This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Computerized Cognition Testing in Participants With Mild Alzheimer's Disease (AD) Treated With Donepezil (MK-0000-318) (NCT02064920), a randomized phase 2 study. The primary outcome is One-card Learning (OCL) Measurement Over 12 Weeks of Treatment.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02064920 |
| Phase | Phase 2 |
| Condition | Alzheimer's Disease |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Triple |
| Enrollment | 36 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Merck Sharp & Dohme LLC |
| Results posted | Yes |
2. Clinical Question
Population
Participants with mild Alzheimer's disease who received study medication and had at least one measurement for the primary endpoint. The trial enrolled 36 participants. The registry record does not report further eligibility details, such as age criteria or disease-duration requirements.
Intervention
Donepezil was given orally. During the first 4 weeks, participants received placebo for donepezil; during the following 2 weeks, they received donepezil at 5 mg per day; during the remaining 10 weeks, the dose was titrated to up to 10 mg per day based on tolerability.
Comparator
The comparison group received a placebo capsule for donepezil hydrochloride orally once a day for 16 weeks. The registry record describes randomized allocation and a parallel-group design.
Primary question
The trial aimed to assess cognitive function through repeated computerized testing, with One-card Learning (OCL) identified as the primary outcome. The registry summary states a hypothesis concerning whether the standard deviation associated with change from baseline in OCL measurements after 12 weeks of donepezil and placebo treatment is ≤0.1. The registry record does not report the observed OCL results or a treatment comparison estimate.
The central statistical question is therefore about the performance of a computerized cognitive measure during treatment, as well as the registered OCL outcome. These are related but not identical questions: showing that a test can measure change is not the same as establishing a difference between treatment groups. The posted record does not provide the outcome values needed to judge either question from the observed data.
3. Trial Design
This was a randomized, triple-masked, parallel phase 2 trial sponsored by Merck Sharp & Dohme LLC. The registry lists Alzheimer's disease as the condition and describes the study as an assessment of cognitive function by repeated computerized testing in participants with mild Alzheimer's disease treated with donepezil. The registry status is completed. The recorded start date is 2014-04-22, and the primary completion date is 2016-07-13.
Participants were assigned to a placebo group or a donepezil group. The placebo group received placebo for donepezil hydrochloride once a day for 16 weeks. The donepezil regimen included a placebo period during the first 4 weeks, followed by 5 mg per day for 2 weeks, then titration to up to 10 mg per day during the remaining 10 weeks based on tolerability. The registry lists donepezil twice among the interventions, but does not provide separate arm descriptions for those entries.
The OCL outcome is identified as primary, with measurements scheduled at weeks 4, 8, 12 and 16. The registry also lists percentage of correct responses in the OCL task over 12 weeks as a secondary outcome. The stated primary hypothesis refers to change from baseline after 12 weeks, while the primary outcome timeframe lists multiple measurement times. The registry record does not explain how these elements were reconciled in the analysis.
The primary endpoint population is described as participants who received study medication and yielded at least one measurement for that endpoint. Other analysis populations, eligibility details, allocation procedures, masking procedures, and the planned statistical model are not reported in the registry record. The enrollment figure is a registry figure; it should not be treated as the number contributing to every endpoint or analysis.
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
| Stage | Placebo | Donepezil |
|---|---|---|
| STARTED | 6 | 30 |
| Treated | 6 | 28 |
| COMPLETED | 5 | 21 |
| NOT COMPLETED | 1 | 9 |
Reasons for not completing this period:
| Reason | Placebo | Donepezil |
|---|---|---|
| Adverse Event | 0 | 4 |
| Physician Decision | 0 | 1 |
| Technical Problems | 0 | 1 |
| Withdrawal by Subject | 1 | 3 |
Analysis population for the primary outcome, as defined in the registry: All participants who received study medication and yielded at least one measurement for that endpoint.
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.
| Characteristic | Unit | Placebo | Donepezil | Total |
|---|---|---|---|---|
| Participants analysed | 6 | 30 | 36 | |
| Age, Continuous | Years | 76.7 (SD 8.6) | 72.7 (SD 8.7) | 73.4 (SD 8.7) |
| Sex: Female, Male: Female | Participants | 2 | 15 | 17 |
| Sex: Female, Male: Male | Participants | 4 | 15 | 19 |
6. Primary Endpoint
The registry lists the primary outcome as One-card Learning (OCL) Measurement Over 12 Weeks of Treatment, measured over Weeks 4, 8, 12 and 16.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | Standard Deviation |
| Unit | Score on a scale |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is the One-card Learning (OCL) measurement over 12 weeks of treatment, with the registered timeframe listing weeks 4, 8, 12 and 16. OCL is one test in the Cogstate battery. The registry describes it as a continuous visual recognition task that assesses visual recognition, memory and attention using a pattern separation algorithm.
The OCL score is defined as the arcsine transformation of the square root of the proportion of correct responses to 80 OCL questions. Its possible range is 0 to 1.571, and higher scores indicate better performance. The score is therefore a transformed measure based on task responses, rather than a direct count of correct answers. The registry also lists percentage of correct responses in the OCL task as a secondary outcome.
Repeated measurement can be useful when the research question concerns whether cognitive performance changes over time or whether a test can detect change during treatment. A continuous score retains more gradation than a simple classification of response, but interpretation depends on the scale, measurement variability, and analysis approach. Those considerations are especially relevant because the registered hypothesis concerns the standard deviation of change, not only the mean score.
The registry does not report observed OCL means, changes from baseline, standard deviations, confidence intervals, or a treatment-effect estimate. It therefore does not allow readers to determine the size or precision of any observed change.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record establishes the trial question, design, endpoint definition, and treatment schedules, but it does not contain the primary outcome results. In particular, it does not report the OCL measurements, change from baseline, variability estimates, or a comparison between donepezil and placebo. The completed status indicates that the study is recorded as completed; it does not supply evidence about the direction, size, or precision of the outcome.
The primary hypothesis in the brief summary concerns the standard deviation associated with change from baseline in OCL measurements after 12 weeks of treatment and gives a threshold of ≤0.1. This is a statement about variability in the measurement or change, not itself a reported treatment effect. Without the corresponding observed standard deviation and a description of the statistical test or estimation procedure, the registry record does not show whether the hypothesis was met.
The registry identifies several OCL measurement times: weeks 4, 8, 12 and 16. It also names a secondary outcome based on percentage of correct responses over 12 weeks. No values or analysis results for these outcomes are reported. The record therefore does not show whether the score changed over time, whether any change differed between groups, or how estimates varied across measurement times.
General interpretation: an estimate describes the observed effect in the data, while a confidence interval describes the range of values reasonably compatible with the data under the analysis assumptions. A small or large point estimate alone does not convey its precision. Neither an estimate nor its precision can be assessed in this case because neither is reported in the registry record.
The adverse-event figures are descriptive counts for the listed groups. They do not replace the absent cognitive outcome results, and they do not establish that a treatment caused an event. The registry record does not provide the methods used to compare adverse events or a detailed account of their timing and severity.
Accordingly, the posted information supports a description of what was intended to be measured and how treatment was scheduled, but not a conclusion about the observed cognitive outcomes. Any assessment of the study's primary result would require outcome estimates and the analysis details, which are not reported in the registry record.
9. Statistical Methods Explained
Statistical interpretation depends on separating the trial's design, its endpoint definition, the estimand of interest, and the analysis actually reported. The registry describes some of these elements but does not report an analysis plan or outcome estimates. The concepts below are general explanations rather than claims about procedures used in this trial.
What does the randomized parallel design tell us?
Randomized allocation is intended to make treatment groups comparable, on average, with respect to factors that could influence outcomes. A parallel design means each participant is assigned to a group and followed under that group's assigned regimen, rather than crossing between treatment regimens. These design labels alone do not establish that allocation was concealed, that groups remained comparable in the analyzed population, or that all randomized participants contributed data. Those details are not reported in the registry record.
Triple masking indicates that masking was planned for three parties, but the registry record does not identify those parties or describe how masking was maintained. Masking can reduce the influence of expectations on participant experience, outcome assessment, or analysis. It cannot eliminate all sources of bias, and its success cannot be inferred just from the masking label.
What does a mean and standard deviation describe?
For a continuous endpoint, the mean describes the average observed score in a group or at a time point. The standard deviation describes how dispersed individual scores are around that mean. A change-from-baseline analysis instead focuses on within-participant differences between a later measurement and baseline. The average change summarizes the direction and size of those differences, while the standard deviation of change describes their spread.
The brief summary's hypothesis focuses on the standard deviation associated with OCL change. That makes it important to distinguish variability from average treatment response. A low standard deviation may indicate that change scores are relatively tightly grouped, but it does not by itself show that the average score improved or that the groups differed. The registry record does not report the observed standard deviation or its estimation method.
Why can repeated measurements matter?
Repeated measurements can describe how scores evolve over time and can help separate a time pattern from a single end-of-treatment observation. Measurements from the same participant are generally related, so an analysis should account for that dependence rather than treating every measurement as independent. General approaches may model time and treatment together or focus on a prespecified time point, but the registry record does not report which approach was used.
The listed primary timeframe includes weeks 4, 8, 12 and 16, while the summary's stated hypothesis refers to change after 12 weeks. This difference in wording makes it important to know which time point and contrast were used for the primary analysis. That information is not reported in the registry record, so no particular time-point analysis should be assumed.
What does a treatment contrast mean?
A treatment contrast compares outcomes between assigned groups, or compares changes between groups, depending on the prespecified estimand. A contrast in means answers a different question from a contrast in standard deviations. Interpretation also depends on whether the analysis follows randomized assignment, includes only participants with observed data, or uses another defined population. The registry provides an endpoint population description—participants who received study medication and yielded at least one measurement for that endpoint—but does not report the analysis population used for the results.
General explanation: a confidence interval conveys uncertainty around an estimate under a specified model and sampling framework. It is not a guarantee that the true value lies in the interval, nor is it a measure of clinical importance. No confidence interval for the OCL outcome or treatment contrast is reported in the registry record.
Why does over-dispersion matter?
General explanation: over-dispersion occurs when observed variability is greater than a statistical model assumes. If it is ignored, standard errors can be too small and confidence intervals too narrow, making results appear more precise than they are. The concern applies most directly to outcomes modeled as counts or proportions, and the appropriate response depends on the outcome distribution and model. OCL is described as a continuous transformed score; the registry record does not report a model or a finding of over-dispersion, so no such issue can be attributed to this trial.
The registry lists a secondary percentage-correct outcome. Proportions are bounded and can have variability that depends on their mean; analytic choices should reflect the outcome's definition and distribution. The record does not provide the model, estimate, or uncertainty for that secondary outcome.
What can planning calculations say?
General explanation: sample-size calculations based on an observed effect are planning illustrations. They ask how many participants per arm might be needed under stated assumptions to detect a chosen effect with a chosen significance level and power. They are not observed power, do not establish what the completed trial could detect, and do not substitute for an analysis of the observed data. The registry record does not report the trial's own sample-size justification or design calculations.
In general, planning values depend on assumptions such as allocation, outcome variability, significance level, and the effect considered worth detecting. If those assumptions differ from the trial's actual design or data, the calculation may not describe the study's operating characteristics. No claim about the trial's planned power, interim analyses, or analysis model can be made from the registry record because these are not reported.
10. Multiplicity, Missing Data and Analysis Populations
The primary outcome lists multiple OCL measurement times, and a secondary OCL outcome is also registered. When several outcomes, time points, or comparisons are examined, the chance of a false-positive finding can increase if analyses are selected or interpreted without accounting for multiplicity. General approaches include prespecifying a primary contrast or adjusting inference across a family of tests. The registry record does not report a multiplicity strategy, a hierarchy of tests, or which time point was designated as the primary comparison for analysis.
The trial's registered hypothesis refers to change after 12 weeks, while the primary endpoint timeframe lists weeks 4, 8, 12 and 16. The record does not explain how the repeated measures were handled, whether all time points were analyzed, or whether the primary hypothesis was tested at a particular time point. It is therefore not possible to determine how repeated testing may affect interpretation.
The endpoint population is described as participants who received study medication and yielded at least one measurement for that endpoint. The registry record does not report missing OCL data, reasons for missingness, how missing values were handled, or whether sensitivity analyses were performed. It also does not report whether the stated endpoint population was the population used for the primary analysis.
These gaps matter because complete-case analysis, imputation, and longitudinal models can rely on different assumptions about missing data. General explanation: when data are missing for reasons related to outcomes or treatment, results based only on observed measurements may not represent the full randomized groups. The registry record does not provide enough information to assess that possibility for this trial.
Randomization is a feature of the design, but the record does not give the allocation process or analysis details needed to assess whether the results, if reported elsewhere, preserve the randomized comparison. No interim analyses or prespecified statistical plan are reported in the registry record.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Placebo | 0 | 6 | 0.0% |
| Donepezil | 1 | 28 | 3.6% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Placebo | 3 of 6 |
| Donepezil | 9 of 28 |
Most frequent serious adverse event terms
| Event term | Organ system | Placebo | Donepezil |
|---|---|---|---|
| Chest pain | General disorders | 0 of 6 | 1 of 28 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Placebo | Donepezil |
|---|---|---|---|
| Diarrhoea | Gastrointestinal disorders | 1 of 6 | 4 of 28 |
| Headache | Nervous system disorders | 1 of 6 | 2 of 28 |
| Dizziness | Nervous system disorders | 0 of 6 | 2 of 28 |
| Insomnia | Psychiatric disorders | 0 of 6 | 2 of 28 |
| Nasopharyngitis | Infections and infestations | 1 of 6 | 0 of 28 |
| Blood urea increased | Investigations | 1 of 6 | 0 of 28 |
Serious events with the highest arm-level rate: chest pain (3.6%).
The registry lists serious and other adverse-event figures for the placebo and donepezil groups. In the placebo group, 0 of 6 participants at risk were recorded as affected by a serious adverse event, and 3 of 6 were affected by other adverse events. In the donepezil group, 1 of 28 participants at risk was recorded as affected by a serious adverse event, and 9 of 28 were affected by other adverse events.
These are descriptive group counts. They do not by themselves establish that a treatment caused an event, and they do not provide a statistical comparison adjusted for any differences between groups. The registry lists chest pain at 3.6% among the top serious terms, but does not provide a group attribution for that term in this record. Death counts are not reported in the registry record.
The figures should be interpreted in light of the small group counts and the limited detail available. The registry record does not report event timing, duration, recurrence, severity grading, relationship to treatment, or the method used to collect and classify events. The listed adverse-event figures also do not provide an account of treatment discontinuations or of all possible safety outcomes.
These safety data describe reported events in the registry. They should not be read as a complete safety profile or as a causal comparison between regimens.
12. What the Result Does, and Does Not, Mean
The trial was designed to assess cognitive function through repeated computerized testing in participants with mild Alzheimer's disease who received donepezil or placebo. The registered primary outcome was OCL, a continuous score derived from correct responses on a visual recognition task. The record also states a hypothesis about the standard deviation associated with change from baseline after 12 weeks.
What the registry record does not report is the observed primary outcome result. There are no posted group means, changes from baseline, standard deviations, treatment contrasts, confidence intervals, or primary-outcome significance tests in the record. Consequently, the record cannot establish whether OCL scores changed, whether the groups differed, or how precisely any difference was estimated.
A study's completion status is not a result. Likewise, a stated hypothesis is not evidence that the hypothesis was supported. The endpoint's score range and direction explain how to read the measure, but do not indicate what participants scored or how scores changed.
General explanation: measurement performance and treatment effect are distinct statistical questions. A test may be evaluated for its ability to detect change, while a treatment comparison asks whether outcomes differ between assigned groups. The registered summary emphasizes the testing battery and a variability hypothesis, but the record does not provide outcome data or analysis details sufficient to answer either question.
The adverse-event counts are also limited to descriptive information. They do not establish a causal treatment effect, and the registry record does not include the details needed for a complete assessment of safety. Overall, the record supports a careful description of the study's design and measurement plan, but not a conclusion about cognitive benefit, lack of benefit, or the precision of a treatment effect.
13. Why This Trial Matters Statistically
This trial raises a useful statistical question about how repeated computerized cognitive measurements should be evaluated in a clinical study. OCL is described as a continuous score constructed from task accuracy and transformed onto a defined scale. The registry's focus on the standard deviation of change highlights that the variability of a measure can be part of the research question, not just its average value.
That distinction matters when interpreting clinical research. A mean change describes average movement in scores, whereas variability describes how consistently participants' changes cluster around that average. Neither alone establishes whether a measure is sensitive to change or whether one assigned treatment differs from another. A clear analysis must connect the endpoint, time point, estimand, and uncertainty measure to the research question.
The repeated OCL time points also illustrate the importance of prespecification. Multiple observations can provide a richer picture of change, but the analysis must account for their relationship within participants and define which comparison is primary. The registry record lists multiple time points and a 12-week hypothesis without reporting the analysis details that would resolve how these elements were handled.
Finally, the difference between a completed trial and a reported result is important for evidence interpretation. The registry record documents a randomized, masked design and a specific measurement plan, but does not report the outcome estimates needed to evaluate the primary question. This makes clear why design information, analysis methods, and outcome reporting each contribute distinct information to a statistical assessment.
14. 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 main limitation of the registry record for interpreting this trial is that it does not report the primary outcome results. There are no OCL summary values, change estimates, treatment contrasts, confidence intervals, or test results. The stated primary hypothesis therefore cannot be evaluated from the posted information.
The primary outcome timeframe lists weeks 4, 8, 12 and 16, while the brief summary describes a hypothesis about change after 12 weeks. The registry record does not explain which time point or comparison was used for the primary analysis. It also does not report the statistical model, the analysis population, a multiplicity approach, or the handling of missing outcomes.
The endpoint population is limited to participants who received study medication and yielded at least one measurement. The record does not describe how many participants contributed to the endpoint analysis, whether participants discontinued, or whether missing data differed between groups. Enrollment does not establish how many observations were available for each outcome.
Further trial details—including eligibility criteria beyond mild Alzheimer's disease, allocation procedures, the parties included in triple masking, and the rationale for the planned sample size—are not reported in the registry record. The safety information is also limited: event timing, severity, treatment relationship, and death counts are not reported.
These limitations restrict statistical interpretation of the registry record. They do not establish that the study itself lacked these procedures or data; they indicate only that the relevant information is not reported in the registry record.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Computerized Cognition Testing in Participants With Mild Alzheimer's Disease (AD) Treated With Donepezil (MK-0000-318) — NCT02064920.
- Registry record last updated 2018-10-12; status last verified 2018-09.
- Results first posted 2017-08-01.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT02064920 is a completed randomized, triple-masked, parallel phase 2 trial sponsored by Merck Sharp & Dohme LLC. It enrolled 36 participants with mild Alzheimer's disease and assessed cognitive function using repeated computerized testing. The primary outcome was OCL measurement, with a listed timeframe of weeks 4, 8, 12 and 16. The secondary outcome was percentage of correct responses in the OCL task over 12 weeks.
The placebo regimen was once daily for 16 weeks. The donepezil regimen included an initial placebo period, then 5 mg per day, followed by titration to up to 10 mg per day based on tolerability. The registry record describes the endpoint population as participants who received study medication and yielded at least one endpoint measurement.
The record does not report primary outcome estimates, statistical analysis details, or missing-data methods. It lists descriptive adverse-event counts, but does not report death counts. Accordingly, it documents the planned design and endpoint but does not permit interpretation of the primary treatment comparison.