The Effect of Donepezil on Gait and Balance in Parkinson's Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Effect of Donepezil on Gait and Balance in Parkinson's Disease (NCT01521117), a randomized phase 4 study. The primary outcome is Sensory Organization Test - Composite Score.

2. Clinical Question

The study is listed as a Phase 4 randomized, triple-masked crossover trial. In this design, participants receive both treatments in an assigned sequence, rather than remaining in a single treatment group throughout. The two sequences were donepezil followed by placebo and placebo followed by donepezil. This makes the within-participant comparison central to the design: each participant’s outcomes under the two periods can be compared while accounting for the fact that the measurements come from the same person.

The registry summary frames the study as an investigation of which aspects of gait and balance might change during donepezil treatment. The posted primary endpoint is specifically a balance measure. Although the study title mentions gait and balance, the primary result is not a broad measure of all walking or balance function. It is the change in a composite score from a standardized balance test. The registry lists Trails B - A as a secondary outcome.

The group summaries show average changes for donepezil and placebo, and an accompanying unadjusted comparison can describe their numerical contrast. However, the registry record does not report a crossover analysis that uses the paired observations, nor does it describe how treatment sequence or period was handled. The question can therefore be stated from the design and endpoint, but the posted figures alone do not resolve every analytical issue raised by that design.

3. Trial Design

This is a randomized, triple-masked, placebo-controlled crossover study listed as Phase 4. The registry describes two treatment sequences: donepezil followed by placebo, and placebo followed by donepezil. Each participant was scheduled to receive both interventions, with a washout period of 4 weeks between treatment phases. The listed treatment schedule used 5 mg a day for weeks 1 - 3 and, if tolerated, 10 mg a day for weeks 4 - 6 in each phase.

The primary outcome timeframe is change from Day 1 of each treatment phase to Day 42 of that phase. This phase-specific measurement is consistent with comparing change while participants receive each intervention. It is distinct from a parallel-group design, in which different participants would remain in different treatment groups. Because participants contribute measurements under both treatments, observations within a participant are related; an analysis should account for that pairing. The registry record does not report whether or how the posted comparison accounts for it.

Enrollment is listed as 21. The outcome summaries list 10 participants for donepezil and 10 for placebo. Those outcome counts should not be interpreted as two independent sets of participants in this crossover study: both treatment periods may involve the same people. The record does not explain the relationship between enrollment and the outcome counts, or why the posted outcome summaries include those counts. Further details about participant flow and missing outcomes are not reported in the registry record.

The study status is completed. The registry lists a start of 2011-12 and primary completion of 2012-04. It identifies Oregon Health and Science University as sponsor. The original sample-size calculation, the assumptions used in the study’s design, and the planned statistical analysis are not reported in the registry record. The posted sample-size illustrations are based on the observed comparisons and are not evidence of the original design rationale.

Masking is recorded as triple, but the registry record does not specify which groups of people were masked. Details such as how the randomization sequence was generated or concealed, and how the washout was selected, are also not reported in the registry record. These omissions limit how fully the conduct of the trial can be assessed from the registry alone.

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.

First Intervention (6 Weeks)

Reasons for not completing this period:

Washout (4 Weeks)

Second Intervention (6 Weeks)

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.

Population: Baseline characteristics are reported on all study completers (N = 10).

6. Primary Endpoint

The registry lists the primary outcome as Sensory Organization Test - Composite Score, measured over Change from Day 1 of each treatment phase to Day 42 of each treatment phase.

The primary endpoint is the change in Sensory Organization Test composite score from Day 1 to Day 42 of each treatment phase. The test assesses balance across six conditions involving eyes open or closed and a moving or stable visual surround, with testing on a stable or moving platform. Center of Pressure was calculated from recordings taken during the test. The registry describes the score as an interval scale from 0 to 100: 100 indicates no sway, while 0 indicates that the trial was stopped because of an impending fall. Higher scores indicate better balance or greater stability.

A change score represents the difference between a participant’s score at the beginning and end of a treatment phase. This focuses the outcome on change during that phase rather than on the end-of-phase score alone. Comparing changes under donepezil and placebo is the stated primary question. In a crossover study, the two phase-specific changes can be paired within each participant; accounting for that pairing is important for statistical precision and interpretation.

Balance testing is used because the study concerns gait and balance difficulties in Parkinson’s disease, and the registry summary says the investigators sought to examine which aspects might change during treatment. The endpoint is a measured balance score, not a direct count of falls or a complete assessment of gait. The registry does not report whether a particular size of change on this score was regarded as clinically important, or provide a threshold for interpreting a change as meaningful to participants.

7. Results Posted in the Registry

• Donepezil: mean 7.70 versus 0.60 on control; difference 7.10 (95% CI -7.38 to 21.58), standardized effect d = 0.43; z-test p = 0.337.

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.

Trails B - A

• 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:
  • Donepezil: mean 9.40 versus 1.40 on control; difference 8.00 (95% CI -56.21 to 72.21), standardized effect d = 0.11; z-test p = 0.807.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Donepezil and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists Trails B - A as a secondary outcome, measured as a change in seconds from Day 1 to Day 42 of each treatment phase. Its posted summaries show a numerical contrast between treatment periods, but the estimates are imprecise. The secondary outcome does not provide a clear, independent confirmation of the primary balance result. Both outcomes are presented as changes over treatment phases, but they measure different aspects of performance and use different scales.

The overall pattern is therefore one of uncertain estimates rather than a consistent set of precise findings. The primary outcome’s unadjusted interval includes zero, and the secondary outcome’s interval also includes zero and is broad. Such a pattern does not establish that the outcomes agree or disagree in a substantive way; uncertainty is substantial for each comparison. The registry does not define a clinically important threshold for either measure, so statistical contrasts alone cannot establish whether a difference would matter to participants.

The registry lists one secondary outcome and reports no analyses in its analysis listing. No adjustment for testing multiple outcomes is reported in the registry record. General statistical explanation: when several outcomes are examined, the chance of seeing at least one apparently notable result by chance can increase. For that reason, secondary results without a reported multiplicity plan generally carry less confirmatory weight than a prespecified primary endpoint. Here, the registry identifies the primary endpoint, but details of the analysis plan and any multiplicity strategy are not reported.

Secondary findings can still be informative, particularly when interpreted as exploratory evidence that may guide later research. They should not be used to overturn the uncertainty in the primary comparison or to imply a conclusion about an unmeasured outcome. The registry record does not report a complete analysis set, the handling of missing secondary observations, or a hierarchy for interpreting the secondary outcome. These gaps further limit how much weight can be placed on the secondary result.

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

• Trails B - A

9. Statistical Interpretation

10. Statistical Methods Explained

The posted endpoint is continuous: it is a numerical score, and the registry reports a mean and standard deviation for each treatment. A mean summarizes the average change, while a standard deviation describes how dispersed individual change scores are around that average. The mean difference compares the two posted average changes. Here, the unadjusted difference is expressed as donepezil minus placebo, so a positive value means a larger mean change in the donepezil summaries.

What does the confidence interval describe?

General statistical explanation: a confidence interval gives a range of values that are reasonably compatible with the data under the method used to calculate it. It reflects uncertainty in the estimate; it is not a range containing a fixed proportion of individual participant outcomes. A narrower interval indicates greater precision, while a wider interval indicates less precision. The interval in this analysis is wide and contains zero, so it does not isolate a single direction or a precise size of difference. Its interpretation also depends on the calculation being appropriate for the design. The posted interval is an unadjusted recalculation from summary statistics, not a documented paired crossover interval.

Why does the crossover design matter?

General statistical explanation: in a crossover study, each participant receives more than one treatment, so measurements from the same person are usually correlated. A paired analysis uses the within-person differences between treatment periods. If that correlation is available and the assumptions of the analysis are appropriate, pairing can provide a more relevant comparison and may improve precision. A calculation that treats the treatment summaries as though they came from separate, independent groups does not use the within-person pairing. The registry record does not provide the individual paired differences, their variability, or a sponsor-reported crossover analysis. As a result, the posted unadjusted recalculation cannot show how the pairing would affect the estimated difference or its uncertainty.

Other elements can matter in crossover analysis. Treatment period and sequence may be relevant because outcomes can differ over time, and the effects of one treatment period may not be fully separated from the next. The registry describes a 4-week washout and two treatment sequences, but it does not report how period, sequence, or possible carryover was handled statistically. This is not evidence that such issues occurred; it means the record does not give enough information to evaluate them.

How should the p-value be read?

General statistical explanation: a p-value measures how unusual data at least as discrepant as those observed would be under a specified null 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 a difference. The posted unadjusted p-value of 0.337 is associated with the summary-statistic comparison. It should not be treated as the result of a documented crossover model, because no trial analysis is reported in the registry. Nor does it establish equivalence: an equivalence conclusion requires a defined margin and an analysis designed for that question, neither of which is reported in the registry record.

The confidence interval and p-value summarize related aspects of this unadjusted comparison. The interval shows the range of mean differences compatible with the calculation, and the p-value reflects the evidence against a zero difference under its assumptions. Neither can resolve the limitations caused by missing paired-data information. The interval is often more informative than a significance label because it shows the range and direction of uncertainty, but its scope is still determined by the analysis that produced it.

What does variability tell us?

General statistical explanation: standard deviations describe spread among observations, not uncertainty in the mean by themselves. Standard errors describe uncertainty in an estimate and depend on both variability and the amount of information. The posted summaries include standard deviations of 21.6 for donepezil and 8.9 for placebo, along with recalculated standard errors of 6.831 and 2.814 for the two means. Those standard errors are derived for the posted summaries; they are not a substitute for the standard error of a paired treatment contrast. In a crossover trial, the variability of within-participant differences is central to estimating uncertainty in the treatment comparison.

General statistical context: over-dispersion means that observations vary more than a statistical model expects. When a model underestimates variability, standard errors and confidence intervals can be too small, making results appear more precise than they are. The registry record does not report an over-dispersion assessment or model diagnostics for the primary outcome. The visible spread in the score changes is a reason to attend to precision, but it does not by itself establish that a particular model was over-dispersed.

What is and is not reported?

The registry provides the outcome definition, group means, standard deviations, and a comparison calculated from the posted summaries. It reports no analyses in the analysis listing. The statistical analysis plan, handling of repeated measurements, treatment of incomplete data, and any adjustment for period or sequence are not reported in the registry record. The calculation is therefore best understood as a transparent, simple summary comparison. It should not be confused with a sponsor-reported inferential analysis of the randomized crossover design.

11. Confidence Intervals and Precision

Precision concerns how narrowly the data locate the estimated difference. The posted unadjusted mean difference in primary change scores is 7.1, with a 95% confidence interval from -7.379 to 21.58. The interval extends across zero, and it allows both a negative difference and a positive difference. It therefore does not determine the direction of the underlying difference with high precision. The values in the interval also cover a broad range of possible magnitudes relative to the point estimate.

General statistical explanation: a confidence interval is not a guarantee that the true value lies within that particular interval. Rather, under the assumptions of the method, the interval procedure is designed to capture the true value at its stated confidence level over repeated applications. The interpretation relies on the method’s assumptions, including the suitability of the analysis for the trial design. In this case, the interval is an unadjusted recalculation from posted summary statistics and is not a paired crossover interval reported by the trial.

The standard deviations of the treatment-period changes are 21.6 for donepezil and 8.9 for placebo. These describe the spread of the change scores, not the uncertainty in the difference between treatments in a paired analysis. In a crossover study, the correlation between a participant’s responses in the two periods affects the precision of the treatment contrast. Because the registry does not report the paired differences or their variability, the posted group summaries do not allow that aspect of precision to be assessed.

The outcome summaries list 10 participants under each treatment, while the registry lists enrollment of 21. Small amounts of information can lead to imprecise estimates, especially when outcome values are variable. The record does not explain the outcome counts or provide enough participant-level information to examine the impact of missing observations. The interval should consequently be read as uncertainty around a simple summary comparison, not as a complete statement of precision for the trial’s planned crossover analysis.

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 sample-size table and curve are planning illustrations based on the observed effect; they do not describe the original enrollment plan. The figures are participants per arm for 80% power and participants per arm for 90% power. They are not total event counts. The registry’s planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect.

In plain language, the 80% power column shows the number of participants per arm in the illustration when the aim is an 80% chance of detecting the specified effect under the stated assumptions, if that effect is the true one. The 90% power column uses a higher target chance of detection, so its participant requirements are larger. These are planning quantities tied to the assumptions, not guarantees about what a future study would find.

The rows consider planning for 100%, 75%, and 50% of the observed effect. A smaller target effect is harder to distinguish from random variation, so the illustrated requirement rises as the target effect gets smaller. The posted table makes that pattern visible. It also shows why a sample-size illustration based on the observed effect should be treated cautiously: the observed effect in a small trial can be imprecise, and using it as the planning target can produce unstable estimates of future requirements.

The original trial’s sample-size justification and design assumptions are not reported in the registry record. The posted illustrations use a two-arm equal-allocation approximation, whereas the trial itself is a crossover study in which participants receive both treatments. Therefore, these figures are not a replacement for a design-specific calculation based on paired outcomes and the original analysis plan. This is not observed power; it is a planning illustration under the stated assumptions.

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 = 7.100; pooled standard deviation σ = 16.519.
2. n = 2 × (1.96 + 0.842)² × 16.519² / 7.100² = 85.0, rounded up to 85 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 <- 7.1                      # observed difference in means
sigma <- 16.5192312169786                     # 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(1521117)
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 = 85 (80% power), 114 (90% power)
 75% of observed effect: n per arm = 152 (80% power), 203 (90% power)
 50% of observed effect: n per arm = 340 (80% power), 456 (90% power)
power.t.test: 85.9 
Simulated power at n = 85 per arm: 0.795 (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

Multiplicity arises when a study evaluates more than one outcome, comparison, time point, or analysis. The registry identifies the Sensory Organization Test composite score as primary and lists Trails B - A as a secondary outcome. It also states that there are 3 outcome measures, though only one secondary outcome is listed in the posted secondary results. No multiplicity adjustment or testing hierarchy is reported in the registry record. Without such a plan, the secondary findings are more appropriately treated as supportive or exploratory rather than as independent confirmation.

The registry lists no reported analyses. It does not provide the prespecified statistical analysis plan, the model used for the crossover design, or a detailed account of which outcomes and comparisons were planned. It is not reported in the registry record whether analyses were adjusted for treatment period or sequence, whether the within-participant pairing was used, or whether any interim analyses were planned. These details should not be inferred from the summary statistics.

Missing-data handling is also not reported in the registry record. The registry lists enrollment of 21, while the primary outcome summaries show 10 participants for each treatment, and the adverse-event risk sets differ from those outcome counts. The record does not explain these differences, say how many participants completed both treatment periods, or identify the analysis population. It does not say whether all randomized participants were included, how incomplete phases were handled, or whether any imputation was used.

General statistical explanation: in a crossover trial, missing measurements in one period can complicate a paired comparison because the within-person treatment contrast may not be available for every participant. The effect on estimates depends on why measurements are missing and how the analysis handles them. The posted marginal summaries cannot establish whether missingness biased the comparison or how sensitive the result is to alternative assumptions.

The registry’s limited analytical detail does not demonstrate that the trial used an inappropriate method; it means the record does not permit that assessment. A full statistical analysis plan and participant flow would be needed to understand the intended analyses, multiplicity control, and treatment of incomplete observations.

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 adverse-event section reports serious adverse events in risk sets of 18 for donepezil and 16 for placebo. The number affected by a serious adverse event is 0 in each group. Deaths are also reported as 0 in each group. These figures describe what is posted for the listed risk sets; they do not establish that no serious events could occur outside the observed participants or period.

For other adverse events, the registry reports 6 affected participants in the donepezil group and 1 in the placebo group, with risk sets of 18 and 16, respectively. The posted summary does not list the specific non-serious event terms. The adverse-event risk sets are not the same as the outcome-summary counts, and the record does not explain the difference. Because this is a crossover design, the registry’s group-level safety summaries should not automatically be read as independent parallel groups.

These counts are descriptive. They do not provide a formal comparison of risk, account for time exposed, or show how events were distributed across treatment periods. The registry record does not report event severity details beyond the serious-event classification, or a statistical safety analysis. The counts should not be interpreted as proof of safety or as a basis for treatment decisions.

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

The posted primary summaries show a larger mean change in the Sensory Organization Test composite score during the donepezil phase than during the placebo phase. Since the score is defined so that higher values indicate better balance, this is a numerical pattern in the direction of greater improvement during donepezil. The unadjusted mean difference is 7.1, but its 95% confidence interval ranges from -7.379 to 21.58 and the unadjusted p-value is 0.337. The uncertainty is therefore substantial, and the calculation does not establish a difference between treatment phases.

The result does not demonstrate equivalence or rule out a meaningful difference. A non-significant unadjusted comparison is not evidence that the treatments have identical effects. The registry does not report an equivalence margin or a trial analysis designed to assess equivalence. Nor can the result be taken as proof that there is a difference: the interval includes zero, and the posted comparison is not a documented crossover analysis.

The outcome is a balance-test score, not a direct measure of falls, daily activities, or every aspect of gait. The registry summary says the study set out to examine aspects of gait and balance, but the primary result should be interpreted specifically as a change in the Sensory Organization Test composite score. The record does not report a threshold for a clinically important change on this measure, so the practical importance of the numerical estimate cannot be determined from the posted data alone.

The study description states that donepezil is not approved to treat Parkinson’s disease in the United States or other countries because more is not known about it. The findings on this page do not change that context or provide a basis for medical decisions. They describe a small completed study and its posted summaries, with substantial statistical uncertainty and missing information about the crossover analysis.

More broadly, the record is useful for seeing the distinction between an observed mean contrast and a well-supported conclusion. A contrast can point in one direction while its interval remains compatible with other possibilities. Interpretation requires attention to the study design, the precision of estimates, and the methods used to analyze repeated measurements. On these points, important details are not reported in the registry record.

17. Why This Trial Matters Statistically

This trial illustrates why the design behind an estimate matters. A crossover study compares treatments within participants, so the most relevant treatment contrast is based on paired outcomes. A summary that presents a mean and standard deviation for each treatment can communicate the observed averages, but those marginal summaries do not reveal the variability of the within-person differences. That missing information is central to understanding how precisely the treatment contrast can be estimated.

It also shows why a confidence interval is more informative than reading a result only as statistically significant or not. The primary unadjusted interval covers values below and above zero and spans a broad range. That conveys uncertainty about both the direction and size of the difference. A p-value alone cannot show that range, establish equivalence, or quantify clinical importance.

The sample-size illustrations offer another statistical lesson. Planning numbers depend on the effect size one aims to detect, the statistical assumptions, and the study design. Smaller target effects require more participants in the posted illustrations. Because the posted planning calculations use a two-arm approximation based on an observed effect, they are not a substitute for an original design-specific calculation for a crossover study. The original sample-size rationale is not reported in the registry record.

Finally, this record demonstrates the value of complete reporting. Details of the paired analysis, missing-data handling, multiplicity strategy, and analysis population are not reported. Without these, readers can describe the posted results but cannot fully reconstruct the planned inferential analysis. Transparency about methods and participant flow is necessary for judging whether the analysis matches the design and how robust the conclusions are.

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 most direct limitation is that the registry record does not report the trial’s analysis of the crossover data. The simple recalculated comparison uses posted group means and standard deviations, rather than a documented analysis of paired treatment-period outcomes. It therefore does not account, in a demonstrated way, for within-participant correlation. The effect of using a paired analysis on the estimate and precision cannot be assessed from the available summaries.

The record also does not report how period or sequence was handled, or whether carryover was assessed. A 4-week washout is listed, but the registry does not describe the rationale for its length or provide information to evaluate its adequacy. This is a reporting limitation, not evidence that carryover occurred.

The registry lists enrollment of 21, but only 10 participants are included in each primary treatment summary, and the adverse-event risk sets are 18 and 16. The record does not explain participant flow, the reasons for differences in counts, or how missing observations were handled. Without that information, the completeness of the outcome data and the analysis population cannot be assessed.

The primary endpoint is a balance-test composite score. The record does not give a clinically important threshold for change, so the meaning of the estimated difference in practical terms remains unclear. The trial summary also discusses gait and falls, but the primary endpoint is not itself a fall count or a comprehensive measure of daily mobility.

Finally, the available outcome summaries are limited in scope. The registry lists one secondary result, no reported analyses, and no multiplicity strategy. The original sample-size rationale and planned statistical analysis are not reported in the registry record. These limitations make the posted estimates useful as descriptive evidence, but restrict the strength and completeness of inference that can be drawn from them.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: The Effect of Donepezil on Gait and Balance in Parkinson's Disease — NCT01521117.
• Registry record last updated 2021-10-05; status last verified 2021-09.
• Results first posted 2021-08-27.
• 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

This completed Phase 4 study is listed as a randomized, triple-masked, placebo-controlled crossover trial in Parkinson’s disease. Participants received donepezil and placebo in one of two sequences, with a 4-week washout between treatment phases. The primary endpoint was change in Sensory Organization Test composite score from Day 1 to Day 42 of each phase.

The posted means were 7.7 during donepezil and 0.6 during placebo. The unadjusted recalculated mean difference was 7.1, with a 95% confidence interval from -7.379 to 21.58 and an unadjusted p-value of 0.337. The registry record lists no reported analyses, so these figures should not be mistaken for a reported paired crossover analysis.

The record does not report the original sample-size justification, detailed crossover analysis, missing-data approach, or multiplicity plan. The posted results are descriptive and uncertain; they do not establish equivalence, clinical importance, or a treatment conclusion.