Brain Biomarkers of Response to Treatment for Apraxia of Speech: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Brain Biomarkers of Response to Treatment for Apraxia of Speech (NCT02046941), a na na study. The primary outcome is Change From Baseline in Percent of Trained Items Correctly Repeated.

2. Clinical Question

This design can describe how measured performance changed during the treatment period among enrolled participants. Without a separate comparison group, it cannot distinguish treatment-associated change from other changes over time or establish how participants would have performed without the intervention. The registry lists an enrollment of 17, but does not report participant-level outcomes, a group summary for the primary measure, or an estimate of uncertainty around any treatment-associated change.

3. Trial Design

NCT02046941 is a completed, single-group behavioral study titled Brain Biomarkers of Response to Treatment for Apraxia of Speech. The listed enrollment is 17. The conditions are aphasia and stroke, and the brief description focuses on MRI brain imaging and speech-language treatment for speech difficulties after stroke. The sponsor is the VA Office of Research and Development.

The intervention is Speech Production Treatment. Its description specifies a response-contingent hierarchy involving verbal modeling and repetition, graphic cueing, integral stimulation, and articulatory placement instruction. The record does not report the number or duration of treatment sessions, how treatment fidelity was assessed, or other details needed to reconstruct the delivered treatment schedule.

The study model is single group, masking is listed as none, and allocation is listed as NA. There is no parallel control arm described. The primary outcome compares each participant’s performance during pre-intervention probe trials with performance at the end of intervention. This baseline-to-end comparison is an observational contrast within the treated group, not a randomized comparison between treatment and control groups.

The primary outcome timeframe is 8 weeks. The study began on 2014-03-01, reached primary completion on 2018-01-08, and has completed status. The record lists three outcome measures and identifies percentage change in fractional anisotropy as a secondary outcome. The available record does not report the full details of every outcome measure, the timing of each MRI assessment, or results for the listed outcomes.

Although the title emphasizes brain biomarkers of response, the primary outcome described is a behavioral measure of trained-item repetition. The registry record does not report analyses linking individual changes in brain imaging measures to changes in speech performance. It also does not report a protocol-level statistical analysis plan or an explanation of the planned analysis population.

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 Change From Baseline in Percent of Trained Items Correctly Repeated, measured over 8 weeks.

The primary endpoint is change from baseline in the percent of trained items correctly repeated, expressed as a d2 treatment effect size. For each participant, three target sounds were selected from single consonants, vowels, or clusters at the word level. Ten trained word items were generated for each target sound. Participants completed probe trials using these trained items during the pre-intervention stage and during treatment.

The endpoint compares mean performance across the pre-intervention phase with mean performance across the end-of-intervention phase. The registry describes the end phase as the final 3 probe trials and the baseline phase as 5 probe trials. The d2 statistic divides the difference in mean percent correct between those phases by the pooled standard deviation of the two phases. The registry states that a larger d2 represents a larger treatment effect for the individual.

Standardizing a change by variability can make the size of a change easier to interpret relative to the spread of measurements across phases. The measure is intended to summarize each participant’s change on trained items, rather than performance on an independently described set of untrained words. The outcome is therefore specific to the trained materials and the repeated probe procedure. The registry record does not report observed d2 values, a group-level summary, or uncertainty estimates for this endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a continuous measure summarized through an individual d2 treatment effect size. In general, a continuous outcome can take a range of values rather than belonging to a small set of categories. The study’s underlying performance measure is the percentage of trained items correctly repeated on probe trials. The d2 statistic expresses the difference between phase means relative to the pooled standard deviation of those phases, as described in the registry record.

In general, standardization can help describe a change in relation to the variability of the measurements used to calculate it. It does not remove design limitations, and it does not by itself establish that an intervention caused a change. Here, the calculation is described for each participant, using the pre-intervention phase and the final intervention phase. The registry does not report the individual effect sizes or a group summary of them, so their distribution and precision cannot be assessed from the record.

The probes were repeated over time. Repeated measurement can provide information about a participant’s performance across phases, but the registry’s description does not include the underlying trial-by-trial data or a statistical model for those repeated observations. The endpoint description says that baseline performance was established during repeated pre-intervention sessions and that performance was assessed during every other treatment session. It then defines the phase comparison using the baseline probes and final probes. How variation across all intervening observations was used in any analysis is not reported in the registry record.

A standard deviation describes spread among measurements. A pooled standard deviation combines variability from the phases used in the calculation. If variability differs substantially between phases, that pooled value may not represent each phase equally well. This is a general statistical consideration; the registry does not report phase-specific standard deviations or enough outcome data to assess whether this issue arose in the study.

Confidence intervals are general tools for expressing the uncertainty around an estimate. They indicate a range of values compatible with the data and a specified method, rather than proving that a result is true or false. No confidence interval for the primary outcome is reported in the registry record. Similarly, no inferential test, p-value, or group-level estimate is reported. The absence of these results means the precision and statistical evidence for change cannot be evaluated from the posted record.

The study has one treatment group and no listed concurrent control group. In general, a within-group change compares participants with their own earlier measurements, while a controlled comparison contrasts outcomes between groups. The former can describe a temporal pattern; the latter can help account for changes that might also occur without the intervention, depending on the design and analysis. Since this study is listed as single group, an observed baseline-to-end change, if reported, would not alone identify the treatment’s causal effect.

The registry’s sample-size illustration is described as planning based on an observed effect, with two-sided significance level 0.05, equal allocation between two hypothetical arms, and a normal approximation. The effect to detect is specified as a fraction of the observed effect. These are planning assumptions, not details of the completed study’s actual design. The illustration does not represent observed power, does not establish the study’s original sample-size justification, and does not provide a direct analysis of the single-group data. The original sample-size justification is not reported in the registry record.

For general interpretation, planning calculations depend on the target effect, variability, significance level, allocation, and statistical assumptions. A planning illustration can show how a hypothetical design might be sized under its assumptions, but it cannot resolve whether the assumptions are appropriate or whether the actual study produced a precise estimate. The registry does not report an original statistical analysis plan, an interim analysis plan, or the methods used for any completed analysis.

Multiplicity arises when several outcomes or statistical tests are considered. In general, multiple testing can increase the chance of at least one apparently notable result when each test is considered separately. This record lists three outcome measures, but does not report a multiplicity strategy, a hierarchy of outcome testing, or completed analyses. It is therefore not possible to determine whether or how multiplicity was addressed.

Finally, the title refers to biomarkers of response, but a biomarker analysis would require reported measurements and an analysis connecting those measurements to clinical or behavioral response. The registry lists fractional anisotropy percentage change as a secondary outcome, but does not report those values or a model relating imaging changes to speech performance. The analysis methods and results for such a relationship are not reported in the registry record.

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("Speech Production Treatment")
n   <- c(10)
m   <- c(6.67)
sp  <- c(4.43)
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  Speech Production Treatment  10  6.670  4.430  3.501  9.839

11. Multiplicity, Missing Data and Analysis Populations

The registry lists three outcome measures, including the primary trained-item repetition endpoint and a secondary outcome of percentage change in fractional anisotropy. It does not report a complete account of all outcome definitions in the available record, a hierarchy for testing outcomes, or a method for controlling multiplicity. Because no analyses are listed, there are no posted hypothesis tests from which to assess how multiple outcomes or comparisons were handled. Whether an adjustment or other prespecified strategy was used is not reported in the registry record.

Missing-data handling is also not reported. The record lists enrollment of 17, but does not state how many participants contributed primary outcome data, whether any observations were incomplete, or how missing measurements were treated. No imputation approach, sensitivity analysis, or missingness assessment is described in the available information.

The analysis population is not defined in the registry record. It is not stated whether the reported outcome, if any, was intended to include all enrolled participants, only participants who completed treatment, or another subset. Participant flow and reasons for any non-completion are not reported.

These omissions matter because outcome availability and analysis choices can affect the estimate and its uncertainty. They prevent a reader from checking how the enrolled group relates to any analyzed group, whether incomplete observations could influence interpretation, and whether outcome testing followed a prespecified plan. Since the registry lists zero analyses and gives no numerical outcome results, these issues cannot be resolved from the posted record.

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 registry’s adverse-event figures list 17 participants at risk in the Speech Production Treatment group. The serious adverse-event count is 0, the other adverse-event count is 0, and deaths are listed as 0. These are the event counts recorded for the group; they do not provide a detailed description of event definitions, collection procedures, or the period of safety monitoring.

A zero count in a registry record means that no events in that category are recorded in these figures. It does not, by itself, establish that no event could occur or that every possible event was captured. The record does not provide event narratives, severity gradings beyond the serious and other categories, relatedness assessments, or participant-level exposure information.

The study is a single-group treatment study, so these figures do not support a comparative safety estimate against an untreated or alternative-treatment group. They also do not establish whether safety monitoring was identical across participants. The registry lists no serious adverse-event terms. The figures should therefore be read as the posted counts, with the limits of the available reporting in mind, rather than as a controlled estimate of risk.

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

The record establishes that the study assessed a behavioral speech-treatment approach in a single group and planned to quantify change in trained-item repetition from a pre-intervention baseline to the end of treatment. It also identifies brain imaging as part of the study’s focus and lists percentage change in fractional anisotropy as a secondary outcome. These details show what the study set out to measure, not what the measurements showed.

No numerical primary outcome results, secondary outcome results, uncertainty intervals, or analysis results are reported in the registry record. Consequently, the direction, size, consistency, and precision of any observed change cannot be described from the posted information. The absence of reported results is not evidence that the outcome was unchanged, nor is it evidence that a particular result occurred.

Even if a baseline-to-end change were available, this single-group design would describe change over time among treated participants rather than isolate the effect of treatment from other possible influences. There is no separate comparator group in the registry record. The design therefore cannot establish what would have happened to the same participants over the same period without treatment.

The endpoint is based on trained word items and repeated probes. It does not, from the description alone, establish whether any change would generalize to untrained speech materials or everyday communication. The registry does not report such outcomes in the available information. Likewise, although imaging is part of the study’s rationale, no imaging result or association between imaging and speech outcomes is reported.

The enrollment figure and safety counts provide context for the study, but do not substitute for outcome estimates or an analysis of uncertainty. The record is informative about the study question and design, while leaving the central empirical result unavailable. No conclusion about the magnitude of treatment-associated change or its causal interpretation can be drawn from the registry record alone.

14. Why This Trial Matters Statistically

This study is statistically notable because its primary measure is built around repeated observations within each participant rather than a comparison between randomized groups. The design makes the baseline-to-end change central: each participant serves as a reference for their own pre-intervention performance. That structure can be useful for describing individual patterns, especially when the endpoint is defined around a specific set of trained items.

The use of a standardized d2 effect size also makes the measurement strategy explicit. The registry describes the numerator as the difference in mean performance between phases and the denominator as the pooled standard deviation. This connects the observed change to the variability of the measurements. It also makes clear why reporting only a group mean, without information about variability and individual patterns, would not fully communicate the endpoint’s behavior.

At the same time, a within-participant baseline is not equivalent to a control group. Repeated observations may help characterize performance over phases, but they do not by themselves rule out other reasons for change over time. The single-group design therefore places an important boundary on causal interpretation. The issue is not unique to this study; it is a general distinction between documenting change and estimating a comparative treatment effect.

The study also connects behavioral measurement with brain imaging. That combination can motivate questions about whether changes in imaging measures track changes in speech performance. However, the registry record does not report numerical imaging results or analyses connecting the two domains. The statistical value of the design’s biomarker question cannot be judged without those results and methods.

Finally, the record illustrates why a clearly described endpoint is not the same as a reported result. The endpoint and calculation are described, but the registry lists zero analyses and provides no numerical outcome estimates. This limits assessment of both the finding and its precision. The study remains informative as a description of a measurement approach and research question, while its posted results do not support a quantitative conclusion.

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 most direct limitation for interpreting this trial is that numerical results for the primary endpoint are not reported in the registry record. The registry also lists zero analyses. Without an estimate, uncertainty measure, or analysis output, the size and precision of the outcome cannot be evaluated.

The single-group design has no separate comparator. A pre-intervention baseline supports a description of change over time, but cannot alone distinguish intervention-associated change from other possible influences. Masking is listed as none. The record does not describe procedures to reduce assessment or measurement bias.

Enrollment is listed as 17, but the number of participants with complete primary outcome data is not reported. Participant flow, treatment completion, missing observations, analysis populations, and missing-data methods are not described in the available record. The treatment schedule, dose, and adherence or fidelity information are also not reported.

The primary measure focuses on trained items and a comparison of pre-intervention with end-of-intervention probes. The record does not report whether outcomes on untrained items or broader communication measures were assessed. It does not provide individual d2 values or a group-level summary that would show how consistent changes were across participants.

Although brain imaging is emphasized in the study description, imaging results and analyses connecting brain changes with speech outcomes are not reported. The registry entries also do not describe the full set of three outcome measures or a plan for handling multiple outcomes. The original sample-size justification, interim analyses, and statistical analysis plan are not reported in the registry record. These omissions limit both reproducibility and interpretation of the study’s intended statistical claims.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Brain Biomarkers of Response to Treatment for Apraxia of Speech — NCT02046941.
• Registry record last updated 2019-08-28; status last verified 2019-07.
• Results first posted 2019-08-28.
• 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

NCT02046941 is a completed, single-group behavioral study sponsored by the VA Office of Research and Development, with enrollment of 17. It evaluated Speech Production Treatment in a study focused on brain biomarkers of response to treatment for apraxia of speech. The treatment description includes verbal modeling and repetition, graphic cueing, integral stimulation, and articulatory placement instruction.

The primary outcome is change from baseline in percent of trained items correctly repeated, expressed as an individual d2 treatment effect size over an 8-week timeframe. Percentage change in fractional anisotropy is listed as a secondary outcome. The study began on 2014-03-01 and reached primary completion on 2018-01-08.

The registry lists no comparator group, no masking, and zero analyses. Numerical outcome results and uncertainty estimates are not reported in the registry record. The posted safety figures list 0 serious adverse events, 0 other adverse events, and 0 deaths among 17 participants at risk in the treatment group.