Improvement In Scratch Behavior And Sleep In Patients With Atopic Dermatitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Improvement In Scratch Behavior And Sleep In Patients With Atopic Dermatitis (NCT05200403), a randomized phase 4 study. The primary outcome is Number of Children's Nighttime Scratching Episodes.

2. Clinical Question

The trial’s stated aim was to evaluate nighttime itch and scratching, and sleep, in children with atopic dermatitis. The primary endpoint focuses specifically on recorded nighttime scratching episodes. The study description also identifies caregiver quality of life and sleep as topics of interest, but results for those outcomes are not reported in the registry record.

The reported primary endpoint values are descriptive: they give a mean scratch count per night and a standard deviation for each group. They do not, by themselves, establish whether the groups differed beyond what might arise from sampling variation. The registry record does not report an inferential comparison, confidence interval, or p-value for this endpoint.

The question is therefore narrower than whether treatment changes itch, sleep, or quality of life overall. It concerns device-recorded nighttime scratching in the available endpoint population during the specified 2-week period. It does not directly answer whether participants or caregivers experienced a perceived change, or whether any change persisted beyond that period.

3. Trial Design

The registry describes a randomized, single-blind, parallel-group Phase 4 study sponsored by Boston University. It lists enrollment of 72 and the condition as atopic dermatitis. The study began on 2022-07-26, has a primary completion date of 2024-02-15, and is listed as terminated. The reason for termination is not reported in the registry record.

In a parallel-group design, participants are assigned to one of the treatment groups and followed in that group rather than crossing over between treatments. Random allocation is intended, in general, to reduce systematic differences between groups at the point of assignment. The registry record does not report the method used to generate or conceal the allocation sequence.

The two treatment groups were crisaborole 2% and vehicle treatment. Caregivers were asked to apply the assigned treatment twice daily for 2 weeks. The registry labels the study single blind, but which people were masked is not reported in the registry record. It also does not report whether masking was assessed or how treatment application and adherence were documented.

Accelerometry devices were used to record the children’s nighttime scratching. The record identifies devices for children and adult caregivers, but only the children’s nighttime scratching endpoint has posted results in the material described. The primary endpoint analysis population was limited by device data availability: the record says some data were excluded because of device malfunction, non-wear, or missing data.

The registry reports 3 outcome measures and 0 posted secondary outcomes. It does not report a statistical analysis plan or an explanation of the study’s termination. The registered enrollment should not be treated as the number contributing to the primary endpoint: the posted endpoint population is smaller because processable data were unavailable for some participants.

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: For the crisaborole group, there were 11 participants that had sufficient processable data. For the Vehicle group, there were 8 participants that had sufficient processable data. The rest were not included due to device malfunction, non-wear, or missing data.

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 Number of Children's Nighttime Scratching Episodes, measured over 2 weeks.

The primary endpoint is the number of children’s nighttime scratching episodes, measured as scratch count per night over 2 weeks. The registry says episodes were recorded by accelerometry. This is a count-based measure of scratching behavior, rather than a direct report of itch intensity or a measure of sleep quality.

Nighttime scratching is relevant to the study’s stated focus because itch and scratching can occur during sleep hours, and the study description identifies relief of nighttime itch and scratching as a central question. A device-based measure can record movement patterns without relying entirely on a child’s or caregiver’s recollection. However, accelerometry does not by itself establish why a movement occurred. The registry record does not describe the device-processing rules used to classify an episode or distinguish scratching from other movement.

The endpoint is summarized using a mean and standard deviation for each group. In general, the mean describes the average recorded count, while the standard deviation describes how spread out individual observations are around that average. A group mean does not show the full distribution of counts or whether a small number of observations strongly influenced it.

The registry reports endpoint data only for participants with sufficient processable device data. The resulting values therefore describe that available-data population, not necessarily every randomized participant. The record does not report a separate estimate of sleep, a subjective itch score, or a caregiver outcome alongside the primary endpoint results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry reports a continuous endpoint summary based on a count per night. Although the underlying quantity is a count, the posted summary uses a mean and standard deviation. Those summaries are useful for describing the observed data, but they do not replace a specified comparison model or an uncertainty estimate for the difference between groups.

What does the mean describe?

In general, a mean is the arithmetic average of the measurements in a group. For this endpoint, it represents the average number of device-recorded nighttime scratching episodes per night among participants included in the group’s endpoint summary. It does not mean that every child had that number of episodes, nor does it convey the range or shape of the observations.

The reported means are 93.04 for crisaborole and 116.8 for vehicle. They are summaries of the observed groups with processable data. The registry does not report an estimated treatment effect, so these two values should not be treated as a formal estimate of the between-group effect. A valid inferential comparison would need to account for the study’s analysis plan and data structure; those details are not reported in the registry record.

What does the standard deviation describe?

In general, a standard deviation describes how dispersed measurements are around a group mean. Larger spread can make an average less informative about the experience of a typical participant and can make a group comparison less precise. The reported standard deviations are 70.35 for crisaborole and 62.17 for vehicle. The registry does not report confidence intervals for the group means or for a between-group comparison.

Because the endpoint is a count, its distribution may not be well described by a normal curve in every setting. In general, count outcomes can be skewed, and their variability can differ from what a simple normal model assumes. Over-dispersion is a general term for variability greater than a model expects; when present, it can make standard errors from an unsuitable model too small. Whether the endpoint data show over-dispersion, and what model was used to assess it, are not reported in the registry record.

Why does the analysis population matter?

The endpoint summaries include participants with sufficient processable accelerometry data: 11 in the crisaborole group and 8 in the vehicle group. The registry says the remaining participants were not included because of device malfunction, non-wear, or missing data. This is important because randomization applies at assignment, whereas the posted endpoint summaries describe the subset with usable measurements.

In general, an analysis restricted to participants with observed endpoint data can be biased if the reasons data are missing are related to the outcome or to treatment assignment. Missingness can also reduce precision. The record does not report how missing data were handled beyond excluding unavailable or unusable endpoint data, whether any imputation was used, or whether sensitivity analyses were performed.

What can be concluded without a comparison analysis?

The registry lists 0 analyses and does not post a treatment-effect estimate, p-value, or confidence interval for the primary endpoint. A confidence interval is a general statistical tool that conveys uncertainty around an estimate under a specified method and assumptions. Without a reported interval or test, the precision of a group contrast cannot be evaluated from the registry summaries alone.

The record’s sample-size description characterizes its calculations as planning illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. It also states that the effect to detect is a stated fraction of the observed effect. Such calculations are planning illustrations, not observed power and not evidence that a study was adequately powered. The trial’s own design justification and the specific numerical planning results are not reported in the registry record.

In general, a prespecified analysis plan would clarify the intended estimand, model, handling of missing data, and approach to uncertainty and multiplicity. These details are not reported in the registry record. As a result, the means and standard deviations support a description of the recorded endpoint, but not a complete inferential assessment of the randomized comparison.

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("Crisaborole 2%", "Vehicle Arm")
n   <- c(11, 8)
m   <- c(93.04, 116.85)
sp  <- c(70.35, 62.17)
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  Crisaborole 2%  11  93.040  70.350  45.778  140.302
CARM  Vehicle Arm  8  116.850  62.170  64.875  168.825

11. Multiplicity, Missing Data and Analysis Populations

The study description includes several areas of interest: nighttime scratching, itch, sleep, and caregiver quality of life and sleep. The registry lists 3 outcome measures, but reports 0 posted secondary outcomes. Results for those other outcomes are not reported in the registry record. It does not identify a multiplicity procedure or explain how conclusions across multiple outcomes were intended to be controlled.

In general, testing many outcomes can increase the chance of a misleading positive finding if each is treated as an independent opportunity to declare a difference. A prespecified hierarchy or adjustment can address that issue, but the registry record does not report whether either approach was planned. Since no inferential primary analysis is posted, the record also does not provide a family of p-values to interpret.

Missing endpoint data are a prominent limitation. The primary analysis population excludes participants without sufficient processable accelerometry data because of device malfunction, non-wear, or missing data. The registry does not report the number in each exclusion category, the timing or pattern of missingness, or whether missingness differed by group. It also does not describe imputation, sensitivity analyses, or an intention-to-treat analysis for this endpoint.

In general, an intention-to-treat analysis aims to retain participants in their randomized groups, but it still requires a defined strategy for outcomes that are not observed. The posted summaries instead describe the participants with usable device data. Whether those summaries differ from results under another missing-data approach cannot be assessed from the registry record.

The record does not report the statistical analysis plan, the definition of analysis populations beyond the processable-data description, or any interim analyses. The primary endpoint findings should therefore be read as descriptive results from available measurements, not as a complete account of the planned analyses.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent non-serious adverse event terms

The registry reports serious adverse events for the crisaborole group, vehicle group, and caregivers. There were 0 serious adverse events among 21 participants at risk in the crisaborole group, 0 among 15 at risk in the vehicle group, and 0 among 36 caregivers at risk. These are group-specific counts; they should not be combined into a single figure.

For other adverse events, the record reports 1 affected participant among 21 at risk in the crisaborole group, 0 among 15 at risk in the vehicle group, and 0 among 36 caregivers at risk. The registry reports 0 deaths in each group. No serious adverse-event terms are listed.

These figures describe events recorded in the registry. They do not show whether an event was caused by treatment, and the record does not provide details about the non-serious event. A count of 0 observed serious events does not establish that an event is impossible or rule out uncommon harms; interpretation depends on the amount and duration of observation and the completeness of reporting. The registry record does not provide enough detail to characterize these aspects or to compare safety with precision.

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

The posted results show group-level descriptive summaries of nighttime scratching recorded by accelerometry over 2 weeks. The mean was 93.04 scratch counts per night in the crisaborole group and 116.8 in the vehicle group, among participants with sufficient processable data. Their standard deviations were 70.35 and 62.17. The record does not report a statistical comparison or an estimate of uncertainty around a group contrast.

These summaries do not establish that the observed group means differ beyond chance, nor do they establish the direction or size of a causal treatment effect. The groups were randomized, but usable endpoint data were available for only a subset of the enrolled study population. Missing device data could affect how well the summaries represent the randomized groups, and the registry does not report analyses that address that concern.

The endpoint is a device-recorded scratch count. It is not itself a direct measure of itch intensity, sleep quality, caregiver burden, or quality of life. While these topics appear in the study description, results for secondary outcomes are not reported in the registry record. The posted endpoint therefore cannot answer all of the questions motivating the study.

The trial is listed as terminated, but the reason is not reported in the registry record. The status alone does not establish whether termination affected the endpoint data or why the study ended. The safety record reports no serious adverse events and no deaths in the listed groups, with one other adverse event in the crisaborole group; these counts are descriptive and do not establish comparative safety.

Overall, the record supports a limited description of available endpoint observations. It does not provide the inferential analysis, missing-data assessment, or broader outcome results needed for stronger conclusions about group differences or the study’s wider aims.

14. Why This Trial Matters Statistically

This trial illustrates a common distinction in clinical research: an outcome can be measured with a device, yet still require careful interpretation of who contributed usable data and how the measurements were analyzed. The primary endpoint is based on accelerometry, but the registry states that device malfunction, non-wear, or missing data prevented some participants from contributing processable results.

The available-data group means and standard deviations convey useful descriptive information, but they are not a substitute for a between-group estimate with an uncertainty interval. Reporting an explicit contrast and its precision would help readers assess the range of effects compatible with the data. Those inferential results are not reported in the registry record.

The endpoint also raises a measurement question. In general, a movement sensor records signals that must be processed and interpreted to identify a behavior. A clear definition of how nighttime scratching episodes were identified would help readers judge the measure’s reliability and meaning. The registry says episodes were recorded by accelerometry, but the processing and classification details are not reported in the registry record.

Finally, the study description reaches beyond the primary endpoint to sleep, itch, and caregiver outcomes. Distinguishing those outcomes from device-recorded scratching matters because a change in one measure cannot automatically stand in for a change in another. No secondary results are posted in the registry record.

These features make the record useful for understanding the limits of descriptive trial reporting. It shows why endpoint definitions, transparent missing-data handling, and a prespecified analysis strategy matter alongside randomization. They help readers judge not only what values were observed, but also how securely those values answer the clinical question.

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 primary endpoint summaries are based on participants with sufficient processable accelerometry data, not everyone listed as enrolled. The registry attributes exclusions to device malfunction, non-wear, or missing data, but does not report the count for each reason or how missingness was distributed across groups. The potential effect of missing data on the observed summaries cannot therefore be assessed.

The registry reports means and standard deviations but no inferential comparison, confidence interval, or p-value. It does not report the analysis model, an analysis plan, or whether the assumptions for the method used were evaluated. The methods used to process accelerometry signals and identify scratching episodes are also not reported in the registry record.

Only the primary endpoint results are posted in the information described. Although the study description discusses itch, sleep, and caregiver quality of life and sleep, results for those outcomes are not reported in the registry record. The record therefore does not show whether the primary endpoint aligns with changes in those broader experiences.

The registry labels the trial single blind but does not identify who was masked. It does not report how randomization was generated or concealed, whether treatment adherence was assessed, or why the study was terminated. These omissions limit assessment of study conduct and potential sources of bias.

The safety record provides event counts but little detail about the non-serious event, its timing, or attribution. The available counts are descriptive and do not provide a precise comparative safety assessment. Taken together, these limitations mean the registry supports a narrow description of observed device data rather than a complete evaluation of efficacy, safety, or the full study objectives.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Improvement In Scratch Behavior And Sleep In Patients With Atopic Dermatitis — NCT05200403.
• Registry record last updated 2025-03-14; status last verified 2025-02.
• Results first posted 2025-03-14.
• 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

The registry describes a randomized, single-blind, parallel-group Phase 4 study of crisaborole 2% versus vehicle treatment in children with atopic dermatitis. Caregivers applied assigned treatment twice daily for 2 weeks. The study is listed as terminated; its reason for termination is not reported in the registry record.

The primary endpoint was children’s nighttime scratching episodes per night recorded by accelerometry. Among participants with sufficient processable data, the mean was 93.04 with a standard deviation of 70.35 for crisaborole, and 116.8 with a standard deviation of 62.17 for vehicle. The endpoint summaries include 11 and 8 participants, respectively.

No inferential primary comparison is posted, and no secondary outcomes are reported. The record notes endpoint exclusions because of device malfunction, non-wear, or missing data. It reports 0 serious adverse events and 0 deaths in each listed group, and 1 other adverse event in the crisaborole group.