Cooperative Studies Program #563 - Prazosin and Combat Trauma PTSD: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Cooperative Studies Program #563 - Prazosin and Combat Trauma PTSD (NCT00532493), a randomized phase 3 study. The primary outcome is CAPS Recurrent Distressing Dreams Item.

2. Clinical Question

The registry also describes a longer-term objective: assessing outcomes during maintenance treatment over 26 weeks. It lists other outcomes addressing sleep quality, PTSD symptoms, clinical change, depression, quality of life, and physical functioning. The primary numerical comparison is the week 10 nightmare-item change; a week 26 result for that primary measure is not reported in the registry record.

3. Trial Design

This was a randomized, double-blind, parallel-group phase 3 trial comparing prazosin with placebo. The registry describes it as a 26 week, multi-site study in Veterans with war zone trauma-induced PTSD. Enrollment is recorded as 304. The study is marked completed; its listed start date is 2010-01-06 and its primary completion date is 2013-02-28. The sponsor was the VA Office of Research and Development.

The primary outcome was administered at baseline and week 10, with the change from baseline reported. The posted arm results include 135 participants in the prazosin group and 136 in the placebo group for this outcome. These outcome-specific counts are not the same as the overall enrollment figure. The registry states that 17 participants in the prazosin group and 16 in the placebo group had missing data on this item at week 10.

The trial’s stated objectives went beyond the week 10 primary comparison. They included evaluating nightmare symptoms, sleep disturbance, and global clinical status, as well as longer-term maintenance outcomes. Listed secondary outcomes include the Pittsburgh Sleep Quality Index, recurrent distressing dreams, clinical global change, total CAPS score, PTSD Checklist-Military Version score, PHQ-9, and SF-12 physical and mental scores.

The registry record does not report the treatment dose, the method used to generate or conceal the randomization sequence, adherence information, or the trial’s original sample-size justification. It also does not report the original analysis plan in enough detail to establish which covariates or sensitivity analyses were prespecified.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: 17 participants in the prazosin group have missing data on this item at week 10; 16 participants in the placebo group have missing data on this item at week 10

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: PTSD clinicians ascertain interest and eligibility among their patients with nightmare. Informed consent obtained; screening assessments to determine eligibility

6. Primary Endpoint

The registry lists the primary outcome as CAPS Recurrent Distressing Dreams Item, measured over This primary outcome measure was administered at baseline and week 10. The change of the 10-week from baseline was reported..

The primary endpoint was change from baseline to week 10 on the CAPS Recurrent Distressing Dreams item. The item assesses the frequency and/or severity of combat trauma-related nightmares. Its possible score range is 0-8, with higher scores indicating more severe PTSD symptoms. The outcome is expressed as a mean score change, with a standard deviation describing variation among participants in each group.

A change score expresses how the recorded value at follow-up differs from the participant’s baseline value. In the posted results, negative mean changes indicate that the average score was lower at week 10 than at baseline. The prazosin group’s mean change was -1.9, and the placebo group’s was -1.7. The posted difference is the prazosin mean change minus the placebo mean change, reported as -0.2.

This endpoint focuses on a specific symptom rather than overall PTSD status. It is relevant to the study’s stated focus on trauma-related nightmares, but it does not by itself represent every aspect of PTSD, sleep, daily functioning, or quality of life. The item’s scale and direction help interpret the observed changes, but the registry record does not report a threshold for a clinically important difference on this measure. It also does not report whether the endpoint was analyzed using an adjusted model or how missing week 10 observations were handled.

7. Results Posted in the Registry

• Prazosin Group: mean -1.90 versus -1.70 on control; difference -0.20 (95% CI -0.72 to 0.32), standardized effect d = -0.09; z-test p = 0.455.

8. Statistical Interpretation

9. Statistical Methods Explained

Randomization assigns participants to study groups by chance. In general, this helps make the groups comparable on average at the start of a trial, including with respect to measured and unmeasured characteristics. It does not guarantee that every characteristic is balanced, and interpretation still depends on how participants were analyzed and how missing observations were handled. The registry identifies this study as randomized, double-blind, and parallel-group, but does not report the sequence-generation or allocation-concealment procedures.

What does the mean change represent?

The primary measure is continuous: it records a score on a scale rather than a yes-or-no event. The registry reports a mean change for each treatment group and a standard deviation. A mean is a measure of the group’s average change. A standard deviation describes how spread out individual scores are around that average; it is not the uncertainty interval for the mean. In the posted results, the mean changes are -1.9 for prazosin and -1.7 for placebo, with standard deviations of 2.1 and 2.3, respectively.

Negative changes correspond to lower scores at follow-up than at baseline. Because higher scores on this item indicate more severe symptoms, lower scores point in the direction of fewer or less severe symptoms on this measure. A group average does not mean that every participant changed by that amount, and it does not describe how many participants improved, remained unchanged, or worsened. Individual-level distributions are not reported in the registry record.

How is the between-group estimate read?

The mean difference compares the groups’ average changes. The posted estimate is -0.2, calculated as the prazosin group’s mean change relative to the placebo group’s mean change. Its negative sign indicates that the average change was more negative in the prazosin group. The estimate is a contrast between group means, not a percentage change and not the change experienced by a typical individual.

The comparison also reports a pooled standard deviation of 2.203 and a standardized effect estimate of -0.091. Standardizing expresses a mean difference relative to the spread of scores, which can assist comparisons across measures that use different scales. It does not replace the original score-scale estimate, and it does not establish whether the difference is important to patients. The registry record does not report a clinical-importance threshold for this item.

What does the confidence interval add?

A confidence interval describes uncertainty around an estimate under the assumptions of the statistical method. In general, a wider interval indicates less precision, while a narrower interval indicates more precision. The interval for the mean difference runs from -0.724 to 0.324. Since it includes zero and values on either side, the observed estimate does not pin down the direction of the underlying group difference with precision. The interval is not a range containing every individual participant’s outcome, nor is it a statement that a particular probability attaches to the fixed true effect.

The unadjusted p-value is 0.455. In general, a p-value measures how compatible the observed data are with a specified null model, assuming that model and the analysis assumptions hold. It is not an effect-size measure, and it does not indicate whether an effect is clinically meaningful. The registry record does not report a prespecified significance threshold for the trial’s primary analysis, so the planning significance level should not be mistaken for a confirmed analysis rule.

What role do analysis assumptions play?

Any statistical comparison relies on assumptions. For continuous outcomes, common approaches may rely on how the outcome distribution behaves and on how the observations relate to one another. The registry record does not identify the sponsor’s statistical model, whether it adjusted for baseline score, or whether it used a repeated-measures approach. The recalculated comparison is an unadjusted comparison based on posted summary statistics, not a reconstruction of unreported participant-level analyses.

The reported arm counts for the primary outcome are 135 and 136, while the record also notes missing week 10 item data for 17 participants in the prazosin group and 16 in the placebo group. These registry entries make the handling of missing observations important. General approaches can include analyzing observed outcomes, imputing missing outcomes under assumptions, or conducting sensitivity analyses that examine departures from those assumptions. Which approach the trial used is not reported in the registry record.

The record lists 13 outcome measures and 10 posted secondary outcomes. When a study examines multiple outcomes, there is a general possibility that some results may appear unusual by chance. Methods for handling multiplicity can include prioritizing a primary endpoint, adjusting inference across a family of tests, or interpreting secondary findings as exploratory. The registry record does not report whether multiplicity adjustments were used, how outcome families were defined, or what prespecified hierarchy governed interpretation. The posted primary comparison should therefore be kept distinct from any unreported secondary analyses.

Finally, the recalculated estimate is based on posted summary statistics and is explicitly unadjusted. It provides a transparent comparison of the reported means and their variability, but it cannot reproduce analyses requiring individual-level data or details about the original statistical plan. Where a discrepancy is flagged, the sponsor’s own analysis takes precedence; no discrepancy is flagged for this comparison.

10. Confidence Intervals and Precision

The estimated mean difference in change was -0.2, with a confidence interval from -0.724 to 0.324. The interval communicates the uncertainty around the estimate, rather than the range of changes experienced by participants. Its inclusion of zero means that the posted summary statistics do not sharply distinguish a difference favoring prazosin from no difference or a difference in the other direction.

The interval’s span is substantial in relation to the point estimate: the data permit effects more negative than the estimate and effects that are positive. This means the point estimate alone should not be treated as a precise measurement of the underlying group contrast. The interval does not demonstrate that all values within it are equally plausible, and its interpretation depends on the assumptions of the method used to calculate it.

The p-value of 0.455 is consistent with the uncertainty conveyed by the interval. It should not be read as evidence that the groups are identical. A non-small p-value can arise when the estimated difference is near zero, when data are variable, or when information is limited; the registry result does not separate these explanations. The posted group standard deviations, 2.1 and 2.3, describe variability in scores and help explain why the estimate is not exact, but they are not themselves confidence limits.

The interval is based on the posted outcome comparison, and details of the sponsor’s confidence-interval method are not reported in the registry record. Missing week 10 data are also relevant: the record reports 17 missing item observations in the prazosin group and 16 in the placebo group but does not state how those observations were incorporated. Accordingly, the interval describes statistical uncertainty under its calculation assumptions; it does not account for every possible source of uncertainty, including uncertainty introduced by unreported missing-data methods or analysis choices.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total events. They illustrate the sample size that a future study might require under the stated assumptions, using the observed mean difference as a reference and considering planning for 100%, 75%, or 50% of that observed effect. They are not the original trial’s enrollment target or a reconstruction of its design.

The assumptions for the illustration are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect. The original trial’s own sample-size assumptions and justification are not reported in the registry record.

The 80% power column gives the number of participants per arm that the planning calculation estimates would be needed for an 80% chance of detecting the specified effect under those assumptions, if that effect were the true one. The 90% power column describes the corresponding planning target for a 90% chance. These are design-planning concepts, not probabilities that the trial’s result is correct or that a treatment effect exists. This is not observed power.

The listed planning figures rise as the target effect becomes smaller: detecting 100% of the observed effect requires fewer participants per arm than planning for 75% or 50% of it. This is a general feature of study planning: smaller differences are harder to distinguish from random variation and therefore usually require more information. The figures are illustrative because they depend on assumptions about the effect, variability, statistical test, and allocation. They should not be treated as a recommendation or as a formal assessment of the completed trial’s design.

12. 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 = -0.200; pooled standard deviation σ = 2.203.
2. n = 2 × (1.96 + 0.842)² × 2.203² / -0.200² = 1,904.0, rounded up to 1,904 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 <- -0.2                      # observed difference in means
sigma <- 2.20264288701831                     # 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(532493)
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 = 1,904 (80% power), 2,549 (90% power)
 75% of observed effect: n per arm = 3,385 (80% power), 4,532 (90% power)
 50% of observed effect: n per arm = 7,616 (80% power), 10,196 (90% power)
power.t.test: 1905 
Simulated power at n = 1,904 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.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists 13 outcome measures and 10 posted secondary outcomes, in addition to the primary nightmare-item measure. Evaluating multiple outcomes creates a multiplicity issue: the more comparisons examined, the greater the general chance that at least one may appear notable through random variation. Trials can address this by defining a primary endpoint, adjusting significance thresholds, specifying a testing hierarchy, or treating secondary analyses as exploratory. Whether any of these approaches was used is not reported in the registry record.

The primary comparison has an unadjusted p-value of 0.455. The record does not say whether this value was adjusted for other analyses, or whether the sponsor’s analysis used a multiplicity procedure. The list of outcomes alone does not establish how results were selected, ordered, or interpreted. Detailed results for the secondary measures are not included in the registry record for this analysis, so no conclusions about those outcomes can be drawn from their names alone.

Missing primary-outcome data are explicitly noted for week 10: 17 participants in the prazosin group and 16 in the placebo group lacked data on the item. The outcome-specific arm counts reported are 135 and 136. The registry record does not state whether these counts refer to a particular analysis population, nor whether participants were analyzed according to randomized assignment, treatment received, or another rule.

Missing-data methods can affect estimates if participants with missing observations differ from those with observed outcomes. General options include using only observed outcomes, modeling missing values, or testing how conclusions change under alternative assumptions. The record does not report which method was used, whether reasons for missingness were compared between groups, or whether sensitivity analyses were conducted. It also does not report an imputation strategy or a prespecified analysis population. These omissions limit assessment of the robustness of the posted comparison, without establishing that the result is biased.

14. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: suicidal ideation (1.3%), anaemia (0.7%), coronary artery disease (0.7%), myocardial infarction (0.7%), arteriovenous malformation (0.7%).

The registry reports serious adverse events affecting 18 participants in the prazosin group and 17 in the placebo group. The listed at-risk count is 152 in each group. These are group-level counts from the registry’s safety reporting and should be read separately from the outcome-specific counts used for the primary nightmare-item analysis. They do not describe the severity, duration, or causal relationship of every reported event.

Other adverse events were reported for 142 participants in the prazosin group and 139 in the placebo group, with 152 listed at risk in each group. The registry also lists serious-event terms and their reported percentages: suicidal ideation, 1.3%; anaemia, 0.7%; coronary artery disease, 0.7%; myocardial infarction, 0.7%; arteriovenous malformation, 0.7%; and retinal detachment, 0.7%. The terms and percentages do not by themselves show whether events were considered related to treatment.

Deaths are not reported in the registry record. The serious-event and other-event counts are not a complete description of the timing, recurrence, or clinical context of events. The record does not provide enough detail to compare event rates with adjusted estimates or to attribute events to either study group’s treatment. These figures describe safety reporting; they do not establish a causal treatment effect.

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

The posted primary result is a small estimated difference in mean change between groups, with an uncertainty interval that includes zero and extends in both directions. It supports a careful description of the observed data: average scores decreased from baseline in both groups, and the estimated difference in those changes was -0.2. The interval and p-value show that the comparison does not estimate the group difference precisely.

This result does not establish that the groups are exactly equivalent. Demonstrating equivalence or non-inferiority requires a prespecified margin and an analysis designed for that question; such a margin and plan are not reported in the registry record. Nor does the p-value establish that there is no effect. The evidence is compatible with several possible underlying differences, including none, within the interval’s limits and the assumptions used to calculate it.

The finding concerns one symptom item at week 10. It does not by itself answer whether there were differences in overall PTSD symptoms, sleep quality, depression, global clinical change, quality of life, or physical functioning. Those were included among the listed outcomes, but their detailed results are not available in the record details used. The longer-term objective concerned 26 week maintenance; a primary outcome result at that time is not reported in the registry record.

The comparison is also limited by incomplete week 10 data and the absence of reported detail about missing-data handling, analysis populations, and the sponsor’s statistical model. These are uncertainties about how the result was produced and should be interpreted, not proof that the reported comparison is invalid.

No medical conclusion or treatment recommendation follows from the statistical result alone. The registry data describe a randomized comparison and its uncertainty for a defined outcome and time point. They do not establish effects beyond the measure, participants, and follow-up represented in the posted result.

16. Why This Trial Matters Statistically

This trial is statistically informative because its primary outcome targets a specific symptom in the study’s stated clinical context: combat trauma-related nightmares among Veterans with PTSD. A randomized, double-blind, placebo-controlled design is intended to support a fair comparison between assigned groups while limiting some sources of bias. The registry identifies these design features, though it does not report all operational details needed to assess their implementation.

The result also illustrates why an estimate, interval, and p-value should be read together. The point estimate is -0.2, but the confidence interval ranges from -0.724 to 0.324, and the unadjusted p-value is 0.455. The point estimate gives the direction and scale of the observed contrast; the interval shows the uncertainty around it; the p-value summarizes compatibility with a null model under statistical assumptions. None of these quantities alone establishes clinical importance.

The trial’s wider objectives encompassed sleep, symptoms, clinical status, and functioning over a longer maintenance period. That breadth makes careful distinction between the primary endpoint and secondary outcomes important. Multiple outcomes can provide a fuller picture, but they also require transparent reporting of prespecified priorities and multiplicity methods. Those details are not reported in the registry record.

Finally, outcome-specific missingness and unreported analysis choices matter when interpreting precision. The posted counts and summary statistics permit a clear unadjusted comparison, but they do not reveal every decision made in the original analysis. Recognizing both the strengths of the randomized comparison and these reporting limits helps keep the statistical interpretation proportional to the evidence.

17. Limitations

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

The primary numerical result is for the week 10 change in a single CAPS item. Although the trial’s objectives included longer-term maintenance and broader clinical outcomes, the primary week 26 result and detailed secondary-outcome results are not reported in the registry record. The result therefore cannot summarize all of the trial’s stated questions.

The registry reports missing week 10 data for 17 participants in the prazosin group and 16 in the placebo group, but does not explain the reasons or analysis handling. The outcome-specific analysis counts are 135 and 136. Without information about missingness assumptions, analysis populations, imputation, or sensitivity analyses, the effect of missing data on the estimate cannot be assessed from the record.

The posted comparison is unadjusted, and the registry record does not describe the sponsor’s model, baseline adjustment, or other analysis details. It does not report the original sample-size justification, the trial’s own significance threshold, multiplicity procedures, or a clinically important difference for the primary item. The sample-size planning figures are illustrations based on the observed effect rather than the trial’s original design.

Further details of treatment delivery, including dose and adherence, are not reported in the registry record. The record identifies the study as randomized and double-blind but does not provide enough information to evaluate sequence generation, allocation concealment, or masking success. These reporting gaps limit assessment of implementation and analytic robustness; they do not, on their own, show that the trial was conducted improperly.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Cooperative Studies Program #563 - Prazosin and Combat Trauma PTSD — NCT00532493.
• Registry record last updated 2018-05-01; status last verified 2018-03.
• Results first posted 2014-06-18.
• 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.

21. Record Summary

The registry describes a completed phase 3 randomized, double-blind, parallel-group comparison of prazosin and placebo in Veterans with war zone trauma-related PTSD. Enrollment is recorded as 304. The listed start date is 2010-01-06, and primary completion is dated 2013-02-28.

The primary outcome was change from baseline to week 10 on the CAPS Recurrent Distressing Dreams item. Mean changes were -1.9 for prazosin and -1.7 for placebo; the estimated mean difference was -0.2, with a confidence interval from -0.724 to 0.324 and an unadjusted p-value of 0.455. The registry notes missing week 10 item data in both groups.

The study’s stated objectives also covered maintenance over 26 weeks and a range of secondary outcomes. Detailed results for those questions, the original sample-size rationale, and key analysis details are not reported in the registry record.