Live Long Walk Strong (LLWS) Rehabilitation Program: What Features Improve Mobility Skills: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Live Long Walk Strong (LLWS) Rehabilitation Program: What Features Improve Mobility Skills (NCT04026503), a randomized na study. The primary outcome is Walking Speed.

2. Clinical Question

3. Trial Design

The registry describes a randomized, double-masked, parallel-group study. It lists an enrollment of 151 and a completed status. The sponsor was the VA Office of Research and Development. The listed start date was 2021-01-21, and the primary completion date was 2024-09-30. The phase is listed as not applicable.

The two arms relevant to the initial primary-outcome comparison were Live Long Walk Strong and an 8 week wait list control. The intervention arm received an 8 week rehabilitation program. The wait list arm had an 8 week wait and was then followed by 8 weeks of the rehabilitation program. The primary outcome timeframe was randomization to 8 weeks, so the posted primary comparison concerns the initial period before the wait list group began its program.

Randomization can help balance groups at the point of assignment, but the registry record does not report the randomization method or allocation concealment details. It labels masking as double, but does not identify which people were masked or describe how masking was maintained. The record also does not report the original statistical analysis plan or the trial’s original sample-size assumptions. Those details should not be inferred from the design labels or from the planning illustrations based on observed effects.

The primary result is based on 51 participants in the intervention arm and 58 in the wait list arm. Counts contributing to the secondary outcomes vary by outcome. The registry describes endpoint data differences associated with dropouts, COVID restricted study visits that resulted in remote assessment only, equipment failure, and assessor error. It does not provide a full participant-flow account or explain how each of these circumstances affected each analysis.

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.

Randomize to 8 Weeks (LLWS PT or Wait)

Wait List Control Receives LLWS PT

Post 1 Both Groups Post LLWS PT

Post 2 Both Groups 8 Weeks Post LLWS PT

Analysis population for the primary outcome, as defined in the registry: Differences due to study drop outs, COVID restricted study visits that resulted in remote assessment only, equipment failure, and assessor error.

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: Some data lost to equipment failure and assessor error.

6. Primary Endpoint

The registry lists the primary outcome as Walking Speed, measured over Randomization to 8 weeks (comparing intervention to wait list).

The primary endpoint was walking speed in meters per second. It was assessed as usual walking pace over a 4-meter course, with the primary timeframe running from randomization to 8 weeks. The registry reports the mean and standard deviation for each group and a mean difference between groups.

Walking speed is a continuous measure: it can take a range of values rather than placing each participant into a category. A group mean gives a summary of the observed speeds, while the standard deviation describes how dispersed individual measurements were around that mean. Comparing group means provides a direct summary of the average difference in this measured outcome during the specified period.

For this study, gait speed was relevant to the stated focus on mobility skills and to the study summary’s interest in features associated with improved gait speed. The endpoint description specifies usual walking pace rather than a different walking task or a longer-distance test. The registry does not report a threshold for a meaningful change, a minimal clinically important difference, or a justification for selecting this endpoint. Statistical evidence about a difference in group averages should therefore be kept distinct from claims about its practical importance.

The recorded endpoint data were affected by dropouts, COVID restricted visits that resulted in remote assessment only, equipment failure, and assessor error. The registry does not describe the procedures used to handle these circumstances in the analysis.

7. Results Posted in the Registry

• Live Long Walk Strong: mean 0.86 versus 0.91 on control; difference -0.05 (95% CI -0.06 to -0.04), standardized effect d = -2.50; 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.

Lower Extremity Power

• 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:
  • Live Long Walk Strong: mean 8.20 versus 7.30 on control; difference 0.90 (95% CI 0.03 to 1.77), standardized effect d = 0.41; z-test p = 0.042.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Live Long Walk Strong than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Trunk Muscle Endurance

• 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:
  • Live Long Walk Strong: mean 0.70 versus 0.59 on control; difference 0.11 (95% CI -0.11 to 0.34), standardized effect d = 0.20; z-test p = 0.329.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Live Long Walk Strong and the control. The interval gives the range of effects that remain compatible with the data.

Gait Variability

• 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:
  • Live Long Walk Strong: mean 0.04 versus 0.05 on control; difference -0.01 (95% CI -0.02 to -0.00), standardized effect d = -0.50; z-test p = 0.028.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Live Long Walk Strong than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes were lower extremity power, trunk muscle endurance, and gait variability, measured over randomization to 8 weeks. Across them, the pattern is mixed rather than a consistent parallel to the primary walking-speed result. Some estimates favor the intervention direction and have intervals that exclude zero; another is less precise and has an interval that includes zero. The primary point estimate, meanwhile, is in the opposite direction from the stated improvement focus. Together, these results do not form a simple, uniformly aligned picture across mobility-related measures.

The secondary measures capture different aspects of movement and physical performance, so a result on one should not be assumed to stand in for another. A difference in power, endurance, or variability is not itself a direct substitute for the walking-speed endpoint. The registry does not report a model linking these measures or an analysis of whether changes in one explain changes in another. It also does not report which program features might be responsible for any observed differences.

These comparisons carry less weight than a single primary result when they are unadjusted and considered among multiple outcomes. Repeated testing across outcomes increases the opportunity to observe apparently notable results by chance. The registry lists eight outcome measures, posts three secondary outcomes, and does not report a multiplicity adjustment or a hierarchy for interpreting secondary tests. The secondary estimates are useful as descriptive evidence and as a basis for future questions, but they should not be treated as independent confirmation of a treatment effect.

The differences in participant counts across secondary outcomes also matter for interpretation. Each estimate reflects the participants contributing to that measure, and the registry does not fully explain the changing counts or the handling of unavailable assessments. That uncertainty, together with unadjusted testing, argues for a cautious reading of the overall secondary pattern rather than focusing on any isolated result.

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

• Lower Extremity Power
• Trunk Muscle Endurance
• Gait Variability

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous, and the registry reports group means and standard deviations. A mean is the arithmetic average of the observed measurements in a group. A standard deviation describes how much individual observations typically vary around that group average. These summaries help describe both the location of the group results and their spread, but they do not show the full distribution of individual values.

What does a mean difference represent?

The reported mean difference compares the intervention mean with the wait list mean. For walking speed, the group means were 0.86 m/s and 0.91 m/s, and the contrast was -0.05 m/s. A negative contrast indicates that the first group’s mean was lower than the comparator’s mean under this ordering. The contrast is expressed in the original endpoint unit, making it directly interpretable as a difference in average walking speed. It does not describe how every participant changed or how much participants within either group varied.

What do the p-value and confidence interval tell us?

General explanation: a p-value describes how incompatible the observed data are with a specified null hypothesis, under the assumptions of the statistical test. It is not the probability that the null hypothesis is true, and it does not measure the size or practical importance of a difference. The primary unadjusted p-value is < 0.001. The registry does not report the test details or a complete analysis plan, so interpretation should remain tied to the stated unadjusted comparison.

General explanation: a confidence interval describes a range of values compatible with the data and the interval procedure. It communicates uncertainty around an estimate; it is not a range containing a fixed proportion of individual participant measurements. The primary interval excludes zero, consistent with the small unadjusted p-value for a test of no mean difference. The interval’s limits are available with the posted comparison, while the plain-language interpretation is that the estimate is below zero throughout the reported interval.

Why do the standard deviations and standard errors matter?

The standard deviations describe between-participant spread within each group. Standard errors describe the uncertainty of estimated quantities such as a group mean, under the analysis assumptions. In general, standard errors depend both on variability and on the amount of information contributing to an estimate. They are not interchangeable with standard deviations: one describes spread among observations, while the other describes precision of an estimate.

The registry includes standard errors for the primary group means and a pooled standard deviation for the contrast. It does not describe the full statistical procedure used to obtain the confidence interval or p-value. The posted group means and dispersion estimates should therefore be treated as the reported summaries, rather than as a complete account of the analysis model.

Why can over-dispersion matter?

General explanation: in count outcomes, over-dispersion means the observed variability is greater than a model assumes. When that is ignored, uncertainty can be understated and statistical evidence can appear stronger than warranted. That issue is not the central feature of this primary endpoint, which is continuous walking speed summarized by means and standard deviations. The registry does not report a count-model analysis for this endpoint.

What can be inferred from the posted analysis information?

The registry lists no reported analyses, and the secondary outcomes do not have reported analyses listed. It does post arm-level summaries and comparison statistics. A statistical analysis plan, model specifications, covariate adjustments, the exact testing procedure, and any analysis-population definitions are not reported in the registry record. Consequently, it is not possible to establish from the record whether the posted comparisons were adjusted for baseline values or other variables, or how departures from the planned assessment were handled.

General explanation: random assignment supports a comparison between groups by reducing systematic differences at assignment, but it does not remove uncertainty due to incomplete data, measurement problems, or chance. The trial is labeled randomized and double-masked, but the registry does not supply enough detail to assess the implementation of those features. The numerical comparison is best read together with the stated endpoint, the available participant counts, and the record’s notes about data differences.

11. Confidence Intervals and Precision

Precision concerns how much uncertainty surrounds an estimated quantity. For the primary outcome, the mean difference is -0.05 m/s, and its reported confidence interval lies below zero. The interval is narrow around the estimate, indicating that the reported statistical uncertainty for this contrast is limited under the calculation used. This does not address possible bias from missing measurements, assessment changes, or other departures that the interval may not capture.

General explanation: a confidence interval is more informative than a point estimate alone because it shows a range of values compatible with the data under a specified procedure. If an interval includes zero for a mean difference, the data are compatible with no difference as well as with values on either side, according to that interval. If the interval excludes zero, it indicates that zero is not among the values supported by that procedure at its stated confidence level. The registry does not state the confidence level in the record summarized for these results.

Precision also depends on what the analysis includes. The primary outcome includes 51 participants in the intervention group and 58 in the wait list group. Secondary outcome counts differ. The registry notes dropouts, remote assessment resulting from COVID restricted visits, equipment failure, and assessor error, but does not explain how these factors were handled in estimating uncertainty. The confidence interval should therefore be understood as uncertainty under the posted analysis, not as a complete measure of every source of uncertainty in the study.

For secondary endpoints, the degree of precision varies: some reported intervals exclude zero and another spans zero. The secondary findings should not be treated as equally conclusive simply because each has a point estimate. Their interpretation also depends on the fact that multiple outcomes were posted and no adjustment for multiple comparisons is reported.

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 illustrations give participants per arm for 80% power and participants per arm for 90% power. They are not total event counts: these outcomes are continuous, and the figures represent participants in each of two equally allocated arms. The assumptions listed for the illustrations are a two-sided significance level of 0.05, equal allocation between the arms, and a normal approximation. Each scenario uses the stated fraction of the observed effect as the effect to detect. The original trial’s own sample-size assumptions are not reported in the registry record.

In general, the 80% power column describes the number of participants per arm that the planning calculation associates with an 80% chance of detecting the specified effect under its assumptions, if that effect is present. The 90% power column uses a higher chance of detection and therefore generally requires more participants per arm for the same assumed effect. These are planning quantities, not descriptions of what happened in the trial.

The planning rows consider 100%, 75%, and 50% of the observed effect. A smaller assumed effect is harder to distinguish from random variation, so the calculation calls for more participants per arm. The figures are based on the observed effect rather than a separately stated target selected before the trial. They should therefore be read as illustrations of how sample requirements change when the assumed effect changes, not as confirmation that the original study was designed with these numbers.

This is not observed power. The calculations do not indicate the probability that the study’s actual result is correct, nor do they repair limitations from missing data or uncertain analysis choices. Their meaning depends on the stated assumptions and the effect size used in each planning scenario.

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 = -0.050; pooled standard deviation σ = 0.020.
2. n = 2 × (1.96 + 0.842)² × 0.020² / -0.050² = 2.5, 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 <- -0.05                      # observed difference in means
sigma <- 0.02                     # 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(4026503)
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), 4 (90% power)
 75% of observed effect: n per arm = 5 (80% power), 6 (90% power)
 50% of observed effect: n per arm = 11 (80% power), 14 (90% power)
power.t.test: 3.8 
Simulated power at n = 3 per arm: 0.634 (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 eight outcome measures and posts three secondary outcomes in addition to the primary walking-speed outcome. Testing multiple outcomes creates a multiplicity issue: even when there is no true difference on any measure, a collection of tests can produce apparently strong findings by chance. General explanation: methods such as prespecifying a primary outcome, ordering secondary tests, or adjusting significance thresholds can help manage this issue. The registry record does not report whether such procedures were planned or used, and no multiplicity adjustment is reported for the posted secondary comparisons.

The record lists no reported analyses. It provides arm summaries and comparison statistics, but a full statistical analysis plan is not reported in the registry record. Specific details that are not reported include model specifications, any adjustment for baseline measurements or other covariates, how analysis populations were defined, and how the primary endpoint was tested. Without those details, the posted unadjusted results should not be assumed to represent a prespecified adjusted analysis.

Missing-data handling is also not reported. The endpoint note refers to differences related to dropouts, COVID restricted study visits that resulted in remote assessment only, equipment failure, and assessor error. The record does not say whether unavailable outcomes were excluded, imputed, handled using another method, or assessed in sensitivity analyses. It also does not specify whether remote assessments were considered comparable to in-person assessments.

The primary outcome counts are 51 and 58 across the two arms, while the listed enrollment is 151. This information alone does not establish why the numbers differ or which analysis population was intended. Counts for the secondary outcomes vary as well. The registry does not provide enough participant-flow or missingness detail to determine whether availability of an outcome was related to participant characteristics or observed results. These omissions limit assessment of potential bias as well as statistical precision.

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.

Most frequent non-serious adverse event terms

The registry lists three safety groups: Live Long Walk Strong, Waitlist Control-Live Long Walk Strong, and Waitlist Control-No Intervention (8 Weeks). The corresponding numbers at risk for serious adverse events were 74, 76, and 76. The record lists 0 serious adverse events related to the study intervention in each group and 0 deaths in each group. It does not list serious-event terms.

For other adverse events, the registry lists 2 related events in the Live Long Walk Strong group, 2 in the Waitlist Control-Live Long Walk Strong group, and 0 in the Waitlist Control-No Intervention (8 Weeks) group. The numbers at risk for these events were 74, 76, and 76, respectively. The record does not provide event descriptions in the information, so the nature and severity of those other events cannot be characterized.

These figures should be read as the registry’s event counts, not as a complete safety profile. They do not establish that no unlisted or unrecorded events occurred, and they do not by themselves support comparisons of risk between groups. The safety groups include different study periods, including a wait list period followed by the rehabilitation program, so the counts should be interpreted in light of those group descriptions.

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

The main statistical result is a difference in average walking speed at the 8 week comparison. The intervention group’s posted mean was 0.86 m/s, compared with 0.91 m/s in the wait list control group. The intervention-minus-control difference was -0.05 m/s, with an unadjusted p < 0.001 and a reported confidence interval below zero. Thus, the posted primary result points to a lower average walking speed in the intervention group for this comparison, despite the study’s stated interest in improved gait speed.

This statistical description does not establish why the difference arose. The record does not report enough detail about analysis methods, missing data, assessment circumstances, or program components to isolate an explanation. Nor does the result establish what would happen with a different program, a different comparison group, a different assessment period, or a population outside the Veterans 50 years and older described in the study summary.

A statistically distinct average is not automatically a clinically important difference. The registry does not report a threshold for meaningful change or other context for deciding how important a difference of -0.05 m/s is. The confidence interval and p-value address statistical uncertainty under the analysis, not the practical consequences for participants. The posted summary also does not describe individual responses, so it cannot show whether all participants experienced similar results or whether experiences varied.

The secondary outcomes add context but do not resolve these limitations. They cover several mobility-related measures, and their results are mixed. Because the secondary comparisons are unadjusted and multiple outcomes were measured, they should be interpreted cautiously rather than used to override or explain the primary result.

The most defensible conclusion is narrow: the posted summaries show a statistically distinct difference in mean walking speed between the two groups during the initial randomization-to-8-week comparison, with the intervention group mean lower. The registry record does not establish a broader causal account or the practical meaning of that difference.

17. Why This Trial Matters Statistically

This study illustrates why statistical interpretation should begin with the direction and size of an estimate, not with a p-value alone. The study summary describes an interest in improved gait speed, but the posted primary contrast is negative when expressed as intervention minus wait list. Reading the sign and the group means prevents a small p-value from being mistaken for evidence that the result points in the hoped-for direction.

The record also shows how precision and analysis context work together. A confidence interval that excludes zero provides evidence against a zero mean difference under the interval procedure, but it does not address endpoint meaning, missing-data bias, or the effect of multiple outcomes. The registry’s notes about dropout and assessment conditions are therefore part of the statistical interpretation, not administrative details to ignore.

The secondary outcomes offer another useful lesson. Multiple measures can provide a broader view of mobility, but mixed unadjusted findings are not equivalent to a coherent, confirmatory pattern. Without a reported testing hierarchy or multiplicity approach, individual secondary results warrant restraint. The changing participant counts across those outcomes further underscore that each estimate may rest on a different subset of observations.

Finally, planning calculations based on an observed effect are illustrations, not a substitute for a trial’s original design rationale. They can show how assumed effect size and desired power affect planning requirements, but the original sample-size justification is not reported. Keeping that distinction clear prevents retrospective calculations from being mistaken for the study’s prespecified plan or for evidence about the probability that its result is correct.

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 registry reports a completed randomized, double-masked, parallel study, but does not describe the randomization process, allocation concealment, or who was masked. Those omissions prevent a detailed assessment of how the design features were implemented. The original sample-size justification and statistical analysis plan are also not reported, so the posted comparison cannot be fully assessed against a prespecified plan.

The endpoint note identifies dropouts, COVID restricted study visits that resulted in remote assessment only, equipment failure, and assessor error as sources of differences in the data. The record does not explain the number of measurements affected by each issue, how remote and in-person assessments were combined, or how missing outcomes were handled. Primary and secondary outcome counts vary, and a full participant-flow description is not reported.

The wait list comparison applies to the initial 8 week period. The wait list group was then followed by the rehabilitation program, so the registry description does not provide a continuing untreated comparison after that period. The posted primary result therefore does not address a longer-term comparison between ongoing program participation and no intervention.

Further limitations include the lack of reported participant characteristics beyond the study summary’s description of Veterans 50 years and older, the absence of a reported threshold for meaningful walking-speed change, and the lack of detail about the program features under study. The record does not report adjustment for multiple outcomes. These limits constrain generalization, causal interpretation, and the practical interpretation of the differences.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Live Long Walk Strong (LLWS) Rehabilitation Program: What Features Improve Mobility Skills — NCT04026503.
• Registry record last updated 2026-02-20; status last verified 2026-02.
• Results first posted 2026-02-20.
• 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

The registry record describes NCT04026503, a completed randomized, double-masked, parallel-group study of the Live Long Walk Strong rehabilitation program in Veterans 50 years and older. It lists an enrollment of 151 and compares an 8 week program with an 8 week wait list control for the primary randomization-to-8-week walking-speed assessment.

The primary posted means were 0.86 m/s for Live Long Walk Strong and 0.91 m/s for wait list control. The mean difference was -0.05 m/s, with an unadjusted p < 0.001. Three secondary outcomes are also posted, with mixed results. The registry notes several sources of endpoint data differences, but does not report detailed missing-data methods, a full statistical analysis plan, or the trial’s original sample-size assumptions.