Enhancing Consent for Alzheimer's Research: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Enhancing Consent for Alzheimer's Research (NCT00453544), a randomized na study. The primary outcome is Level of Understanding.

2. Clinical Question

The outcome was assessed during the session, approximately 2-3 minutes after completion of the simulated consent process. This timing makes the endpoint a measure of immediate understanding after a consent exercise, rather than a measure of later recall or decisions about real research participation. The posted results give means and standard deviations for groups defined by diagnosis, consent format, and hypothetical risk. The registry record does not post a formal analysis of the central enhanced-versus-routine consent question, nor does it report results for the proposed predictor analyses. The arm summaries and recalculated comparisons therefore describe the recorded scores, but do not by themselves establish whether one consent approach produces a different level of understanding within each diagnostic group.

3. Trial Design

This was a completed, randomized, parallel-group study with single masking. The registry lists the phase as not applicable. Enrollment was 252, and the record identifies the study as sponsored by the University of California, San Diego. The study started in 2006-04 and its primary completion was 2012-09.

Participants were studied in diagnostic groups: people with Alzheimer disease and non-neuropsychiatric comparison subjects. Within those groups, the registry describes conditions involving enhanced or routine consent and hypothetical protocols at moderate or higher risk. The enhanced procedure included a multimedia aid; the routine procedure included a printed consent document. The protocols were hypothetical, so the posted outcome concerns a simulated consent exercise rather than enrollment in the described research protocols.

Although the record identifies randomization and single masking, details about the randomization procedure, allocation concealment, and who was masked are not reported in the registry record. Details of how participants were assigned among the diagnosis, consent-format, and risk combinations are also not reported in the registry record. The arm summaries list the number of observations contributing to each outcome group, but they should not be treated as a substitute for a complete account of participant flow or allocation.

The registry lists a primary outcome and secondary outcomes, but the number of posted registry analyses is 0. That means the results available in the record are group-level summaries rather than a registry-posted inferential analysis plan with reported adjusted tests. The study’s original sample-size assumptions, planned analytic model, and any interim analyses are not reported in the registry record. The displayed recalculated comparisons are unadjusted comparisons based on the posted summaries, and they do not recreate an unavailable original 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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: 114 people with Alzheimer's disease (AD); 134 non-neuropsychiatric comparison (NC) subjects; the effects of routine vs. enhanced consent and moderate vs. higher risk were analyzed within diagnostic groups separately

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: 259 subjects met inclusion/exclusion criteria, but 8 withdrew prior to the randomization (only two due to study - i.e. 2 found assessments too frustrating); 3 additional subjects had missing data key outcomes; resulting in 248 subjects in the final data analysis..

6. Primary Endpoint

The registry lists the primary outcome as Level of Understanding, measured over Within session - administered approximately 2-3 minutes after completion of the simulated consent process..

The primary endpoint is the level of understanding, measured as a score on the modified MacArthur Competence Assessment Tool for Clinical Research (MacCAT-CR) Understanding subscale. The score ranges from 0 to 26, with higher scores representing better performance. It was administered within the study session, approximately 2-3 minutes after the simulated consent process was completed.

Understanding is relevant to a consent study because the intervention is intended to communicate information about a research protocol. A measure of comprehension can capture whether participants correctly understand material presented during the consent process. It is distinct from satisfaction with the process and from willingness to participate: a participant’s score on understanding does not itself establish how satisfied that person felt or what participation decision they would make.

The measure was modified for this study and applied to both people with Alzheimer disease and non-neuropsychiatric comparison subjects. The registry says that the effects of consent format and hypothetical protocol risk were analyzed within diagnostic groups separately, but it does not report a formal analysis of those effects. The posted means therefore describe performance in the listed groups; they are not, on their own, a complete estimate of the causal effect of enhanced versus routine consent.

Because assessment followed shortly after a simulated consent exercise, the endpoint reflects immediate performance in that setting. The registry does not report a later assessment of retention, understanding in an actual research decision, or whether the score predicts subsequent participation behavior. These distinctions define what this endpoint can describe and what conclusions would require additional evidence.

7. Results Posted in the Registry

• AD Enhanced Consent - Higher Risk: mean 16.80 versus 17.80 on control; difference -1.00 (95% CI -5.06 to 3.06), standardized effect d = -0.14; z-test p = 0.629.
• AD Routine Consent - Mod Risk: mean 15.10 versus 17.80 on control; difference -2.70 (95% CI -6.28 to 0.88), standardized effect d = -0.39; z-test p = 0.139.
• AD Routine Consent - Higher Risk: mean 15.70 versus 17.80 on control; difference -2.10 (95% CI -5.92 to 1.72), standardized effect d = -0.28; z-test p = 0.281.
• NC Enhanced Consent - Mod Risk: mean 25.20 versus 17.80 on control; difference 7.40 (95% CI 4.68 to 10.12), standardized effect d = 1.58; z-test p = <0.001.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Level of Satisfaction With Consent Process

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • AD Enhanced Consent - Higher Risk: mean 2.40 versus 1.30 on control; difference 1.10 (95% CI -0.23 to 2.43), standardized effect d = 0.55; z-test p = 0.105.
  • AD Routine Consent - Mod Risk: mean 2.00 versus 1.30 on control; difference 0.70 (95% CI -0.55 to 1.95), standardized effect d = 0.33; z-test p = 0.271.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between AD Enhanced Consent - Higher Risk and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between AD Routine Consent - Mod Risk and the control. The interval gives the range of effects that remain compatible with the data.

The registry identifies satisfaction with the consent process and expressed willingness to participate in the hypothetical protocol as secondary outcomes. The posted secondary summaries available in the record concern satisfaction, while the results for expressed willingness to participate are not reported in the registry record. Satisfaction is a distinct construct from the primary understanding score: a process can be experienced differently without that difference necessarily indicating better or poorer comprehension.

Across the listed satisfaction groups, the group means are relatively close, and the recalculated comparisons displayed for the Alzheimer disease groups have confidence intervals that include zero. The summaries therefore do not establish a clear difference in satisfaction for those comparisons. As with the primary endpoint, these are unadjusted calculations against a reference arm and are not a formal registry-reported analysis of the main consent-format question.

The secondary results do not resolve the uncertainty in the primary result. They measure a different aspect of the consent experience, and the registry does not report a joint analysis linking satisfaction to understanding or explaining whether satisfaction predicts willingness. Nor does the record post inferential results for willingness to participate. It is therefore not possible to use the secondary outcomes to confirm, qualify, or explain the primary pattern in a formal way.

Secondary outcomes and multiple comparisons generally carry less confirmatory weight than a prespecified primary analysis, especially when the analysis plan and multiplicity approach are unavailable. Testing several outcomes or contrasts increases the chance that some results will appear notable by chance. The registry lists 0 posted analyses, so the original hierarchy of hypotheses and any adjustment strategy are not reported in the registry record. These outcomes remain informative as descriptive measures, but should not be treated as independent confirmation of a treatment effect.

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

• Level of Satisfaction With Consent Process
• Expressed Willingness to Participate in the Hypothetical Protocol

9. Statistical Interpretation

10. Statistical Methods Explained

The registry posts continuous outcomes as group means and standard deviations. A mean gives the average score in a group; a standard deviation describes how dispersed individual scores are around that average. Together, these summaries convey both the typical score and the variability among participants. They do not show the full distribution, and they do not indicate whether the same individuals contributed to every outcome.

What does a mean difference represent?

A mean difference compares the average score in one listed group with the average score in a reference group. Positive values indicate a higher mean in the compared group than in the reference; negative values indicate a lower mean. For example, the recalculated contrast between AD Enhanced Consent - Higher Risk and AD Enhanced Consent - Mod Risk is -1.0, with a 95% confidence interval from -5.06 to 3.06. This is an unadjusted comparison of the posted summaries, not a result reported as a formal sponsor analysis in the registry.

Several displayed comparisons use AD Enhanced Consent - Mod Risk as the reference, including comparisons with non-neuropsychiatric groups. Those contrasts combine differences in diagnosis with differences in consent or hypothetical risk. They should not be interpreted as isolated effects of consent format. A direct comparison intended to isolate consent format would need to compare the relevant randomized groups within the same diagnostic and risk context, with an analysis that respects the study design.

How should confidence intervals be read?

A confidence interval describes the uncertainty around an estimated difference under the assumptions of the analysis. A narrower interval indicates greater precision; a wider interval leaves a broader range of values compatible with the data and method. When an interval for a mean difference includes zero, the data do not rule out no difference at the confidence level used for that interval. This does not prove the groups are identical. Establishing equivalence or non-inferiority requires a prespecified margin and a suitable analysis; those details are not reported in the registry record.

The recalculated comparisons include standard errors derived from the posted sample sizes and standard deviations. They are unadjusted: they do not account for a family of multiple comparisons, covariates, or a more complex model unless such adjustment is explicitly described. The registry lists 0 posted analyses, and the study’s planned statistical methods are not reported in the registry record. Thus, these recalculations help readers understand the summaries but should not be mistaken for the trial’s original analysis.

What does a p-value tell the reader?

A p-value describes how compatible the observed data are with a specified null hypothesis under the test assumptions. It is not the probability that the null hypothesis is true, and it does not measure the size or practical importance of an effect. The p-values attached to the recalculated comparisons are unadjusted. With multiple outcomes and contrasts, an isolated small p-value has less confirmatory weight unless the analysis plan and approach to multiplicity are known. Those details are not reported in the registry record.

The non-neuropsychiatric groups have means close to the scale maximum of 26. When observations cluster near the upper bound of a measure, a ceiling effect can limit the measure’s ability to distinguish groups that may differ in underlying understanding. That is a general measurement concern, not proof that the score fails to measure understanding. The registry provides means and standard deviations, but not score distributions or item-level results that would clarify how much the upper bound affected the observed scores.

Why do diagnostic groups need separate interpretation?

The study description says that consent and hypothetical-risk effects were analyzed within diagnostic groups separately. This is important because the study includes populations with notably different score distributions. A cross-group comparison does not isolate an intervention effect; diagnosis and other group characteristics may differ as well. A suitable within-group analysis would compare consent conditions for participants with the same diagnosis and the same hypothetical risk, while accounting for the allocation structure. A model could also assess whether patterns differ between diagnostic groups, but the registry does not report such a model or its results.

General principles of trial analysis favor using a prespecified approach that matches the randomization and outcome structure. The record does not report the allocation details needed to evaluate that structure fully, nor does it report adjustments for covariates or missing outcomes. These omissions do not invalidate the descriptive means, but they limit how far the recalculated comparisons can support causal interpretation.

11. Confidence Intervals and Precision

Precision refers to how tightly an estimate is determined by the available observations and the variability in scores. The confidence intervals for the Alzheimer disease comparisons are broad relative to their estimated differences. For example, the recalculated mean difference for AD Enhanced Consent - Higher Risk versus AD Enhanced Consent - Mod Risk is -1.0, with a 95% confidence interval from -5.06 to 3.06. The interval spans values in both directions, so the data are compatible with a lower mean in the higher-risk group, little difference, or a higher mean.

The other displayed comparisons among the AD groups also have confidence intervals that include zero: -2.7 with an interval from -6.275 to 0.875 for AD Routine Consent - Mod Risk, and -2.1 with an interval from -5.919 to 1.719 for AD Routine Consent - Higher Risk, each compared with AD Enhanced Consent - Mod Risk. These intervals indicate uncertainty; they do not show that the groups are equivalent. The record does not report an equivalence margin or a formal equivalence analysis.

For comparisons involving non-neuropsychiatric groups, the recalculated intervals are separated from zero, but these are cross-diagnostic contrasts against an Alzheimer disease reference group. Their precision does not solve the interpretive problem that diagnosis differs between groups. Statistical precision and causal relevance are separate considerations.

Group sizes and score variability both affect precision. The posted summaries show more variable scores in the Alzheimer disease groups than in the non-neuropsychiatric groups. The registry does not report a formal power calculation for the original study, its target precision, or its original sample-size justification. Consequently, the confidence intervals are useful for describing uncertainty in these recalculations, but they cannot establish whether the study was designed to detect a particular consent-format difference.

12. Planning a Similar Trial: Sample Size

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

The planning figures are participants per arm, not total events. They are illustrative calculations based on the observed effect, rather than the study’s original design assumptions. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. For each scenario, the effect to detect is the stated fraction of the observed effect.

The 80% power column gives the estimated participants per arm for a design with an 80% chance of detecting the specified effect under those assumptions, if that effect and the calculation model were correct. The 90% power column gives the corresponding per-arm planning figure for a 90% chance. These are planning probabilities, not statements about whether a particular result is true. Smaller effects need more participants to be detected with the same planned power, which is why the displayed requirements increase when planning for a smaller fraction of the observed effect.

The numbers use effects estimated from the observed comparisons, so they should not be treated as universal requirements for future studies. In particular, an observed effect can be unstable, and a planning illustration based on a cross-diagnostic contrast may not represent the effect of consent format that a future study intends to estimate. A future design would need assumptions matched to its target comparison, outcome variability, allocation, and analysis plan. The original design assumptions are not reported in the registry record.

This is not observed power. It is a prospective planning illustration under stated assumptions, not a post-study measure of what the trial achieved. The registry record does not report the original sample-size justification, so these figures do not explain why the study enrolled its participants.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -1.000; pooled standard deviation σ = 7.294.
2. n = 2 × (1.96 + 0.842)² × 7.294² / -1.000² = 835.1, rounded up to 836 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 <- -1                      # observed difference in means
sigma <- 7.29366163185543                     # 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(453544)
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 = 836 (80% power), 1,118 (90% power)
 75% of observed effect: n per arm = 1,485 (80% power), 1,988 (90% power)
 50% of observed effect: n per arm = 3,341 (80% power), 4,472 (90% power)
power.t.test: 836 
Simulated power at n = 836 per arm: 0.797 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome and secondary outcomes, but it reports 0 posted analyses. It does not describe how many hypotheses or contrasts were planned, which comparisons were primary, or whether a multiplicity adjustment was intended. This matters because testing multiple consent conditions, risk levels, diagnostic groups, and outcomes can generate a collection of results that need to be interpreted together rather than one p-value at a time.

The recalculated p-values are unadjusted. They do not account for the number of comparisons considered. A small unadjusted p-value can be less persuasive when it arises among many tests, while a non-small p-value does not establish that there is no meaningful difference. Without a stated hypothesis hierarchy or multiplicity strategy, the strength of confirmatory claims is limited.

Participant flow and missing outcome data are also not described in enough detail to assess their potential influence. The arm summaries show different numbers of participants contributing to particular outcomes. The record does not explain why these numbers vary, how missing outcomes were handled, whether any participants were excluded from analyses, or whether the analyses followed an intention-to-treat approach. It also does not report imputation or sensitivity analyses for missing data.

The registry describes separate analyses within diagnostic groups, but does not post their methods or results. Details about covariate adjustment, model specification, handling of the risk conditions, and assessment of interactions are not reported in the registry record. The stated interest in factors and patient characteristics that might predict benefit from enhanced consent is not accompanied by posted predictor analyses.

These missing details do not change the arithmetic of the posted group means, but they constrain assessment of bias, robustness, and the intended statistical interpretation. Readers can describe the observed summaries and the unadjusted recalculations, while recognizing that the registry does not document the analysis decisions needed to judge whether the original hypotheses were tested as planned.

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

The results describe understanding scores recorded shortly after participants completed a simulated consent process for a hypothetical protocol. They show how the listed groups performed on a modified understanding measure, with higher scores representing better performance. The Alzheimer disease groups have means in a lower range than the non-neuropsychiatric comparison groups, while the latter have scores close to the measure’s upper limit.

Those patterns do not establish that enhanced consent improves understanding compared with routine consent. The registry does not post a formal analysis of the consent-format effect within each diagnostic group, and several recalculated contrasts compare groups that differ in diagnosis as well as consent or hypothetical risk. The unadjusted comparisons among the Alzheimer disease groups have confidence intervals that include zero, but that is not proof that the consent procedures are equivalent or that any difference is absent.

The score is an immediate measure in a simulated setting. It does not show whether participants retained the information later, understood every aspect of a real research protocol, felt satisfied, or chose to participate. Satisfaction is listed as a secondary outcome, and willingness to participate is named as a secondary outcome, but willingness results are not reported in the registry record.

The study description also refers to identifying characteristics that predict whether enhanced consent is more beneficial than routine consent. Such predictor results are not reported in the registry record. As a result, the posted evidence cannot identify which participants might experience different effects or explain why group scores vary.

In short, the record offers descriptive data that can inform questions about measurement and study design. It does not provide a complete, formally reported estimate of the central causal comparison. Conclusions should stay close to the immediate scores and the uncertainty in the recalculated comparisons, rather than extending to real-world consent decisions or later outcomes.

16. Why This Trial Matters Statistically

This study raises a useful statistical problem: a consent intervention is intended to affect understanding, yet the outcome is measured in populations whose baseline score distributions differ substantially. An analysis needs to distinguish the effect of consent format from differences associated with diagnosis and from the stated risk level of the hypothetical protocol. Comparisons across diagnostic groups can look numerically large while answering a different question from the randomized contrast of interest.

The upper end of the understanding scale is also relevant. The non-neuropsychiatric group means are near the maximum score of 26, leaving limited room to represent further differences. A ceiling can compress score variation and make small distinctions harder to detect. This is a general measurement issue that matters when selecting an endpoint and interpreting apparent similarities between high-scoring groups.

The record also illustrates why effect estimates need uncertainty intervals and a clear comparison definition. A mean difference without its reference group, confidence interval, and population context can be misleading. Here, the cross-diagnostic comparisons are precise relative to some of the AD comparisons, but they do not isolate the consent procedure. Precision cannot compensate for a comparison that does not answer the intended causal question.

Finally, registry transparency about the planned analysis, missing data, multiplicity, and sample-size rationale is essential for evaluating how confirmatory results should be. The group summaries permit useful descriptive reading, but the lack of posted formal analyses leaves the principal inferential question unresolved in the registry record. This distinction is central to responsible interpretation of clinical research data.

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 findings concern a simulated consent process for hypothetical protocols, not an actual decision to enroll in the protocols described. The primary endpoint was administered approximately 2-3 minutes after the exercise, so it reflects immediate understanding rather than longer-term retention. The registry does not report later assessments or whether the scores correspond to understanding in other consent settings.

The primary measure is described as a modified version of the MacCAT-CR Understanding subscale. The record gives the score range and direction but does not report item-level results, reliability information, or details sufficient to assess the effect of the modification. Group-level means and standard deviations also do not reveal score distributions, which is particularly relevant when scores cluster near the scale maximum.

The registry reports randomized allocation and single masking, but does not describe the allocation procedure or identify who was masked. It does not report participant flow, reasons for varying outcome counts, missing-data handling, or the analysis population. These omissions limit assessment of selection, attrition, and robustness.

The posted comparisons are unadjusted recalculations, and the registry lists 0 posted analyses. The original statistical plan, multiplicity strategy, covariate adjustments, and sample-size justification are not reported in the registry record. Some displayed contrasts cross diagnostic groups and therefore do not isolate consent format. Predictor analyses described as an objective of the study are also not reported.

Finally, the registry’s secondary outcomes include willingness to participate, but results for that outcome are not reported in the registry record. The posted satisfaction summaries do not substitute for willingness or establish a relationship between satisfaction and understanding. Together, these limitations restrict conclusions to descriptive patterns in the recorded outcomes.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Enhancing Consent for Alzheimer's Research — NCT00453544.
• Registry record last updated 2023-06-06; status last verified 2023-06.
• Results first posted 2023-06-06.
• 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

NCT00453544, Enhancing Consent for Alzheimer's Research, is listed as a completed randomized, parallel-group study with single masking and enrollment of 252. It was sponsored by the University of California, San Diego. The registry lists a start date of 2006-04 and primary completion in 2012-09.

The primary outcome was understanding measured with a modified MacCAT-CR Understanding subscale after a simulated consent process. The study compared enhanced consent using a multimedia aid with routine consent using a printed document, for hypothetical protocols described as moderate or higher risk. The registry describes separate analyses within Alzheimer disease and non-neuropsychiatric comparison groups.

Posted results include group means and standard deviations, but the registry lists 0 posted analyses. The displayed comparisons are unadjusted recalculations, and the original analysis plan, missing-data methods, multiplicity approach, and sample-size justification are not reported in the registry record. Results for willingness to participate are also not reported in the registry record.