Home-based Telehealth Stroke Care: A Randomized Trial for Veterans: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Home-based Telehealth Stroke Care: A Randomized Trial for Veterans (NCT00384748), a randomized phase 2 study. The primary outcome is Physical Function as Measured by Telephone Version of FIM.

2. Clinical Question

3. Trial Design

This was a randomized phase 2 trial with a factorial model and no masking. The registry reports enrollment of 48 participants and a completed status. The study start is recorded as 2008-11, and primary completion as 2010-12. It was sponsored by the US Department of Veterans Affairs and concerned cerebrovascular accident.

The intervention description focuses on tele-rehabilitation after discharge home. It describes tele-video visits, exercise targeting stroke-related impairment, and adaptive strategies for safe functional mobility in the home. An in-home messaging device is also listed. The comparator is described as routine Veterans Affairs care, with therapy tracked using weekly diaries over the six-month study period.

The registry record labels the study model factorial, but the factors, their levels, and the planned comparisons among them are not reported in the registry record. The record does not clearly map the listed intervention descriptions to Arm 1 and Arm 2 in the outcome results, so the numerical comparison is best described by the arm labels. The record also does not report the randomization procedure or allocation concealment details. Masking is listed as none.

The primary outcome was a continuous measure: mean physical function on the telephone version of FIM at six months. The posted results include 25.0 participants in Arm 1 and 23.0 in Arm 2. The registry record lists one outcome measure and no analyses; detailed analysis procedures are not reported in the registry record. These results therefore describe the posted arm summaries and comparison, rather than a fully specified statistical analysis plan.

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

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 Physical Function as Measured by Telephone Version of FIM, measured over 6-month.

The primary endpoint was physical function at six months, measured with the telephone version of the Functional Independence Measure, or FONEFIM. This instrument was developed as a telephone alternative to the in-person performance-based FIM. The registry description says it has good concordance with the in-person measure.

The reported Motor FONEFIM covers 13 items across self-care, sphincter control, transfers, and locomotion. Each item is scored on an ordinal scale from 1, total dependence, to 7, total independence. The possible total scores range from 13 to 91; higher scores indicate greater independence. Scoring accounts for adaptive equipment and the personal assistance or supervision needed to complete tasks.

A continuous score allows the trial to compare average measured function between groups. The measure is relevant to the stated aim of assessing physical function after discharge home, while its telephone format is suited to assessment without an in-person performance test. The posted result is a group mean comparison at the six-month timeframe, not a report of each participant’s trajectory or of other outcomes such as disability, falls-related self-efficacy, or satisfaction. Results for those secondary outcomes are not reported in the registry record.

7. Results Posted in the Registry

• Arm 2: mean 80.90 versus 83.70 on control; difference -2.80 (95% CI -9.05 to 3.45), standardized effect d = -0.26; z-test p = 0.380.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry reports a continuous primary outcome summarized by a mean and standard deviation. The outcome is the six-month Motor FONEFIM score. Arm 1 has a mean of 83.7 and a standard deviation of 9.9 among 25.0 participants; Arm 2 has a mean of 80.9 and a standard deviation of 12.0 among 23.0 participants. A mean describes the arithmetic average of the measured scores in an arm. A standard deviation describes how spread out individual scores are around that average.

What does the mean difference represent?

The posted comparison is expressed as Arm 2 minus Arm 1. Its mean difference is -2.8 points. A negative value means that the observed average in Arm 2 is lower than the observed average in Arm 1; it does not, on its own, indicate whether the difference is distinguishable from random variation. The registry reports a confidence interval and p-value alongside the estimate to describe uncertainty and compatibility with a no-difference value.

How should the confidence interval be understood?

A confidence interval gives a range of values that are reasonably compatible with the observed data and the statistical method used. The interval for the mean difference runs from -9.054 to 3.454. It includes zero, the value corresponding to no difference between the arm means. It also includes values on either side of zero, so the data do not pin down the direction of the underlying difference precisely. A confidence interval is not a statement that the true value has a particular probability of lying in this one calculated interval.

What does the p-value say?

The unadjusted p-value is 0.38. In general, a p-value describes how unusual data at least as discordant with a specified null model would be if that model were correct, under the analysis assumptions. It is not the probability that the null model is true, nor a measure of the size or importance of a difference. Here, the value is not small by commonly used thresholds, and the confidence interval also includes zero. The registry record does not report a multiplicity adjustment, so this is an unadjusted result.

What role do the arm standard deviations play?

Standard deviations describe participant-to-participant variability, while the standard error describes uncertainty in an estimated mean or contrast. In general, greater variability makes it harder to estimate a group mean difference precisely when sample size is unchanged. The arm standard deviations are 9.9 and 12.0. The registry comparison also reports a pooled standard deviation of 10.95, but the record does not report the complete analysis method used for the comparison. The comparison should therefore be read as posted, without inferring unreported model details.

Why does the factorial label matter?

A factorial design can examine more than one intervention factor and may allow assessment of their separate effects or interaction. This is a general description of factorial designs, not a claim about the exact planned analysis in this trial. The registry record labels the model factorial, but does not report the factors, their assignments, or whether interaction effects were planned or estimated. The posted comparison is between Arm 2 and Arm 1; the record does not fully explain how those arm labels correspond to the listed tele-rehabilitation and messaging interventions.

What is not specified about the analysis?

The registry record lists no analyses. It does not report whether the comparison was adjusted for baseline scores or other covariates, how missing outcome values were handled, which analysis population was used, or how the factorial structure was incorporated. The original statistical analysis plan, including any interim analyses or multiplicity procedures, is not reported in the registry record. These omissions limit the ability to assess analytic choices beyond the posted summary statistics.

In general, over-dispersion means that observed variability exceeds what a statistical model assumes. It matters because a model that understates variability can produce confidence intervals that are too narrow and p-values that are too small. The registry does not report an over-dispersion assessment or indicate that such a model was used for this continuous outcome; no such assessment should be inferred from the posted results.

The planning estimates associated with the observed effect are separate from the trial’s outcome analysis. They use stated planning assumptions and illustrate how required enrollment changes with the effect size chosen for planning. They are not a reconstruction of the trial’s original design justification, which is not reported in the registry record.

10. Confidence Intervals and Precision

The mean difference of -2.8 points is accompanied by a 95% confidence interval from -9.054 to 3.454. This interval conveys substantial uncertainty about the comparison: it includes zero and values both below and above zero. Thus, the posted data do not determine the direction of the underlying mean difference with precision. The interval is more informative than the point estimate alone because it displays a range of differences compatible with the data and method.

Precision is also shaped by the number of observations and the variability of scores. The comparison includes 25.0 participants in Arm 1 and 23.0 in Arm 2, with standard deviations of 9.9 and 12.0. These summaries provide context for the uncertainty but do not remove it. The confidence interval should not be interpreted as a range containing every individual’s score; it concerns the arm difference in means.

The interval includes both no difference and nonzero values, so the data are not precise enough to rule out either direction. The unadjusted p-value of 0.38 is consistent with that uncertainty, but it does not quantify the probability that the groups are the same. Nor does the confidence interval identify whether any possible difference would be important to patients. The registry record does not report a threshold for a clinically important difference.

These precision statements apply to the posted six-month comparison. The registry record does not report results from secondary outcomes or other time points, and it does not describe missing-data methods or analysis-population details that could further inform interpretation.

11. 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 participants per arm, not total event counts. They are planning illustrations based on the observed effect, not the trial’s own design assumptions; those original assumptions are not reported in the registry record. The stated calculation uses a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It considers effects set at 100%, 75%, and 50% of the observed effect.

For each effect assumption, the 80% power column gives the estimated number of participants per arm for a design with an 80% chance of detecting that specified effect under the stated assumptions, if the effect exists and the assumptions hold. The 90% power column gives the corresponding per-arm planning number for a 90% chance. These are design-planning concepts, not guarantees about any future study’s result. The displayed figures are not observed power.

The estimates are 241 and 322 participants per arm when planning for 100% of the observed effect; 428 and 572 when planning for 75% of the observed effect; and 962 and 1,287 when planning for 50% of the observed effect, for 80% and 90% power respectively. The pattern reflects a general statistical principle: smaller effects are harder to distinguish from random variation, so planning to detect a smaller effect requires more participants under otherwise unchanged assumptions.

These figures are illustrations only. They do not establish what enrollment this trial originally intended, and they should not be mistaken for the trial’s actual sample size, which was 48 participants overall. The original sample-size justification is not reported in the registry record.

12. 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 = -2.800; pooled standard deviation σ = 10.955.
2. n = 2 × (1.96 + 0.842)² × 10.955² / -2.800² = 240.3, rounded up to 241 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 <- -2.8                      # observed difference in means
sigma <- 10.9546892877573                     # 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(384748)
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 = 241 (80% power), 322 (90% power)
 75% of observed effect: n per arm = 428 (80% power), 572 (90% power)
 50% of observed effect: n per arm = 962 (80% power), 1,287 (90% power)
power.t.test: 241.2 
Simulated power at n = 241 per arm: 0.796 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry record describes a primary aim focused on physical function and mentions secondary aims concerning disability, falls-related self-efficacy, and patient satisfaction. It lists one outcome measure and no analyses. Results for the secondary aims are not reported in the registry record, so no conclusions about those outcomes can be drawn from the posted primary comparison.

Multiplicity arises when several outcomes, time points, groups, or hypotheses are tested. In general, testing many questions without an adjustment can increase the chance of at least one apparently notable result arising by chance. The registry record does not report a multiplicity strategy or specify which outcomes or comparisons were designated for formal testing beyond identifying physical function as the primary aim. The posted p-value is unadjusted.

Missing-data handling is also not reported in the registry record. The record gives arm sample sizes for the primary outcome, but does not specify whether every randomized participant contributed a six-month score, how unavailable measurements were handled, or whether any imputation or sensitivity analyses were used. The analysis population is not reported in the registry record, so it is unclear whether the posted comparison follows an intention-to-treat, per-protocol, or another approach.

The record labels the model factorial but does not report the factor structure or analysis of interactions. It also does not report baseline adjustment, interim analyses, or the statistical analysis plan. These are limitations of the available reporting, not evidence that a particular procedure was or was not used. Without those details, interpretation should remain focused on the posted arm summaries and unadjusted comparison.

14. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

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 25 participants at risk in Arm 1 and 23 in Arm 2 for the adverse-event summaries. In each arm, zero participants are listed as affected by a serious adverse event, and zero are listed as affected by another adverse event. No serious-event terms are listed. These are the counts recorded in the registry; they do not describe event types that were not posted.

Zero affected participants in these summaries means no events were recorded in the corresponding fields for the listed groups. It does not establish that the underlying risk is absent, particularly in a study with 48 participants overall. The record does not report the adverse-event collection definitions, the observation window for these summaries, or additional detail about ascertainment. Death counts are not reported in the registry record.

These safety figures should be read separately from the physical-function comparison. They do not provide evidence about the primary outcome, and they do not resolve the uncertainty in the six-month mean difference.

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

The result describes the six-month physical-function scores recorded for two randomized arms. Arm 1 had a mean FONEFIM score of 83.7, and Arm 2 had a mean of 80.9. The mean difference, expressed as Arm 2 compared with Arm 1, was -2.8 points. The 95% confidence interval, from -9.054 to 3.454, includes zero, and the unadjusted p-value is 0.38. Together, these show an uncertain estimate rather than a precisely determined difference.

The result does not show that the groups are equivalent. A failure to distinguish a difference in this comparison is not proof that the underlying means are identical. The confidence interval includes multiple possible values, including values in either direction. The registry does not report a threshold for a clinically important difference, so the clinical importance of values in the interval cannot be determined from the posted information.

The primary result also does not answer the study’s secondary questions about disability, falls-related self-efficacy, or satisfaction. Those results are not reported in the registry record. Nor does a single six-month measure describe how participants changed over time. No additional time-point results are reported in the registry record.

The findings should be interpreted with the trial’s reporting limitations in mind. The record identifies a randomized factorial design, but does not describe the factors or fully connect the arm labels to the intervention components. It does not report analysis-population or missing-data methods, adjustment procedures, or a detailed statistical plan. The result is therefore best understood as the posted comparison of arm means, with its stated uncertainty, rather than as a complete account of all possible analyses or outcomes.

16. Why This Trial Matters Statistically

This trial illustrates why a point estimate, confidence interval, and p-value should be read together. The mean difference is -2.8 points, but the confidence interval extends from -9.054 to 3.454 and includes zero. The point estimate alone could suggest a direction, while the interval makes clear that uncertainty remains about both direction and size. The p-value of 0.38 provides a related but distinct summary of compatibility with a no-difference model.

It also highlights the difference between a trial’s observed enrollment and a planning illustration. The trial enrolled 48 participants, while the displayed planning estimates vary according to the fraction of the observed effect selected for future planning and the desired power. Smaller assumed effects require more participants under the stated assumptions. Such illustrations are not a substitute for the original design justification, which is not reported in the registry record.

Finally, statistical interpretation depends on reporting as well as calculation. The registry identifies a factorial design but does not explain its factors, analysis population, missing-data approach, or multiplicity strategy. Those details matter because they clarify what a comparison estimates and how robustly it addresses the research question. The posted numbers support a focused interpretation of the six-month mean comparison, while leaving important design and analysis details unresolved.

17. 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 record provides limited detail about eligibility, baseline characteristics, and participant flow. It identifies a population of stroke patients after discharge home and a total enrollment of 48, but further inclusion and exclusion criteria are not reported in the registry record. The primary outcome summaries are listed by Arm 1 and Arm 2, without a clear mapping in the posted results to each intervention description.

The design is labeled factorial, but the factors and their levels are not reported in the registry record. This makes it difficult to determine what the two-arm contrast represents in relation to the tele-rehabilitation intervention and in-home messaging device. The record also does not report allocation concealment, the randomization procedure, or detailed information about usual-care exposure.

Analysis details are limited. The registry lists no analyses and does not report the analysis population, missing-data handling, baseline adjustment, or any sensitivity analyses. It does not report the original sample-size justification, multiplicity plan, or interim analyses. Secondary results are not reported in the registry record, despite the stated interests in disability, falls-related self-efficacy, and patient satisfaction.

The safety summaries list zero affected participants for serious and other adverse events in both arms, but event definitions and collection details are not reported in the registry record. Death counts are not reported in the registry record. These reporting gaps restrict what can be inferred about outcomes beyond the posted primary measure and comparison.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Home-based Telehealth Stroke Care: A Randomized Trial for Veterans — NCT00384748.
• Registry record last updated 2014-11-24; status last verified 2014-11.
• Results first posted 2014-11-10.
• 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.

21. Record Summary

NCT00384748 is a completed randomized phase 2 trial with a factorial model and no masking, sponsored by the US Department of Veterans Affairs. It enrolled 48 participants and studied tele-rehabilitation after stroke discharge to home, with usual care as a comparator. The primary outcome was six-month physical function measured by the telephone version of FIM.

The posted means were 83.7 in Arm 1 and 80.9 in Arm 2. The mean difference for Arm 2 compared with Arm 1 was -2.8, with a 95% confidence interval from -9.054 to 3.454 and an unadjusted p-value of 0.38. The interval includes zero, indicating uncertainty about the direction and size of the difference. Secondary outcome results and important analysis details are not reported in the registry record.