Brain Machine Interface Control of an Robotic Exoskeleton in Training Upper Extremity Functions in Stroke: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Brain Machine Interface Control of an Robotic Exoskeleton in Training Upper Extremity Functions in Stroke (NCT01948739), a na na study. The primary outcome is Change From Baseline in Fugl-Meyer Arm (FMA) Motor Score.

2. Clinical Question

The study also aimed to identify robotic and electrophysiological measures of arm motor impairment and recovery after stroke. It describes robotic kinematic data acquired through device sensors and EEG-based electrophysiological measures. The record lists additional measures of arm and hand function, including the Action Research Arm Test, the Jebsen-Taylor Hand Function Test, grip strength, and pinch strength. It does not provide their results in the record.

Thus, the question is exploratory and within-group in structure: the primary outcome concerns change over time among people receiving the described training. Without a listed control group, any change cannot be separated from other changes that may occur over time or from influences associated with study participation. The registry record does not report a comparative estimate against usual care, another device, or no intervention.

3. Trial Design

The registry record describes a completed, single-group study of the MAHI EXO-II exoskeleton augmented with a brain-machine interface. Enrollment is listed as 18. The study start date is 2013-09-24, and the primary completion date is 2018-04-28. Its phase is listed as NA. Allocation is listed as NA, and masking is listed as NONE.

The intervention is a device: MAHI EXO-II exoskeleton augmented with BMI system. The description says it was intended to include the patient in the control loop and make therapy active. The brief summary identifies three broad purposes: augmenting the exoskeleton with a non-invasive brain-machine interface, determining appropriate robotic and electrophysiological measures, and demonstrating the feasibility and effectiveness of BMI-controlled robotic arm training. These are statements of study aims, not evidence that the aims were achieved.

The record does not list a parallel control arm. A single-group design can describe measurements over time in the treated group, but it cannot by itself provide a controlled estimate of how outcomes would have changed without the intervention. In particular, a baseline-to-follow-up change is not equivalent to a difference between randomized groups. The registry record does not report random assignment, a comparator intervention, or masking procedures beyond listing masking as NONE.

The primary outcome is assessed at baseline and at time points after treatment: immediately after the end of treatment, within a week; 2 weeks after the end of treatment; and 12 weeks after the end of treatment. The record does not report the treatment schedule, follow-up procedures, eligibility criteria, or analysis populations in further detail. It also does not report the statistical plan, including the approach used to estimate change or account for repeated measurements.

Although the study status is COMPLETED, the status alone does not indicate that every planned outcome was collected or analyzed. The registry record notes missing primary-outcome data at some follow-up assessments and does not report numerical primary-outcome results.

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: 18 were enrolled and allocated to the intervention, but only 10 received the intervention.

Overall Study

Analysis population for the primary outcome, as defined in the registry: 2 weeks after end of treatment, data for this outcome measure were not collected for 3 participants. 12 weeks after end of treatment, data for this outcome measure was not collected for 1 participant.

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.

Population: Baseline characteristics are reported for those who received the intervention.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Fugl-Meyer Arm (FMA) Motor Score, measured over Baseline, immediately after end of treatment (within a week), 2 weeks after end of treatment, 12 weeks after end of treatment.

The primary endpoint is change from baseline in the Fugl-Meyer Arm motor score. The registry record describes the Fugl-Meyer assessment as a stroke-specific, performance-based impairment index. Its upper-limb component scores 33 items using an ordinal scale: 0 means unable to perform, 1 means able to perform partially, and 2 means able to perform fully. The total range is 0-66, with 0 described as poor and 66 as normal.

This measure is used to quantify arm motor impairment in the context of stroke recovery. Because it is based on performance and an ordered scoring scale, a score represents assessed function rather than a direct measure of how a person performs every activity in daily life. The record’s focus on change from baseline allows the outcome to be considered at several post-treatment assessments, but the registry record does not report the observed means, standard deviations, changes, or uncertainty estimates.

The scheduled assessments are baseline, immediately after the end of treatment within a week, 2 weeks after the end of treatment, and 12 weeks after the end of treatment. The record notes that primary-outcome data were not collected for 3 participants at the 2-week assessment and for 1 participant at the 12-week assessment. The registry record does not explain the reasons for these missing measurements or report how they were handled in the analysis.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The outcome is a continuous score, and the registry record specifies a mean as the summary parameter and a standard deviation as the measure of dispersion. A mean describes the central level of observed scores or changes. A standard deviation describes how spread out individual values are around that mean. These summaries answer different questions: the mean concerns the group’s average, while the standard deviation concerns variability among observations. Neither is reported numerically for the primary outcome in the registry record.

What does a change score represent?

Change from baseline compares a follow-up measurement with a participant’s baseline measurement. In general, an average change can describe the direction and size of within-group movement in the measured score. It does not, on its own, identify why that change occurred. In a single-group study, there is no concurrent comparator estimate to distinguish intervention-related change from other changes that may happen over time. The record names change from baseline as the outcome but does not describe the exact analysis used to estimate it.

How should uncertainty be understood?

In general, a confidence interval expresses the range of estimates compatible with the data under a specified statistical procedure. A narrower interval indicates greater precision for that estimate; a wider interval indicates greater uncertainty. A p-value, when reported, addresses compatibility with a specified null hypothesis under the analysis assumptions, not the clinical importance of an effect. The registry record does not report a confidence interval or p-value for the primary outcome. Without an estimate and its uncertainty, readers cannot judge the precision of the primary finding.

What does the score scale imply?

The Fugl-Meyer upper-limb score is described as an ordinal score made from 33 items, each scored 0, 1, or 2. The total is described as ranging from 0-66. The registry record specifies a mean and standard deviation for this continuous outcome. As a general statistical point, use of a mean summarizes scores numerically, while the underlying item categories are ordered levels. Interpretation should keep both features in view: a summary score can be useful for comparing group averages, but the scale’s numeric labels do not necessarily mean every step has identical clinical meaning.

Why do repeated assessments matter?

The primary outcome is scheduled at baseline and at several follow-up times: immediately after the end of treatment within a week, 2 weeks after the end of treatment, and 12 weeks after the end of treatment. Repeated measures can describe how a group’s scores vary over time, but they also raise analysis questions. For example, measurements from the same participant are generally related rather than independent. Statistical methods for repeated observations may account for this dependence. The registry record does not report whether such a method was used, how each time point was analyzed, or whether one follow-up time was designated as the main inferential comparison.

What do the planning assumptions mean?

The record describes planning illustrations based on the observed effect, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The illustrations consider detecting a stated fraction of the observed effect. In general, a sample-size calculation estimates the number of participants needed under its assumptions to reach a chosen target for statistical power. Those assumptions describe a hypothetical planning exercise; they do not describe the single-group design listed for this study. The calculations are planning illustrations, not observed power, and their assumptions do not establish that a two-arm comparison occurred.

The registry record does not report the study’s own sample-size justification or a statistical analysis plan. It therefore does not establish whether the planning illustrations reflect the original design, what effect was anticipated before enrollment, or whether any prespecified analysis was followed. The registry record also does not report whether analyses adjusted for baseline score or other variables. These omissions limit the ability to assess the analysis strategy independently.

What can recalculation establish?

In general, a recalculation from posted summary statistics can check a simple unadjusted estimate based on those summaries. It cannot reproduce analyses requiring participant-level data, adjustments, or assumptions that are not documented. The registry record does not include primary-outcome summary values or a reported analysis to compare against, so no recalculated primary estimate can be interpreted from the information recorded. Any calculations based on posted statistics should be distinguished from the sponsor’s own analysis; where a discrepancy is identified, the sponsor’s own analysis takes precedence.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists the Fugl-Meyer Arm motor score as the primary outcome and names secondary outcomes for the Action Research Arm Test, Jebsen-Taylor Hand Function Test, grip strength, and pinch strength. It lists 11 outcome measures overall. The primary outcome is scheduled at multiple assessments, creating several possible time-specific comparisons. Considering multiple outcomes or time points can increase the chance of finding an apparently notable result by chance if many comparisons are tested. In general, statistical plans may address this through prespecified primary comparisons or multiplicity adjustments. The registry record does not report whether any such approach was planned or used.

Missing primary-outcome data are explicitly noted at follow-up: data were not collected for 3 participants at 2 weeks after the end of treatment, and data were not collected for 1 participant at 12 weeks after the end of treatment. These are the missingness details reported for this outcome. The record does not report why those measurements were not collected, whether other outcome data were missing, or how missing observations were handled. It also does not give the analysis population or the number of participants analyzed at each assessment.

In general, missing-data methods rely on assumptions about why measurements are absent and how the missing values relate to observed information. Different assumptions can lead to different estimates. Because reasons and handling are not reported, the robustness of any time-point estimate to missing data cannot be assessed from the registry record. The listed enrollment should not be interpreted as the number analyzed at every outcome assessment.

The record does not report whether analyses followed an intention-to-treat approach, used available observations, or applied another population definition. It also does not report prespecified rules for multiplicity or missing data. These are not details that can be inferred from the completed status or from the outcome descriptions.

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.

For the listed group, the registry record reports 10 at risk for serious adverse events and 0 affected. It also reports 10 at risk for other adverse events and 0 affected, with 0 deaths. No serious adverse event terms are listed.

These are registry counts, not estimates of comparative risk. The study has a single group and no listed control group, so the counts cannot show whether events were more or less common than they would have been under another treatment or without the intervention. A count of 0 affected in the reported group does not establish that the intervention is free of risk. It describes only what the registry record lists for the recorded safety group.

The enrollment is listed as 18, while the safety fields list 10 at risk for each adverse-event category. The registry record does not explain this difference or specify the safety population in further detail. The listed counts should therefore be read using their stated at-risk group rather than treated as covering every enrolled participant. The record does not provide event-level descriptions, severity details beyond the serious-event category, timing, or attribution information.

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

The study record establishes that investigators evaluated a BMI-augmented MAHI EXO-II exoskeleton in a single-group training study and designated change in the Fugl-Meyer Arm motor score as the primary outcome. It also describes additional aims involving measures of arm impairment and recovery, and lists several secondary outcomes. These details clarify the topic and intended measurements.

The registry record does not report the numerical primary-outcome results. It therefore does not show whether average scores increased, decreased, or remained similar, nor does it show how variable or precise any observed change was. Although the record lists a completed status, completion does not supply missing outcome estimates or establish that every objective was answered.

Even if a within-group change were reported, the design would not by itself establish that the intervention caused that change. There is no separate comparator group listed, and there is no randomized comparison in the record. Change over time can reflect multiple influences, and a single-group observation does not estimate the counterfactual outcome that would have occurred without the intervention. The registry record does not report enough information to separate those influences.

Likewise, the safety counts should not be read as proof of no risk. They report zero affected in the listed at-risk group, but the record does not explain the difference between enrollment and the safety denominators or provide comparative safety information. Nor do the counts establish feasibility or effectiveness; those were among the study’s stated aims, not results reported in this record.

The most defensible reading is therefore narrow: the registry record documents a completed single-group study, its intervention, its outcome framework, some missing follow-up measurements, and listed adverse-event counts. It does not provide the numerical efficacy results or detailed analysis information needed to judge the size, precision, or cause of outcome changes. The planning illustrations, if considered, are hypothetical calculations under stated assumptions rather than evidence of a controlled treatment effect.

13. Why This Trial Matters Statistically

This study is statistically notable because its main question concerns change over time, while its design is single-group. That structure makes the distinction between observed change and causal effect especially important. A baseline-to-follow-up comparison can describe what happened to measured scores among participants, but it cannot supply a concurrent estimate of what would have happened under a comparator condition. Interpreting change as an intervention effect would require evidence beyond the design features reported in the registry record.

The measurement framework also spans multiple outcomes and assessment times. The primary measure is the Fugl-Meyer Arm motor score, and the record lists additional arm and hand function measures. Multiple outcomes can provide a broader view of function, while multiple assessment times can describe its course. At the same time, each additional comparison creates choices about which result is primary and how to account for multiple testing. The registry record does not report a multiplicity strategy or identify the analysis used for each time point.

Missing follow-up measurements matter for the same reason. The record states that primary-outcome data were not collected for some participants at specified follow-up assessments, but does not report why or how the analysis handled the missing values. This leaves uncertainty about the analyzed populations and whether the reported observations, if any, represent all enrolled participants.

Finally, the planning illustrations use assumptions for a two-arm design, while the listed study design is a single group. That distinction is useful when interpreting sample-size calculations: calculations under a hypothetical design do not retroactively create a comparison or demonstrate observed power. The record’s limitations are therefore not merely numerical; they concern which question the design can answer and what information is available to support interpretation.

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 main trial-specific limitation is the absence of a listed comparator group. The intervention was evaluated in a single-group design, so the study cannot make a controlled estimate of the effect relative to another condition. The registry record does not report randomization or a comparator intervention. This constrains causal interpretation of any change from baseline.

The registry record does not report the primary outcome values, summary statistics, confidence intervals, or p-values. It therefore does not permit assessment of the direction, magnitude, or precision of the primary outcome. The record is marked COMPLETED, but that status does not replace the missing numerical results.

Missing primary-outcome data are recorded at 2 weeks and 12 weeks after the end of treatment. Reasons for missing measurements, the analysis sample at each time point, and the method used to address missing data are not reported in the registry record. The analysis populations and statistical plan are also not reported. These gaps limit assessment of whether the results could be sensitive to the way incomplete measurements were handled.

The record lists several secondary outcomes and multiple primary-outcome assessment times but does not report a plan for prioritizing comparisons or handling multiplicity. It also lists 18 enrolled participants and a smaller at-risk group in the safety fields, without explaining how the safety population was defined. The safety counts therefore require attention to their listed denominator and should not be generalized to all enrolled participants without further information.

Additional details—including eligibility criteria, treatment schedule, and analysis procedures—are not reported in the registry record. The brief description of longitudinal training does not establish a specific dose or schedule. Taken together, these limitations mean the record describes the study question and framework more clearly than its results or the methods used to analyze them.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Brain Machine Interface Control of an Robotic Exoskeleton in Training Upper Extremity Functions in Stroke — NCT01948739.
• Registry record last updated 2021-06-29; status last verified 2021-06.
• Results first posted 2021-06-29.
• 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

This completed study evaluated BMI control of the MAHI EXO-II exoskeleton in a single-group setting for upper-extremity training after stroke. Enrollment is listed as 18. The registry record gives the study start date as 2013-09-24 and primary completion date as 2018-04-28. Allocation is listed as NA, masking as NONE, and phase as NA.

The primary outcome is change from baseline in the Fugl-Meyer Arm motor score, assessed at baseline and after treatment. The registry record notes missing data at two follow-up assessments but does not report numerical outcome results, uncertainty estimates, or the analysis method. It lists additional arm and hand function outcomes and reports zero affected for serious and other adverse events in the listed at-risk group, with zero deaths.

The record’s single-group design supports describing measured outcomes over time but does not provide a controlled comparison. The registry record does not report enough statistical results to assess the magnitude or precision of the primary outcome change.