Evaluation of Diphenhydramine Hydrochloride Effects in Subjects With Occasional Sleeplessness: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Evaluation of Diphenhydramine Hydrochloride Effects in Subjects With Occasional Sleeplessness (NCT02578186), a randomized phase 4 study. The primary outcome is Mean Latency to Persistent Sleep.

2. Clinical Question

3. Trial Design

This was a randomized, double-blind, placebo-controlled, crossover, in-home study. The registry classifies it as Phase 4, with randomized allocation, quadruple masking, and a crossover model. Its registered title describes an assessment of diphenhydramine hydrochloride in subjects with occasional sleeplessness. Procter and Gamble is listed as the sponsor. The study started in 2013-11, had primary completion in 2014-12, and is marked completed.

The two interventions were diphenhydramine hydrochloride and placebo. Both are described as elixirs taken when subjects had trouble falling asleep; the diphenhydramine description specifies 50 mg. In a crossover study, treatment comparisons can draw on measurements from the same participants during different treatment periods. This structure can be statistically useful because participants can serve as their own controls, but interpretation depends on appropriate handling of within-participant measurements and any period or sequence effects. The registry record does not report the treatment sequence, period lengths, washout arrangements, or statistical handling of those features.

Enrollment was 33. The primary efficacy description specifies a Per Protocol population that completed the treatment crossover and had evaluable data for mean latency to persistent sleep for both treatment periods, subject to other defined Per Protocol criteria. The arm summaries report 22.0 for each treatment. These are the posted efficacy-analysis counts, not a count that should be combined to infer the number of unique participants in a crossover study. The registry record does not report the reasons that some enrolled participants are not represented in these summaries.

The record identifies one primary outcome measure and one sponsor-reported analysis. Further details of the study’s original design assumptions, including any sample-size justification, are not reported in the registry record. The registry also does not identify which parties were masked despite classifying the study as quadruple-masked.

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: 43 subjects screened. 33 subjects received product (ie, enrolled, for the purposes of this reporting), of whom 2 withdrew but re-enrolled (per protocol amendment). These 2 subjects are counted as starting each sequence in the table below, bringing total started to 35 and for AEs to 30 and 29 (DPH and placebo, respectively.) 25 subjects completed.

Period 1 (First of Crossover)

Reasons for not completing this period:

Period 2 (Second of Crossover)

Analysis population for the primary outcome, as defined in the registry: Subjects with evaluable data for mean latency to persistent sleep for both treatment periods were included in the efficacy analysis provided they meet the other defined Per Protocol criteria. All data for each treatment and endpoint were averaged for all days in which the study medication was taken in a given treatment period.

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 Mean Latency to Persistent Sleep, measured over 4 weeks.

The primary endpoint is mean latency to persistent sleep, measured in minutes. Latency describes the time from the relevant sleep-onset reference point until persistent sleep is reached. The endpoint therefore addresses the study’s stated interest in the ability to initiate sleep, rather than measuring sleep duration or another aspect of sleep. The registry specifies a 4-week timeframe.

The outcome is a continuous measure, and the posted treatment summaries are means with standard errors. A mean provides a summary of the central value across observations, while the standard error describes uncertainty in the estimated mean. The endpoint description states that the efficacy analysis used subjects who completed the crossover, had evaluable latency data for both treatment periods, and met other defined Per Protocol criteria. It also says that data for each treatment and endpoint were averaged across all days on which study medication was taken.

This endpoint is useful for comparing the time to persistent sleep between the two treatment conditions. The treatment means were 19.1 minutes with diphenhydramine hydrochloride and 27.1 minutes with placebo. Those summaries describe the analyzed observations; by themselves, they do not establish the reliability of the contrast or address the paired structure of a crossover. Interpretation also depends on the sponsor’s analysis and its assumptions, as well as the precision of the estimated difference.

The registry record does not give a more detailed operational definition of persistent sleep or the specific measurement instrument. It also does not report how missing daily measurements were handled beyond describing the Per Protocol efficacy population and the averaging of days on which medication was taken.

7. Results Posted in the Registry

• Diphenhydramine Hydrochloride: mean 19.10 versus 27.10 on control; difference -8.00 (95% CI -16.59 to 0.59), standardized effect d = -0.55; z-test p = 0.068.

8. Statistical Interpretation

9. Statistical Methods Explained

The endpoint is continuous, and the registry reports mean latency to persistent sleep with standard errors. Statistical interpretation involves separating the observed summaries, the uncertainty around them, and the model used to compare treatment conditions. In this trial, that distinction is especially important because the design is crossover and the sponsor’s reported inferential result differs from a simple unadjusted recalculation.

What does the mean describe?

A mean is a measure of the center of a set of observations. The reported treatment means are 19.1 minutes for diphenhydramine hydrochloride and 27.1 minutes for placebo. The simple contrast is reported as -8.0 minutes, with the diphenhydramine mean subtracted from the placebo mean according to the listed comparison. The negative sign indicates a lower observed mean in the diphenhydramine condition. A mean does not show how individual observations are distributed, whether a small number of observations strongly influence the average, or how much participant-level responses vary.

The standard error is a measure of the uncertainty in an estimated mean, rather than a direct description of the spread of individual values. The registry lists a standard error of 3.1 for each treatment. It does not provide the raw observations or a distributional plot in the registry record, so the shape and range of individual latency values cannot be assessed from these summaries alone.

Why does the crossover design matter?

In a crossover design, participants receive more than one treatment across different periods. Measurements from the same person are generally related, so a comparison can use within-person information. Accounting for that relationship may affect the estimated uncertainty and the comparison. A simple unadjusted recalculation from posted group summaries does not reproduce a model that uses participant-level pairing, period effects, sequence effects, or other covariates. The registry lists ANCOVA as the sponsor’s method, but does not report its covariates or specify how crossover features were represented.

General statistical context: ANCOVA compares outcomes while adjusting for one or more covariates. Adjustment can change an estimated treatment contrast or its uncertainty, depending on the variables and model. Since the registry does not report those details, it is not possible to explain precisely why the sponsor’s p-value differs from the unadjusted recalculation. The sponsor’s own analysis takes precedence when the posted result and a recalculation disagree.

What does a p-value tell the reader?

A p-value describes how unusual data at least as discrepant as the observed data would be under a specified statistical model and a null hypothesis. 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 sponsor reported p = 0.031 from ANCOVA. The simple unadjusted recalculation gives p = 0.068. These values answer questions under different analyses and should not be treated as competing estimates from an identical method. The model specification for the sponsor’s result is not reported in the registry record.

General statistical context: a significance threshold is a decision convention chosen for a particular analysis plan; it does not make an effect more or less important. The sample-size planning illustration lists a two-sided significance level of 0.05, but that planning assumption does not establish the trial’s original analysis plan. The original design assumptions and statistical plan are not reported in the registry record.

How should uncertainty be understood?

The unadjusted confidence interval for the mean difference extends from -16.59 to 0.593 minutes. It includes zero, which corresponds to no difference between treatment means. In general, a confidence interval describes the range of parameter values compatible with the observations and the method’s assumptions. Its width conveys precision: narrower intervals indicate greater precision, while wider intervals indicate less. The posted interval is based on the simple recalculation, not the sponsor’s ANCOVA. No sponsor-adjusted interval is reported, so the adjusted estimate’s precision cannot be evaluated from the registry result.

The standard errors, confidence interval, and p-value are related aspects of inferential uncertainty, but each depends on the analysis method. Because this is a crossover study, the treatment-period relationship can matter. The posted summaries do not provide the within-participant covariance needed to independently reconstruct an adjusted crossover comparison. The unadjusted recalculation is therefore illustrative rather than a replacement for the sponsor’s analysis.

What can be inferred from the analysis population?

The primary efficacy analysis is described as Per Protocol. It included subjects who completed the treatment crossover, had evaluable data for both treatment periods, and met other defined criteria. This focuses the result on participants who satisfied those conditions; it does not show the result for every enrolled subject. In general, restricting analysis to participants who completed treatment and met protocol criteria can affect how broadly the result represents the enrolled population, particularly if exclusions are related to outcomes or treatment experience. The registry record does not report the exclusions, reasons for missing data, or an intention-to-treat analysis.

Only one primary outcome measure and one sponsor analysis are reported in the record details. No secondary outcome results are posted. The registry record does not report whether additional analyses were conducted but not posted, whether any interim analyses occurred, or what adjustment, if any, was planned for multiple testing. Those absences limit assessment of the full analysis process; they do not establish that such analyses or procedures did or did not occur.

10. Confidence Intervals and Precision

The simple unadjusted recalculation estimates a mean difference of -8.0 minutes, with a confidence interval from -16.59 to 0.593 minutes. The interval crosses zero, so the posted unadjusted calculation is compatible with a range of values that includes no mean difference. It also includes negative values, representing a lower mean latency under diphenhydramine hydrochloride according to the stated contrast. The interval is not a statement that every value inside it is equally likely, nor does it quantify the probability that the true difference lies within those bounds.

In general, a confidence interval conveys both the direction and precision of an estimate. A relatively wide interval indicates that the data are compatible with a range of effects rather than pinning the estimate down closely. The posted interval reaches from a potentially substantial negative contrast to a value slightly above zero. This limits the precision of the unadjusted estimate: its direction is not settled by the interval alone, and its magnitude is uncertain.

The interval belongs to the simple recalculation from posted summaries. The sponsor’s ANCOVA result is reported with a p-value but without an adjusted estimate or confidence interval. Therefore, the precision of the sponsor-adjusted contrast cannot be judged from the registry record. The p-value alone does not supply the missing effect estimate or confidence bounds.

Interpretation is further limited by the crossover design and the Per Protocol efficacy population. The registry summaries do not provide participant-level data, period-specific estimates, or the within-person covariance needed to independently assess a crossover model. The posted treatment means and standard errors are useful summaries, but cannot establish how the sponsor’s ANCOVA treated the repeated observations. The registry record does not report sensitivity analyses that would indicate whether the estimate remains similar under alternative assumptions.

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 sample sizes under the stated planning assumptions: a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set as a stated fraction of the observed effect. These are planning illustrations based on the observed contrast, not observed power, and they do not report the trial’s original design assumptions.

The 80% power column gives the illustrative number of participants per arm for a design that would have 80% power under those assumptions to detect the effect listed in that row. The 90% power column gives the corresponding per-arm count for 90% power. Power is a design-stage probability of detecting a specified effect under a statistical model if that effect is present; it is not the probability that a result is true, and it does not describe the probability that this completed trial detected an effect.

The rows scale the effect being targeted relative to the observed effect. Planning for 100% of the observed effect gives 52 participants per arm for 80% power and 70 participants per arm for 90% power. Planning for 75% of the observed effect gives 93 participants per arm for 80% power and 124 participants per arm for 90% power. Planning for 50% of the observed effect gives 208 participants per arm for 80% power and 278 participants per arm for 90% power. The pattern illustrates a general design principle: smaller effects are harder to detect and need more participants to achieve the same target power.

These figures are not a retrospective assessment of the trial’s success and are not a substitute for a protocol-based design calculation. The original sample-size justification and design assumptions are not reported in the registry record. In particular, the planning illustration uses equal allocation between two arms, while this trial used a crossover structure; the illustration should not be mistaken for a reconstruction of the original 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 = -8.000; pooled standard deviation σ = 14.540.
2. n = 2 × (1.96 + 0.842)² × 14.540² / -8.000² = 51.9, rounded up to 52 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 <- -8                      # observed difference in means
sigma <- 14.5402888554526                     # 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(2578186)
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 = 52 (80% power), 70 (90% power)
 75% of observed effect: n per arm = 93 (80% power), 124 (90% power)
 50% of observed effect: n per arm = 208 (80% power), 278 (90% power)
power.t.test: 52.8 
Simulated power at n = 52 per arm: 0.798 (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. Analyses Reported by the Sponsor

The sponsor’s reported analysis is ANCOVA and gives p = 0.031. The simple unadjusted recalculation from posted summary statistics gives a mean difference of -8.0 minutes, a confidence interval from -16.59 to 0.593 minutes, and p = 0.068. Because the sponsor’s result and the recalculation differ, the sponsor’s own analysis takes precedence.

The two results should not be combined or treated as though they came from the same model. The recalculation is unadjusted, while ANCOVA is an adjusted analysis. In a crossover trial, a participant’s measurements across treatment periods are related; an analysis may also account for period, sequence, or other covariates. The registry identifies ANCOVA but does not report its covariates, model specification, adjusted estimate, or confidence interval. It is therefore not possible to reproduce the sponsor result or identify from the registry record exactly why the p-values differ.

The sponsor p-value alone does not give the size or precision of the adjusted treatment contrast. It should be read as the inferential result of the sponsor’s reported method, with the limitation that important model details are not reported in the registry record. The unadjusted calculation remains a transparent descriptive check, not a replacement for that analysis.

14. Multiplicity, Missing Data and Analysis Populations

The registry record reports one primary outcome measure, mean latency to persistent sleep, and one sponsor-reported analysis. It reports no secondary outcome results. That provides a limited view of the outcome and analysis record, but does not establish whether other outcomes or analyses were considered during the study.

Multiplicity arises when several hypotheses are tested, such as multiple endpoints, treatment contrasts, subgroups, or repeated looks at accumulating data. If many tests are conducted, the chance of at least one apparently unusual result can increase unless the analysis plan accounts for that process. General statistical context: multiplicity procedures can control a chosen error rate, but the appropriate approach depends on which tests were planned and how they relate to each other. The registry record does not report whether multiplicity adjustments were planned or used, or whether interim analyses were conducted.

The primary efficacy population is described as Per Protocol. Subjects needed to complete the treatment crossover, have evaluable data for both treatment periods, and satisfy other defined Per Protocol criteria. The registry does not report the criteria in detail, the number excluded, or reasons for exclusion. It also does not report an intention-to-treat analysis. As a result, the posted primary result cannot be compared with a result based on all randomized participants.

The record says the treatment and endpoint data were averaged across days when study medication was taken. It does not specify the handling of missing daily observations, missing treatment-period measurements, or incomplete crossover participation beyond the analysis-population description. It does not report imputation, sensitivity analyses, or assumptions about why data might be missing. Those details matter because an analysis restricted to evaluable completers can be affected if missingness is associated with treatment or outcomes.

The sponsor ANCOVA’s covariates and treatment of crossover features are not reported. The registry record also does not provide a statistical analysis plan, so it is not possible to check whether the posted result followed a prespecified model or whether any analysis choices were made after observing the data. These are limitations of the available record, not proof that the study lacked a plan or did not consider these issues.

15. 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 non-serious adverse event terms

The registry reports serious adverse events in 0 of 30 participants at risk in the diphenhydramine hydrochloride group and 0 of 29 participants at risk in the placebo group. These figures mean that no serious adverse events were recorded among the participants at risk in either treatment group in the posted safety results. The safety denominators differ from the efficacy summaries, which use the Per Protocol population and report 22.0 for each treatment; they describe a separate safety reporting set.

Zero recorded serious events does not establish that the risk is absent. Safety counts from a study are bounded by the participants observed, the observation period, and how events were collected and classified. The registry does not report the serious-event definitions or detailed event surveillance in the registry record. Death counts are not reported in the registry record.

The record also lists other adverse events affecting 8 of 30 in the diphenhydramine hydrochloride group and 7 of 29 in the placebo group. These are separate from the serious adverse-event counts and should not be interpreted as serious events. The posted figures do not by themselves establish whether the groups differ reliably in safety, and no comparative safety analysis is reported in the registry record.

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

The posted efficacy summaries show a mean latency to persistent sleep of 19.1 minutes for diphenhydramine hydrochloride and 27.1 minutes for placebo. The simple unadjusted recalculation expresses the difference as -8.0 minutes. These figures describe the analyzed Per Protocol data and indicate that the observed mean was lower in the diphenhydramine condition. They do not, on their own, establish the size of an adjusted treatment effect or its practical importance.

The unadjusted confidence interval crosses zero, and its p-value is 0.068. The sponsor’s ANCOVA analysis reports p = 0.031. Because these analyses disagree, the sponsor’s own result takes precedence; however, the registry does not provide the adjusted estimate, confidence interval, or enough model detail to fully interpret or reproduce it. Statistical significance is not a measure of how important a difference is, and a p-value is not the probability that a treatment effect is real.

The result applies to the population described for the efficacy analysis: subjects who completed the treatment crossover, had evaluable data in both periods, and met other Per Protocol criteria. The registry record does not report an intention-to-treat result, detailed missing-data handling, or sensitivity analyses. The findings therefore do not describe every enrolled participant in the same way, and generalization beyond the analyzed population requires caution.

The study was a randomized, masked, placebo-controlled crossover trial, which allows comparisons across treatment periods. Yet sequence, period, and washout details are not reported, and the ANCOVA specification is also unavailable. Those omissions prevent a reader from assessing how the sponsor’s model accounted for important features of the design. The posted data do not establish whether a difference would be similar under another population, analysis method, endpoint, or study setting.

The safety record lists no serious adverse events in either group among participants at risk, and also lists other adverse events in both groups. These counts are descriptive, not proof that serious events cannot occur or that the treatments have equivalent safety. Taken together, the record supports a careful reading of the reported means, the sponsor analysis, and the uncertainty, rather than a broad or definitive claim about treatment effects.

17. Why This Trial Matters Statistically

This trial illustrates why a statistical result should be read alongside the design and analysis population. A crossover study measures participants under different treatment conditions, so the relationship between observations from the same person can be important. A summary-based unadjusted comparison and an ANCOVA result need not agree, particularly when the analysis details are unavailable. Recognizing that distinction prevents a recalculation from being mistaken for the sponsor’s primary analysis.

The example also shows why effect estimates and uncertainty matter alongside p-values. The posted means and recalculated difference describe the direction and magnitude of the observed contrast, while the unadjusted confidence interval communicates substantial uncertainty and includes zero. The sponsor reports a different p-value but no adjusted estimate or confidence interval. Without those effect-size details, the sponsor result cannot fully describe the magnitude or precision of the adjusted contrast.

Analysis-population definitions shape what the result represents. The primary efficacy analysis was Per Protocol and required evaluable data for both treatment periods. Enrollment and efficacy summaries therefore describe different aspects of the record; the efficacy summaries should not be treated as covering every enrolled participant. Missing-data methods and sensitivity analyses are not reported, limiting evaluation of how exclusions or incomplete data might influence the finding.

Finally, planning calculations are not retrospective evidence of power. The displayed planning figures illustrate how the required per-arm sample size changes as the target effect becomes smaller or the desired power increases. They are based on the observed effect and stated assumptions, not on the original design justification, which is not reported. This distinction is useful when interpreting trial results and planning future studies without confusing a planning scenario with an outcome observed in the completed trial.

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.

The registry record leaves important design details unreported. It identifies a randomized, quadruple-masked crossover study but does not identify which parties were masked, the treatment sequence, the duration of each treatment period, or any washout arrangement. Those details could help assess the possibility of period or carryover effects and understand how the crossover comparison was implemented.

The sponsor analysis is identified as ANCOVA, but its covariates, model specification, adjusted estimate, and confidence interval are not reported. The sponsor’s p-value differs from the simple unadjusted recalculation, and the missing model details prevent an independent explanation of that discrepancy. The sponsor’s own analysis takes precedence, but the available record does not allow a full assessment of its estimated effect or precision.

The primary efficacy population was Per Protocol and required completion of the treatment crossover, evaluable data in both periods, and other defined criteria. The record does not specify those other criteria, the reasons for exclusions, or the handling of missing daily and period-level measurements. An intention-to-treat analysis and sensitivity analyses are not reported. These omissions limit assessment of how the analysis population and missingness could affect the result.

The endpoint description says that data were averaged across days on which study medication was taken, but the registry does not provide individual-level data, outcome distributions, or a detailed operational definition of persistent sleep. The summary statistics therefore do not permit assessment of variability among participants or daily patterns. The registry also does not report secondary outcome results, multiplicity procedures, interim analyses, or the original sample-size justification. These are limitations of the registry record and do not establish whether the study team addressed them elsewhere.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Evaluation of Diphenhydramine Hydrochloride Effects in Subjects With Occasional Sleeplessness — NCT02578186.
• Registry record last updated 2017-03-15; status last verified 2017-01.
• Results first posted 2017-03-15.
• 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

The completed Phase 4 study was randomized, double-blind, placebo-controlled, quadruple-masked, and crossover, and was conducted in-home. It enrolled 33 subjects with occasional sleeplessness. The intervention was diphenhydramine hydrochloride elixir, described as 50 mg taken when subjects had trouble falling asleep; the comparator was placebo elixir.

The primary endpoint was mean latency to persistent sleep in minutes over 4 weeks, analyzed in a Per Protocol population. Posted means were 19.1 minutes for diphenhydramine hydrochloride and 27.1 minutes for placebo. The sponsor reported an ANCOVA p-value of 0.031. A simple unadjusted recalculation gave a mean difference of -8.0 minutes, a confidence interval crossing zero, and p = 0.068; because of the discrepancy, the sponsor’s analysis takes precedence.

The registry lists no serious adverse events in either treatment group among participants at risk. Important details, including the ANCOVA specification, missing-data approach, and original design assumptions, are not reported in the registry record.