Efficacy, Safety and Tolerability of AVP-786 for the Treatment of Agitation in Patients With Dementia of the Alzheimer's Type: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy, Safety and Tolerability of AVP-786 for the Treatment of Agitation in Patients With Dementia of the Alzheimer's Type (NCT02442765), a randomized phase 3 study. The primary outcome is Stage 1 and Stage 2: Change From Baseline in the Cohen-Mansfield Agitation Inventory (CMAI) Composite Score to Week 6 and Week 12.

2. Clinical Question

The clinical question concerns the frequency of agitated behaviors, measured using the CMAI. Its score is based on ratings of listed behaviors, with higher total scores indicating worsening of the condition. The study record describes separate Stage 1 and Stage 2 assessment periods, so the Week 6 and Week 12 results would relate to different stage baselines: Stage 1 baseline and Stage 2 baseline at Week 6.

The randomized, double-blind, placebo-controlled design is intended to support comparison of assigned regimens while limiting some sources of bias. However, the registry record does not give the primary outcome estimates, their uncertainty, or the statistical test for the comparison. The clinical question can therefore be identified, but the posted record does not establish the direction or size of a primary-outcome difference.

3. Trial Design

This was a completed Phase 3, randomized, quadruple-masked, parallel-design study sponsored by Otsuka Pharmaceutical Development & Commercialization, Inc. The registry lists enrollment of 387. The study started on 2015-07-23, and primary completion was recorded as 2019-01-30. The condition listed was agitation in patients with dementia of the Alzheimer's type.

The trial had two stages. In Stage 1, participants assigned to placebo received matching placebo orally twice daily for 6 weeks, Days 1-42. One AVP-786-18 schedule began with 18 mg once daily in the morning and placebo once daily in the evening for 7 days, then used AVP-786-18 twice daily for the remaining 5 weeks. The AVP-786-28 description begins with AVP-786-18 once daily and placebo once daily for 7 days, followed by AVP-786-18 twice daily for 2 weeks, then AVP-786-28 twice daily from Day 22 for the remaining 3 weeks.

Participants who completed Stage 1 were eligible to participate in Stage 2. The descriptions state that Stage 1 placebo recipients were re-randomized in Stage 2 to study drug or placebo from Day 43 to Day 85. Other descriptions say participants who received AVP-786 in Stage 1 continued their assigned dose in Stage 2. The registry also describes Stage 1 placebo participants receiving AVP-786-18 or AVP-786-28 schedules in Stage 2. Further details of Stage 2 assignment and the analysis comparing these pathways are not reported in the registry record.

The study is described as double-blind in its official title and quadruple-masked in the design fields. Which specific roles were masked is not reported in the registry record. The record identifies randomized allocation and a parallel model, but the allocation procedure, concealment details, and operational masking procedures are not reported.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: This study was conducted in 2 stages. 387 participants were randomized to treatment in Stage 1 and all the participants completed Stage 1 and were re-randomized into Stage 2 to compare AVP-786 28 mg, AVP-786 18 mg and placebo in parallel group (PG) design. In Stage 2, participants randomized to active treatment in Stage 1 (AVP-786-18 or AVP-786-28) continued to receive the same treatment and those randomized to placebo in Stage 1 were re-randomized to AVP-786-18, AVP-786-28, or placebo.

Only the first 6 of 8 groups are shown.

Stage 1- (Days 1 to 42)

Stage 2- (Days 43 to 85)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Stage 1 Modified Intent-to-Treat (mITT) Population: all participants randomized in Stage 1 who had at least 1 post-Baseline efficacy assessment. Stage 2 mITT Population: all participants randomized in Stage 2 who had at least 1 efficacy assessment in Stage 2. Data were reported for only those participants contributing data to the analysis.

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 data were collected in members of the Stage 1 Modified Intent-to-Treat (mITT) Population, comprised of all participants randomized in the study who had at least one post-Baseline efficacy assessment.

6. Primary Endpoint

The registry lists the primary outcome as Stage 1 and Stage 2: Change From Baseline in the Cohen-Mansfield Agitation Inventory (CMAI) Composite Score to Week 6 and Week 12, measured over Stage 1: Baseline, Week 6; Stage 2: Baseline (Week 6), Week 12.

The primary endpoint was change from baseline in the Cohen-Mansfield Agitation Inventory composite score, assessed at Week 6 and Week 12. The CMAI is used to assess the frequency of manifestations of agitated behaviors. It includes 29 behaviors, each rated on a 7-point frequency scale. The scale anchors range from 1, never, to 7, several times an hour; the intermediate anchors describe increasing frequency. The total score ranges from 29 to 203, and higher scores indicate worsening of the condition.

Change from baseline was defined as the post-baseline score minus the baseline score. This makes the outcome a difference in score rather than the raw score at follow-up. For Stage 1, the stated timeframe is baseline to Week 6. For Stage 2, baseline is Week 6 and the follow-up is Week 12. These stage-specific baselines matter when interpreting changes: a Week 12 change is measured from the Stage 2 baseline, not necessarily from the original study baseline.

Using a frequency measure allows the outcome to represent change in reported agitated behaviors over time. The registry specifies a mean and standard deviation as the outcome summary, but does not report the arm-level means, changes, standard deviations, or a between-group estimate. The clinical meaning and precision of the observed primary outcome therefore cannot be assessed from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: it is a numerical score on the CMAI, and the registry says results were summarized as a mean with a standard deviation. A mean describes the central value of a group’s measurements. A standard deviation describes how dispersed individual scores are around that group mean. These summaries answer different questions: the mean concerns the group’s average score, while the standard deviation concerns variability among participants. Neither summary, by itself, gives the precision of a between-group estimate.

What does change from baseline mean?

The registry defines change as the post-baseline score minus the baseline score. This subtraction is the stated endpoint definition, not a recalculation of trial results. Since higher CMAI scores indicate worsening, the sign of a change is interpreted relative to that scale direction. A change score also depends on the value used as baseline. Stage 1 uses the initial baseline, whereas Stage 2 uses the Week 6 score as its baseline. Comparisons across stages should therefore retain the stage-specific definition rather than treating both follow-ups as changes from one common starting point.

What would a between-group estimate describe?

In general, a randomized trial’s between-group estimate expresses how the outcome differs between assigned groups under a specified analysis. For a continuous change outcome, it may be expressed as a difference in mean changes. The estimate’s sign depends on the order in which groups are contrasted. Its magnitude describes the observed contrast in the score’s units. The registry record does not report a primary between-group estimate or specify the contrast used, so no direction or magnitude can be assigned to this study’s primary result.

How should uncertainty be read?

In general, a confidence interval gives a range of values compatible with the data and the statistical model at a stated confidence level. Its width conveys precision: a narrower interval indicates greater precision, and a wider interval indicates less precision. A p-value addresses compatibility of data with a specified null hypothesis under a statistical model; it does not measure the size or clinical importance of an effect. Neither a confidence interval nor a p-value for the primary comparison is reported in the registry record. Statistical significance and precision therefore cannot be evaluated from the posted primary outcome information.

What do randomization and masking contribute?

In general, random assignment aims to make treatment groups comparable on average at baseline, including with respect to factors that may not be measured. It does not guarantee that every group characteristic is identical, and it does not replace careful follow-up and analysis. Masking can reduce the influence of expectations on care, reporting, or assessment, depending on which people are masked and how masking is maintained. This trial is identified as randomized and quadruple-masked, and its official title describes it as double-blind. The specific masking roles and procedures are not reported in the registry record, so the practical operation of masking cannot be evaluated.

Why do the two stages matter?

The registry describes a Stage 1 period through Week 6 and a Stage 2 period from Day 43 to Day 85. Participants who completed Stage 1 were eligible for Stage 2; Stage 1 placebo recipients were described as re-randomized for Stage 2. Those design features mean a Week 12 analysis may address a different treatment assignment and population question from the initial Stage 1 comparison. In general, the interpretation of a staged design depends on which participants enter each stage, how they are assigned, and which contrast is prespecified. The registry record does not report the detailed statistical model or the primary comparison for each stage.

What planning calculations can and cannot show

In general, a sample-size planning illustration asks how many participants per arm would be needed under stated assumptions to detect a specified effect with a chosen power and significance level. Such an illustration is not observed power and does not tell readers how likely the trial was to detect an effect after results are known. The planning assumptions described for this record use a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. These are planning assumptions, not a report of the trial’s original design justification. The record does not report the sponsor’s sample-size justification. Because a primary comparison is not reported, the planning illustration cannot be interpreted as evidence about the study’s observed primary result.

10. Multiplicity, Missing Data and Analysis Populations

The record lists 21 outcome measures and 84 analyses. Multiple outcomes, time points, stages, or treatment contrasts can create multiplicity: when many statistical tests are considered, the chance of at least one apparently unusual result can increase. In general, a prespecified testing hierarchy or adjustment method can help control the interpretation of multiple tests. The statistical plan, including any multiplicity adjustment or hierarchy, is not reported in the registry record. It is therefore not possible to determine which analyses were confirmatory, whether the listed tests were adjusted, or how secondary results were intended to be interpreted.

The primary outcome itself includes Stage 1 and Stage 2 assessments, and the secondary outcomes also include assessments at Week 6 and Week 12. The record lists measures including the mADCS-CGIC agitation score, NPI agitation/aggression, caregiver distress, aberrant motor behavior, irritability/lability and total scores, ZBI, and CGIS-agitation. The number of analyses in the record underscores why the analysis plan and the status of each comparison matter; it does not establish that any particular result is statistically significant.

The record defines the Stage 1 mITT population as participants randomized in Stage 1 who had at least one post-baseline efficacy assessment. The Stage 2 mITT population is defined as participants randomized in Stage 2 who had at least one efficacy assessment in Stage 2. These definitions exclude randomized participants without the specified assessment from those populations. The registry record does not report the counts included in each analysis, reasons for missing assessments, methods used to handle missing data, sensitivity analyses, or whether additional analysis populations were used. The primary population description is truncated in the registry record. The effects of missing data and the robustness of the results therefore cannot be assessed.

11. 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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: urinary tract infection (1.3%), alcohol poisoning (1.3%), cerebrovascular accident (1.3%), atrial fibrillation (1.0%), abdominal pain (0.6%).

The registry lists serious adverse event counts, counts for other adverse events, and deaths for three groups. For Placebo, 194 participants were listed at risk for serious events, with 6 affected; 194 were listed at risk for other events, with 84 affected; and 1 death was listed. For AVP-786-18, the corresponding figures were 156 at risk for serious events, 17 affected, 156 at risk for other events, 78 affected, and 3 deaths. For AVP-786-28, the figures were 159 at risk for serious events, 7 affected, 159 at risk for other events, 71 affected, and 1 death.

These are registry counts, not adjusted comparisons. The record does not report the event definitions, duration of exposure by group, or a statistical analysis of these safety outcomes. Counts alone do not establish whether an event was caused by treatment, and they should not be treated as directly comparable measures of risk without the relevant design and follow-up details.

The registry lists urinary tract infection and alcohol poisoning at 1.3% each, cerebrovascular accident at 1.3%, atrial fibrillation at 1.0%, abdominal pain at 0.6%, and non-cardiac chest pain at 0.6% among the top serious terms. The denominators for these percentages are not reported in the registry record. The listed terms and percentages describe what was recorded; they do not establish causality or a comparative treatment effect.

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

The record establishes the question the trial was designed to address, the main measure used to assess agitation, and the broad randomized, placebo-controlled design. It also describes separate treatment and assessment stages and gives some adverse event counts. These details help readers understand the intended comparison and the scale on which the primary outcome was defined.

What the record does not establish is the primary outcome result. There are no arm-level CMAI changes, no difference between groups, and no confidence interval or p-value for that comparison. Without these data, it is not possible to say whether the measured scores changed in either direction, how large a difference was observed, or how precisely any difference was estimated. The absence of a reported estimate should not be interpreted as evidence of no difference.

Nor do the adverse event counts alone establish a causal safety effect. A count of affected participants is a descriptive quantity. Interpreting it comparatively requires details such as the relevant follow-up, exposure, event ascertainment, and any planned statistical analysis. The registry record does not report enough information to assess those elements for the listed safety figures.

In general, a trial result applies most directly to its defined population, interventions, comparator, outcomes, and follow-up. This trial concerned participants with agitation secondary to dementia of the Alzheimer's type and a CMAI frequency outcome. The record describes eligibility and treatment schedules only in part, and the primary population description is truncated. It does not provide enough primary efficacy results to judge the observed effect in this study or its precision.

Accordingly, the responsible interpretation is limited: the trial’s design and endpoint are described, but the posted registry record does not report the primary comparative result. The design alone cannot answer whether the outcome differed between arms, and the listed safety counts cannot substitute for the missing efficacy analysis.

13. Why This Trial Matters Statistically

This trial is statistically notable for its staged structure. Stage 1 and Stage 2 use different baseline-to-follow-up intervals, and the registry describes Stage 1 placebo participants being re-randomized for Stage 2. Such a design can address more than one treatment question, but each question depends on the participants eligible for that stage, the assignments being compared, and the analysis population. Clear reporting of those details is necessary to distinguish a continued-treatment comparison from a comparison involving a new assignment.

The outcome also illustrates why the endpoint definition matters. CMAI scores represent frequency ratings across agitated behaviors, and the record states that higher scores indicate worsening. The primary measure is change from baseline, rather than a follow-up score alone. In a staged study, the reference baseline differs by stage, so interpreting the estimate requires knowing which stage and baseline it represents.

The record lists 21 outcome measures and 84 analyses. In general, a broad set of outcomes and time points raises questions about multiplicity and prespecification. Without a reported statistical plan, readers cannot tell how the study distinguished primary from secondary claims or handled repeated testing. This is a reporting issue even when the trial design and measurement instruments are described.

Finally, the primary outcome and analysis results are not reported in the registry record. This limits statistical interpretation more than the design description alone can resolve. A randomized design can support a carefully defined comparative estimate, but that estimate, its uncertainty, and the analysis details are needed to evaluate what the study observed. The record therefore serves as a useful description of the question and measurement plan, while leaving the central comparative result unavailable.

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 most important limitation for interpreting this record is that it does not report the primary outcome estimates or the statistical comparison. The group means and standard deviations, between-group effect, confidence interval, and p-value are not reported in the registry record. As a result, the primary result’s direction, magnitude, and precision cannot be assessed.

The primary population description is incomplete, ending before the full text is available. The record does define Stage 1 and Stage 2 mITT populations, but does not report the numbers analyzed or the missing-data methods. There is no reported account of withdrawals, reasons for missing outcome measurements, or sensitivity analyses. These omissions prevent assessment of how incomplete follow-up may have affected the estimates.

The staged treatment descriptions are also not fully detailed. One Stage 1 AVP-786-28 description is truncated, and the registry record does not explain all Stage 2 assignment procedures or provide the analysis contrasts. It describes randomization and masking but not the allocation procedure or the people whose roles were masked. These limits constrain assessment of implementation and interpretation across stages.

The statistical plan, including any multiplicity strategy, is not reported in the registry record, despite the record listing multiple outcomes and analyses. The sponsor’s sample-size justification is also not reported. The planning assumptions described for sample-size illustrations do not replace the original design rationale or the missing primary result.

For safety, the record gives group counts and selected serious-event terms, but not event definitions, exposure-adjusted comparisons, or the denominators behind the selected term percentages. These limitations do not demonstrate a particular bias or treatment effect; they identify information needed to evaluate the trial’s findings more fully.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Efficacy, Safety and Tolerability of AVP-786 for the Treatment of Agitation in Patients With Dementia of the Alzheimer's Type — NCT02442765.
• Registry record last updated 2023-02-13; status last verified 2023-01.
• Results first posted 2023-02-13.
• 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.

18. Record Summary

NCT02442765 is a completed Phase 3 randomized, quadruple-masked, parallel study of AVP-786 and placebo for agitation in patients with dementia of the Alzheimer's type. Enrollment was 387. The study started on 2015-07-23, with primary completion on 2019-01-30. The design included Stage 1 and Stage 2, with the primary outcome defined as change in CMAI composite score at Week 6 and Week 12.

The registry describes the CMAI scoring direction and stage-specific baselines, and gives adverse event counts for Placebo, AVP-786-18, and AVP-786-28. It lists 21 outcome measures and 84 analyses. However, the primary comparative estimates, their uncertainty, the statistical plan, and missing-data methods are not reported in the registry record. The design and endpoint are identifiable, but the primary efficacy result cannot be evaluated from the posted registry information.