Effectiveness of Virtual Reality Gaming Therapy Versus CI Therapy for Upper Extremity Rehabilitation: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effectiveness of Virtual Reality Gaming Therapy Versus CI Therapy for Upper Extremity Rehabilitation (NCT02631850), a randomized na study. The primary outcome is Wolf Motor Function Test.

2. Clinical Question

The trial addresses a delivery question as well as a treatment question: whether a home-based gaming format, with or without added remote therapist contact, can be compared with clinic-based therapy and a lower-dose occupational or physical therapy approach. The registry describes a multi-site randomized controlled trial and lists four interventions. Traditional clinic-based constraint-induced movement therapy is the primary comparison arm in the posted results.

The primary endpoint is the change in the natural-log-transformed time needed to complete standardized upper-extremity tasks. The registry’s timeframe is 0 to 1 months. The outcome is intended to reflect proportional change in performance time; negative change represents improvement on this scale, while positive change represents worsening. The registry describes a change of -0.17 as clinically meaningful.

The registry summary also describes a six-month follow-up and a crossover for people assigned to standard occupational or physical therapy. The primary results concern the one-month outcome, not that later follow-up or crossover. Outcomes beyond the one-month timeframe are not reported in the registry record.

3. Trial Design

NCT02631850 is listed as a completed, randomized, parallel-group trial with single masking. The registry lists an enrollment of 193 and identifies Ohio State University as the sponsor. The study start is recorded as 2015-11 and primary completion as 2019-09. The registry does not identify who was masked, and further details of the masking procedure are not reported in the registry record.

Four behavioral interventions were compared. Traditional clinic-based constraint-induced movement therapy involved 35 therapist/client contact hours. The gaming approach used 15 hours of independent in-home game practice and 5 hours of clinic contact. A second gaming arm added 2.6 hours of therapist contact by video conference. The occupational or physical therapy arm involved 5 hours of therapist/client contact. These are the treatment descriptions in the registry summary; further details of delivery and adherence are not reported in the registry record.

The brief summary says participants were randomized to one of the four interventions. It also describes a six-month follow-up, after which the standard occupational or physical therapy group would cross over to a modified gaming therapy condition. The registry record does not provide results for that later crossover period in the posted primary outcome data.

The primary measure was the Wolf Motor Function Test, assessed from baseline to one month. The posted outcome population is described as modified intent-to-treat, defined in the registry as those who started treatment. The arm-level outcome counts differ across measures, so the count for the primary outcome should not be assumed to represent every randomized participant or every secondary outcome. Reasons for missing outcome data and the handling of missing measurements are not reported in the registry record.

Although the registry describes the trial as randomized and multi-site, details of the randomization procedure, allocation concealment, site-level analysis, and the original statistical analysis plan are not reported in the registry record. The trial’s original sample-size justification is also 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: 193 participants met enrollment criteria and signed a consent form during the screening visit. 14 withdrew shortly thereafter, citing logistical challenges (e.g., scheduling, transportation). 5 no-showed to the first treatment session and could not be reached. 4 experienced medical events that prompted them to withdraw prior to beginning treatment. 2 could not be contacted to schedule participation. The randomization assignment was recycled back into pool for those who did not begin treatment.

Treatment Period

Reasons for not completing this period:

6 Month Follow up

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Modified intent-to-treat (those who started treatment)

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: Includes all eligible participants that began treatment.

6. Primary Endpoint

The registry lists the primary outcome as Wolf Motor Function Test, measured over 0 to 1 months.

The primary endpoint is mean change in Wolf Motor Function Test performance time from baseline to one month, expressed on the natural-log scale. The test assesses time to complete 15 standardized tasks, including examples such as folding a towel, stacking checkers, and placing a hand on a box. The registry says that tasks a participant cannot accomplish receive a score of 120 seconds.

Taking the natural logarithm of performance time is intended to represent proportional rather than simply absolute change and to address skew in task-completion times. On the posted change scale, -4.79 is described as the best possible improvement, 0 as no improvement, and positive values as worsening. The registry says a mean change of -0.17 corresponds to a proportional improvement of 16% and is considered clinically meaningful. These descriptions help interpret direction and scale; they do not establish that any particular arm achieved a clinically meaningful effect.

A timed functional test gives a standardized way to summarize task performance across participants and treatment arms. The mean describes the average change in the measured scale, while the standard deviation describes how dispersed individual changes are around that mean. The registry outcome population includes people who started treatment, rather than specifying all randomized participants. The endpoint is reported at one month; later primary-outcome measurements are not reported in the registry record.

7. Results Posted in the Registry

• Gaming CI Therapy: mean -0.24 versus -0.38 on control; difference 0.14 (95% CI -0.00 to 0.28), standardized effect d = 0.41; z-test p = 0.058.
• Gaming CI Therapy With Additional Contact Via Video Conference: mean -0.29 versus -0.38 on control; difference 0.09 (95% CI -0.05 to 0.23), standardized effect d = 0.26; z-test p = 0.221.
• Traditional Occupational Therapy/Physical Therapy: mean -0.22 versus -0.38 on control; difference 0.16 (95% CI 0.00 to 0.32), standardized effect d = 0.45; z-test p = 0.045.

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 Neuro-Quality of Life (Neuro-QOL)

• 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:
  • Gaming CI Therapy: mean -0.05 versus 0.69 on control; difference -0.74 (95% CI -1.56 to 0.08), standardized effect d = -0.39; z-test p = 0.078.
  • Gaming CI Therapy With Additional Contact Via Video Conference: mean -0.72 versus 0.69 on control; difference -1.41 (95% CI -2.39 to -0.43), standardized effect d = -0.61; z-test p = 0.005.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Gaming CI Therapy and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Gaming CI Therapy With Additional Contact Via Video Conference than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Bilateral Activity Monitors

• 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:
  • Gaming CI Therapy: mean 0.00 versus 0.12 on control; difference -0.12 (95% CI -0.27 to 0.03), standardized effect d = -0.51; z-test p = 0.119.
  • Gaming CI Therapy With Additional Contact Via Video Conference: mean 0.05 versus 0.12 on control; difference -0.07 (95% CI -0.21 to 0.07), standardized effect d = -0.34; z-test p = 0.310.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Gaming CI Therapy and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Gaming CI Therapy With Additional Contact Via Video Conference and the control. The interval gives the range of effects that remain compatible with the data.

Change in Brief Kinesthesia Test (BKT)

• 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:
  • Gaming CI Therapy: mean -0.90 versus -1.80 on control; difference 0.90 (95% CI -1.49 to 3.29), standardized effect d = 0.16; z-test p = 0.461.
  • Gaming CI Therapy With Additional Contact Via Video Conference: mean 0.80 versus -1.80 on control; difference 2.60 (95% CI 0.15 to 5.05), standardized effect d = 0.46; z-test p = 0.037.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Gaming CI Therapy and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Gaming CI Therapy With Additional Contact Via Video Conference than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary endpoints broaden the outcome picture beyond timed motor-task performance. They include Neuro-QOL, bilateral arm-use monitoring, the Brief Kinesthesia Test, the Semmes-Weinstein Monofilament Test, and the 9 Hole Peg Test. Posted results are available for some, but not all, of the listed secondary measures. The available outcomes cover different constructs and use different scales, so their numerical estimates are not interchangeable with the primary endpoint.

Across the posted comparisons, secondary findings are not uniform. Some estimates are near zero or have intervals that include zero; others have intervals that exclude zero. This mixed pattern does not supply a single consistent account that either confirms or overturns the primary-outcome comparisons. Secondary measures can be useful for generating questions about particular aspects of function, but they should be interpreted alongside their own outcome definitions, participant counts, and uncertainty.

The primary outcome measures change in standardized task-completion time, whereas the secondary endpoints address quality of life, spontaneous arm use, sensory or kinesthetic measures, and dexterity. Differences in the constructs and scales mean that a result on one measure need not move in parallel with another. The secondary outcomes therefore complement, rather than directly validate, the Wolf Motor Function Test result.

Unadjusted secondary results carry less weight as confirmatory evidence when many outcomes and comparisons are examined. Without a reported multiplicity strategy, nominal p-values do not account for the full set of opportunities to observe an apparently unusual result. The registry record does not report a prespecified hierarchy of secondary outcomes or adjusted analyses. Interpret secondary findings as exploratory descriptions of the posted data, not as independent proof of a broader treatment effect.

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

• Change in Neuro-Quality of Life (Neuro-QOL)
• Bilateral Activity Monitors
• Change in Brief Kinesthesia Test (BKT)
• Semmes-Weinstein Monofilament Test
• 9 Hole Peg Test

9. Statistical Interpretation

10. Statistical Methods Explained

What does the mean change represent?

The primary outcome is a continuous measure: each participant has a value for change in natural-log-transformed task-completion time, and the arm mean averages those changes. The scale’s direction matters. More negative change indicates greater improvement in performance time; a positive value indicates worsening. Because the measure is log transformed, the registry describes it as reflecting proportional improvement, rather than a fixed number of seconds saved. This interpretation follows the registry’s description of the endpoint.

The arm-level standard deviation describes variability among participants’ changes. It is not the uncertainty around the mean itself. A standard error, by contrast, describes sampling uncertainty in an estimated mean. The recalculated comparisons use the posted means, standard deviations, and arm counts to estimate a difference between means and its uncertainty.

How should the comparisons be read?

Each posted primary comparison uses traditional clinic-based constraint-induced movement therapy as the reference and compares its mean change with one other arm. The reported mean difference is the compared arm’s mean minus the reference arm’s mean. For this outcome, a positive difference means the compared arm’s mean change is less negative, and thus reflects less average improvement on the scale than the reference mean. A negative difference would indicate a more negative mean change in the compared arm.

The confidence interval expresses uncertainty around the estimated difference. If an interval includes zero, the data are compatible with no mean difference as well as with differences in either direction within the interval. If an interval excludes zero, it indicates that zero is outside that interval under the calculation used. This is not, by itself, a measure of clinical importance. The size and direction of the estimate, the interval’s width, the outcome scale, and the registry’s clinical-meaningfulness reference all matter to interpretation.

The p-values attached to these comparisons are unadjusted. In general, a p-value describes how unusual the observed result, or a more extreme result, would be under a specified no-difference model and its assumptions. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of an effect. Multiple unadjusted comparisons increase the chance of seeing a nominally unusual result by chance across a collection of tests. The registry record does not report a multiplicity plan or an adjusted analysis.

Why is the log scale used?

General statistical explanation: time measurements often have a right-skewed distribution, with some much longer task times than others. A logarithmic transformation can reduce the influence of large values and make proportional changes more suitable for analysis. It also changes how differences are interpreted: a difference on the log scale is not a raw difference in seconds. The registry explicitly describes this transformation as a way to reflect proportional improvement and correct for skew.

The outcome rules also matter. A task that cannot be accomplished is assigned 120 seconds before the times are transformed, according to the registry description. This creates a defined value for inability to complete a task rather than leaving the task time unrecorded. The registry does not report additional details about how such task scores were combined or how incomplete test records were handled.

What does the recalculation add?

The comparison estimates and intervals are simple unadjusted recalculations from the posted arm-level summaries. They are useful for describing the data in a common pairwise format, but they do not reproduce an unknown prespecified model. The registry does not report a detailed primary analysis method, covariate adjustments, site effects, stratification factors, handling of repeated measures, or a missing-data method. Those features can affect estimates and uncertainty, so the recalculations should not be presented as the sponsor’s analysis.

The primary comparison intervals use the posted outcome counts and standard deviations. No discrepancy is flagged between the recalculations and a sponsor analysis; however, no sponsor-reported primary analysis is listed in the record. The recalculated p-values therefore remain descriptive, unadjusted results based on the available summaries.

What does power mean in the planning illustration?

General statistical explanation: power is the probability, under a specified assumed effect and statistical model, that a study will reject a no-difference hypothesis when that assumed effect is present. The planning illustrations describe the number of participants per arm that would be needed under stated assumptions. They are not evidence about the probability that this completed trial detected an effect. A smaller assumed effect generally requires more participants to achieve the same target power, because it is harder to distinguish from random variation.

The posted planning illustration uses a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It considers the full observed effect and smaller fractions of that observed effect. These are illustrative assumptions, not the trial’s original design assumptions. The original sample-size justification and original statistical design assumptions are not reported in the registry record.

How should the secondary outcomes inform interpretation?

The secondary outcomes cover quality of life, arm use, kinesthesia, sensation, and dexterity. Their measures and posted counts differ, and several are available for fewer participants than the primary outcome. They can add descriptive context, but they do not replace the primary endpoint or resolve uncertainty in its comparisons. The record does not list reported analyses for the secondary endpoints, nor does it report a plan for controlling the overall chance of false-positive findings across endpoints and comparisons.

11. Confidence Intervals and Precision

Precision concerns how much uncertainty surrounds an estimated difference. The recalculated confidence intervals for the primary comparisons provide a range of values compatible with the posted summaries under the calculation used. The gaming-only interval includes zero, as does the interval for gaming therapy with added video-conference contact. The occupational or physical therapy interval excludes zero. The record’s intervals are unadjusted and are not results of a reported sponsor analysis.

The intervals are not equally informative simply because each is a confidence interval. Their width reflects the estimated variability and the number of participants with outcome data in the compared arms. Wider intervals permit a broader range of possible differences; narrower intervals indicate greater precision. The primary outcome standard deviations are similar across arms, while arm counts range from 38.0 to 45.0. These sample sizes leave uncertainty around each difference, as reflected in the intervals.

For the two gaming comparisons, the intervals include both zero and values on either side of zero. Thus, the posted summaries do not sharply distinguish no mean difference from a difference favoring either arm on the scale. The interval for occupational or physical therapy excludes zero, but this is an unadjusted recalculation among multiple comparisons. An interval excluding zero does not by itself establish clinical importance, and the registry does not report an adjusted confirmatory analysis.

The registry’s clinical-meaningfulness reference is a mean change of -0.17 for improvement. That reference applies to the scale’s change, not directly to the pairwise mean differences. The precision of the interval and its relationship to zero should therefore be considered separately from whether an arm’s mean change meets that reference. The registry does not report confidence intervals for the arm-level means themselves.

These intervals capture statistical uncertainty under the recalculation, not every source of uncertainty. They do not account for unreported choices in the original analysis plan, missing-data handling, site structure, or multiplicity. More detailed sponsor-reported analyses are not reported in the registry record.

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 participants per arm for 80% power and participants per arm for 90% power. They are not total sample sizes. The registry’s planning illustration assumes a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It uses the observed effect as a reference and also considers planning for 75% and 50% of the observed effect.

The 80% power column gives the illustrative number of participants per arm under which the assumed design would have an 80% chance of detecting the specified effect, if that effect and the planning assumptions held. The 90% column uses a higher target probability of detection and therefore generally calls for more participants per arm. For example, the primary gaming-only comparison lists 93 participants per arm for 80% power and 124 for 90% power when planning for 100% of the observed effect. The other rows show how the illustrative requirements change with the comparison and assumed effect fraction.

Smaller effects are harder to distinguish from random variation, so the planning figures increase when the target is a smaller fraction of the observed effect. These values are planning illustrations based on the observed comparisons, not the original trial design. The trial’s original sample-size assumptions and justification are not reported in the registry record. This is not observed power: the figures do not measure the chance that the completed trial detected its true effect.

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 = 0.140; pooled standard deviation σ = 0.340.
2. n = 2 × (1.96 + 0.842)² × 0.340² / 0.140² = 92.5, rounded up to 93 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 <- 0.14                      # observed difference in means
sigma <- 0.339785545406908                     # 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(2631850)
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 = 93 (80% power), 124 (90% power)
 75% of observed effect: n per arm = 165 (80% power), 221 (90% power)
 50% of observed effect: n per arm = 370 (80% power), 496 (90% power)
power.t.test: 93.4 
Simulated power at n = 93 per arm: 0.800 (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

Several intervention arms are compared with the clinic-based constraint-induced therapy arm, and the registry also lists multiple secondary outcomes. Testing many comparisons creates a multiplicity problem: even if there is no real difference for any individual comparison, a collection of tests can produce one or more nominally unusual results by chance. Adjustment methods or a prespecified testing hierarchy can address this issue, but the registry record does not report whether such a plan was used.

The primary comparison p-values are unadjusted recalculations from posted summaries. The secondary comparisons are also presented as unadjusted recalculations, and no reported secondary analyses are listed. The registry does not state which outcomes or comparisons were considered confirmatory, whether any adjustment was planned, or how the overall type I error was controlled. Accordingly, nominal p-values should not be treated as if they reflect a single, isolated test.

Missing-data methods are another important unreported part of the analysis. The primary outcome population is described as modified intent-to-treat, meaning participants who started treatment. This differs from a population defined as everyone randomized. The record does not explain how many randomized participants did not start treatment, why outcome observations are absent, whether missingness differed by arm, or what method was used to handle missing measurements. Outcome counts also vary across measures, reinforcing that the available sample is not identical for every analysis.

The registry does not report the primary statistical model, covariate adjustments, handling of site or repeated measurements, sensitivity analyses, or a detailed analysis plan. The posted pairwise estimates and intervals are simple unadjusted recalculations rather than a reported sponsor analysis. These omissions limit the ability to assess how robust the estimates are to alternative analytic choices. They do not show that a particular method was or was not used; those details are not reported in the registry record.

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 posted serious adverse event figures show zero serious adverse events in each of the four intervention groups, and zero deaths in each group. The numbers at risk for serious events are 41, 45, 44, and 38 across the listed groups. No serious adverse event terms are listed.

For other, non-serious adverse events, one event is listed in the gaming constraint-induced movement therapy group, with zero in each of the other groups. The corresponding numbers at risk are 45, 44, 44, and 38 across the groups as posted for other adverse events. These safety counts are not the same as the primary outcome analysis counts in every arm.

These figures describe the events recorded in the registry, not a complete safety comparison. The registry record does not provide event descriptions, severity details beyond the serious/non-serious classification, attribution, timing, exposure-adjusted rates, or the method of event collection. Zero recorded serious events does not establish that an intervention has no risk. The counts should be read as reported observations within this trial record, not as a basis for broader safety conclusions.

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

The primary outcome data describe mean change in a timed functional measure over the interval from baseline to one month. Each arm’s mean change is negative, which the registry defines as improvement on the natural-log scale. The recalculated comparisons show that the two gaming arms had less negative mean changes than traditional clinic-based constraint-induced movement therapy, while the traditional occupational or physical therapy arm also had a less negative mean change. The uncertainty differs across these comparisons, and the estimates are unadjusted.

These results do not establish that one approach is better or worse in a clinical sense. The comparisons are based on posted arm-level summaries rather than a sponsor-reported primary analysis, and the record does not report a multiplicity strategy, detailed missing-data methods, or the original statistical plan. One unadjusted interval excludes zero, but this finding must be interpreted in the context of several arms and outcomes. Statistical compatibility with a difference is not equivalent to evidence that the difference is large enough to matter to participants.

The registry’s stated clinically meaningful reference is a change of -0.17 on the primary outcome scale. That threshold helps frame arm-level changes, but it does not by itself answer whether the difference between two arms is clinically meaningful. Mean change and a between-arm contrast are distinct quantities. Individual experiences may also vary around each arm mean, as indicated by the posted standard deviations.

The results apply to the participants included in the posted one-month outcome population, described as those who started treatment. They do not provide a complete account of longer-term outcomes, adherence, the six-month crossover, or outcomes outside the registry’s reported measures. The data are best read as a description of short-term outcome patterns and the precision of unadjusted comparisons, rather than a definitive ranking of treatment strategies.

17. Why This Trial Matters Statistically

Statistically, this trial is useful because it puts distinct therapy-delivery approaches into a randomized, parallel-group comparison. The interventions vary in setting, therapist contact, and use of independent home gaming. That structure makes clear that a comparison of “gaming” with “clinic therapy” can involve more than the technology itself: treatment format and contact time differ across the trial arms.

The primary outcome is a continuous, transformed measure with a stated direction and clinical-meaningfulness reference. This offers a chance to distinguish the average change within an arm from the difference between arms, and to consider whether an estimate is precise enough to rule out a range of plausible effects. The posted results illustrate why looking only at arm means or p-values is incomplete; the confidence intervals and scale interpretation also matter.

The record also illustrates the limits of registry-level results. The posted pairwise estimates are unadjusted recalculations, while key details about the original statistical plan, multiplicity, and missing-data handling are not reported. Those omissions constrain how confidently nominal findings can be treated as confirmatory. Variation in outcome counts across measures also matters when comparing the apparent strength of results.

Finally, the trial’s multiple outcomes help frame different dimensions of recovery, but they require careful interpretation rather than a search for a single favorable result. The combination of randomized design, several active approaches, a clinically interpretable primary scale, and incomplete reporting of analytic details makes the study informative for understanding both the potential and the limits of registry-level statistical evidence.

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 primary outcome analyses include participants who started treatment, according to the registry’s modified intent-to-treat definition. The record does not report how this group relates to all randomized participants, why some participants may be absent from outcome data, or how missing measurements were handled. These gaps limit assessment of possible bias from incomplete outcome data.

The posted comparisons are unadjusted recalculations from group summaries, not a detailed sponsor-reported analysis. The registry record does not report the original primary analysis model, covariate adjustments, site-level handling, or sensitivity analyses. Although the study is described as multi-site, the posted summaries do not show whether or how site differences were considered.

There are several arms and secondary outcomes, but a multiplicity plan or hierarchy is not reported. The unadjusted comparisons therefore should not be treated as though they were the only tests conducted. Secondary outcome counts are smaller than or different from primary outcome counts, and posted results are not available for every listed secondary measure.

The registry summary describes a six-month follow-up and crossover from standard occupational or physical therapy to a modified gaming condition, but the primary results discussed are limited to the one-month timeframe. Later follow-up results and details about treatment adherence are not reported in the registry record. The original sample-size rationale and assumptions are also not reported; the available planning figures are illustrative calculations from observed effects, not the trial’s design basis.

Finally, the reported endpoint is an average on a transformed scale. Arm means and standard deviations do not show the full distribution of individual responses, and the registry does not provide arm-level confidence intervals or detailed task-level results. These limitations constrain conclusions about heterogeneity and clinical importance.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Effectiveness of Virtual Reality Gaming Therapy Versus CI Therapy for Upper Extremity Rehabilitation — NCT02631850.
• Registry record last updated 2021-10-05; status last verified 2021-09.
• Results first posted 2021-09-14.
• 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

NCT02631850 is a completed randomized, parallel-group trial of four behavioral approaches for chronic upper-extremity hemiparesis after stroke. The registry lists an enrollment of 193 and an outcome timeframe of baseline to one month. The primary endpoint is mean change in natural-log-transformed Wolf Motor Function Test performance time, with negative values indicating improvement.

The posted arm means are negative in all four groups. Unadjusted recalculated comparisons with traditional clinic-based constraint-induced movement therapy have differing precision; the occupational or physical therapy comparison’s interval excludes zero, while both gaming comparisons’ intervals include zero. The record does not list a sponsor-reported primary analysis.

Secondary measures address quality of life, arm use, kinesthesia, sensation, and dexterity. The record does not report a multiplicity plan, detailed missing-data methods, or the original sample-size justification. Serious adverse events and deaths are listed as zero in all four groups; one other adverse event is listed in the gaming-only group.