Comparison of Two High-intensity Gait Training Interventions on Contraversive Pushing Behaviors in Individuals Poststroke: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparison of Two High-intensity Gait Training Interventions on Contraversive Pushing Behaviors in Individuals Poststroke (NCT04550039), a non-randomized na study. The primary outcome is Change in Scale for Contraversive Pushing.

2. Clinical Question

The study also planned to evaluate functional mobility, strength, balance, endurance, and therapist burden. The listed secondary outcomes include walking tests, balance and sitting tests, functional independence, quality indicators, manual muscle testing, and passive range of motion. Those outcomes may describe different aspects of rehabilitation, but their results are not reported in the registry record.

The main statistical question is therefore not answerable from the posted registry results: there is no reported estimate of the change in pushing behavior for either group, nor an estimate of the difference between groups. Even if group changes were available, the nonrandomized allocation would make interpretation as a causal intervention comparison difficult. Differences could reflect participant or care characteristics as well as the training approach.

3. Trial Design

This completed study is registered as a parallel-group, nonrandomized comparison of body-weight-supported treadmill training and EksoNR exoskeleton training. The registry lists no masking. It reports an enrollment of 14 and identifies Shirley Ryan AbilityLab as the sponsor. The study start is listed as 2021-01-01, primary completion as 2023-12-31, and status as completed. The phase is listed as NA.

Both intervention descriptions specify a prescribed gait training program for at least three weeks or until participants discharge. The primary outcome timeframe is baseline and 3 weeks. The registry does not explain whether all participants were expected to complete the same number of sessions, how training intensity or other treatment components were standardized, or how discharge before the three-week assessment was handled.

Because allocation was nonrandomized, the interventions were not assigned by chance according to the registry record. The particular allocation procedure and the characteristics used to determine which intervention a participant received are not reported in the registry record. As a result, the groups may have differed before treatment began. No masking was used, which means the registry does not describe blinded assessment or treatment delivery.

The registry record gives an overall enrollment but does not report the number enrolled in each arm for the primary outcome, the number completing follow-up, or the analysis populations. It lists 15 outcome measures and reports zero analyses. Numerical primary and secondary outcome results, baseline group summaries, and a statistical analysis plan are not reported in the registry record. Consequently, the design can be described, but its planned and completed statistical procedures cannot be reconstructed in detail.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Outcome performance data was collected at baseline and at 3 weeks. Average score for each group was calculated at week 0 and week 3. The change scores from baseline to week 3 are used for this analysis. They were calculating by subtracting the average score at baseline from the average score at week 3. Due to noticeable differences in outcome performance for both groups at baseline, percent change was used to provide a more accurate comparison of changes.

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 in Scale for Contraversive Pushing, measured over Baseline, 3 weeks.

The primary endpoint is change in the Scale for Contraversive Pushing, described in the registry as percent change in points and summarized using a mean and standard deviation. The scale is a three-item test of lateropulsion, also called contraversive pushing. It assesses a participant’s action or reaction when maintaining or changing position in sitting and standing.

Each item uses a three point ordinal scale. The registry states that a score >0 in each section indicates the presence of contraversive pushing behaviors. The total scale ranges from a minimum score of 0, indicating absence of pushing behaviors, to a maximum score of 6; lower scores are described as better. These features make the measure relevant to the study’s stated focus on pushing behavior during rehabilitation.

The outcome timeframe is baseline and 3 weeks. The registry describes using change scores based on average group scores at baseline and week 3, but the description of the calculation ends incompletely and does not fully explain how the change was calculated or analyzed. Participant-level score distributions, group means, standard deviations, and a between-group effect estimate are not reported in the registry record. Without those results, the direction and magnitude of the endpoint changes cannot be assessed.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical methods are interpretable only when linked to the outcome data, analysis population, and comparison being estimated. The registry record identifies the primary outcome as a continuous measure summarized by a mean and standard deviation, and describes change from baseline to a later assessment. It does not report the outcome values or the statistical test or model used. The explanations below are general statistical context, not a description of confirmed procedures for this study.

What does a mean change describe?

For a continuous measure, a mean change represents the average difference between measurements over time in the group being summarized. It is a compact description of the group’s overall movement, but it does not show whether all participants changed similarly. The standard deviation describes the spread of individual observations around the mean. A large spread can coexist with a small or large average change, so both summaries help characterize the data.

The registry says the analysis uses change scores derived from average Scale for Contraversive Pushing scores at baseline and week 3. Its description of the calculation is incomplete, so it is not clear from the record whether each participant’s own change was calculated and then averaged, or whether the difference between group averages was used. These approaches need not produce the same uncertainty estimate. The exact analysis method is not reported in the registry record.

What would a confidence interval add?

In general, a confidence interval expresses uncertainty around an estimated effect. It helps readers assess which effect sizes are reasonably compatible with the data and the statistical assumptions. The interval’s width reflects precision: estimates based on more information and less variability often have narrower intervals, all else equal. A confidence interval does not remove bias from nonrandom allocation, and it cannot correct for unmeasured differences between groups.

No confidence interval for the primary outcome is reported in the registry record. No p-value or other inferential result is reported either. As a result, readers cannot evaluate statistical compatibility with a null difference, or distinguish a precise estimate from an uncertain one. General statistical principles cannot fill in those missing study results.

Why does nonrandom allocation matter?

In a randomized study, chance assignment is intended to reduce systematic differences between groups on average. In a nonrandomized study, the reason a participant receives one intervention rather than another may be connected to prognosis or outcome. For example, factors affecting treatment selection can also affect later mobility or pushing behavior. This is a general concern, not a claim that a particular imbalance occurred in this study. The allocation procedure and baseline characteristics are not reported in the registry record, so the possible influence of confounding cannot be assessed.

Statistical adjustment can sometimes account for measured baseline differences, but it depends on knowing which variables were collected and how they were modeled. The registry does not report an adjustment strategy or an analysis model. It is therefore not appropriate to describe the comparison as adjusted, or to assume that any potential confounding was controlled.

How should planning illustrations be read?

General sample-size calculations ask how many participants would be needed under stated assumptions to detect an effect with a chosen level of statistical power. The record’s planning assumptions describe a two-sided significance level of 0.05, equal allocation between arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. Such calculations are planning illustrations based on an effect estimate; they are not evidence that the trial was designed with those assumptions, and they do not show the probability that the completed study detected a true effect.

In particular, a planning calculation based on an observed effect should not be called observed power. Sample-size illustrations are not a substitute for reporting the effect estimate and its confidence interval. The registry’s own sample-size justification, if any, is not reported in the registry record. Numerical planning outputs are not stated in the record details, so no sample-size target beyond the reported enrollment can be described.

What can be learned from the posted analysis status?

The registry record reports zero analyses. This is distinct from the number of participants enrolled and does not establish whether analyses were performed outside the registry. It means that the registry record does not present an analysis result to interpret. The method actually used for the primary outcome, any sensitivity analysis, and any handling of repeated measurements are not reported in the registry record. Accordingly, methodological explanations should be read as general guidance rather than as a reconstruction of the investigators’ analysis.

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("Body-weight-supported Treadmill Training", "EksoNR Exoskeleton")
n   <- c(3, 3)
m   <- c(-47.01, -59.72)
sp  <- c(47.65, 8.67)
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  Body-weight-supported Treadmill Training  3  -47.010  47.650  -165.379  71.359
CARM  EksoNR Exoskeleton  3  -59.720  8.670  -81.257  -38.183

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary outcome and multiple secondary outcome measures, including walking tests, balance and sitting measures, functional independence, quality indicators, manual muscle testing, and passive range of motion. It also states an intention to evaluate endurance and therapist burden. Testing many outcomes can create a multiplicity issue: when several hypotheses are assessed, the chance of at least one apparently noteworthy result can increase. General approaches include identifying a primary outcome in advance, adjusting for multiple testing, or interpreting secondary findings as exploratory. The registry record does not report a multiplicity strategy or a statistical analysis plan, so it is not possible to determine how this issue was handled.

Missing-data handling is also important because the interventions were planned to continue for at least three weeks or until discharge, while the primary outcome timeframe includes a three-week assessment. A participant discharged before that assessment might have a different follow-up pattern from a participant who remained in the setting. The registry does not report how early discharge, missed assessments, or incomplete outcome measures were handled, or how many participants had primary outcome data at each time point.

The analysis populations are not reported in the registry record. It is not stated whether analyses included all enrolled participants, only those who completed training, or another defined group. The record reports enrollment of 14, but does not provide arm-specific numbers for the primary outcome or follow-up completeness. These omissions prevent assessment of attrition, missingness, or whether the analyzed participants differed from those enrolled.

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 reports serious adverse event figures separately for the two intervention groups. For Body-weight-supported Treadmill Training, 3 participants are listed as at risk, with 0 serious adverse events, 0 other adverse events, and 0 deaths. For EksoNR Exoskeleton, 3 participants are listed as at risk, with 0 serious adverse events, 0 other adverse events, and 0 deaths. No serious adverse event terms are listed.

These counts describe the safety entries in the registry record; they should not be read as a complete account of every possible safety outcome. The at-risk counts are not the same as the overall enrollment of 14, and the record does not explain why only 3 participants per group appear in these entries. The adverse-event observation period, event-collection procedures, and definitions used for these entries are not reported in the registry record.

A count of zero is a report of no events recorded in the stated group entry, not proof that an event could not occur. With the information available, the counts do not establish a comparative safety profile or support conclusions about risk beyond the reported entries.

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

The registry record establishes that a completed, nonrandomized study compared two high-intensity gait training interventions in an acute inpatient rehabilitation setting, with change in the Scale for Contraversive Pushing identified as the primary outcome. It also describes a group of secondary outcome domains and reports safety counts for the two intervention groups.

It does not report the numerical primary outcome results. The size and direction of change in either arm, the between-group difference, and the precision of any estimate are therefore unknown from the registry record. The record also does not report numerical results for the listed secondary outcomes. No conclusion that one intervention was better, worse, or equivalent to the other can be drawn from these posted results.

The nonrandomized design limits causal interpretation independently of the missing numerical results. If outcome differences were available, they could reflect intervention effects, baseline differences, treatment selection, or other factors. The allocation method, baseline characteristics, and any adjustment for confounding are not reported in the registry record. Thus, even a numerical contrast would need to be interpreted with attention to how participants entered each group.

The reported enrollment of 14 and the safety entries do not fill the outcome-data gap. The safety record lists zero serious adverse events and zero other adverse events among 3 participants at risk in each group, but it does not explain the limited at-risk counts or observation period. These figures should be read as registry entries, not as a broad estimate of safety.

In general, a treatment-effect estimate is most informative when accompanied by an uncertainty interval, clear analysis methods, and details about missing observations and group comparability. Those elements are not reported for the primary outcome in the registry record. The appropriate interpretation is therefore limited: the study question and design are documented, while the posted results do not permit a quantitative assessment of comparative effectiveness.

14. Why This Trial Matters Statistically

This study raises a useful statistical question about measuring change in a rehabilitation setting. The primary endpoint targets contraversive pushing behavior, while the listed secondary measures cover walking, balance, sitting, strength, functional independence, and range of motion. These outcomes reflect different dimensions of recovery and are not interchangeable. A statistical analysis should make clear which outcome is primary and how other outcomes are interpreted, particularly when many measures are considered.

The nonrandomized design also makes the distinction between description and causal comparison important. A change within each group describes what happened among its measured participants; it does not by itself show that the intervention caused the change. Comparing group changes is more informative about differences between groups, but it remains vulnerable to confounding when treatment assignment is not random. Baseline data and a transparent analysis plan are important for evaluating that risk.

Precision matters alongside the estimated direction of change. Means and standard deviations can describe group outcomes, while confidence intervals can communicate uncertainty around comparisons. The registry record does not report those primary outcome results, so readers cannot assess either magnitude or precision. This makes the distinction between a completed study and a reported, interpretable result especially salient.

The record also illustrates why follow-up timing and missing-data rules matter when a treatment program is described as lasting at least three weeks or until discharge. Clear reporting of who was assessed, when they were assessed, and how incomplete follow-up was handled would help readers judge how representative the analyzed data are. Those details are not reported in the registry record.

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.

Several trial-specific limitations restrict interpretation of the registry record. First, allocation was nonrandomized, and the record does not explain how participants entered either intervention group. Without that information and baseline group characteristics, the possibility of systematic differences between groups cannot be assessed.

Second, the registry does not report numerical results for the primary Scale for Contraversive Pushing outcome or the listed secondary outcomes. It therefore gives no effect estimates, standard deviations, confidence intervals, or inferential results for evaluating change or comparing interventions. The record reports zero analyses.

Third, the primary outcome description is incomplete about the change calculation. It notes group averages at baseline and week 3 and describes a change from baseline, but does not fully specify the calculation or analysis. The registry also does not report whether individual-level changes were used, whether baseline values were adjusted for, or how repeated measurements were modeled.

Fourth, follow-up and analysis details are absent. The interventions were to continue for at least three weeks or until discharge, but handling of discharge before the three-week assessment, missing observations, and analysis populations is not reported in the registry record. Arm-specific enrollment and outcome completion are also not reported.

Finally, the registry does not report a statistical analysis plan, a multiplicity approach for the numerous outcomes, or the basis for the study’s sample-size design. The safety entries list 3 participants at risk in each arm, but the relationship of those entries to the overall enrollment and the adverse-event observation period is not explained. These limitations concern the completeness of the registry record and should not be mistaken for proof that the information is unavailable elsewhere.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Comparison of Two High-intensity Gait Training Interventions on Contraversive Pushing Behaviors in Individuals Poststroke — NCT04550039.
• Registry record last updated 2024-03-01; status last verified 2024-02.
• Results first posted 2024-03-01.
• 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

NCT04550039 is a completed, nonrandomized, parallel-group study sponsored by Shirley Ryan AbilityLab. It compares body-weight-supported treadmill training with EksoNR exoskeleton training for individuals poststroke with contraversive pushing behaviors in acute inpatient rehabilitation. The registry reports enrollment of 14, no masking, and a program lasting at least three weeks or until discharge.

The primary outcome is change in the Scale for Contraversive Pushing from baseline to the three-week timeframe. The registry lists 15 outcome measures and reports zero analyses. Numerical primary and secondary outcome results, analysis methods, missing-data handling, and analysis populations are not reported in the registry record.

Safety entries list 0 serious adverse events, 0 other adverse events, and 0 deaths among 3 participants at risk in each group. The record does not explain the at-risk counts or observation period. The design and outcome question are documented, but the posted results do not permit assessment of the comparative treatment effect.