Prazosin Treatment for Disruptive Agitation in Alzheimer's Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Prazosin Treatment for Disruptive Agitation in Alzheimer's Disease (NCT01126099), a randomized na study. The primary outcome is Alzheimer's Disease Cooperative Study - Clinical Global Impression of Change.

2. Clinical Question

The study concerns disruptive agitation and other difficult behaviors in people with Alzheimer’s disease or a related dementia. Its stated primary outcome is a rater’s overall impression of improvement or worsening, assessed at the last visit. The registry record reports group means and standard deviations for that outcome, rather than a reported estimate of the difference between the assigned groups.

The central comparison is between participants assigned to prazosin and those assigned to placebo during the double-blind phase. The registry record describes the study as randomized, parallel-group, and quadruple-masked. It does not identify which roles were masked. The primary outcome timeframe is 12 weeks after baseline. The record also lists change in Brief Psychiatric Rating Scale Total Score and days in study as secondary outcomes, but their results are not reported in the registry record.

The open-label phase matters to interpretation: all participants took prazosin in that phase, so it is not a continued randomized comparison with placebo. The posted primary results concern the 12-week assessment, not a reported head-to-head comparison during the open-label period. The clinical question is therefore about the descriptive outcome values recorded at the end of the double-blind phase, and about how much uncertainty the registry record leaves around any treatment contrast.

3. Trial Design

The registry record describes a randomized, parallel-group study with quadruple masking. It lists prazosin and placebo as the interventions and identifies placebo as the control group. The record does not state which participants, clinicians, assessors, or other roles were included in the masking designation, so the meaning of “quadruple” cannot be specified further from the registry record.

During the double-blind phase, participants were assigned to prazosin or placebo for 12 weeks. The arm descriptions then state that all participants took prazosin during the 12-week open-label phase. Thus, only the initial phase offers the randomized prazosin-versus-placebo comparison. Outcomes from the open-label phase would not by themselves preserve the same randomized contrast, because there was no continuing placebo group described for that phase.

The registry lists enrollment as 20 and status as completed. The recorded study dates are 2010-03 for the start and 2014-03 for primary completion. The study phase is listed as NA. The sponsor is the Seattle Institute for Biomedical and Clinical Research. These registry entries describe the trial’s administrative record; they do not establish why the study enrolled the number it did or how its design choices were justified.

For the primary outcome, the registry reports 11 participants in the prazosin group and 8 in the placebo group. The outcome population is described as participants with at least one follow-up assessment. One participant in the placebo group did not return for follow-up and was not included in that outcome analysis. The adverse-event reporting uses different group counts: 11 for prazosin and 9 for placebo. These are outcome-specific reported denominators and should not be treated as interchangeable.

The study also lists two secondary outcomes: change in Brief Psychiatric Rating Scale Total Score and days in study. Their results are not reported in the registry record. The record does not report a detailed statistical analysis plan, a sample-size justification for the trial, or the identities of any additional outcome measures beyond those named in the 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.

Double Blind Phase

Open-label Extension

Analysis population for the primary outcome, as defined in the registry: Any participant who had at least one follow-up assessment was included in the analysis. One participant in the Placebo group did not return for follow-up, and therefore was not included in 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.

6. Primary Endpoint

The registry lists the primary outcome as Alzheimer's Disease Cooperative Study - Clinical Global Impression of Change, measured over 12 Weeks after Baseline.

The primary endpoint is the Alzheimer’s Disease Cooperative Study – Clinical Global Impression of Change at 12 weeks after baseline. The registry describes it as a rater’s overall impression of improvement or worsening, assessed at the last visit. It is reported in units on a scale, using a mean and standard deviation for each group.

In general, a global-impression measure can bring together a rater’s overall judgment into a single score that is easier to interpret than a list of separate symptoms. It can be useful when the clinical question concerns overall change rather than one isolated behavior. That general purpose does not make the measure objective or remove the influence of judgment: the rating depends on the rater’s assessment, and its interpretation depends on clear scale anchors and consistent assessment.

The registry’s scale description lists values from 1 to 7 and associates higher values from 4 onward with no change and worsening. However, it assigns the value 1 to both “Marked improvement” and “Moderate improvement.” That duplicated anchor is an ambiguity in the record. The reported group means are 2.5 for prazosin and 2.43 for placebo. Without a corrected scale description, the lower-end categories cannot be interpreted with complete confidence. The registry record does not report a confidence interval or a statistical comparison for this endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary result is a continuous-outcome summary: the registry reports a mean and standard deviation for each group. The outcome itself is a rating scale, but the posted analysis treats its scores numerically. The entries describe what was observed in the analyzed groups; they do not include a direct inferential comparison between prazosin and placebo.

What does a mean describe?

In general, a mean is the arithmetic average of observed scores. It gives a compact description of the group’s central value, but it does not show how many participants had each particular score. For an ordered rating scale, the categories have a meaningful order, while the numerical spacing between adjacent categories may not represent equal changes in the underlying clinical state. A mean can still be reported, as it is in this registry record, but its interpretation depends on the scale and the assumptions used to treat its scores numerically.

The posted means are 2.5 for prazosin and 2.43 for placebo. The registry also reports standard deviations of 0.25 and 0.28, respectively. In general, a standard deviation describes dispersion around a group mean. It is not the uncertainty of the mean, and it is not a confidence interval for a treatment contrast. Without individual-level outcomes or additional analysis, it cannot show the exact pattern of responses within either group.

What would a confidence interval add?

In general, a confidence interval gives a range of values that are compatible with the observed data under a specified statistical method. For a treatment comparison, an interval around the estimated between-group effect would help readers assess both the direction and precision of that estimate. A narrow interval would indicate greater precision than a wide interval, although interpretation would still depend on the endpoint’s clinical meaning and the analysis assumptions. No confidence interval for the primary comparison is reported in the registry record.

A confidence interval is not the same thing as the range of individual participant scores. It concerns uncertainty in an estimated quantity, such as a group mean or a difference between groups. The posted standard deviations concern variation among observed scores. Because the record gives group-level means and standard deviations but not an interval for the contrast, the precision of the treatment comparison cannot be reconstructed from the registry record without performing an additional analysis. No such recalculation is reported.

Why does the randomized comparison matter?

In general, random assignment is intended to make the groups comparable, on average, with respect to factors that could affect outcomes. Masking can reduce the influence of expectations on participant experience, clinical care, or outcome assessment. Those design features support an interpretable comparison during the randomized double-blind phase, but they do not replace reporting the comparison’s estimate, uncertainty, analysis population, and methods.

The registry identifies the design as randomized, parallel, and quadruple-masked. It does not specify which roles were masked or describe the allocation procedure in further detail. It also does not report an analysis model for the primary outcome. Therefore, the record supports describing the study’s design at a high level, but not a more detailed account of how the primary comparison was estimated.

How should the rating scale be interpreted?

The scale is described as a global judgment of improvement or worsening at the last visit. In general, a scale with ordered categories provides a ranking of responses, but the category definitions are essential to interpreting its scores. The registry description places “No change” at 4 and lists worsening categories at higher values. It also assigns 1 to both “Marked improvement” and “Moderate improvement.” Because those two labels share a value in the record, the scale’s lower-end labels are ambiguous. The record does not clarify whether the duplicated value is intentional or a transcription issue, so the description should not be silently corrected.

For a continuous outcome with substantial extra variation beyond a model’s assumptions, over-dispersion can make standard errors too small and confidence intervals too narrow if it is ignored. That is a general statistical concern, not a finding about this study. The registry record does not report a model, a dispersion assessment, or evidence that over-dispersion was relevant to this outcome.

What do the planning assumptions mean?

In general, a sample-size planning illustration asks how many participants might be needed under specified assumptions to detect an effect with a chosen statistical power and significance level. Such an illustration is not an observed result and is not the same as the trial’s design justification. The registry’s planning assumptions describe a two-sided significance level of 0.05, equal allocation, a normal approximation, and an effect defined as a stated fraction of the observed effect. The record explicitly distinguishes this illustration from the trial’s own design. The registry does not report the trial’s sample-size justification or a detailed statistical plan.

Planning calculations based on an observed effect are especially assumption-dependent: the observed effect is uncertain, and choosing a fraction of it changes the planning target. In general, calculations under equal allocation and a normal approximation also depend on whether those assumptions adequately represent the intended study. Such calculations do not establish the probability that this completed trial detected an effect, and they should not be described as observed power.

What analyses are posted?

The registry record reports no analysis records. It does not post a p-value, confidence interval, adjusted comparison, or unadjusted comparison for the primary outcome. It also does not report results for the two named secondary outcomes. Consequently, statistical significance and the precision of the group contrast cannot be evaluated from the registry record. This absence is not evidence that a difference is absent; it means that the relevant inferential results are not reported in the registry record.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Prazosin", "Placebo")
n   <- c(11, 8)
m   <- c(2.5, 2.43)
sp  <- c(0.25, 0.28)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Prazosin  11  2.500  0.250  2.332  2.668
CARM  Placebo  8  2.430  0.280  2.196  2.664

11. Multiplicity, Missing Data and Analysis Populations

The registry names the Alzheimer’s Disease Cooperative Study – Clinical Global Impression of Change as the primary outcome and lists change in Brief Psychiatric Rating Scale Total Score and days in study as secondary outcomes. Results for the secondary outcomes are not reported in the registry record. The record does not describe a multiplicity adjustment or explain how testing across outcomes was handled. Without that information and without posted analyses, the chance of false-positive findings across multiple tests cannot be assessed from the registry record.

The primary outcome population is defined as participants with at least one follow-up assessment. One participant in the placebo group did not return for follow-up and was excluded from that endpoint analysis. The registry reports primary-outcome group counts of 11 for prazosin and 8 for placebo. It does not say whether the absent follow-up was handled using imputation, a sensitivity analysis, or another method. It also does not report the reason for the missing assessment, so its relationship to the participant’s outcome cannot be judged.

In general, missing outcome data can affect a comparison if the likelihood of missingness is related to treatment assignment or to the outcome itself. The consequences depend on why data are missing and on the assumptions used in the analysis. Those assumptions and any missing-data analyses are not reported in the registry record.

The adverse-event denominators differ from the primary endpoint denominators: the safety reporting lists 11 participants in the prazosin group and 9 in the placebo group. These outcome-specific counts should be kept separate. The registry does not report a general statistical analysis plan, an analysis population for all outcomes, or a multiplicity strategy.

12. 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 5% reporting threshold.

Most frequent non-serious adverse event terms

The serious adverse-event entries report 0 affected participants among 11 at risk in the prazosin group and 0 among 9 at risk in the placebo group. These are counts within the safety denominators reported by the registry. They describe the serious adverse events recorded for this study and should not be interpreted as a comparative risk estimate.

A zero count does not establish that the underlying risk is zero. In general, when no events are observed, the data still leave uncertainty about event rates, particularly in a study with a limited number of participants. The registry does not report confidence intervals or a statistical comparison for serious adverse events. It also does not list serious adverse-event terms in the posted record.

The record separately indicates that other adverse events affected all participants in the reported safety denominator for each group: 11 in the prazosin group and 9 in the placebo group. Those entries are distinct from the serious adverse-event counts and should not be conflated with them. The deaths fields are not populated, so deaths are not reported in the registry record; a missing entry should not be read as a reported zero.

These figures are a limited account of safety reporting in this trial. They do not provide event descriptions for non-serious adverse events, comparative uncertainty, or information beyond the recorded study participants and reporting period.

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

The posted primary outcome values are group-level descriptive summaries at 12 weeks after baseline: a mean of 2.5 with a standard deviation of 0.25 for prazosin, and a mean of 2.43 with a standard deviation of 0.28 for placebo. They describe the analyzed participants’ ratings. The registry record does not report a between-group effect estimate, confidence interval, or p-value, so the precision and inferential meaning of any contrast are unknown from this record.

The findings do not, by themselves, establish that prazosin changes disruptive agitation, nor do they establish that the treatments have the same effect. A conclusion about a difference requires an appropriate comparison and an assessment of uncertainty; a conclusion about clinical importance also requires a clear understanding of what the scale scores mean. The duplicated improvement anchor in the scale description makes the lower end difficult to interpret without clarification.

The results should also be understood in the context of the study’s phases. The randomized comparison was during the 12-week double-blind phase. During the 12-week open-label phase, all participants took prazosin, so that phase does not provide a concurrent placebo comparison as described in the registry. The primary outcome is specified at 12 weeks after baseline, and no later comparative result is reported in the registry record.

Follow-up was incomplete for one placebo-group participant, who was excluded from the primary outcome population. The record does not report an analysis of how this missing outcome might affect the comparison. The safety counts are also not a substitute for efficacy analysis: zero serious adverse events were recorded in the listed safety denominators, while deaths are not reported in the registry record.

The most defensible reading is therefore limited: the registry posts outcome summaries for the analyzed groups and certain safety counts, but not the inferential results needed to quantify a treatment contrast or its uncertainty. It is not possible from these registry entries alone to determine the statistical strength, precision, or clinical importance of a difference between groups.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of how a randomized design and a reported result are not the same thing. Random assignment and masking are important features of a comparative study, but readers also need an explicit effect estimate and an account of its uncertainty. The registry describes the design and provides group summaries, yet it does not report an inferential comparison for the primary endpoint.

The endpoint illustrates a further issue: summary statistics must be read alongside the measurement definition. The global-impression score is intended to represent a rater’s overall judgment of change, but the scale description assigns the same value to two different degrees of improvement. That ambiguity limits the clarity of the reported means. Statistical analysis cannot compensate for unclear outcome anchors; good interpretation requires both a usable measure and transparent reporting.

The record also distinguishes endpoint analysis from safety reporting. The primary outcome excludes one placebo-group participant without follow-up, while safety denominators include a different placebo count. This is a reminder that denominators can vary across outcomes and that each analysis population should be identified. Treating one reported group count as universal could misrepresent what was analyzed.

Finally, the registry lists secondary outcomes but does not report their results or a multiplicity approach. In general, multiple outcomes raise questions about which results were prespecified and how the chance of false-positive findings was managed. The absence of posted analyses leaves those questions unanswered. These features make the record useful for teaching careful separation of design, descriptive results, analysis populations, and inferential conclusions.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record leaves several trial-specific questions unanswered. It identifies the study as quadruple-masked but does not say which roles were masked. The record also does not provide an allocation procedure beyond stating that assignment was randomized, or detail the statistical model used for the primary outcome. These omissions limit how specifically the analysis can be assessed.

The primary endpoint’s scale description is internally ambiguous: it assigns 1 to both marked and moderate improvement. The record does not explain or correct this duplication. Because the endpoint is a rater’s overall impression, interpretation also depends on how ratings were made and whether assessment procedures were consistent; those details are not reported in the registry record.

The primary outcome analysis includes participants with at least one follow-up assessment, and one placebo-group participant did not return for follow-up. The reason is not reported, and no missing-data method or sensitivity analysis is described. The record does not report individual scores or response-category counts, which would help readers understand the distribution behind the means and standard deviations.

No inferential comparison is reported for the primary endpoint, and results for the named secondary outcomes are not reported in the registry record. A multiplicity approach, statistical analysis plan, and trial-specific sample-size justification are also not reported. The planning assumptions described in the record are explicitly an illustration based on an observed effect, rather than the trial’s own design rationale.

Safety reporting is limited to the listed counts. Serious adverse-event terms are not listed, deaths are not reported in the registry record, and no comparative uncertainty is given. Finally, the open-label phase assigned prazosin to all participants, so it does not offer a concurrent randomized placebo comparison as described. These limitations concern what can be established from the registry record and do not resolve questions that may be addressed elsewhere.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Prazosin Treatment for Disruptive Agitation in Alzheimer's Disease — NCT01126099.
• Registry record last updated 2015-05-21; status last verified 2015-05.
• Results first posted 2015-05-21.
• 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.

19. Record Summary

NCT01126099 is a completed randomized, parallel-group, quadruple-masked study sponsored by the Seattle Institute for Biomedical and Clinical Research. The registry lists enrollment as 20, a start date of 2010-03, primary completion in 2014-03, and phase NA. Participants with Alzheimer’s disease or a related dementia and difficult behaviors were assigned to prazosin or placebo during a 12-week double-blind phase; all participants took prazosin during the 12-week open-label phase.

The primary outcome was the Alzheimer’s Disease Cooperative Study – Clinical Global Impression of Change at 12 weeks after baseline. The posted group means and standard deviations are 2.5 and 0.25 for prazosin, and 2.43 and 0.28 for placebo. The registry record does not report a comparative analysis or confidence interval. One placebo-group participant did not return for follow-up and was excluded from the primary outcome analysis. Secondary outcomes are named, but their results are not reported in the registry record.