SCI Acute Intermittent Hypoxia and Non-Invasive Spinal Stimulation Combined With Gait Training: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for SCI Acute Intermittent Hypoxia and Non-Invasive Spinal Stimulation Combined With Gait Training (NCT03922802), a randomized na study. The primary outcome is Change in 6 Minute Walk Test.

2. Clinical Question

3. Trial Design

NCT03922802 is recorded as a randomized, single-blind, sham-controlled crossover trial. In a crossover design, participants receive more than one condition in sequence, so comparisons can be made within the same participant as well as across treatment conditions. This can help account for stable differences between participants, but it also makes the timing and sequence of conditions important to interpretation.

The registry lists three conditions: AIH with non-invasive spinal cord stimulation and gait training; sham AIH with active stimulation and gait training; and sham AIH with sham stimulation and gait training. Thus, the listed conditions distinguish the combined active approach from conditions that retain gait training while changing whether AIH and stimulation are active or sham. The record describes the trial as single blind, but does not report who was blinded or how blinding was maintained.

The listed enrollment is 14, and the trial status is completed. The primary-outcome description says that 10 participants completed and were analyzed, with 1 removed from analysis as an outlier, leaving 9 in the analysis. These enrollment and analysis figures describe different stages; the record does not explain the status or outcome data for each enrolled participant.

The record identifies assessments at baseline, intervention session 5, 3 days after intervention sessions were complete, and 1 week after sessions were complete, within each arm. It does not report the sequence assigned to participants, the duration between conditions, whether a washout period was used, or the number and schedule of all intervention sessions. These details matter in crossover studies because treatment effects may carry over, and changes over time may be confused with condition effects if sequence and timing are not adequately addressed.

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: All participants completed all arms of the study. The order of arm completion was randomly assigned using a block style approach.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: 10 participants completed and analyzed. 1 was dropped from analysis as an outlier (difference of greater than 1.5\*IQR (interquartile range) for all gait outcomes) = 9 participants in analysis total.

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: 10 participants completed and analyzed. 1 was dropped from analysis as an outlier (difference of greater than 1.5\*IQR (interquartile range) for all gait outcomes) = 9 participants in analysis total.

6. Primary Endpoint

The registry lists the primary outcome as Change in 6 Minute Walk Test, measured over Baseline (initial visit within an arm), Intervention session 5 within each arm (T5), 3 days after intervention sessions complete within each arm (POST), 1 week after intervention sessions are complete within each arm (1-week follow up).

The primary endpoint is change in the 6MWT, expressed as distance in meters. The test measures how far a participant can walk indoors on a flat, hard surface during 6 minutes, using assistive devices if needed. Participants may stop and rest, but the timer continues; if a participant cannot finish the test, the time and reason for stopping are recorded.

The 6MWT is described in the registry as a reliable and valid assessment of functional exercise capacity. It is self-paced and used as a sub-maximal assessment of aerobic capacity and endurance. Because it records distance walked during a set period, it captures a functional walking task rather than measuring only a single movement or laboratory capacity. Its use of the participant's own pace also allows assessment of functional capacity while permitting rests.

Measurements were planned at baseline before each condition, at intervention session 5, 3 days after that condition's intervention sessions were complete, and at a 1-week follow-up. Those repeated measurements can describe changes during and after each condition. The registry record does not report the numerical values at these time points or the treatment-effect estimates comparing conditions, so the endpoint's observed pattern and precision cannot be summarized from the posted results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies a randomized crossover design and a continuous primary measure summarized as a mean with a 95% confidence interval. It does not report the statistical analysis plan or the specific model used for the primary outcome. The general concepts below explain how these design and summary choices are usually interpreted; they are not claims about analyses performed in this trial.

What does randomization contribute?

Randomization assigns the order or allocation of conditions according to a chance process rather than a participant or investigator choice. In a crossover study, the assigned sequence is important because each participant experiences multiple conditions. Random allocation can reduce systematic differences in sequence assignment, but it does not, by itself, prevent carryover effects, missing measurements, or bias from departures from the assigned schedule. The registry record says the trial was randomized but does not report its sequence-generation method, allocation concealment, or sequence assignments.

How does a crossover comparison work?

In general, crossover analyses can compare measurements within each participant, since the same person contributes data under more than one condition. This can account for stable personal characteristics that might otherwise make participants' outcomes differ. A valid interpretation also depends on whether outcomes are affected by the order of conditions, by changes over time, or by lingering effects from a preceding condition. The registry record does not report the crossover sequence, washout arrangements, or an analysis accounting for period or carryover effects. It is therefore not possible to determine from the record how these issues were addressed.

What does a mean and confidence interval convey?

A mean describes the arithmetic average of the observed values in the analysis population. It does not show how values are distributed among participants, and can be influenced by extreme observations. A 95% confidence interval is an uncertainty interval around an estimate under the assumptions of the analysis. In general, a narrower interval indicates greater precision and a wider interval indicates less precision. The interval is not a guarantee that the true effect lies within its bounds. The registry specifies this summary format for the primary outcome but does not report the numerical means or intervals for treatment comparisons.

Why does the outlier exclusion matter?

The primary-outcome description says one participant was dropped from analysis as an outlier, using a rule based on a difference greater than 1.5 times the interquartile range for all gait outcomes. The interquartile range is the span between the middle half of ordered observations; it is commonly used to describe spread and identify values far from the central observations. Excluding an extreme observation can make a summary less sensitive to that value, but can also change the treatment estimate and its uncertainty. Such exclusion is most interpretable when the rule is specified in advance and when results are checked with and without the excluded observations. Whether the rule was prespecified and whether sensitivity analyses were performed are not reported in the registry record.

What can be inferred from the reported analysis status?

The registry record lists 0 analyses and identifies the primary outcome as having several categories or time points that were not reduced to one comparison. This means no single numerical comparison can be used to describe the primary result from the posted information. It does not establish that no analyses were conducted outside the record. The statistical plan, model specification, handling of repeated observations, treatment contrast, and inferential results are not reported in the registry record.

The record also describes planning illustrations based on a fraction of the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. Such illustrations are prospective planning calculations, not measures of the trial's observed power and not evidence that the trial was designed around those assumptions. The sample-size justification for the actual trial is not reported in the registry record. No numerical planning values are stated in the information available, so their implications cannot be assessed.

These statistical distinctions are important: design features, descriptive summaries, uncertainty intervals, and hypothesis tests answer related but different questions. A trial can be randomized and have a defined endpoint without the registry reporting the estimate required to quantify a treatment contrast. Here, the information describes the design and measurement schedule, but not a numerical result or analysis sufficient to evaluate the primary comparison.

10. Multiplicity, Missing Data and Analysis Populations

The primary outcome is measured at several time points within each condition: baseline, intervention session 5, 3 days after the intervention sessions are complete, and 1-week follow-up. The registry record notes that the primary outcome has several categories or time points and was not reduced to one comparison. It does not identify which time point or contrast was intended as the principal inferential test. It also does not report a multiplicity adjustment, a hierarchy of outcomes, or a prespecified method for interpreting multiple time-point comparisons. These details are not reported in the registry record.

Multiplicity is a general statistical concern when several outcomes, contrasts, or time points are tested. As the number of opportunities to declare a difference increases, the chance of at least one apparently notable result can increase if tests are interpreted independently. Approaches such as prespecifying a primary contrast, controlling error rates, or using a planned hierarchy can help clarify interpretation. The registry information does not establish whether such approaches were used in the trial.

The primary-outcome description reports 10 participants completing and analyzed, with 1 excluded as an outlier, leaving 9 in the analysis. It does not report how many participants had missing observations at individual time points, how missing measurements were handled, or whether the analysis included all randomized participants. The reason for the difference between enrollment and the reported completed analysis is not detailed. The statistical plan and the analysis population definitions are not reported in the registry record.

These gaps limit assessment of whether results could be affected by selective availability of measurements, outcome selection, or multiple testing. They do not prove that such problems occurred. Without the planned analysis and results at each time point, readers cannot determine whether the reported data represent all prespecified comparisons or how uncertainty was handled across them.

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.

The registry lists 10 participants at risk in each of the three conditions. For each condition, the posted counts show 0 serious adverse events, 0 other adverse events, and 0 deaths. These are the event counts recorded for the groups; they are not estimates of a comparative safety effect.

Zero events in a group means that the registry record lists no events of that category for the participants counted at risk. It does not prove that an intervention cannot cause an adverse event, and it does not establish that the conditions have identical safety profiles. Safety interpretation depends on the completeness of event collection, the observation period, definitions of events, and how exposure is counted. The registry record does not report the event-monitoring period, detailed collection procedures, or additional context for these counts.

The three conditions share gait training, while the AIH and stimulation components differ. The counts therefore should be read as group-specific registry entries, not as a direct estimate of the effect of one component. The record does not list serious adverse event terms. These figures describe what was posted and should not be taken as medical advice or as evidence of safety beyond the trial's recorded observations.

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

The record is useful for understanding the trial's question, design, primary measure, and reported analysis population. It describes a randomized, single-blind crossover comparison of AIH and non-invasive spinal stimulation combined with gait training against two sham configurations that also include gait training. It identifies change in 6MWT distance as the primary outcome and lists repeated measurement points during and after each condition.

What the record does not establish is the magnitude or direction of a treatment contrast. Numerical outcome means, treatment-effect estimates, confidence intervals, and p-values are not reported in the available information. Without those estimates, the data cannot be used to judge how large a difference was observed or how precise it was. The completed status alone does not supply the missing efficacy result.

The reported primary-outcome analysis population is small: 10 participants completed and were analyzed, and 1 was excluded as an outlier, leaving 9 participants in the analysis. A small analysis population generally limits precision and makes estimates more sensitive to individual observations. This is a general statistical limitation, not a numerical conclusion about the unreported treatment effect. The registry record also does not report a sensitivity analysis addressing the excluded participant.

The posted safety counts record no serious adverse events, other adverse events, or deaths among 10 participants at risk in each condition. These counts are descriptive and do not prove an absence of risk or establish comparative safety. The event-monitoring context is not reported.

Accordingly, the appropriate reading is bounded: the record documents a completed trial and its planned measurement framework, but does not provide the numerical primary comparison needed to interpret efficacy. It should not be read as evidence that the conditions are equal, or that a particular condition improves or fails to improve walking. Those conclusions require results and analysis details not reported in the registry record.

13. Why This Trial Matters Statistically

This trial is statistically notable because it combines a crossover design with repeated functional measurements and several treatment conditions. A crossover can make within-participant comparisons informative by reducing the influence of stable differences between people. At the same time, it makes sequence, timing, and possible carryover important. The registry identifies the design but does not report the details needed to assess those features.

The endpoint is also measured at multiple points within each condition, from baseline through follow-up. This creates a distinction between describing change over time and estimating a treatment contrast at a specified time. A participant's walking distance may vary across the measurement schedule, and each time point can answer a different question about timing and persistence. The record notes that the primary outcome was not reduced to a single comparison, leaving the inferential target unclear.

The analysis population calls attention to the influence of outlier handling. One participant was excluded using a stated rule, leaving 9 participants in the analysis. In small datasets, individual observations can have substantial influence on estimates; transparent reporting of prespecified exclusion criteria and sensitivity analyses helps readers assess robustness. Whether those checks were conducted is not reported in the registry record.

Finally, this record illustrates the difference between having a registered design and having interpretable posted results. The design and endpoint are described, but the numerical outcome summaries and treatment comparisons are not reported. Without estimates and uncertainty intervals, readers cannot assess effect size or precision. This is a limitation of the information in the registry record, not proof about what analyses investigators may have performed elsewhere.

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 limitation for interpreting this trial is the absence of numerical primary-outcome results and treatment comparisons in the registry record. Although the record specifies a mean and 95% confidence interval, it does not report the estimates or intervals. It also lists 0 analyses and indicates that multiple time points were not reduced to one comparison.

The population description reports 10 participants completing and analyzed, with 1 excluded as an outlier, leaving 9 participants in the analysis. The record does not explain the status of all enrolled participants, provide participant-level data, or report a sensitivity analysis including the excluded participant. The basis for the outlier rule is stated, but whether that rule was prespecified is not reported in the registry record.

The crossover features require details not available in the record: assigned sequences, timing between conditions, washout arrangements, and any method for accounting for period or carryover effects. The record also does not identify who was blinded or describe how blinding was maintained. These omissions prevent a detailed appraisal of the design's implementation.

The repeated endpoint assessments raise questions about the primary contrast and multiplicity. The record does not report a hierarchy of outcomes, a plan for handling the multiple time points, or a statistical analysis plan. It also does not report how missing observations were handled or define the analysis populations beyond the primary-outcome description.

Finally, the adverse-event counts lack reported details about collection methods and observation periods. The counts can be described as posted, but their broader safety implications cannot be assessed from these figures alone.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: SCI Acute Intermittent Hypoxia and Non-Invasive Spinal Stimulation Combined With Gait Training — NCT03922802.
• Registry record last updated 2026-01-13; status last verified 2025-12.
• Results first posted 2026-01-13.
• 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

NCT03922802, titled “SCI Acute Intermittent Hypoxia and Non-Invasive Spinal Stimulation Combined With Gait Training,” is recorded as a completed randomized, single-blind crossover trial sponsored by Shirley Ryan AbilityLab. Listed enrollment is 14. The trial concerns spinal cord injury and compares AIH, non-invasive tSCS, and gait training with sham configurations that retain gait training.

The primary outcome is change in the 6MWT, measured in meters at baseline, intervention session 5, 3 days after intervention completion, and 1-week follow-up within each condition. The outcome description reports 10 participants completing and analyzed, with 1 excluded as an outlier, leaving 9 in the analysis. Secondary outcomes listed are change in fast-velocity 10 Meter Walk Test and Timed Up and Go Test.

The registry record does not report numerical primary treatment-effect estimates, confidence intervals, or p-values. It lists 0 serious adverse events, 0 other adverse events, and 0 deaths in each condition among 10 participants at risk.