This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Improving Gait Performance in Individuals With Spinal Cord Injuries: an Intervention Using Robotic Exoskeletons (NCT03011099), a randomized na study. The primary outcome is Change in Over Ground Gait Speed as Assessed by the 10 Meter Walk Test.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03011099 |
| Phase | NA |
| Condition | Spinal Cord Injuries |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Single |
| Enrollment | 12 |
| Arms in the primary outcome | 2 |
| Lead sponsor | The University of Texas Health Science Center, Houston |
| Results posted | Yes |
2. Clinical Question
Population
The study concerns individuals with spinal cord injuries. The registry record lists an enrollment of 12, but does not describe participant characteristics or eligibility criteria in the information available for this analysis. It also does not state how many participants contributed data to each outcome beyond the arm-level outcome summaries.
Intervention
The intervention was robotic exoskeleton training. The description says participants wore a lower-extremity robotic walking device during individualized sessions. Possible activities included sit-to-stand practice, standing balance, weight shifting, walking, turning, and stand-to-sit practice. The registry description is truncated, so it does not give a complete account of the intervention.
Comparator
The comparison was conventional physical therapy designed to facilitate or promote gait. Individualized sessions could involve stretching, strengthening, balance training, standing, and gait training. The registry description is truncated, and details such as the amount and schedule of therapy are not reported in the registry record.
Primary question
The primary question was whether change in over-ground gait performance, assessed by the 10 Meter Walk Test, differed between the two assigned training approaches. The listed endpoint is the change in time taken to walk 10 meters, measured at baseline and week 4. The registry record posts arm-level means and standard deviations but no treatment contrast, confidence interval, or inferential test for this endpoint.
3. Trial Design
The registry describes a randomized, single-masked, parallel-group study comparing robotic exoskeleton training with conventional physical therapy. The listed phase is NA. The sponsor is The University of Texas Health Science Center, Houston. The study is marked completed; the recorded start is 2014-01 and primary completion is 2016-03. The registry enrollment is 12.
In a randomized design, assignment is intended to allocate participants to study arms by chance rather than by a clinician’s or participant’s choice. This can help limit systematic differences between groups at the start, although randomization does not guarantee balance, particularly in a small study. The record does not report the randomization sequence, allocation concealment, or the procedures used to implement assignment.
Single masking means one party or group was masked, but the identity of who was masked is not reported in the registry record. The record also does not describe whether outcome assessors were masked, or how masking was maintained. The parallel model indicates that the study compared separate groups assigned to the two interventions; the posted results are presented separately by arm.
The study description frames the comparison as an investigation of whether robotic exoskeleton training can improve walking performance after spinal cord injury compared with conventional physical therapy. The primary outcome was change in time on the 10 Meter Walk Test from baseline to week 4. Five secondary outcomes are listed, covering gait characteristics, lower-extremity strength, and user feedback. The registry record does not report the session frequency, session duration, overall treatment period details beyond the primary outcome timeframe, adherence, or co-interventions.
The endpoint summaries list 6.0 participants in the robotic exoskeleton training arm and 3.0 in the conventional physical therapy arm. The reason these arm-level outcome counts differ from the registered enrollment is not reported in the registry record. They should therefore be read as the counts associated with the posted endpoint summaries, not as a complete account of participant flow.
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: One participant dropped out of the study after signing the consent before randomization.
Overall Study
| Stage | Robotic Exoskeleton Training | Conventional Physical Therapy |
|---|---|---|
| STARTED | 6 | 5 |
| COMPLETED | 6 | 3 |
| NOT COMPLETED | 0 | 2 |
Reasons for not completing this period:
| Reason | Robotic Exoskeleton Training | Conventional Physical Therapy |
|---|---|---|
| Withdrawal by Subject | 0 | 2 |
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.
| Characteristic | Unit | Robotic Exoskeleton Training | Conventional Physical Therapy | Total |
|---|---|---|---|---|
| Participants analysed | 6 | 3 | 9 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 5 | 3 | 8 |
| Age, Categorical: >=65 years | Participants | 1 | 0 | 1 |
| Sex: Female, Male: Female | Participants | 2 | 1 | 3 |
| Sex: Female, Male: Male | Participants | 4 | 2 | 6 |
| Region of Enrollment: United States | participants | 6 | 3 | 9 |
6. Primary Endpoint
The registry lists the primary outcome as Change in Over Ground Gait Speed as Assessed by the 10 Meter Walk Test, measured over baseline, week 4.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | Standard Deviation |
| Unit | seconds |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is titled change in over-ground gait speed as assessed by the 10 Meter Walk Test. The measure is recorded in seconds and described as the time taken to walk 10 meters, with measurements at baseline and week 4. It therefore focuses on performance over a fixed walking distance, expressed as elapsed time. In general, for a fixed distance, a shorter time corresponds to faster walking.
Measuring change over time can describe how the recorded outcome differs between an initial assessment and a later assessment. However, the registry record does not state the precise calculation used to derive change, including whether it is week 4 minus baseline or the reverse. That matters for interpreting the sign: the posted mean changes are negative, but the direction those values represent cannot be established from the record alone.
A timed walking test is used to quantify walking performance in a standardized way. Reporting the outcome in seconds makes the measured quantity clear, but it does not by itself establish whether a change is important to participants or whether it reflects a difference attributable to treatment. Interpretation also depends on the observed variability, the completeness of the measurements, and an appropriate comparison between randomized groups. The registry record does not report a threshold for a clinically meaningful change or an inferential comparison of the arms.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry posts a mean change of -6.815 seconds (standard deviation 9.58) for robotic exoskeleton training, based on 6.0 participants in that endpoint summary. For conventional physical therapy, it posts a mean change of -0.5 seconds (standard deviation 4.02), based on 3.0 participants. These are arm-level descriptive summaries. They do not, by themselves, quantify the uncertainty in the difference between the assigned approaches.
The primary outcome is labelled as change in gait speed but is recorded in seconds for the time taken to walk 10 meters. Because the registry record does not specify the direction of subtraction used to calculate change, the negative values cannot safely be translated into an increase or decrease in walking time. In general, a lower time over the same distance corresponds to faster walking, but applying that interpretation to a change score requires knowing how the score was calculated.
The standard deviations describe the spread of participant-level measurements around each arm’s mean change. The posted values indicate that observations varied within both arms; a standard deviation is not a confidence interval and does not state how precisely the mean or between-arm contrast has been estimated. With only the posted arm summaries, the registry does not provide a confidence interval, p-value, or adjusted estimate for the primary comparison. Those inferential results are not reported in the registry record.
The arm-level means can be read as the average changes recorded among the participants included in each summary. They should not be treated as a direct estimate of the causal effect of assignment, since that requires an explicit between-group contrast and its uncertainty. Nor do the summaries establish whether any observed change would be important in practice. The registry record does not report a clinically meaningful-change criterion.
The outcome counts are small: the robotic exoskeleton summary includes 6.0 participants and the conventional physical therapy summary includes 3.0. Small groups make estimates sensitive to individual observations and generally leave substantial uncertainty. The standard deviations help show within-arm spread, but without a reported interval for the contrast, the precision of the comparative result is not quantified.
The record lists an enrollment of 12, while the posted primary endpoint summaries include the arm-specific counts above. It does not explain the difference between the enrollment and the endpoint counts. Reasons for missing endpoint observations, exclusions, or other differences in analysis populations are not reported in the registry record. Consequently, the arm summaries should be interpreted as results for the posted outcome data, not as evidence that all enrolled participants contributed to that endpoint.
Overall, the posted primary outcome information is descriptive rather than a complete comparative analysis. It gives a mean and standard deviation for each arm, but no reported statistical test or interval for the difference. This limits how confidently the two groups can be compared using the registry record alone.
9. Statistical Methods Explained
The registry presents a continuous primary outcome: change in the time taken to walk 10 meters. The posted arm summaries use the mean as the measure of central tendency and the standard deviation as the measure of dispersion. These statistics describe the observed outcome data, but are not a full treatment-effect analysis.
What does the mean describe?
In general, the mean is the arithmetic average of the observations in a group. It provides a summary of the group’s central value, but can be influenced by unusually high or low observations. For this study, the registry reports a mean change for each intervention arm. The exact method used to calculate each participant’s change is not reported in the registry record, so the direction represented by a negative change cannot be determined from the posted information alone.
What does the standard deviation describe?
In general, the standard deviation describes how spread out observations are around their group mean. A larger standard deviation indicates more variation among the observed values; it does not directly measure the precision of the mean or the difference between groups. Precision depends on both variability and the amount of information available. The registry reports standard deviations for the two arm summaries, but it does not report a confidence interval for either mean or for the between-arm contrast.
Why is an arm comparison needed?
A randomized two-arm trial is designed to compare outcomes between assigned groups. In general, a treatment contrast expresses how the outcomes differ between those groups, while an uncertainty interval indicates the range of values compatible with the data and analysis assumptions. A p-value, when reported, addresses how compatible the data are with a specified null hypothesis under a statistical model; it does not measure the size or importance of an effect. The registry record does not post a primary between-arm contrast, confidence interval, or p-value, and lists no analysis entries. The available means and standard deviations therefore do not support a registry-reported inferential conclusion.
What can the participant counts tell us?
The posted endpoint summaries list 6.0 participants for robotic exoskeleton training and 3.0 for conventional physical therapy. The registry enrollment is 12, and the reason the outcome counts differ from that enrollment is not reported in the registry record. Counts attached to an outcome summary indicate how much participant-level information is represented, but do not explain why observations may be absent. Without participant flow or missing-data information, it is not possible to assess whether the posted endpoint data represent all randomized participants or whether exclusions could affect the comparison.
How should planning illustrations be understood?
In general, a sample-size calculation estimates the amount of information a future study might need to detect an effect under specified assumptions. The planning assumptions described for this study use a two-sided significance level of 0.05, equal allocation, a normal approximation, and an effect defined as a stated fraction of the observed effect. Such planning illustrations are not observed power and do not describe what the completed trial demonstrated. A calculation based on an observed effect can also be unstable when the estimate comes from a small study. Numerical planning results and the fractions used are not reported in the registry record.
“Participants per arm for 80% power” describes a planning target under an assumed effect and statistical model; it is not a count of participants who were actually observed in this trial. Planning quantities should not be confused with the posted arm counts or treated as evidence that the trial had a particular probability of detecting an effect. The registry record does not report its own sample-size justification, target effect, or statistical planning details beyond the assumptions stated for the planning illustration.
Why does over-dispersion matter?
In general, over-dispersion means that observed variability is greater than a statistical model expects. It can make standard errors and inferential results unreliable if the model does not account for that extra variability. The posted endpoint is continuous and the record gives standard deviations, but it does not report a model-based analysis or evidence of over-dispersion. This concept is relevant as a general consideration, not as a finding about this trial.
More broadly, statistical interpretation depends on a clearly defined endpoint, a transparent analysis population, an appropriate comparison, and uncertainty estimates. The registry record identifies the endpoint and posts descriptive arm summaries, but the analysis model, change-score definition, missing-data approach, and inferential results are not reported in the registry record. Those absences constrain interpretation rather than indicating that any particular analysis was or was not performed outside the registry.
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.
| Arm | Participants | Mean | Standard deviation | 95% CI for the mean |
|---|---|---|---|---|
| Robotic Exoskeleton Training | 6 | -6.815 | 9.580 | -16.869 to 3.239 |
| Conventional Physical Therapy | 3 | -0.500 | 4.020 | -10.486 to 9.486 |
R code and output
arm <- c("Robotic Exoskeleton Training", "Conventional Physical Therapy")
n <- c(6, 3)
m <- c(-6.815, -0.5)
sp <- c(9.58, 4.02)
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 Robotic Exoskeleton Training 6 -6.815 9.580 -16.869 3.239
CARM Conventional Physical Therapy 3 -0.500 4.020 -10.486 9.48611. Multiplicity, Missing Data and Analysis Populations
The registry lists five secondary outcomes: changes in cadence, step length, and step time assessed by the GAITRite Walkway; change in lower-extremity strength assessed by the ASIA Lower Extremity Motor Score; and exoskeleton user feedback assessed by questionnaire. The record lists 10 outcome measures in total. The presence of multiple outcomes raises a general multiplicity issue: when many outcomes are tested, the chance of at least one apparently notable result can increase. Whether and how this was addressed is not reported in the registry record.
The record does not report a hierarchy for primary and secondary testing, an adjustment for multiple comparisons, or whether the secondary outcomes were exploratory. It also does not provide inferential analyses for the primary endpoint. The posted results should therefore be read as outcome summaries, not as a set of statistically controlled tests.
Missing-data information is also limited. The registry enrollment is 12, while the primary endpoint arm summaries list 6.0 and 3.0 participants. The reasons for the difference are not reported in the registry record. The record does not describe withdrawals, incomplete assessments, exclusions, imputation, or sensitivity analyses. Without that information, it is not possible to determine how missing outcome values were handled or whether the posted summaries reflect a particular analysis population.
The registry record does not state whether the primary analysis followed intention-to-treat principles, used a per-protocol population, or applied another definition. It also does not report details of the statistical analysis plan, including covariate adjustment or interim analyses. Those elements should not be inferred from the randomized design or from the descriptive summaries.
12. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Robotic Exoskeleton Training | 0 | 6 | 0.0% |
| Conventional Physical Therapy | 0 | 3 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Robotic Exoskeleton Training | 0 of 6 |
| Conventional Physical Therapy | 0 of 3 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Robotic Exoskeleton Training | 1 of 6 |
| Conventional Physical Therapy | 0 of 3 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Robotic Exoskeleton Training | Conventional Physical Therapy |
|---|---|---|---|
| swelling and pain in foot | Musculoskeletal and connective tissue disorders | 1 of 6 | 0 of 3 |
The registry reports serious adverse events for the two arms using the number at risk and the number affected. In the robotic exoskeleton training arm, 6 participants were listed as at risk, with 0 serious adverse events. In the conventional physical therapy arm, 3 participants were listed as at risk, also with 0 serious adverse events. No deaths are listed in either arm.
For other adverse events, the record lists 1 event in the robotic exoskeleton training arm among 6 at risk and 0 in the conventional physical therapy arm among 3 at risk. These are registry counts, not estimates of comparative risk. The record does not provide event descriptions in the listed serious-event terms, and the top serious-event terms list is empty.
Zero recorded serious events in these small groups does not establish that serious events cannot occur or that the approaches have equivalent safety. In general, absence of a recorded event during a study is not proof of absence of risk, especially when the observed groups are small and the period and completeness of safety follow-up are not described. The registry record does not report those follow-up details or an inferential safety comparison. The counts should be read as the safety events posted for this trial, not as a definitive assessment of safety.
13. What the Result Does, and Does Not, Mean
The posted result describes mean change in the timed 10-meter walking measure separately for each study arm. The robotic exoskeleton training arm has a mean change of -6.815 seconds, with a standard deviation of 9.58 and an endpoint count of 6.0. The conventional physical therapy arm has a mean change of -0.5 seconds, with a standard deviation of 4.02 and an endpoint count of 3.0. These values describe the posted outcome summaries; they are not a reported estimate of the difference between treatments.
The registry record does not define the subtraction used for the change score. Therefore, the negative signs cannot be translated into a direction of change in walking time from the posted information alone. The record also does not provide a confidence interval, p-value, or other inferential comparison for the primary endpoint. The degree of uncertainty around a comparative estimate is consequently not quantified in the registry results.
The result does not establish that either approach is better or worse. It does not establish that a change is clinically important, and it does not show how the outcome relates to participants’ daily function. No threshold for meaningful change or broader functional outcome is reported in the registry record. Although gait characteristics and lower-extremity strength are listed as secondary outcomes, the registry entries available for this analysis do not include numerical results for those outcomes.
These summaries also do not establish that all enrolled participants contributed outcome data. Enrollment is listed as 12, whereas the endpoint summaries list 6.0 and 3.0 participants in the respective arms. The record does not explain this difference or describe missing-data handling. That limits conclusions about the completeness and representativeness of the posted outcome data.
The appropriate reading is narrow: the registry reports descriptive changes and their variability in the endpoint summaries for each arm. It does not report enough information to make a precise comparative or causal conclusion from the primary outcome. The results should be understood in that context rather than as proof of benefit, lack of benefit, or equivalence.
14. Why This Trial Matters Statistically
This trial is statistically instructive because a randomized comparison can still yield limited evidence when the posted outcome information is sparse. The record identifies a clinically interpretable task, the time taken to walk 10 meters, and presents mean changes and standard deviations by arm. Yet those descriptive statistics do not replace a direct comparison with an uncertainty estimate.
The difference between an arm summary and a treatment-effect estimate is central. A mean in each arm describes what was recorded among participants contributing to that summary. An analysis comparing the randomized groups is needed to quantify how the outcomes differ, and an interval is needed to show the precision of that estimate. Neither a comparative estimate nor its interval is reported in the registry record. This distinction helps prevent readers from treating separate group averages as a definitive test of treatment effect.
The record also makes clear why participant counts matter. The primary endpoint summaries list 6.0 and 3.0 participants, while enrollment is listed as 12. Without an account of participant flow or missing outcomes, readers cannot tell why these numbers differ or how the analysis population was defined. Such reporting is important because incomplete data can affect both the precision and interpretation of a comparison.
Finally, a continuous endpoint measured in seconds requires a clear definition of its change score. Here the registry identifies the assessment times but does not state the subtraction convention. The sign of a change is therefore not self-interpreting. Transparent endpoint definitions, analysis methods, and uncertainty reporting are essential complements to an outcome mean, especially in small studies.
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 primary limitation for interpreting this trial is that the registry posts arm-level descriptive statistics but no inferential comparison for the primary outcome. There is no reported treatment contrast, confidence interval, or p-value. The posted means and standard deviations therefore do not quantify the precision of a comparative effect.
The endpoint is described as change in over-ground gait speed, but its unit is seconds and the assessment is the time taken to walk 10 meters. The registry record does not state the change-score formula or its subtraction direction. This makes the sign of the posted mean changes ambiguous. A clinically meaningful-change threshold is also not reported.
The outcome summaries include 6.0 participants in the robotic exoskeleton training arm and 3.0 in conventional physical therapy, compared with a recorded enrollment of 12. Reasons for the different counts, participant flow, missing-data handling, and analysis-population definitions are not reported in the registry record. The small endpoint counts also mean that individual observations may have substantial influence on group summaries, while the posted standard deviations show variation within the arms.
Details needed to assess trial conduct and analysis are limited. The identity of the masked party, randomization procedures, intervention schedule, adherence, statistical model, covariate adjustment, multiplicity approach, and sample-size justification are not reported in the registry record. The intervention descriptions are truncated. Although secondary outcomes are listed, numerical results for them are not included in the registry record for this analysis.
Safety information is similarly limited to posted event counts. Follow-up duration for adverse events and a comparative safety analysis are not reported. These limitations do not establish that unreported procedures or analyses were absent; they indicate that the registry record does not provide those details for evaluation.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Improving Gait Performance in Individuals With Spinal Cord Injuries: an Intervention Using Robotic Exoskeletons — NCT03011099.
- Registry record last updated 2018-04-03; status last verified 2018-03.
- Results first posted 2018-04-03.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
19. Record Summary
NCT03011099 is a completed randomized, single-masked, parallel-group study sponsored by The University of Texas Health Science Center, Houston. It compared robotic exoskeleton training with conventional physical therapy in individuals with spinal cord injuries. The registry lists enrollment of 12 and a study start of 2014-01, with primary completion in 2016-03.
The primary endpoint was change in the time taken to walk 10 meters, assessed at baseline and week 4. The registry posts arm-level means and standard deviations, with endpoint counts of 6.0 for robotic exoskeleton training and 3.0 for conventional physical therapy. It does not report the change-score direction or an inferential comparison for the primary outcome. Five secondary outcomes are listed, covering gait characteristics, lower-extremity strength, and user feedback.
The registry reports no serious adverse events or deaths in either arm, and lists one other adverse event in the robotic exoskeleton arm and none in the conventional physical therapy arm. Interpretation is limited by the absence of a reported primary contrast, uncertainty interval, participant-flow explanation, and details of the analysis population and missing-data approach.