The Efficacy of Intensive Nutritional Supplement in Patient With Stroke: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Efficacy of Intensive Nutritional Supplement in Patient With Stroke (NCT01810263), a randomized phase 2, phase 3 study. The primary outcome is Modified Barthel Index (MBI) Score at 6 Months.

2. Clinical Question

3. Trial Design

The record describes a prospective randomized controlled trial, with phase listed as PHASE2 and PHASE3. Its design is recorded as randomized, double-masked, and parallel. The sponsor is Seoul National University Bundang Hospital. Enrollment was 34 participants, allocated to a high protein supplement arm and a control arm described as receiving no intervention.

The listed study conditions are stroke and malnutrition. Further details about the enrolled population and eligibility criteria are not reported in the registry record. Although masking is recorded as double, the record does not specify which participants or study personnel were masked, or how masking was maintained.

The recorded start was 2012-05 and primary completion was 2014-07. The current status is TERMINATED. The reason for termination and whether termination affected enrollment, follow-up, or data collection are not reported in the registry record. The record does not describe the supplement regimen or how adherence was assessed.

The primary outcome is the Modified Barthel Index (MBI) Score at 6 Months. Group results are reported for 16 participants in the high protein supplement arm and 18 in the control arm. Five outcome measures are listed, and Chemical Laboratory Findings is identified as a secondary outcome. Detailed results for that secondary outcome are not included in the registry record. The trial’s original sample-size justification, planned analysis methods, and reasons for choosing its enrollment target are not reported.

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

Reasons for not completing this period:

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 Modified Barthel Index (MBI) Score at 6 Months, measured over 6 months.

The primary endpoint is the Modified Barthel Index (MBI) score at 6 months, reported in units on a scale. The record describes the scale range as 0-100 and states that higher values represent a better outcome. It reports a mean and standard deviation for each arm.

The MBI is used to summarize aspects of functional ability. As a continuous score, it can be described using group averages and the spread of individual scores around those averages. A difference between group means gives a concise comparison of average scores at the recorded follow-up time. It does not show the full distribution of scores, how many participants improved or worsened, or how individual changes varied.

The registry record does not describe the specific MBI version, the assessment procedures, who completed the assessment, or how missing six-month scores were handled. It also does not report a threshold for a clinically important difference. Therefore, the numerical difference can be interpreted as a group-average contrast on the stated scale, but its practical importance cannot be determined from the record alone.

7. Results Posted in the Registry

• High Protein Supplement: mean 23.30 versus 21.90 on control; difference 1.40 (95% CI -0.42 to 3.22), standardized effect d = 0.52; z-test p = 0.131.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: each participant has an MBI score, and the registry summarizes those scores with a mean and standard deviation in each arm. A mean gives the arithmetic average of the recorded scores. A standard deviation describes how dispersed individual scores are around the mean. It is not a measure of uncertainty in the mean itself; uncertainty about a group mean or a difference between group means is reflected in a standard error or an interval estimate.

What does the mean difference represent?

The posted comparison is a mean difference of 1.4 units, calculated in the direction of high protein supplement minus control. A positive value indicates that the observed mean was higher in the supplement arm. Since the scale description says higher values represent a better outcome, the direction of this observed contrast favors a higher average MBI score in that arm. This is a descriptive direction, not a conclusion about whether the difference is reliable or important.

Comparing means is a common way to summarize a continuous outcome when the research question concerns average scores. The difference retains the scale’s original units, which can make it easier to interpret than a standardized measure. However, an average does not convey every feature of the data. Two arms can have similar means while differing in their score distributions, and a mean can be affected by unusually high or low observations. Participant-level scores and distributional summaries are not reported in the registry record.

How should the p-value be understood?

In general, a p-value measures how unusual data at least as discrepant as those observed would be under a specified null hypothesis and a statistical model. Its interpretation depends on the analysis method and its assumptions. The registry lists an unadjusted p-value of 0.131, but the specific test, model, and assumptions used for the trial analysis are not reported. The p-value alone does not indicate the probability that the supplement has no effect, nor does it quantify the size of the observed difference.

The record also does not state whether the comparison was adjusted for baseline MBI score or other characteristics. It does not report whether the analysis followed an intention-to-treat approach, included only participants with observed six-month measurements, or used another analysis population. These details matter because the handling of baseline values, protocol deviations, and missing outcomes can affect an estimate and its uncertainty.

What can an interval estimate add?

In general, a confidence interval expresses uncertainty around an estimated quantity under a statistical procedure and its assumptions. For a difference in means, an interval that includes zero includes no difference among the values compatible with the interval’s construction. The posted interval extends from below zero to 3.218, so it includes zero and positive values. The interval’s confidence level and the method used to calculate it are not reported in the registry record.

An interval is not a range containing a known proportion of individual participants’ scores. It concerns uncertainty in the estimated contrast, not the spread of individual outcomes. It also does not make every value in the interval equally plausible, and it cannot establish clinical importance without a relevant threshold. No such threshold is reported for the MBI endpoint in this record.

Why does variation matter?

In general, greater variability among individual observations makes a group mean less precise when the number of observations is fixed. The standard deviations reported for both arms are 2.7. The groups also have different numbers of participants: 16 in the supplement arm and 18 in control. The interval estimate reflects uncertainty in the comparison, but the record does not identify the exact calculation method that produced it.

Over-dispersion is a general concern when observed variability exceeds what a chosen statistical model expects. If a model fails to account for extra variability, its uncertainty estimates can be too small. This record does not report a count outcome or a model for count data, so over-dispersion is not an identified issue for the primary continuous MBI comparison. The model assumptions and diagnostic checks for the trial’s analysis are not reported.

What is not documented about the analysis?

The registry record does not report the original statistical analysis plan, the test or model used for the primary endpoint, any adjustment for multiple outcomes, or whether interim analyses were conducted. It does not report the planned analysis population, treatment of missing data, handling of protocol deviations, or any prespecified subgroup analyses. These elements should not be inferred from the randomized, double-masked design or from the posted means.

The posted comparison gives an estimate, interval, and unadjusted p-value. Those summaries can be read as the registry’s reported contrast, but they do not replace details of the analysis procedure. Any recalculation based only on summary statistics would be unadjusted and would not recover methods or participant-level information that the registry record does not report.

10. Confidence Intervals and Precision

The mean difference is 1.4 MBI units, with a posted interval that extends from below zero to 3.218. The lower endpoint is negative, so the interval includes zero; its upper endpoint is positive. This means the estimate is not precise enough to exclude both no difference and a positive difference in the direction of a higher mean in the supplement arm, under the interval’s construction.

In general, a confidence interval describes uncertainty in an estimate, conditional on the statistical method and its assumptions. It does not describe the range of individual participant scores. The registry record does not report the confidence level or calculation method for this interval. Accordingly, its endpoints can be described as posted, but the interval should not be attributed to an unreported confidence level or analysis procedure.

The arm standard deviations are both 2.7, while the reported arm sizes are 16 and 18. These summaries indicate that the estimate comes from a small set of observations, but they do not show the underlying score distributions. The interval’s inclusion of zero should not be interpreted as proof that the groups are the same. Nor does the interval alone establish that any positive difference would be clinically meaningful; a threshold for clinical importance is not reported.

Precision and statistical significance are related but distinct ideas. A p-value summarizes compatibility with a null model under a particular analysis; an interval conveys a range of values consistent with its method. Here, the unadjusted p-value is 0.131, and the interval includes zero. The analysis method and its assumptions are not reported in the registry record, limiting how specifically either statistic can be interpreted.

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 events. They are illustrations calculated from the observed mean difference, not the original trial’s design justification. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. The trial’s own sample-size assumptions are not reported in the registry record.

For 100% of the observed effect, the illustration gives 59 participants per arm for 80% power and 79 participants per arm for 90% power. For 75% of the observed effect, it gives 104 participants per arm for 80% power and 139 participants per arm for 90% power. For 50% of the observed effect, it gives 234 participants per arm for 80% power and 313 participants per arm for 90% power.

In general, power is the probability that a study will detect an effect of a specified size under the assumptions of the calculation. The 80% and 90% columns represent different target probabilities for detecting the specified effect, if that effect exists and the planning assumptions hold. The 90% target calls for more participants per arm than the 80% target at each listed effect fraction. Smaller effects require more participants because they are harder to distinguish from random variation.

These figures are planning illustrations, not observed power and not evidence that the trial was designed to meet any of these targets. The recorded arm sizes were 16 and 18. The illustration’s assumptions, including equal allocation, differ from the observed arm counts. The registry record does not report the trial’s original sample-size justification or whether the study was planned around the effect observed in the posted results.

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 = 1.400; pooled standard deviation σ = 2.700.
2. n = 2 × (1.96 + 0.842)² × 2.700² / 1.400² = 58.4, rounded up to 59 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 <- 1.4                      # observed difference in means
sigma <- 2.7                     # 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(1810263)
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 = 59 (80% power), 79 (90% power)
 75% of observed effect: n per arm = 104 (80% power), 139 (90% power)
 50% of observed effect: n per arm = 234 (80% power), 313 (90% power)
power.t.test: 59.4 
Simulated power at n = 59 per arm: 0.798 (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 record lists five outcome measures and identifies Chemical Laboratory Findings as a secondary outcome. It reports the MBI score at 6 months as the primary outcome. The number of analysis entries is recorded as zero, and the posted registry record does not describe a statistical analysis plan or a multiplicity adjustment. Detailed results for the named secondary outcome are not reported in the registry record.

Multiplicity arises when several outcomes, time points, comparisons, or subgroups are tested. In general, making many statistical tests can increase the chance of at least one apparently unusual result even when none of the underlying effects is present. Methods for addressing this include prespecifying a primary analysis or adjusting interpretation for multiple comparisons. The record does not report whether such methods were planned or used, so no adjustment should be assumed for the reported unadjusted p-value.

Missing outcome data can also affect interpretation. The registry provides arm counts for the posted MBI summaries, but does not explain whether these counts include all randomized participants with available measurements, how many six-month measurements were missing, or why any data were unavailable. It does not report an imputation method, sensitivity analysis, or other missing-data procedure. The effect of missingness on the estimate therefore cannot be assessed from the record.

The analysis population is not specified. In general, an intention-to-treat analysis retains participants in their assigned groups, while other analysis populations can apply different inclusion rules. The record does not state which population produced the reported means and comparison, how adherence or protocol deviations were handled, or whether all randomized participants were included. These are not details that can be inferred from randomization alone.

The registry record also does not report interim analyses, subgroup analyses, or the original statistical analysis plan. Their presence or absence should not be guessed. Without those details, the primary comparison can be described as posted, but its relationship to prespecified analysis decisions and any multiplicity strategy remains unclear.

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 posted safety figures list 0 serious adverse events affecting participants in the high protein supplement arm, among 16 at risk, and 0 in the control arm, among 18 at risk. The figures also list 0 other adverse events affecting participants in each arm, with 16 and 18 participants at risk, respectively.

These are counts in the registry record. They do not establish that no adverse events occurred outside the categories or reporting period represented by these entries. The record does not describe how adverse events were defined, collected, monitored, or attributed, or whether all participants were followed for safety for the same duration.

Deaths are not reported in the registry record; the fields for deaths are blank rather than recorded as zero. No serious adverse event terms are listed. With the available information, the safety entries can be described as posted, but they do not permit a complete assessment of safety or a detailed comparison of risk between arms.

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

The result describes a difference in six-month mean MBI scores among the participants represented in the registry outcome data. The supplement arm’s mean was 23.3 and the control arm’s mean was 21.9; the posted mean difference was 1.4 units. The scale ranges from 0-100, with higher values representing a better outcome. The interval estimate includes zero, and the unadjusted p-value is 0.131.

These statistics do not establish that the arms had the same outcomes, nor do they establish a dependable or clinically important advantage for either arm. A nonzero point estimate is compatible with uncertainty, and an interval including zero does not prove equivalence. The registry record does not report an MBI threshold for clinical importance or an equivalence margin. The reported contrast should therefore be read as an estimate with uncertainty, rather than as a definitive answer about clinical benefit.

The endpoint is one measure at one recorded follow-up time. It does not describe other aspects of health, laboratory outcomes, changes from baseline, or individual trajectories. Although Chemical Laboratory Findings is listed as a secondary outcome, its detailed results are not reported in the registry record. The safety entries list no affected participants in the posted serious and other adverse event categories, but deaths are not reported and the safety collection methods are not described.

The trial record reports 34 enrolled participants and a TERMINATED status. The reason for termination, completeness of follow-up, missing outcome data, and analysis population are not reported. These gaps limit assessment of how representative the reported outcome summaries are of all randomized participants. The record also does not describe eligibility criteria in enough detail to determine how well the results would apply to a broader population.

In sum, the registry reports a modest positive difference in the observed group means, but the estimate is uncertain and the record lacks key analysis and follow-up details. The result should not be extended beyond the measured outcome and the participants represented in the posted summaries.

16. Why This Trial Matters Statistically

This trial illustrates why a point estimate should be read together with its uncertainty. The observed mean difference is 1.4 MBI units, but the interval extends from below zero to 3.218. The estimate alone gives a direction and scale; the interval makes clear that the data do not pin down a narrow range on one side of zero. Reporting both avoids treating a single average contrast as a complete account of the evidence.

The record also shows how sample size and uncertainty are connected. Its planning illustration indicates that detecting smaller fractions of the observed effect would require more participants per arm under the stated assumptions. The observed enrollment was 34, with arm counts of 16 and 18. The illustration is not the trial’s original design rationale, and it cannot establish what the study was intended to detect. The original sample-size justification is not reported in the registry record.

Several design labels are available—randomized, double-masked, and parallel—but these do not answer every statistical question. The masking roles, analysis population, missing-data handling, analysis method, and multiplicity plan are not reported. Clear reporting of these decisions helps readers evaluate whether an estimate reflects a prespecified comparison and how robust it may be to alternative assumptions.

The record is also a reminder that a posted safety count is not a complete safety analysis. Zero affected participants are listed in the serious and other adverse event categories for both arms, while deaths are not reported. Definitions, observation periods, and collection methods are not described. Statistical interpretation depends not only on the numerical result but also on how outcomes and participants were defined and followed.

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 leaves important details about the enrolled population unspecified. It identifies stroke and malnutrition as conditions but does not report eligibility criteria, participant characteristics, or how malnutrition was defined. The degree to which the enrolled participants represent other populations cannot be assessed from the record.

The intervention description is limited to a high protein supplement, and the control is described as no intervention. The supplement’s composition, dose, schedule, treatment duration, adherence, and any co-interventions are not reported. The record identifies double masking but does not say who was masked or how masking was maintained.

The primary outcome summaries are available for 16 participants in the supplement arm and 18 in control, but the record does not explain whether these are all randomized participants with six-month measurements. Missing outcome counts, reasons for missingness, the analysis population, and methods for handling missing data are not reported. No baseline MBI results or adjusted analysis are described, so the posted comparison cannot be interpreted as a change-from-baseline contrast.

The specific statistical method, confidence level for the interval, prespecified significance threshold, and original sample-size justification are not reported. Neither an analysis plan nor an interim-analysis plan is described. The reason for the TERMINATED status is also not reported, leaving the effect of termination on trial conduct and completeness of follow-up unclear.

The registry identifies Chemical Laboratory Findings as a secondary outcome, but does not report its detailed results. It lists zero serious and other adverse events affecting participants, while deaths are not reported. Event definitions, ascertainment procedures, and safety follow-up are not described. These limitations restrict interpretation to the posted MBI summaries and safety counts, without resolving how complete or representative those data are.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: The Efficacy of Intensive Nutritional Supplement in Patient With Stroke — NCT01810263.
• Registry record last updated 2015-07-02; status last verified 2015-07.
• Results first posted 2015-07-02.
• 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

The record describes a prospective randomized controlled trial with PHASE2 and PHASE3 listed, a parallel design, and double masking. The sponsor is Seoul National University Bundang Hospital. Enrollment was 34, and the status is TERMINATED. The recorded start was 2012-05 and primary completion was 2014-07.

The comparison is high protein supplement versus control, described as no intervention. The primary endpoint is the MBI score at 6 months, with higher scores on the 0-100 scale representing a better outcome. The posted means are 23.3 and 21.9, and the mean difference is 1.4. The interval estimate includes zero, and the unadjusted p-value is 0.131.

The record does not report the trial’s original sample-size justification, analysis plan, analysis population, missing-data methods, or reason for termination. The named secondary outcome is Chemical Laboratory Findings, but its detailed results are not reported in the registry record.