The Sensorimotor Locus of Balance Control in Elderly Gait: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Sensorimotor Locus of Balance Control in Elderly Gait (NCT03341728), a randomized na study. The primary outcome is Change in Postural Sway After 10 Min of Walking.

2. Clinical Question

The study addresses a question about the immediate measurement of balance-related movement under two walking conditions. It does not directly test falls, long-term balance, or the prevention of future falls. The registry’s background rationale describes interest in visual contributions to balance and in motor responses to sensory perturbations, but the posted primary endpoint is postural sway after walking, not a clinical fall outcome.

The registry reports a primary outcome comparison and one posted secondary comparison. It names additional secondary outcomes concerning cognitive-motor interference accuracy and response time, but analyses for those outcomes are not reported in the registry record. Consequently, the primary question can be described from the posted sway results, while the full range of questions raised by the study rationale cannot be answered from the posted results alone.

3. Trial Design

This completed study is identified as NCT03341728 and is titled “The Sensorimotor Locus of Balance Control in Elderly Gait.” The sponsor is the University of North Carolina, Chapel Hill. The registry classifies the study as randomized, unmasked, and crossover, with enrollment of 14. Its phase is listed as not applicable. The study start date is 2017-10-30 and the primary completion date is 2018-07-25.

The two conditions are optical-flow perturbations and normal walking. The registry describes one sequence as intervention followed by control and the other as control followed by intervention. In the first sequence, older adults walk during optical-flow perturbations; in the second, they walk normally without those perturbations. Because this is a crossover design, each sequence represents an order of exposure to the conditions rather than a simple description of two independent parallel treatment groups.

The registry reports group summaries with 14 observations for each condition. It does not report the timing or duration of any washout, whether sequence or period effects were assessed, or the exact procedure for handling within-participant dependence in the analysis. Those details matter when interpreting a crossover comparison, because observations from the same participant are related and the order of conditions may be relevant.

The primary outcome is change in postural sway after ten minutes of walking, with assessments described at baseline and ten minutes. The registry lists six outcome measures and three secondary outcomes, but no reported analyses are listed. The status is completed; that status alone does not establish that every planned measurement or analysis is posted. The original design’s sample-size justification and detailed statistical analysis plan are not reported in the registry record.

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.

Session 1

Washout (1 Week)

Session 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.

6. Primary Endpoint

The registry lists the primary outcome as Change in Postural Sway After 10 Min of Walking, measured over Baseline, 10 minutes.

The primary endpoint is change in postural sway after ten minutes of walking. The registry defines the measure as the magnitude of side-to-side sway and gives centimeters as the unit. It describes measurements at baseline and ten minutes, with the endpoint represented as a mean and variability represented by a standard deviation.

Postural sway is a movement measure relevant to the study’s focus on balance control during gait. Comparing change across optical-flow perturbation and normal walking conditions can describe whether the measured sway shifts differently during the two exposures. In this study, the endpoint therefore concerns a balance-related response over the specified walking interval, rather than a direct count of falls or a measure of long-term function.

The registry does not provide a threshold for a meaningful change in sway or explain how a given change should be interpreted in relation to daily function. Without such context, the numerical result can describe the direction, size, and precision of the measured difference, but it cannot by itself establish the practical importance of that difference. Nor does a postural-sway result establish a change in fall risk: falls are part of the study’s stated background, but they are not the posted primary endpoint.

7. Results Posted in the Registry

• Intervention: mean -2.39 versus -0.01 on control; difference -2.38 (95% CI -3.03 to -1.73), standardized effect d = -2.70; z-test p = <0.001.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change in Margin of Stability Variability After 10 Min of Walking

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Intervention: mean -1.02 versus 0.16 on control; difference -1.18 (95% CI -1.72 to -0.64), standardized effect d = -1.63; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Intervention than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary result concerns change in margin-of-stability variability after ten minutes of walking. Its comparison has the same negative direction as the primary postural-sway comparison, and the posted unadjusted p-value is <0.001. Taken together, the two posted balance-related results suggest a consistent direction across the primary measure and this secondary measure. They remain distinct outcomes, however, and should not be treated as interchangeable measurements of one construct.

The registry also names cognitive-motor interference accuracy and response time as secondary outcomes. Their analyses are not reported in the registry record. This means the apparent pattern across the available results cannot be assumed to describe every secondary question named for the study. Nor can the available secondary result establish that changes in balance-related measures were accompanied by changes in cognitive-motor performance.

Secondary outcomes generally carry less evidential weight than a prespecified primary outcome when many outcomes or comparisons are examined. Each additional test creates another opportunity for a chance finding, particularly if results are selected for emphasis after analysis. The registry does not report whether the secondary outcomes were prespecified in a statistical hierarchy, whether multiplicity adjustments were planned, or how missing secondary results should be interpreted. The posted unadjusted result should therefore be read as a secondary finding rather than as a confirmatory conclusion.

The secondary measure can offer supporting context because it relates to balance variability and is measured over the same walking interval. But it comes from the same study, population, and general experimental setting as the primary endpoint. Agreement in direction may make the overall pattern more coherent, while not providing an independent replication or resolving the limitations of the design and reporting.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Change in Cognitive-motor Interference Accuracy After 10 Min of Walking
• Change in Cognitive-motor Interference Response Time After 10 Min of Walking
• Change in Margin of Stability Variability After 10 Min of Walking

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is a continuous measurement summarized by a mean and standard deviation. A mean gives the average observed change in each condition. A standard deviation describes how dispersed individual measurements are around that group mean; it is not the uncertainty interval for the mean and does not indicate the range containing every participant’s value.

What does the mean difference represent?

The posted mean difference compares the intervention mean with the normal-walking mean, using intervention minus control. A negative value therefore means the intervention-condition mean is numerically lower under that subtraction convention. The estimate is expressed in the endpoint’s original unit, centimeters. This makes it possible to interpret the contrast on the measurement scale, but it does not by itself establish the clinical importance of the difference.

The groups’ posted mean changes and their standard deviations describe the condition-level summaries. A comparison between those summaries can be analyzed in different ways. In a crossover design, a participant may contribute measurements under both conditions, so an analysis can use the pairing and within-participant differences. A method that treats the condition observations as independent would make different assumptions. The registry does not report the analysis details needed to determine exactly how the posted comparison accounts for pairing, period, or sequence.

What does the confidence interval describe?

A confidence interval expresses uncertainty around an estimated contrast under the statistical method and assumptions used to construct it. A narrower interval indicates greater precision on the scale of the estimate; a wider interval indicates more uncertainty. The primary interval extends from -3.034 to -1.726 cm. Because both endpoints are below zero, the interval is consistent with a negative mean difference under the posted comparison. A confidence interval does not provide the probability that the true effect lies within that specific interval, and it does not indicate whether the result is practically meaningful.

How should the p-value be read?

The primary unadjusted p-value is <0.001. In general, a p-value describes how incompatible the observed data, or more extreme data, are with a specified null hypothesis under the model used for the test. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of an effect. The registry labels this value unadjusted; the multiplicity strategy and the detailed test specification are not reported in the registry record.

The registry also posts a standardized effect-size statistic for the primary comparison. Standardized measures express a contrast relative to variability rather than in centimeters. They can aid comparisons across measurement scales, but they depend on how variability is defined and may be less directly interpretable than the difference in the endpoint’s original unit. For the clinical question, the mean difference and its interval keep the contrast on the postural-sway scale.

Why does crossover structure matter?

In a crossover study, the same participant can experience both conditions in an assigned order. This can make the within-person contrast informative because stable characteristics of a participant are shared across conditions. At the same time, measurements within a participant are not independent. An analysis should account for that dependence, and interpretation may also depend on possible period effects, order effects, or carryover from an earlier condition. The registry identifies randomized crossover allocation but does not report whether these issues were assessed or the model used to address them.

The registry’s posted comparison includes a mean difference, confidence limits, and unadjusted p-value. Those statistics communicate the reported estimate and its inferential summary, but they do not reveal all analytical choices. In particular, the registry does not report missing-data handling, the analysis population, adjustment variables, or a multiplicity plan. These omissions limit a reader’s ability to reconstruct or fully assess the analysis.

What does variability add?

Standard deviations provide descriptive information about the spread of the observed measurements. If outcomes are more variable than a chosen model allows, a model that assumes less variability can produce uncertainty estimates that are too small. This general issue is called over-dispersion: the data vary more than expected under the model. The posted primary outcome is continuous and has standard deviations reported for both conditions; the registry does not identify over-dispersion as an issue or report a diagnostic assessment. The concept is relevant generally, but it should not be taken as evidence that over-dispersion occurred in this study.

The trial’s planning assumptions are distinct from the inferential analysis of the observed results. The registry’s planning illustration uses a two-sided significance level of 0.05, equal allocation, and a normal approximation, with target effects stated as fractions of the observed effect. It gives participants per arm, even though the actual study is described as a crossover. The original trial’s own sample-size assumptions and justification are not reported in the registry record, so the illustration should not be treated as a reconstruction of the trial design.

Finally, the posted analyses are not a complete account of all outcomes listed in the record. No reported analyses are listed, while the primary comparison and a secondary comparison are posted. The registry names cognitive-motor accuracy and response-time outcomes as well, but their results are not reported in the registry record. The distinction between outcomes named in a record and results available for interpretation is important when assessing how complete the evidence is.

11. Confidence Intervals and Precision

The primary estimate is a mean difference of -2.38 cm, with a confidence interval from -3.034 to -1.726 cm. This interval gives a range of values compatible with the posted analysis under its statistical assumptions. Its position entirely below zero indicates that the posted contrast is negative according to the intervention-minus-control convention. Its width also makes clear that the estimate is not known with infinite precision; the true contrast remains uncertain.

The primary group standard deviations are 1.1 cm for optical-flow perturbations and 0.59 cm for normal walking. These describe the observed dispersion, whereas the confidence interval describes uncertainty around the estimated difference. The two quantities answer different questions. A standard deviation should not be read as a confidence interval, and the interval should not be read as a range of individual participant responses.

The registry does not identify the confidence level, the precise inferential model, or the handling of the crossover pairing for the posted interval. Therefore, the interval is best interpreted as the sponsor-posted uncertainty summary rather than as an independently verified interval from a fully documented model. No discrepancy is flagged in the registry record, so the sponsor’s reported analysis is the operative result.

Statistical precision is not the same as clinical importance. The registry does not provide a threshold for a meaningful change in postural sway or relate the interval to future falls or daily function. The estimate and interval describe the measured endpoint and its statistical uncertainty; they cannot, on their own, establish what size of change would matter to participants or clinicians.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are expressed as participants per arm, not as total event counts. They are illustrations based on the observed effect, not the trial’s original design justification. The registry’s stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is set to the stated fraction of the observed effect. The original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column gives the illustrated number of participants per arm for a design with an 80% chance of detecting the chosen effect, if the planning assumptions hold and the effect is present. The 90% power column gives the corresponding illustration for a 90% chance. These are prospective planning concepts, not descriptions of the probability that the completed study’s finding is true.

For the primary outcome, the illustration lists 3 participants per arm for planning at 100% of the observed effect, 4 per arm at 75% of the observed effect, and 9 per arm at 50% of the observed effect for 80% power. The corresponding 90% power figures are 3, 6, and 12 participants per arm. For the secondary margin-of-stability measure, the corresponding 80% figures are 6, 11, and 24 participants per arm; the 90% figures are 8, 15, and 32 participants per arm.

The pattern illustrates a general principle: smaller effects are harder to detect and require more participants under otherwise comparable assumptions. These figures are not observed power and should not be used to infer the design adequacy of this crossover trial. In particular, the illustration assumes equal allocation between two arms and reports per-arm values, whereas the registry describes an intervention sequence crossover design.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -2.380; pooled standard deviation σ = 0.883.
2. n = 2 × (1.96 + 0.842)² × 0.883² / -2.380² = 2.2, rounded up to 3 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -2.38                      # observed difference in means
sigma <- 0.882638091178938                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(3341728)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 3 (80% power), 3 (90% power)
 75% of observed effect: n per arm = 4 (80% power), 6 (90% power)
 50% of observed effect: n per arm = 9 (80% power), 12 (90% power)
power.t.test: 3.4 
Simulated power at n = 3 per arm: 0.699 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures and identifies three secondary outcomes. It posts a primary result and one secondary comparison, while no reported analyses are listed. The registry does not report a multiplicity plan, such as a hierarchy among outcomes or a correction for testing multiple outcomes. Without that information, the unadjusted p-values should not be interpreted as if they necessarily account for all the outcomes or comparisons considered during the study.

Multiplicity matters because the chance of at least one statistically notable finding can increase when many hypotheses are tested. This is a general statistical concern, not evidence that the posted results are false. A prespecified primary endpoint can help focus interpretation, but the registry does not report the detailed analysis plan needed to establish the endpoint hierarchy, the planned tests, or how secondary findings were to be handled.

The cognitive-motor interference accuracy and response-time outcomes are named, but their results are not reported in the registry record. The record does not explain whether these results were unavailable, not analyzed, or omitted for another reason. Likewise, the registry does not report missing-data counts, reasons for missingness, or the method used to handle missing measurements. No assumptions about imputation, complete-case analysis, or participant exclusions can be made from the posted record.

The analysis population is also not reported in the registry record. Although enrollment and condition-level summaries are listed, the record does not say whether the posted comparison includes every randomized participant, every participant with a usable measurement, or a different analysis set. The registry does not report the statistical model, covariate adjustment, handling of crossover period or sequence, or any sensitivity analyses.

These omissions do not invalidate the reported numbers, but they limit independent assessment of their robustness and completeness. The posted estimates can be described as registry results, while conclusions about the analysis population, missingness, multiplicity control, and model adequacy remain unavailable.

15. 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 no serious adverse events among the 14 participants at risk in the intervention condition and no serious adverse events among the 14 participants at risk in the normal-walking condition. It also reports no other adverse events among those at risk in either condition, and no deaths in either condition. No serious adverse event terms are listed.

These figures should be read as the events recorded in the registry for the stated conditions and groups. Zero recorded events means that none were posted for those categories; it does not establish that the conditions are free of risk or that uncommon events can be ruled out. The record does not provide additional detail about event definitions, monitoring procedures, severity gradations for other events, or whether any events were considered related to a condition.

The safety figures are also limited by the study’s small enrollment and short walking exposure described for the outcomes. They do not provide a basis for characterizing safety in other populations, over longer periods, or under different exposure conditions. They are descriptive registry counts, not a general safety guarantee.

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

The posted result indicates that the mean change in measured postural sway differed between optical-flow perturbations and normal walking after ten minutes. The estimate is negative under the intervention-minus-control convention, and the confidence interval lies below zero. The unadjusted p-value is <0.001. These statistics describe the observed short-term contrast and its uncertainty under the analysis as posted.

They do not show whether a difference of this size is meaningful in everyday life. The registry does not give a threshold for clinically important postural-sway change, nor does it connect the measured change to the likelihood of falling. The study rationale discusses falls and balance, but a short-term sway measurement is not the same outcome as falls, injuries, independence, or quality of life.

The findings also do not establish what happens with longer exposure, whether any response lasts beyond the measured interval, or whether the result applies to older adults beyond the study’s enrolled group. The registry does not provide detailed eligibility criteria or broader population characteristics for evaluating generalizability. Its crossover design may help compare conditions within participants, but the analysis details needed to assess period, sequence, and carryover effects are not reported.

The posted secondary balance-related result is directionally consistent with the primary result, but it is not an independent confirmation. Additional named secondary outcomes lack posted analyses, and the multiplicity plan is not reported. The appropriate meaning is therefore bounded: the registry reports a statistically distinct short-term difference in a continuous balance-related measure under the study conditions, with important questions about clinical relevance, analysis detail, and broader applicability left unresolved.

17. Why This Trial Matters Statistically

This trial is statistically interesting because it combines randomized allocation with a crossover design and measures a continuous balance-related outcome under two walking conditions. In a crossover study, within-participant comparisons can be useful because each participant serves as a reference for their own response. That feature makes the handling of paired observations and condition order central to understanding the uncertainty around the result.

The registry posts a sizeable difference on the outcome’s original scale, an interval that excludes zero, and an unadjusted p-value below 0.001. These are informative summaries, but the registry does not report enough model detail to assess fully how the crossover structure was incorporated. The contrast between a clear numerical signal and limited analysis documentation is a useful reminder that statistical evidence depends both on the observed data and on transparent reporting of the method.

The study also illustrates why a statistical result and a clinical conclusion are not the same. Postural sway is relevant to balance, but a short-term measure does not directly establish a change in falls or daily function. The registry does not define a meaningful-change threshold, making the clinical interpretation less certain than the statistical direction of the estimate.

Finally, the planning illustrations show how required sample sizes change when the target effect is reduced, while not representing the trial’s original design calculation. The difference between a planning scenario and an observed analysis is important: the illustrations cannot substitute for the original rationale, which is not reported. Overall, the record offers a focused example of how to read effect estimates, uncertainty, outcome hierarchy, and incomplete statistical reporting together.

18. 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 principal limitation for interpretation is that the registry does not report the detailed statistical model. It identifies a randomized crossover design, but does not say how the analysis handled within-participant dependence, condition order, period effects, or possible carryover. Those issues can affect both the estimate and its uncertainty in a crossover study.

Enrollment is reported as 14, and the registry provides condition-level summaries. The record does not give detailed participant characteristics or full eligibility criteria, which limits assessment of how broadly the results may apply. It also does not report the original sample-size justification, so the planning illustrations should not be confused with the design rationale for this study.

The endpoint is a short-term change in postural sway after ten minutes of walking. The record does not report a threshold for a meaningful change or a direct measure of falls as the primary outcome. The relationship between the measured sway response and longer-term function or fall outcomes therefore remains unestablished by the posted primary result.

For the secondary outcomes, the registry names cognitive-motor accuracy and response time but does not report their analyses. The multiplicity plan, missing-data handling, analysis population, and sensitivity analyses are also not reported in the registry record. These gaps constrain evaluation of completeness and robustness. The adverse-event figures are descriptive and do not provide detailed event definitions or broader evidence about longer-term safety.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: The Sensorimotor Locus of Balance Control in Elderly Gait — NCT03341728.
• Registry record last updated 2019-07-05; status last verified 2019-02.
• Results first posted 2019-07-05.
• 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.

22. Record Summary

NCT03341728 is a completed randomized, unmasked crossover study titled “The Sensorimotor Locus of Balance Control in Elderly Gait,” sponsored by the University of North Carolina, Chapel Hill. The registry reports enrollment of 14 and compares walking with optical-flow perturbations against normal walking.

The primary outcome is change in side-to-side postural sway after ten minutes of walking, measured in centimeters. The posted mean difference is -2.38 cm, with a confidence interval from -3.034 to -1.726 and an unadjusted p-value of <0.001. A secondary comparison for margin-of-stability variability is also posted; analyses for the named cognitive-motor outcomes are not reported in the registry record.

No serious adverse events, other adverse events, or deaths are reported in either condition. Detailed analysis methods, the original sample-size justification, multiplicity handling, missing-data methods, and the analysis population are not reported in the registry record. The results describe short-term measured outcomes and do not directly establish effects on falls or longer-term function.