Environmental Design for Behavioral Regulation in People With Dementia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Environmental Design for Behavioral Regulation in People With Dementia (NCT04555616), a randomized na study. The primary outcome is Success Rate of Adherence.

2. Clinical Question

The clinical question is a feasibility question centered on protocol adherence, not a direct test of changes in dementia symptoms. The registry lists Change in Occupational Performance and Change in Behavior as secondary outcomes, but their results are not reported in the registry record. The primary outcome therefore speaks to whether participants met the specified adherence threshold, rather than to whether environmental design changed clinical outcomes. The posted counts and analyses should be interpreted in that limited context.

3. Trial Design

The registry describes a randomized, single-masked, parallel-group trial. It lists enrollment as 30 and status as completed. The record gives the start date as 2021-01-25 and the primary completion date as 2021-10-13. The phase is listed as NA. The sponsor is Elizabeth K Rhodus.

Participants were assigned among a control group, a standard environmental design group, and an individualized environmental design protocol group. The brief summary says participants with moderate to severe Alzheimer’s disease as their primary dementia type and their caregivers would be randomly assigned to standard care or one of the two environmental design approaches. The arm descriptions identify Alzheimer’s disease patients and caregivers in each group. The posted outcome counts are 2 in the control group, 6 in the standard environmental design group, and 7 in the individualized group; each arm reports 10 participants for this outcome.

The registry labels masking as single, but who was masked is not reported in the registry record. Further details about randomization procedures, allocation concealment, the protocol delivery, and the care received in the control group are not reported in the registry record. The record also does not report how caregivers’ roles were incorporated into the outcome assessment.

The primary outcome was assessed over 6 weeks and described as the percentage of participants who had 75% adherence to the protocol based on visits and journals. The registered primary outcome is a count of participants, and the posted arm results are counts. The registry lists two secondary outcomes, Change in Occupational Performance and Change in Behavior, but does not report their results. The trial design and posted results therefore permit a narrow reading of adherence counts, rather than a full assessment of all intended feasibility and clinical measures.

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: Of 96 patient-caregiver dyads, 60 dyads were excluded (52 did not meet inclusion criteria; 4 declined to participate; 10 for other reasons.) Patients and their caregivers were enrolled as dyads. 30 dyads were randomized (10 to control, 10 to Standardized, 10 to Individualized) and 25 dyads completed the study.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Difference of success rates among participants

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 Success Rate of Adherence, measured over 6 weeks.

The primary endpoint is the count of participants who achieved at least 75% adherence to the protocol, based on visits and journals, during 6 weeks. Although the outcome description refers to a percentage of participants, the registry identifies the measure as a count of participants, and the posted results give counts for each group. The stated population-level question is the difference in success rates among participants.

An adherence threshold is useful in a feasibility study because it indicates whether participants reached a prespecified level of engagement with the protocol. It does not, by itself, measure changes in occupational performance, behavior, or other health outcomes. The registry lists Change in Occupational Performance and Change in Behavior as secondary outcomes, but their results are not reported in the registry record. As a result, the primary endpoint can describe adherence as recorded through visits and journals, but cannot establish what participants’ adherence meant for those secondary outcomes.

The registry does not report further detail about how visits and journals were assessed, how the threshold was applied when information was incomplete, or whether adherence was adjudicated by a masked assessor. Such details would help clarify how consistently the endpoint was measured. The posted counts should therefore be understood according to the registry’s brief endpoint definition, without assuming additional measurement procedures.

7. Results Posted in the Registry

• Standard Environmental Design: 60.0% (6/10) versus 20.0% (2/10) on control; risk difference 40.0 percentage points (95% CI 0.8 to 79.2), risk ratio 3.00 (0.79–11.44); unadjusted two-proportion z-test p = 0.068.
• Individualized Environmental Design Protocol: 70.0% (7/10) versus 20.0% (2/10) on control; risk difference 50.0 percentage points (95% CI 12.3 to 87.7), risk ratio 3.50 (0.95–12.90); unadjusted two-proportion z-test p = 0.025.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary at the participant level: a participant either met the 75% adherence threshold during 6 weeks or did not. The registry expresses the posted result as a count of participants meeting that criterion. For a binary outcome, a group’s success rate describes the proportion meeting the criterion, while a risk difference compares two such rates by subtracting one from the other. In this record, the risk differences are labeled recalculated and are unadjusted pairwise comparisons against the control group.

What does a risk difference describe?

A risk difference expresses the separation between two groups’ outcome rates on an absolute scale. A value of 0 indicates equal rates; positive values indicate a higher observed rate in the first group in the comparison as presented. This measure can be easier to interpret as an absolute contrast than a ratio, but it remains an estimate based on the observed sample. The recalculated risk difference is 0.4 for standard environmental design versus control, and 0.5 for individualized environmental design versus control. These estimates are not adjusted for other factors.

What do confidence intervals and p-values contribute?

A confidence interval describes a range of values compatible with the data under the method used. It conveys uncertainty around an estimate rather than a guarantee that the true value lies within the reported interval. The recalculated interval for standard environmental design versus control is 0.008 to 0.792, and the interval for individualized environmental design versus control is 0.123 to 0.877. Their breadth indicates substantial uncertainty about the size of the pairwise differences.

A p-value measures how unusual data at least as incompatible with a specified null hypothesis would be, assuming the null hypothesis and test assumptions. It is not the probability that the null hypothesis is true, nor a measure of clinical importance. The recalculated unadjusted p-values are 0.068 and 0.025 for the two pairwise comparisons. The sponsor-reported Fisher Exact p-value is 0.12 across the three groups. The p-values address different comparisons and should not be read as if they were estimates of the same quantity.

Why does the choice of comparison matter?

A three-group test evaluates evidence of a difference somewhere among the groups, without by itself specifying which pair differs. Pairwise comparisons instead examine particular pairs, in this case each environmental design group against the control group. The registry reports the Fisher Exact method for the grouped analysis, while the posted pairwise calculations are recalculations. Because the record marks a discrepancy, the sponsor’s own analysis takes precedence. The registry does not report enough detail to determine whether the pairwise comparisons were part of the planned analysis or how multiplicity was handled.

Fisher’s exact test is a general method for testing association in categorical data, particularly useful when counts are small because it calculates the test probability from the possible tables under the null model rather than relying on a large-sample approximation. This is a general explanation of the method, not an additional claim about the trial’s analysis choices. The registry identifies Fisher Exact as the reported method but does not provide the test statistic, confidence interval, or further analytical details.

What does randomization contribute?

In general, random assignment is intended to make groups comparable on average with respect to factors known and unknown at baseline. In a small trial, chance differences can nevertheless remain, and randomization does not eliminate uncertainty in the outcome estimates. The record states that allocation was randomized, but details of the sequence generation and concealment are not reported in the registry record. A single-masked design is recorded, but the masked party is not identified.

The displayed recalculations use the posted summary counts and compare each intervention arm with the control arm without adjustment. They do not make use of individual-level data or account for covariates. No adjusted analysis, model specification, or alternative handling of missing observations is reported in the registry record. The results are therefore best read as limited summaries of the posted binary adherence outcome, not as a complete account of the trial’s statistical analysis.

10. Confidence Intervals and Precision

The pairwise recalculated estimates have wide confidence intervals. For standard environmental design versus control, the risk difference is 0.4, with a 95% confidence interval from 0.008 to 0.792. For individualized environmental design versus control, the risk difference is 0.5, with a 95% confidence interval from 0.123 to 0.877. These intervals communicate that the estimated differences are not precise: a substantial range of values remains compatible with the observed counts under the recalculation method.

Each arm reports 10 participants for the primary outcome. With this limited number in each group, each participant’s outcome can have a noticeable influence on the group count and the contrast between groups. The confidence intervals should therefore be considered alongside the observed counts, rather than as exact descriptions of the underlying effect. The registry does not report sponsor-derived confidence intervals for the three-group Fisher Exact analysis.

The recalculated intervals concern risk differences for two specific comparisons with control. They do not quantify uncertainty for every possible comparison, and they do not establish causality or clinical importance. The registry’s three-group p-value of 0.12 is a separate reported analysis and does not provide the same interval information. Since the sponsor analysis takes precedence when the results differ, the recalculated intervals are supplementary descriptions of the posted counts rather than a replacement for the sponsor’s analysis.

Precision is also limited by the narrow endpoint definition and the absence of reported results for the secondary outcomes. The record does not describe how missing visits or journals were handled, or the population included in the outcome analysis. Those omissions make it difficult to assess whether the counts reflect all randomized participants or a particular analysis set. The intervals describe statistical uncertainty for the recalculated contrasts, but cannot account for unreported design and data-handling details.

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 event counts. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to a stated fraction of the observed effect. These are planning illustrations based on the observed effect; the trial’s own original design assumptions are not reported in the registry record.

For the comparison of standard environmental design with control, planning for 100% of the observed effect gives 23 participants per arm for 80% power and 30 participants per arm for 90% power. Planning for 75% of the observed effect gives 39 participants per arm for 80% power and 52 participants per arm for 90% power. Planning for 50% of the observed effect gives 82 participants per arm for 80% power and 109 participants per arm for 90% power.

For the comparison of individualized environmental design with control, planning for 100% of the observed effect gives 15 participants per arm for 80% power and 19 participants per arm for 90% power. Planning for 75% of the observed effect gives 26 participants per arm for 80% power and 34 participants per arm for 90% power. Planning for 50% of the observed effect gives 54 participants per arm for 80% power and 72 participants per arm for 90% power.

The 80% and 90% columns describe the planned probability of detecting the specified effect under the stated assumptions if that effect were present. They are not observed power. Smaller effects require more participants in these illustrations, as reflected by the larger planned group sizes for the smaller fractions of the observed effect. Because the calculations use effects estimated from this trial, they should be read as conditional planning examples, not as proof that a future study will achieve those effects or power.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.200; treatment proportion p₁ = 0.600; average p̄ = 0.400.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6928 = 1.3579.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6325 = 0.5323.
4. n = (1.3579 + 0.5323)² / (0.4000)² = 22.3, rounded up to 23 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
p0 <- 0.2                          # control proportion
p1 <- 0.6                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(4555616)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 23 (80% power), 30 (90% power)
 75% of observed effect: n per arm = 39 (80% power), 52 (90% power)
 50% of observed effect: n per arm = 82 (80% power), 109 (90% power)
power.prop.test: 22.3 
Simulated power at n = 23 per arm: 0.825 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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. Analyses Reported by the Sponsor

The registry reports the sponsor’s own analysis as a Fisher Exact test across the control, standard environmental design, and individualized environmental design groups, with p = 0.12. Separately, the posted results include recalculated, unadjusted pairwise risk differences comparing each environmental design group with control. Their unadjusted p-values are 0.068 for standard environmental design and 0.025 for individualized environmental design.

These analyses do not ask precisely the same question. The reported three-group test evaluates the groups together, while the recalculations focus on two separate comparisons with control. The record marks a discrepancy between the reported analysis and the recalculations. Accordingly, the sponsor’s own analysis takes precedence; the recalculated pairwise results are supplementary and should not replace or overrule it.

The registry does not report a detailed analysis plan explaining which contrasts were prespecified, whether pairwise testing was intended, or how any multiple comparisons were handled. It also does not report a sponsor-derived effect estimate or confidence interval for the three-group analysis. Without those details, the difference between the reported and recalculated analyses should be understood as a difference in analysis scope and reported results, not as evidence that one result invalidates the other.

14. Multiplicity, Missing Data and Analysis Populations

The registry reports one primary outcome, Success Rate of Adherence, and lists Change in Occupational Performance and Change in Behavior as secondary outcomes. Results for the secondary outcomes are not reported in the registry record. The record lists one analysis and a Fisher Exact method across the three groups, while the posted recalculations include two pairwise comparisons against control. It does not report a detailed statistical analysis plan specifying the primary contrast, the role of pairwise testing, or whether any multiplicity adjustment was planned or used.

Multiplicity arises when several hypotheses or outcomes are examined. In general, testing more comparisons can increase the chance of finding at least one result that appears notable by chance if each test is considered separately. This general concern is relevant when reading the two unadjusted pairwise p-values alongside a three-group test, but the registry record does not establish which comparisons were prespecified or what adjustment, if any, was intended. The recalculated p-values are explicitly unadjusted and should not be read as if they account for the full set of outcomes or analyses.

Missing-data handling is not reported in the registry record. The record does not state whether every randomized participant contributed an outcome, how missing visits or journals were handled, or whether participants with incomplete adherence information were classified in a particular way. The posted outcome counts are therefore not enough to identify the analysis population with certainty.

The registry record also does not report whether the analysis followed an intention-to-treat population, a per-protocol population, or another definition. It does not describe exclusions, deviations from assigned treatment, or any sensitivity analyses. These omissions do not change the reported counts, but they limit assessment of how the outcome data were selected and how robust the findings might be to missing information or analysis choices.

15. 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.

The adverse-event record reports zero serious adverse events in the control group, zero in the standard environmental design group, and zero in the individualized environmental design protocol group. It reports 1 death in the control group, zero deaths in the standard environmental design group, and 1 death in the individualized group. The record lists 10 participants at risk for serious adverse events and 10 at risk for other adverse events in each group.

These are descriptive counts. The registry’s serious-adverse-event figures do not list any serious adverse events in the groups, but small group counts and limited information constrain interpretation. The death counts are reported separately; the registry does not provide further details about their circumstances or relationship to study participation. No serious adverse event terms are listed.

The record does not report additional detail about how adverse events were collected, how long participants were monitored for safety, or whether events were assessed for relatedness to the interventions. The counts should not be interpreted as establishing comparative safety. They describe the events recorded in the registry, without enough information to draw broader conclusions about risk.

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

The posted results show different counts of participants reaching the adherence threshold: 2 in the control group, 6 in the standard environmental design group, and 7 in the individualized environmental design protocol group. The recalculated pairwise risk differences are positive for both environmental design groups compared with control, but their confidence intervals are broad. The sponsor’s reported Fisher Exact analysis across the three groups has p = 0.12 and takes precedence over the recalculations because the record marks a discrepancy.

These findings describe adherence as measured by visits and journals over 6 weeks. They do not establish that either environmental design approach caused the observed counts, and they do not show whether participants experienced changes in occupational performance or behavior. Although these outcomes are listed as secondary outcomes, their results are not reported in the registry record.

The results also do not establish the best approach for routine care. The registry’s brief descriptions do not provide enough detail about the content of standard care, the protocols, masking, or the handling of missing information to assess all aspects of implementation. The limited group sizes and broad intervals mean the pairwise estimates leave substantial uncertainty about the magnitude of any difference.

The appropriate conclusion is narrow: the registry records adherence counts that differ across the three groups, alongside a sponsor-reported three-group test and separate recalculated pairwise contrasts. These statistics can inform consideration of feasibility and future study planning, but they are not evidence by themselves of clinical benefit, lack of benefit, or comparative treatment effectiveness. Further conclusions would require fuller reporting of the trial’s analyses and outcomes.

17. Why This Trial Matters Statistically

This trial illustrates why the question a statistic answers matters as much as the numerical result. The primary outcome is a threshold-based adherence count, and the sponsor-reported analysis compares the three groups together. The recalculated results instead compare each environmental design group separately with control. Those analyses describe different contrasts, so their p-values should not be treated as competing answers to an identical question.

It also shows how confidence intervals add information beyond a p-value. The recalculated risk differences are accompanied by broad intervals, making clear that the magnitude of the pairwise differences is uncertain. A single p-value cannot convey that range of plausible effect estimates. The registry’s reported Fisher Exact p-value adds a three-group test, but does not provide an effect estimate or interval in the posted record.

Feasibility outcomes require careful distinction from clinical outcomes. Meeting an adherence threshold may be relevant to whether a protocol can be delivered as intended, but it is not itself evidence of changes in behavior or occupational performance. Those secondary outcomes are named in the registry but their results are not reported. Keeping the endpoint’s meaning clear prevents adherence counts from being interpreted as a broader clinical result.

Finally, the record demonstrates the importance of transparent reporting of analysis plans, missing-data decisions, and analysis populations. Such information helps readers understand how reported counts were formed and how multiple comparisons were considered. In their absence, the statistics remain useful but limited descriptions of a small randomized feasibility study.

18. 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.

This trial’s posted primary outcome has limited precision: each arm reports 10 participants, and the recalculated pairwise confidence intervals are broad. The estimates therefore do not pin down the size of the differences. The registry also marks a discrepancy between its sponsor-reported analysis and the pairwise recalculations; the sponsor’s analysis takes precedence, but the record does not give a detailed explanation of the difference in analysis scope.

The outcome definition is concise. It refers to 75% adherence based on visits and journals, but the registry does not report how adherence was evaluated, whether the same procedure was used across groups, how incomplete information was handled, or who assessed the endpoint. The person masked in the single-masked design is not identified. Details of the randomization procedure and allocation concealment are not reported in the registry record.

The record does not describe the content of standard care or provide enough protocol detail to assess how the environmental design approaches differed in practice. It also does not report the original design assumptions, including the rationale for the enrollment target, or a detailed analysis plan. Missing-data procedures, analysis populations, and handling of multiple comparisons are not reported. These gaps limit assessment of the reported endpoint and its statistical interpretation.

Although Change in Occupational Performance and Change in Behavior are listed as secondary outcomes, their results are not reported in the registry record. Safety reporting is limited to counts of serious adverse events and deaths, without further event details. These limitations mean the posted information supports a focused description of adherence counts and reported tests, rather than a full assessment of outcomes, implementation, or safety.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Environmental Design for Behavioral Regulation in People With Dementia — NCT04555616.
• Registry record last updated 2023-04-12; status last verified 2023-03.
• Results first posted 2023-04-12.
• 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.

22. Record Summary

NCT04555616, Environmental Design for Behavioral Regulation in People With Dementia, is a completed randomized, single-masked, parallel-group trial with enrollment listed as 30. It compared a control group with standard and individualized environmental design protocols among people with moderate to severe Alzheimer’s disease and their caregivers. The primary outcome was adherence to the protocol over 6 weeks, defined by meeting the 75% threshold based on visits and journals.

Posted counts meeting the threshold were 2 in control, 6 in standard environmental design, and 7 in individualized environmental design. The registry reports a three-group Fisher Exact p-value of 0.12. Pairwise risk differences and unadjusted p-values are recalculated from the posted counts; because the record flags a discrepancy, the sponsor’s analysis takes precedence. Secondary outcome results are not reported in the registry record.