Rivastigmine Study in Adolescents With Down Syndrome: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Rivastigmine Study in Adolescents With Down Syndrome (NCT01084135), a randomized phase 1, phase 2 study. The primary outcome is Vineland Adaptive Behavior Scales, Second Edition (Survey Interview Form).

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group trial with quadruple masking, registered as phase 1 and phase 2. The record identifies Duke University as the sponsor. The study began in 2009-11, had primary completion in 2014-02, and is listed as completed. Its official title describes a 20-week double-blind placebo-controlled trial in children ages 10–18 with Down syndrome. The registry enrollment is 42; that figure describes enrollment in the overall study, not the number contributing to every outcome analysis.

Participants were assigned to liquid rivastigmine or liquid placebo. The active-arm description gives a dose-titration schedule through the week 10 safety visit. Placebo volume increases were intended to match treatment-arm titration, and the liquid preparations were described as matched in consistency and taste. The registry does not report how randomization was generated or concealed, nor does it describe procedures for assessing whether masking was successful.

The primary outcome was change in the Vineland Adaptive Behavior Composite score from baseline to the final visit at week 20. The primary analysis population is described as people who completed the 20-week period, excluding 2 people whose forms were completed incorrectly. The posted primary group summaries each include 10 participants. The registry does not report why the analysis counts differ from overall enrollment, how many participants discontinued, or how any other missing endpoint data were handled.

The record also lists a secondary measure, the Behavior Rating Inventory of Executive Function-Preschool. Its timeframe is baseline and the final week 20 visit. The record does not report the original sample-size justification, the trial’s planned statistical analysis, interim analyses, or a multiplicity strategy. Those details 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: 42 subjects signed consent, 10 were screen failures, 1 withdrew consent prior to being assigned to an arm, 1 was withdrawn by PI after being assigned an arm for noncompliance, 30 completed study.

12 Week

20 Week

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All subjects who completed the 20 week period were included in analysis except for 2 subjects whose form was completed incorrectly.

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 Vineland Adaptive Behavior Scales, Second Edition (Survey Interview Form), measured over Baseline & Study termination (Week 20).

The primary endpoint is change in the Vineland Adaptive Behavior Scales, Second Edition, Survey Interview Form, Adaptive Behavior Composite score between baseline and the final visit at week 20. This instrument measures adaptive behavior in children, adolescents, and adults. Its overall composite is accompanied by age standard scores in four domains. The registry describes the composite as having a mean of 100 and a standard deviation of 15, with a range of 20 to 160. Higher scores indicate a higher level of adaptive functioning, and a rise from baseline to the final visit indicates improvement.

Using change focuses the comparison on how scores shifted during the study rather than only on the final scores. The measure is relevant to the stated study aim because the registry describes the aim in terms of functional abilities, including language, memory, and executive function, and the Vineland assesses adaptive behavior. This does not mean that the composite directly measures every ability named in the study summary.

The registry reports group means and standard deviations for the change scores. It says the analysis included people who completed the 20-week period, with 2 incorrectly completed forms excluded. It does not describe a separate adjusted model or explain how baseline values were used beyond calculating each participant’s change. Interpretation therefore centers on the posted change-score comparison and its uncertainty, rather than on an unreported analysis method.

7. Results Posted in the Registry

• Rivastigmine- Liquid Form: mean -1.70 versus 2.00 on control; difference -3.70 (95% CI -6.69 to -0.71), standardized effect d = -1.09; z-test p = 0.015.

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.

Behavior Rating Inventory of Executive Function-Preschool (BRIEF-P)

• 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:
  • Rivastigmine- Liquid Form: mean -3.60 versus -6.10 on control; difference 2.50 (95% CI -6.22 to 11.22), standardized effect d = 0.25; z-test p = 0.574.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Rivastigmine- Liquid Form and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists one secondary outcome: the Behavior Rating Inventory of Executive Function-Preschool (BRIEF-P), measured from baseline to the final week 20 visit. Its posted comparison is not clearly distinguishable from zero by its unadjusted confidence interval and p-value. The primary endpoint’s unadjusted comparison, by contrast, has an interval that lies below zero. The results therefore do not form a uniformly clear pattern across the posted primary and secondary outcomes.

The secondary measure addresses executive function, while the primary Vineland measure is an adaptive behavior composite. They are related to the study’s broad interest in functional abilities, but they are not interchangeable outcomes. A difference on one measure does not necessarily imply a similar difference on another, and the registry does not describe a formal joint analysis or a hierarchy linking the endpoints.

Unadjusted secondary results carry less weight when several outcomes are evaluated because each comparison creates an opportunity for a chance finding. General statistical explanation: if many hypotheses are tested, the chance of at least one apparently notable result can increase unless the analysis plan accounts for multiplicity. The record does not report an adjustment strategy, a prespecified hierarchy, or sponsor-reported secondary analyses. This makes it difficult to assess whether the secondary result was intended as confirmatory or exploratory.

The secondary outcome also has its own analysis count and variability, and its posted group counts differ from those for the primary endpoint. Those differences reinforce the need to interpret each measure according to its own participants, scale, and missing-data context. The record does not report why the analysis populations differ. The secondary result can provide context for the broader study question, but it should not be treated as independent confirmation of the primary comparison or as a definitive contradiction of it.

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

• Behavior Rating Inventory of Executive Function-Preschool (BRIEF-P)

9. Statistical Interpretation

10. Statistical Methods Explained

The posted primary endpoint is continuous: it is a numerical change score measured in units on a scale. The registry reports a mean and standard deviation for each group. A mean describes the average change, while a standard deviation describes how much individual change scores vary around that average. General statistical interpretation: the mean captures the group’s central tendency, but does not show the distribution of individual values or whether the data contain unusual observations.

What does the mean difference represent?

The reported mean difference compares the rivastigmine group’s average change with the placebo group’s average change. It is expressed as rivastigmine minus placebo, so a negative value means the average change was lower in the rivastigmine group. For this endpoint, the registry states that a rise in the score indicates improvement. The difference is therefore interpretable in the scale’s original units, while the direction must be read using the endpoint’s stated scoring meaning.

This is a comparison of group averages, not a statement that every participant had the group-average experience. General statistical explanation: a mean difference estimates how far apart two group averages are, but it does not describe the full spread or shape of the individual outcomes. The reported group standard deviations help convey that spread; they are not confidence limits for the mean difference.

How should the confidence interval and p-value be read?

The posted confidence interval for the primary mean difference extends from -6.685 to -0.715. In general, a confidence interval expresses the precision of an estimate by giving a range of values that are reasonably compatible with the data and the method’s assumptions. It is not a range containing a known fraction of individual outcomes, and it does not account for design or analysis issues that are not represented in the calculation.

The unadjusted p-value is 0.015. In general, a p-value measures how unusual the observed data, or data more extreme, would be under a specified null model. It is not the probability that the null hypothesis is true, nor does it measure the size or practical importance of an effect. Here, “unadjusted” means the reported value is not adjusted for multiple comparisons. The registry record does not state whether the trial’s own analysis used a different model or adjustment.

Why does the group size matter?

The primary summaries include 10 participants per group. The number contributing to an endpoint affects the precision of group means and their comparison. General statistical explanation: with fewer observations, estimates can be more sensitive to the particular participants included, and uncertainty can be wider. A small analysis group does not automatically invalidate a result, but it makes it important to consider the interval and the analysis population alongside the point estimate.

The posted comparison also includes a pooled standard deviation and a standardized effect estimate. A pooled standard deviation combines information about spread across the two groups for a standardized comparison. The standardized effect expresses the mean difference relative to that pooled spread, rather than in the Vineland scale’s units. Standardization can help describe magnitude on a scale-independent basis, but it does not make an estimate more reliable or remove uncertainty.

What does over-dispersion mean in general?

Over-dispersion is a general issue for outcomes whose observed variability exceeds what a chosen statistical model expects. It is especially relevant for count data, where a model may assume a particular relationship between the mean and variance. The primary endpoint in this trial is a continuous change score summarized by means and standard deviations, and the record does not report an over-dispersion assessment. This concept should not be taken as a description of a problem demonstrated in this trial.

What analysis details are missing?

The registry record gives the outcome definition, group summaries, and an unadjusted comparison, but does not report the trial’s full statistical analysis plan. In particular, the record does not specify whether the primary comparison was prespecified in the form shown, what model was planned, whether covariates were to be included, or how missing observations beyond the stated exclusions were handled. It also does not describe an adjustment for multiple outcomes. These omissions limit the ability to determine whether the posted summary comparison matches the sponsor’s intended analysis.

General statistical explanation: design and analysis choices influence both estimates and uncertainty. Random allocation can support a fair comparison, but it does not by itself resolve missing outcome data, imbalances arising by chance, or the consequences of evaluating multiple outcomes. Interpretation should therefore distinguish the numerical result from the unreported analytic decisions that would be needed for a complete trial-level assessment.

11. Confidence Intervals and Precision

The primary mean difference is -3.7 units, with a confidence interval from -6.685 to -0.715. This interval is entirely below zero, matching the direction of the observed mean changes: the average change was lower in the rivastigmine group than in the placebo group. The interval gives a sense of statistical precision around the difference, but it does not make the estimate exact. Values throughout the interval remain compatible with the posted comparison under the assumptions used to calculate it.

In general, a narrower confidence interval indicates greater precision, while a wider interval indicates more uncertainty. Precision depends on factors such as the number of observations and variation in outcomes. Here, the primary analysis group summaries include 10 participants per group, and the standard deviations are 3.2 and 3.6. The interval is consequently important context for the point estimate: the observed difference is not the only value compatible with the data.

The interval does not address all uncertainties. It does not show whether the change-score analysis was prespecified, account for any unreported multiplicity adjustment, or correct for possible selection related to completion and form exclusions. The registry record does not report a sponsor analysis that could clarify those issues. Nor does the interval give the probability that a particular treatment effect is true. It describes uncertainty under a statistical calculation, not certainty about causation or future results.

The unadjusted p-value of 0.015 is related to the interval but answers a different question: it quantifies compatibility with a null model under the calculation’s assumptions. It should not replace attention to the interval’s scale, direction, and width. The estimated difference is expressed in the Vineland score’s original units, which helps connect it to the measure, but the registry does not report a threshold for a clinically important difference. The posted precision therefore supports a careful reading of the numerical contrast, not a claim about clinical importance.

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, not a total across the trial. They illustrate sample sizes under a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set to a stated fraction of the observed effect. The registry record does not report the original design assumptions or sample-size justification for this trial; these figures are planning illustrations based on the observed comparison, not a reconstruction of the study’s original design.

The 80% power column means the planning calculation targets an 80% chance of detecting the specified effect under the stated assumptions if that effect is present. The 90% column uses a higher target chance of detection, which requires more participants under the same assumptions. These are prospective planning concepts, not statements about what happened in the completed trial. This is not observed power.

Rows that plan for a smaller fraction of the observed effect require more participants per arm. A smaller difference is harder to distinguish from random variation when outcome variability remains similar, so the planning calculation calls for a larger study. The figures should be interpreted conditionally: they depend on the assumed effect size and the stated statistical setup. If the true effect differs from the observed estimate, or if the trial’s design and analysis depart from those assumptions, the planning numbers may not correspond to the needed sample size.

The curve should be read in the same way as the planning rows: it represents how the planning requirement changes as the target effect changes. Neither the curve nor the tabulated figures are observed results, and neither establishes that the completed trial had adequate power. The registry record does not report the trial’s own planned effect, power target, or sample-size calculation.

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 = -3.700; pooled standard deviation σ = 3.406.
2. n = 2 × (1.96 + 0.842)² × 3.406² / -3.700² = 13.3, rounded up to 14 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 <- -3.7                      # observed difference in means
sigma <- 3.40587727318528                     # 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(1084135)
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 = 14 (80% power), 18 (90% power)
 75% of observed effect: n per arm = 24 (80% power), 32 (90% power)
 50% of observed effect: n per arm = 54 (80% power), 72 (90% power)
power.t.test: 14.3 
Simulated power at n = 14 per arm: 0.790 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

The registry record lists a primary endpoint and a secondary executive-function outcome, but it does not report a statistical plan for controlling multiplicity. It does not say whether the outcomes were tested in a hierarchy, whether any adjustment was planned, or whether the secondary measure was considered exploratory. Without that information, the unadjusted p-values should not be interpreted as though they necessarily represent a family-wise confirmatory analysis. When several outcomes are considered, chance findings become more likely unless the testing strategy is accounted for.

The registry record lists no sponsor-reported analyses. The posted comparisons therefore do not resolve whether the sponsor used a different analysis model, applied adjustments, or drew conclusions using other methods. The primary comparison is an unadjusted comparison based on the posted summary statistics; it should be distinguished from any prespecified sponsor analysis, which is not reported in the registry record.

The primary outcome population is described as participants who completed the 20-week period, except for 2 people whose forms were completed incorrectly. The record does not report the number of randomized participants who completed the study, reasons for non-completion, or how other missing outcome values were treated. It also does not say whether an intention-to-treat analysis was conducted. Restricting the analysis to completers can affect interpretation if completion is related to outcome, but the available record does not show whether or how that occurred.

General statistical explanation: missing data can affect both the composition of the analyzed groups and the uncertainty of estimates. Excluding incorrectly completed forms may be reasonable, but the record does not describe how the forms were judged incorrect or whether those exclusions were planned. The lack of detailed reporting prevents a full assessment of the relationship between randomization, completion, and the final analyzed sample. These are limitations of what can be concluded from the registry record, not evidence that a particular bias occurred.

15. 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.

Most frequent non-serious adverse event terms

The serious adverse event figures are reported separately for the 20-week and 12-week records. At 20 weeks, the registry lists 0 serious adverse events among 12 at risk in the rivastigmine group and 0 among 11 at risk in the placebo group. At 12 weeks, it lists 0 among 4 at risk in each group. These are the registry’s reported counts for the stated timeframes and risk groups; the 12-week and 20-week figures should not be combined because the record does not explain their relationship.

For other adverse events, the record lists 11 affected among 12 at risk in the 20-week rivastigmine group and 8 among 11 at risk in the 20-week placebo group. At 12 weeks, it lists 3 affected among 4 at risk in the rivastigmine group and 4 among 4 at risk in the placebo group. These counts indicate participants recorded as affected, not the number of event episodes or their severity. The registry does not provide top serious-event terms, and deaths are not reported in the registry record. These figures alone do not establish comparative safety.

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

The posted primary result is a difference in average change on the Vineland Adaptive Behavior Composite over the study period. The average change was -1.7 units in the rivastigmine group and 2.0 units in the placebo group; the unadjusted mean difference was -3.7 units. The confidence interval ranges from -6.685 to -0.715. Because the registry says that a rise in the score indicates improvement, the observed group averages do not favor rivastigmine in their direction. This is a description of the posted data, not a medical recommendation.

The result does not establish that rivastigmine caused the difference. Randomization is an important design feature, but full interpretation also depends on the planned analysis, handling of missing outcomes, participant adherence, and other trial details. The registry record does not report enough information to assess all of those issues. The primary analysis includes people who completed the 20-week period, with 2 incorrectly completed forms excluded, so its estimates are not automatically representative of every person enrolled or randomized.

The estimate also does not show the response of every individual. Group averages can conceal variation in individual outcomes, and the standard deviations indicate that participants’ change scores varied. The confidence interval expresses uncertainty in the comparison under its calculation; it is not proof that the true effect lies within the interval with certainty, nor does it describe future outcomes.

The study’s secondary measure addresses executive function, but its unadjusted result does not provide clear additional confirmation. The registry does not report a multiplicity plan or a clinically important difference threshold for either measure. Accordingly, statistical evidence and clinical importance cannot be treated as the same question. The record describes a completed trial and posts outcome summaries, but it does not report the full analysis rationale needed to make stronger claims about efficacy, clinical relevance, or generalizability.

17. Why This Trial Matters Statistically

This trial illustrates why an outcome’s direction and scale must be understood before interpreting a group contrast. The primary measure is a change in a standard score, and the registry explicitly states that a rise indicates improvement. That information makes it possible to connect the negative mean difference with the direction of the group summaries without relying on a p-value alone. The confidence interval then adds information about the precision of that contrast.

It also demonstrates the distinction between an overall enrollment figure and the participants contributing to a particular endpoint. The registry lists enrollment of 42, while the primary group summaries include 10 participants per arm. The record explains that the primary analysis used 20-week completers, except for 2 incorrect forms, but does not account fully for the analysis counts. A transparent account of participant flow and missing data is important for judging how well an endpoint estimate represents the randomized groups.

The contrast between the primary and secondary results is another useful statistical lesson. Different measures can address related aspects of function without producing identical patterns, and a secondary result should not be treated as confirmation simply because it is posted alongside a primary outcome. Multiplicity planning matters when interpreting several outcomes, yet the registry does not report such a plan.

Finally, planning calculations based on an observed effect are not the same as the original design calculation or an assessment of observed power. The trial record does not report its original assumptions, while the posted planning illustrations state their own assumptions. Keeping these separate helps avoid attributing design intentions to calculations made from the observed result. Together, these points show how effect direction, uncertainty, analysis populations, and reporting completeness all contribute to statistical interpretation.

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 endpoint analysis is described as restricted to participants who completed the 20-week period, excluding 2 incorrectly completed forms. The registry does not report the reasons for non-completion, the number of missing outcomes, or how those missing data were handled. This limits assessment of whether the analyzed participants differ from those not included. The primary outcome group summaries each include 10 participants, so the estimate may be sensitive to the observations available for analysis.

The registry record does not report the trial’s original sample-size justification, prespecified statistical analysis, interim analyses, or multiplicity strategy. It also does not provide sponsor-reported analyses. Consequently, the posted unadjusted summary comparison cannot be fully checked against an analysis plan or interpreted in the context of planned outcome testing. No clinically important difference threshold is reported for the Vineland score, so the numerical contrast cannot be classified by such a threshold from this record.

The listed BRIEF-P measure is named as a preschool form, while the study is described as enrolling children ages 10–18. The record does not explain the rationale for using this form in the study population or how its scores were interpreted. The secondary outcome also has a different group count from the primary outcome, and the reason for that difference is not reported.

The safety record gives affected counts at 12 and 20 weeks, but does not explain the relationship between those timeframe groups. It supplies no top serious adverse-event terms, and deaths are not reported in the registry record. The record also does not describe adherence, masking success, or the handling of protocol deviations. These reporting gaps prevent a complete assessment of the study’s conduct and the robustness or generalizability of its outcome estimates.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Rivastigmine Study in Adolescents With Down Syndrome — NCT01084135.
• Registry record last updated 2015-04-06; status last verified 2015-02.
• Results first posted 2015-03-12.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

The registry describes a completed randomized, quadruple-masked, parallel-group phase 1 and phase 2 trial sponsored by Duke University. It compared liquid rivastigmine with liquid placebo in children and adolescents with Down syndrome. The primary outcome was change in Vineland Adaptive Behavior Composite score from baseline to the final week 20 visit.

The posted primary group means were -1.7 units for rivastigmine and 2.0 units for placebo, with an unadjusted mean difference of -3.7 units and a confidence interval from -6.685 to -0.715. The registry also lists one secondary executive-function outcome. The primary analysis used 20-week completers, apart from 2 incorrectly completed forms; analysis details beyond the posted summaries are not reported in the registry record.

The record does not report the original sample-size justification, full statistical plan, multiplicity approach, or complete missing-data handling. Safety counts are listed at separate timeframes, with deaths not reported in the registry record. These limitations call for interpreting the numerical findings as posted summaries rather than as a complete account of the trial’s analysis.