intervals or another reported uncertainty measure, the precision of the estimates cannot be evaluated directly from the registry record.

The record’s design labels make transparent reporting especially relevant: the official title calls the study an incomplete crossover trial, while the registry model lists a parallel design. The record does not reconcile those descriptions. Clear reporting of design, analysis populations, outcome handling, and the planned comparison would help readers understand which observations are compared and how they contribute to an estimate.

Finally, the primary-outcome population is described as 9 active-arm participants available at both follow-up times, while the total enrollment is 24. The registry record does not explain the participant flow or missing-data handling. That information is important for judging how well the reported summaries represent the randomized study population. These are statistical reporting issues; they do not by themselves determine the clinical value of the intervention.

15. 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 main limitation for interpreting the primary outcome is that only active-arm summaries are posted. The registry record does not report the sham-arm Boston Naming Test results, a between-arm estimate, or uncertainty around a comparative estimate. The randomized comparison therefore cannot be evaluated from the posted primary outcome.

The outcome summaries are based on 9 participants at each follow-up time, as described for the Active TMS arm. The record does not provide individual values, baseline scores, or the formula used to calculate percent change. It is not possible to reconstruct raw score changes, examine the shape of the observed values, or verify how the reported percentages were calculated from the group summaries.

There is no reported analysis plan, statistical model, p-value, confidence interval, or method for repeated measurements. The registry lists 4 outcome measures and 0 analyses, but does not describe how multiplicity was addressed. The handling of missing outcomes and the definition of the analysis population are also not reported beyond the description of participants available at follow-up.

The official title and the listed design model differ: the title says incomplete crossover, while the model is parallel. The registry record does not clarify the design or explain how the two descriptions relate. Further details about randomization and masking procedures are not reported in the registry record.

The safety counts are also limited by the small denominators stated for the adverse-event groups and by the lack of event detail. Finally, enrollment of 24 does not establish that the study was planned to detect a particular effect with a particular level of precision. A trial-specific sample-size justification is not reported in the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: TMS in Aphasia Recovery — NCT04777214.
• Registry record last updated 2021-11-26; status last verified 2021-04.
• Results first posted 2021-11-26.
• 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.

19. Record Summary

The registry record describes a completed randomized, double-masked study sponsored by the University of Pennsylvania, with enrollment of 24. Its official title describes an incomplete crossover trial, while the registry’s design model is listed as parallel. The intervention arms are active TMS and sham TMS.

The primary endpoint is percent change from baseline on the Boston Naming Test. The active-arm mean is 10.3 with standard deviation 20.0 at 2 months, and 37.7 with standard deviation 31.2 at 6-months; each summary includes 9 participants. The sham-arm primary outcome and a comparative estimate are not reported in the registry record.

Adverse-event listings report zero serious events, zero other events, and zero deaths in both groups, with 10 at risk in Active TMS and 5 in Sham TMS. The record lists 4 outcome measures and 0 analyses. It does not report the primary comparison, its precision, or a detailed account of missing-data handling.