Translating Intensive Arm Rehabilitation in Stroke to a Telerehabilitation Format: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Translating Intensive Arm Rehabilitation in Stroke to a Telerehabilitation Format (NCT02665052), a randomized na study. The primary outcome is Wolf Motor Function Test Performance Time (WMFT).

2. Clinical Question

3. Trial Design

The registry describes a randomized, parallel-group trial with single masking and an enrollment of 63. The study is listed as completed. Its title and intervention descriptions concern translating intensive arm rehabilitation after stroke into a telerehabilitation format. The record lists stroke and hemiparesis as conditions and identifies the VA Office of Research and Development as sponsor. A phase designation is listed as NA.

Three intervention groups are described: home-based BATRAC plus TTT, lab-based BATRAC plus TTT, and delayed-entry usual care. The home-based program was described as 45 minutes of high-intensity bilateral reaching and rest periods using BATRAC, followed by 15 minutes of video-guided TTT. The lab-based program was described as 45 minutes using BATRAC and 15 minutes of therapist-guided TTT in a 60-minute lab session. The delayed-entry group initially served as a control during the first 6 weeks, receiving study evaluations but no study intervention. Further details of the usual-care experience are not reported in the registry record.

The primary outcome was change in WMFT performance time from baseline to post-training at week 6. The posted primary outcome analysis included 24 participants: 8 in the home-based group, 7 in the lab-based group and 9 in delayed-entry usual care. The record says that participants with no 6-week follow-up data for the primary outcome were excluded. The reason for the difference between total enrollment and the primary analysis population, beyond that stated exclusion, is not reported in the registry record.

The registry reports one inferential test for the primary comparison: a two-sided t-test comparing home-based BATRAC with delayed-entry usual care. Results for the lab-based comparison and details such as the masking role, allocation concealment, randomization procedures, and 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.

Before assignment: Of 63 enrolled participants, 44 were study trial participants and 19 were participant caregivers for the home-based group. Of the 44 study trial participants, 15 did not meet inclusion criteria resulting in 29 individuals randomized into the study.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Only 24 participants were included in the analysis, as we excluded patients who had no 6-week follow-up data on the primary outcome.

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: Caregivers did not receive outcome evaluations.

6. Primary Endpoint

The registry lists the primary outcome as Wolf Motor Function Test Performance Time (WMFT), measured over Final Training (week 6).

The primary endpoint was change in WMFT performance time from baseline to post-training, measured at final training in week 6. The WMFT assesses the time required to complete fifteen timed tasks, with a maximum of 120 seconds. The endpoint is expressed as log (seconds). The registry defines change as the 6-week WMFT log time minus baseline WMFT log time.

The WMFT focuses on task performance speed. The registry states that faster completion, represented by a lower value, indicates a better outcome. Time measurements can be skewed, so the registry says the logarithm-transformed time values were analyzed to reduce skewness. As a general statistical point, transformation changes the scale on which an outcome is analyzed; results on a log scale should therefore be interpreted in relation to that scale rather than treated as raw seconds.

The stated primary study outcome was the between-group difference in post-training WMFT log time for home-based treatment compared with control. The outcome description also says that change from baseline was analyzed. The posted arm summaries are 3.6 (standard deviation 1.0) for home-based BATRAC, 2.9 (standard deviation 1.2) for lab-based BATRAC plus TTT, and 3.1 (standard deviation 1.1) for delayed-entry usual care. The registry does not report a between-group effect estimate or confidence interval alongside those summaries.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the reported test assess?

The registry identifies a two-sided t-test for the comparison of home-based BATRAC and delayed-entry usual care. As a general statistical explanation, a t-test evaluates whether the data are compatible with a null hypothesis of no difference in group means, subject to the assumptions of the chosen test. A two-sided test allows evidence for a difference in either direction rather than specifying only one direction in advance. The registry does not report whether the test was paired or unpaired, whether a variance adjustment was used, or other implementation details.

The primary endpoint is continuous, and the record states that the mean was the summary parameter. The arm summaries are means with standard deviations. A mean describes the average observed outcome in a group; a standard deviation describes the spread of individual observations around that group’s mean. As a general statistical point, a standard deviation is not a confidence interval for the mean and does not itself describe uncertainty in the between-group contrast.

Why analyze log-transformed time?

The registry explains that WMFT performance time was logarithm-transformed to reduce skewness in the time values. As a general statistical point, time measurements may have a long upper tail when some participants take much longer than others. Skewness can make a simple model of untransformed values a poor representation of the data. Transforming values can change the distribution and scale used in the analysis, but it also means that results refer to transformed time rather than directly to seconds. The record gives the endpoint unit as log (seconds).

The registry defines change as 6-week log time minus baseline log time, while also describing the primary study outcome as the between-group difference in post-training log time. These descriptions indicate the outcome and comparison of interest, but the record does not provide enough analysis detail to resolve whether the reported test used change scores, post-training values, or an adjustment for baseline. The exact model specification is not reported in the registry record.

How should the p-value be read?

The reported p-value is 0.64. As a general statistical point, a p-value is calculated under a null model and summarizes how unusual data at least as discrepant as those observed would be under that model. It is not an effect size, and it does not indicate the probability that a result will replicate. It also does not show whether a difference is important to patients. Interpretation is stronger when a p-value is accompanied by a between-group estimate and a confidence interval, but neither is reported for this comparison in the registry record.

A confidence interval, as a general concept, gives a range of effect estimates compatible with the data under the method’s assumptions and conveys uncertainty about the estimate. No confidence interval for the primary comparison is reported. The posted group means and standard deviations are not enough to substitute for a reported interval without making additional calculations, which are not presented as registry results.

What do the analyzed sample sizes imply?

The primary outcome summaries include 8 home-based participants, 7 lab-based participants and 9 delayed-entry participants. The registry says that only 24 participants were included because those without 6-week follow-up data on the primary outcome were excluded. As a general statistical point, smaller analyzed groups usually yield less precise estimates than larger groups, all else being equal. The degree of precision cannot be quantified from the reported p-value alone, especially without a confidence interval or a detailed analysis output.

The record lists an enrollment of 63, which is distinct from the 24 participants included in the primary outcome analysis. It does not report the number randomized to each group, the reasons for missing follow-up beyond the stated exclusion, or whether any missing-data method was used. The analysis population and missing-data handling are therefore important uncertainties when considering the result.

How do planning illustrations differ from observed results?

Any sample-size planning illustration associated with these results is not an observed trial outcome. The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect to detect defined as a stated fraction of the observed effect. As a general statistical point, such calculations describe hypothetical future planning under those assumptions; they are not evidence that the completed trial had a particular power, and they should not be called observed power. The registry’s own sample-size justification and planning assumptions are not reported in the registry record.

The planning illustration is based on an observed effect, but no numeric planning outputs are included in the record details described for the analysis. No values for participants per arm at particular power targets can therefore be stated. More fundamentally, planning based on an observed effect is sensitive to uncertainty in that effect, particularly when the analysis sample is limited. Such an illustration should be read as a scenario, not as a design feature established by the registry.

Why does over-dispersion matter?

As a general statistical concept, over-dispersion arises when variability in observed outcomes is greater than a model assumes. It can make standard errors too small if the extra variability is ignored, leading to intervals that are too narrow and tests that appear more decisive than warranted. This endpoint is analyzed as a continuous measure with means and standard deviations; the registry does not report an over-dispersion assessment or indicate that over-dispersion was a concern. The concept is therefore general context, not a reported finding about this trial.

Overall, the reported statistical information identifies the outcome scale, descriptive group summaries, and a two-sided test for one comparison. It does not report a full model specification, an effect estimate, a confidence interval, a multiplicity procedure, or missing-data methods. Those omissions constrain how much can be concluded about the size and precision of the result.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Home-Based BATRAC", "Lab-based BATRAC Plus TTT", "Delayed Entry Usual Care")
n   <- c(8, 7, 9)
m   <- c(3.6, 2.9, 3.1)
sp  <- c(1, 1.2, 1.1)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Home-Based BATRAC  8  3.600  1.000  2.764  4.436
CARM  Lab-based BATRAC Plus TTT  7  2.900  1.200  1.790  4.010
CARM  Delayed Entry Usual Care  9  3.100  1.100  2.254  3.946

11. Analyses Reported by the Sponsor

The registry reports the sponsor’s own analysis for the primary comparison as a two-sided t-test of home-based BATRAC against delayed-entry usual care, with p = 0.64. No numerical between-group estimate or confidence interval is reported with that test. The record also indicates that values are missing or not estimable for a two-arm comparison of this continuous endpoint, so there is no independent recalculated comparison to set alongside the sponsor’s result.

The arm-level means and standard deviations describe the analyzed groups, but they are not a substitute for a reported contrast and its uncertainty. The record lists 8 analyzed participants in the home-based group and 9 in delayed-entry usual care. It also reports 7 in the lab-based group, for which no inferential comparison is reported in the registry record.

No discrepancy is flagged between a sponsor analysis and a recalculated estimate. Because no recalculated comparison is available, the interpretation rests on the sponsor-reported p-value and the descriptive summaries. The p-value does not quantify the size of the group difference or its precision; those quantities are not reported in the registry record.

12. Multiplicity, Missing Data and Analysis Populations

The registry lists three outcome measures and identifies two secondary outcomes: the Fugl-Meyer Assessment Scale-Upper Extremity and the Stroke Impact Scale (SIS)-Hand Domain. It also lists four analyses, but does not describe the full set of analyses or how they relate to the outcome measures. A multiplicity adjustment, a prespecified hierarchy of outcomes, and a strategy for controlling the chance of false-positive findings across analyses are not reported in the registry record.

As a general statistical point, when several outcomes, time points or comparisons are tested, the chance of obtaining at least one small p-value by chance can increase. Methods such as prespecifying a primary outcome or adjusting for multiple testing may address that issue, but the registry record does not report whether such procedures were used. The single reported primary comparison has p = 0.64; the lack of a reported multiplicity method does not change that value, but it limits assessment of the wider analysis strategy.

The primary endpoint analysis included 24 participants because those without 6-week follow-up data were excluded. The record does not report the number excluded, reasons for missing follow-up beyond its absence, or how missing data were handled. It does not specify whether any imputation, sensitivity analysis, or model-based approach was used. The number randomized to each group and a complete participant flow are also not reported in the registry record.

These details matter because an analysis restricted to participants with observed outcomes can differ from an analysis that includes all randomized participants, depending on why outcomes are missing and on the method used. As a general statistical point, the consequences of missing data depend on the missingness process and assumptions used in the analysis. The registry record does not provide enough information to assess those assumptions or determine whether the analyzed participants represent all randomized participants.

The record states that the primary outcome population excluded participants without week-6 follow-up, but does not report an intention-to-treat analysis or another named analysis population. Further information on the analysis population, missingness, and multiplicity is not reported in the registry record.

13. 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 reports serious adverse event counts of 0 among 15 participants at risk in the home-based BATRAC group, 0 among 14 in the lab-based BATRAC plus TTT group, and 0 among 9 in delayed-entry usual care. It also lists a separate home-based BATRAC caregivers group, with 0 serious adverse events among 19 at risk. Deaths are recorded as 0 for each listed group.

These are counts during the registry’s safety reporting, not estimates of comparative risk. The safety group sizes are not the same as the primary outcome analysis sizes in every group, and the record does not explain the relationship between the caregiver category and the participant groups. No serious adverse event terms are listed. The record also reports other adverse events affecting 4 of 15 in the home-based group, 4 of 14 in the lab-based group, 1 of 9 in delayed-entry usual care, and 0 of 19 in the caregiver group.

As a general statistical point, observing no serious adverse events in a group does not establish that the underlying risk is zero. The counts should be read in the context of the number at risk and the reporting period; further details about the safety collection period and event definitions are not reported in the registry record.

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

The posted result documents a randomized, parallel-group study with a primary outcome focused on change in WMFT performance time at week 6. For the reported comparison of home-based BATRAC with delayed-entry usual care, the registry gives a two-sided t-test p-value of 0.64. It also supplies group-level means and standard deviations for the 24 participants included in the primary analysis.

These results do not provide a numerical estimate of the between-group difference or a confidence interval. Consequently, the size and precision of the primary contrast cannot be determined from the reported inferential result alone. A p-value is not a measure of clinical importance and does not show the probability that either intervention is beneficial. The study’s posted data should therefore be described as a reported comparison and not as a definitive conclusion about treatment effects.

The result applies to the analyzed participants with 6-week primary outcome data. The record says participants without such follow-up were excluded, but does not report the full missing-data process or analysis methods. The trial enrollment was 63, while the primary outcome analysis included 24. This difference is relevant when considering how the analyzed result relates to all enrolled or randomized participants.

The lab-based group has a descriptive mean and standard deviation, but the registry does not report a corresponding primary inferential comparison with delayed-entry usual care. The home-based versus usual-care p-value should not be generalized to that unreported contrast. Nor does the record establish how results would differ under another analysis method, another population, or a different follow-up period.

In short, the registry reports one primary comparison, a p-value, and descriptive summaries. It does not report the effect estimate, uncertainty interval, full analysis plan, or enough missing-data detail to support a more complete quantitative interpretation. These limits are about what can be concluded from the registry record; they are not evidence for or against a particular intervention effect.

15. Why This Trial Matters Statistically

This trial raises a statistical question that is common in rehabilitation research: how to compare an intervention delivered at home with a clinic-based approach and a control period when the primary endpoint is a continuous performance measure. WMFT performance time is task-based, and the study analyzed logarithm-transformed time to address skewness. That makes the outcome and scale of analysis important to interpreting the result.

The record also highlights the distinction between enrollment and the analysis population. Enrollment was 63, while 24 participants were included in the primary outcome analysis because participants without week-6 follow-up data were excluded. For readers assessing an estimate, the number contributing outcome data and the handling of missing observations are central parts of the statistical evidence, not administrative details.

The inferential result illustrates why a p-value should be accompanied by an effect estimate and an uncertainty interval. The reported p-value of 0.64 does not convey the magnitude or precision of the comparison. The arm summaries give descriptive context but do not replace a direct contrast. Likewise, a reported test for the home-based comparison does not answer the separate question about the lab-based group.

Finally, the safety counts and secondary outcome listings form part of the broader record, but the registry does not describe a multiplicity strategy or the complete analysis plan. Statistical interpretation therefore depends not only on the selected endpoint and test, but also on transparent reporting of analysis populations, missing data, and the family of comparisons. Those details are not reported in the registry record.

16. 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 primary outcome analysis included 24 participants, compared with a total enrollment of 63. The registry says that participants with no 6-week follow-up data on the primary outcome were excluded, but does not give the number excluded, a full account of participant disposition, or the reasons for missingness. Missing-data methods and sensitivity analyses are not reported in the registry record.

The reported inferential result is limited to home-based BATRAC versus delayed-entry usual care. The record gives descriptive results for the lab-based group but does not report an inferential comparison involving that group. It also does not report a between-group effect estimate or confidence interval for the primary comparison. As a result, the size and precision of the comparison cannot be assessed directly from the posted inferential statistics.

Although the outcome description defines change from baseline, it also describes the primary study outcome as a between-group difference in post-training log time. The exact statistical model and whether it used change scores, post-training values, or a baseline adjustment are not reported in the registry record. Details of the t-test implementation, including variance assumptions, are also not reported.

The registry lists three outcome measures and four analyses, but does not give a full analysis plan or report a multiplicity procedure. Further trial-specific information about the randomization process, allocation concealment, masking role, eligibility criteria, protocol deviations and the relation between the safety groups and analysis groups is not reported in the registry record. These omissions limit assessment of precision, bias, missing-data assumptions and the scope of the reported comparison.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Translating Intensive Arm Rehabilitation in Stroke to a Telerehabilitation Format — NCT02665052.
• Registry record last updated 2021-03-17; status last verified 2021-02.
• Results first posted 2021-03-17.
• 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.

20. Record Summary

The registry record describes a completed, randomized, parallel-group trial with single masking and enrollment of 63. It compared home-based BATRAC plus TTT, lab-based BATRAC plus TTT, and delayed-entry usual care. The primary endpoint was change in WMFT performance time from baseline to week 6, analyzed on a log-time scale.

The primary outcome analysis included 24 participants: 8 in the home-based group, 7 in the lab-based group and 9 in delayed-entry usual care. The reported home-based versus usual-care analysis was a two-sided t-test with p = 0.64. No between-group effect estimate or confidence interval is reported in the registry record. The registry also lists two secondary outcomes and safety counts, including 0 serious adverse events in each listed group.