Saracatinib and Alcohol Drinking: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Saracatinib and Alcohol Drinking (NCT02955186), a randomized phase 2 study. The primary outcome is Change in the Number of Drinks Consumed From Baseline to Day 8 (Minus Baseline).

2. Clinical Question

3. Trial Design

This completed phase 2 trial, NCT02955186, is titled “Saracatinib and Alcohol Drinking,” with the official title “Fyn Kinase Inhibitors and Alcohol Drinking.” Yale University is listed as the sponsor. The registry records a randomized, triple-masked, parallel-group design and enrollment of 50. The listed study start was 2017-05-09, and the primary completion date was 2019-04-24.

The active arm is described as taking 125 mg of saracatinib daily for 8 days; the placebo arm took placebo daily for 8 days. The registry summary describes a laboratory session in which participants received an initial alcoholic drink and could then choose to drink up to 12 more drinks over a three-hour period. The primary endpoint compares drinking at the Day 8 session, identified as ADP 2, with drinking at baseline, identified as ADP 1.

Those design details establish the broad treatment contrast and setting, but they do not supply all information needed to reconstruct trial conduct. Eligibility criteria, the method used to generate or conceal allocation, the masking procedures, adherence information, and the original sample-size justification are not reported in the registry record. Nor does the record describe the planned statistical model in enough detail to identify covariates or other adjustment choices.

The registry reports outcome summaries for the primary measure and two secondary alcohol-response measures. Participant counts differ across the posted summaries: the primary arm summaries list 26 for saracatinib and 15 for placebo, while the secondary summaries and safety groups list 33 and 17, respectively. The reasons for these differences are not reported in the registry record. These counts should therefore be read in the context of their respective posted outcomes rather than treated as interchangeable analysis populations.

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: 54 participants agreed to the study but 4 did not end up participating in the full study and were never randomized to treatment.

Overall Study

Reasons for not completing this 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 Change in the Number of Drinks Consumed From Baseline to Day 8 (Minus Baseline), measured over Baseline and Day 8.

The primary endpoint is the change in the number of standard drinks consumed during the laboratory drinking session from baseline to Day 8. The registry defines the change as the Day 8 count minus the baseline count. A negative value therefore represents fewer drinks at Day 8 than at baseline for the group summary; it does not by itself establish why the count changed or whether the change is attributable to treatment.

The endpoint is a continuous measure of drinking behavior in the study’s laboratory setting. It captures the change between two drinking sessions, rather than only the Day 8 amount or a measure of craving. The registry’s broader description says participants received an initial alcoholic drink and could choose to drink up to 12 more over a three-hour period. The primary outcome description identifies the sessions as ADP 1 at baseline and ADP 2 after study medication.

Using a change score can focus the comparison on how the measured quantity differs between the two sessions. The posted group estimates are least-squares means with standard errors, and the posted comparison is the saracatinib-minus-placebo mean difference in change. The registry does not report enough detail to determine how the least-squares means were modeled or whether any baseline covariates were included. The primary endpoint is therefore best interpreted as the posted between-arm comparison of change in this particular laboratory measure.

7. Results Posted in the Registry

• 125 mg Saracatinib: mean -1.85 versus -2.33 on control; difference 0.48 (95% CI -1.95 to 2.91), standardized effect d = 0.13; z-test p = 0.699.

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.

Area Under the Curve (AUC) During the Drinking Session Using the Stimulation Responses to Alcohol (BAES)

• 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:
  • 125 mg Saracatinib: mean 1,651 versus 1,607 on control; difference 43.70 (95% CI -1,159 to 1,246), standardized effect d = 0.02; z-test p = 0.943.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between 125 mg Saracatinib and the control. The interval gives the range of effects that remain compatible with the data.

Area Under the Curve (AUC) During the Drinking Session Using the Sedation Responses to Alcohol (BAES)

• 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:
  • 125 mg Saracatinib: mean 1,150 versus 515.29 on control; difference 634.61 (95% CI 61.07 to 1,208), standardized effect d = 0.50; z-test p = 0.030.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for 125 mg Saracatinib than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes address alcohol-related stimulation and sedation responses, rather than the primary endpoint’s change in drinks consumed. Across these outcomes, the estimates do not show a uniform pattern: the stimulation comparison is centered close to zero with a wide interval, while the sedation comparison is positive and its interval lies above zero. These measures therefore give different statistical signals, and neither changes what the primary endpoint directly measures.

The primary comparison concerns change in drinking behavior from baseline to Day 8. The secondary comparisons concern response-scale areas under the curve during the Day 8 session. They may provide related context about responses during the laboratory task, but they are not substitutes for the primary drinking measure. Differences in measurement scale and outcome definition mean that the results should be interpreted separately rather than combined into a single overall treatment estimate.

The registry reports unadjusted p-values for the secondary comparisons and no reported analyses in its analysis listing. It does not report whether the secondary outcomes were prespecified as confirmatory tests, how many total statistical tests were planned, or whether any multiplicity adjustment was used. General statistical explanation: when multiple outcomes are examined without adjustment, an isolated small p-value can be less persuasive because the chance of at least one small value increases across a collection of tests. The secondary results should therefore be treated as exploratory signals unless a fuller analysis plan establishes otherwise.

The registry’s secondary summaries list 33 participants in the saracatinib group and 17 in the placebo group. The primary summary lists different counts, and the reasons for this are not reported in the registry record. The interpretation of the secondary results is consequently limited not only by their unadjusted status but also by the absence of detail about analysis populations and missing observations. The pattern is informative as a description of the posted outcomes, but it does not establish a coherent or definitive conclusion across endpoints.

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

• Area Under the Curve (AUC) During the Drinking Session Using the Stimulation Responses to Alcohol (BAES)
• Area Under the Curve (AUC) During the Drinking Session Using the Sedation Responses to Alcohol (BAES)

9. Statistical Interpretation

10. Statistical Methods Explained

For the primary outcome, the registry posts least-squares mean changes with standard errors. A least-squares mean is a model-based group estimate; unlike a simple arithmetic mean, its value can depend on the statistical model and any included terms. The record does not report enough modeling detail to identify those terms. The standard error describes the estimated sampling uncertainty of each group estimate, rather than the spread of individual participants’ values.

The primary comparison is a mean difference between the saracatinib and placebo groups. Its estimated value is 0.48 drinks. The confidence interval gives a range of values compatible with the data and the stated statistical procedure under its assumptions. The interval is -1.954 to 2.914. Because it includes zero, the summary does not distinguish the observed contrast from no between-arm difference at the conventional two-sided 0.05 level. The unadjusted p-value is 0.699; it describes how unusual data at least this incompatible with a null difference would be under the null model and the procedure used. It is not the probability that the null hypothesis is true, nor a measure of the probability that the result will replicate.

For the secondary endpoints, the registry reports arithmetic means and standard deviations. A mean describes the group’s average measured response, while a standard deviation describes variation among individual measurements around that average. These quantities are not interchangeable with the standard errors posted for the primary least-squares means. The secondary comparisons are reported as mean differences, with confidence intervals and unadjusted p-values. The posted secondary group sizes are 33 for saracatinib and 17 for placebo.

The primary and secondary outcomes also differ in what they measure. The primary outcome is a change in drinks consumed between baseline and Day 8. The secondary measures are areas under the curve for stimulation and sedation responses to alcohol during the Day 8 session. Their units are listed as units on a scale*days. Comparing the numerical sizes of these estimates with the primary drinking measure would not be meaningful because the outcomes use different scales.

How should the mean differences be read?

Each posted comparison is oriented as saracatinib relative to placebo. For the primary endpoint, the difference is based on the change scores. For the secondary endpoints, it is based on the Day 8 group means. A positive difference indicates a larger numerical value in the saracatinib group, and a negative difference would indicate a smaller one. Whether a larger value represents a favorable or unfavorable response depends on the measurement and cannot be inferred from the sign alone. The registry’s hypothesis is a study rationale, not a substitute for interpreting the observed endpoint.

Why do confidence intervals matter alongside p-values?

General statistical explanation: a confidence interval communicates the precision of an estimate as well as the range of effects that remain compatible with the data under the method’s assumptions. A narrow interval can rule out more values than a broad interval. An interval that includes zero may still include effects in both directions that could matter scientifically; the interval’s full range should be considered rather than reducing the result to a binary significant or not-significant label. Here, the primary interval is broad relative to its point estimate and spans both negative and positive differences.

The posted primary confidence interval is paired with a p-value that is explicitly unadjusted. The registry does not report a multiplicity procedure or a prespecified hierarchy for the outcomes. When several outcomes are examined, unadjusted p-values do not account for the opportunity to observe an apparently unusual result among multiple comparisons. General statistical explanation: multiplicity can increase the chance of at least one small p-value when many hypotheses are tested, so secondary findings usually require context about prespecification, the number of tests, and any adjustment strategy.

What does over-dispersion mean in general?

General statistical explanation: over-dispersion occurs when observed variability is greater than a model’s assumed variability. It can make standard errors too small if the model does not accommodate that extra variation. The registry’s primary endpoint is a count of drinks, but the record does not report a count-data model or an over-dispersion assessment. The posted primary summary uses least-squares means and standard errors; it is not possible to determine from the registry record whether over-dispersion was assessed or how the outcome distribution was handled. This is a general issue to consider for count outcomes, not a claim that over-dispersion occurred in this trial.

The registry does not report the original analysis plan, model specification, handling of repeated measurements, covariate adjustment, assumptions checks, or missing-data method. It also reports no analyses in the registry’s analysis listing. As a result, the summaries support a descriptive reading of the posted estimates, but do not permit a full audit of the analysis choices that produced them.

11. Confidence Intervals and Precision

The primary mean difference is 0.48 drinks, with a 95% confidence interval from -1.954 to 2.914. This interval includes zero and allows for differences favoring either arm numerically, as well as a range of magnitudes. It is therefore important not to interpret the point estimate alone as a precise estimate of the contrast. The unadjusted p-value of 0.699 is consistent with the interval’s inclusion of zero, but the interval adds information about the range of values that remain compatible with the data.

The primary group estimates are based on 26 participants in the saracatinib arm and 15 in the placebo arm. Their posted standard errors are 0.75 and 0.99, respectively. A standard error expresses uncertainty in an estimate; it is not the same as the standard deviation of individual outcomes. The registry does not provide confidence limits for the separate primary group estimates.

The secondary estimates also have differing precision. The stimulation-response mean difference is 43.7, with a 95% confidence interval from -1159.0 to 1246. The sedation-response mean difference is 634.6, with a 95% confidence interval from 61.07 to 1208.0. These intervals should be read on their respective outcome scales, not compared directly with the primary interval. The registry does not report adjusted secondary analyses or a multiplicity procedure.

General statistical explanation: confidence intervals reflect sampling uncertainty under the analysis method; they do not account for every possible source of uncertainty, such as outcome measurement choices, missing data, or departures from model assumptions. Those features are not fully described in the registry record. The intervals are consequently useful for assessing the precision of the posted comparisons, but they do not by themselves establish how robust the findings are to other analysis decisions.

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 illustrations give participants per arm for these continuous outcomes, not total events. They use the observed effect as a reference and display planning for 100%, 75%, and 50% of that observed effect. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect targeted in each illustration is the stated fraction of the observed effect. The original trial’s own sample-size assumptions and justification are not reported in the registry record.

The 80% power column represents the number of participants per arm estimated to give an 80% chance of detecting the specified effect under those planning assumptions, if that effect is the true effect and the assumptions hold. The 90% power column gives the corresponding planning target for a 90% chance. These are design-planning probabilities, not the chance that the observed result is correct. The illustrations are not observed power.

Smaller target effects require more participants because a smaller contrast is harder to distinguish from sampling variation. In the posted planning illustrations, the figures rise as the target moves from the full observed effect to a smaller fraction. For example, the primary-outcome planning rows list 999 participants per arm for 80% power at 100% of the observed effect and 3,994 participants per arm at 50% of the observed effect. Those figures are hypothetical planning outputs based on the observed estimate, not the trial’s enrollment target or a claim about the trial’s ability to detect its result.

Any curve should be read using the same distinction: it depicts planning requirements under stated assumptions, not the number enrolled in the completed study or evidence that a result would be statistically significant. The calculation uses the observed effect, which itself is uncertain, so the illustrations should not be treated as a definitive design recommendation.

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 = 0.480; pooled standard deviation σ = 3.828.
2. n = 2 × (1.96 + 0.842)² × 3.828² / 0.480² = 998.3, rounded up to 999 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- 0.48                      # observed difference in means
sigma <- 3.8278533851836                     # 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(2955186)
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 = 999 (80% power), 1,337 (90% power)
 75% of observed effect: n per arm = 1,775 (80% power), 2,376 (90% power)
 50% of observed effect: n per arm = 3,994 (80% power), 5,346 (90% power)
power.t.test: 999.3 
Simulated power at n = 999 per arm: 0.799 (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 two secondary outcomes: area under the curve for stimulation responses to alcohol and area under the curve for sedation responses to alcohol. The primary endpoint is the change in drinks consumed from baseline to Day 8. The registry does not report a statistical testing hierarchy, multiplicity adjustment, or an analysis plan specifying which outcome tests were confirmatory. It also does not report how many hypotheses were planned across all outcomes or time points. The posted secondary p-values are unadjusted, so their interpretation should account for the possibility of multiple testing.

Missing-data procedures are not reported in the registry record. It is not stated how missing measurements were handled, whether any imputation was used, or what assumptions were made about unavailable outcomes. Nor does the record identify the analysis population for each outcome, such as whether results are based on all randomized participants or only participants with observed measurements. These omissions matter because the posted group sizes differ by outcome: the primary summaries list 26 and 15, while the secondary summaries and safety groups list 33 and 17. The reasons for those differences are not reported.

The registry record reports no analyses in its analysis listing. This does not explain how the posted outcome summaries were generated, and it should not be read as evidence that no analysis occurred. The registry does not provide sufficient information to assess adherence to an original analysis plan or whether analyses were changed after data collection.

Important population and conduct details are also absent. Eligibility criteria, participant characteristics, allocation concealment, masking procedures, treatment adherence, and the handling of protocol deviations are not reported in the registry record. Without those details, readers cannot fully judge how representative the analyzed groups were, whether the treatment contrast was maintained, or how sensitive the results might be to exclusions or missingness. These are limitations of what can be established from the registry summary, not evidence that any particular problem occurred.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: car accident (5.9%), hernia (3.0%).

The registry’s safety groups list 33 participants at risk in the saracatinib arm and 17 in the placebo arm. One participant in each group is listed as having a serious adverse event, and deaths are listed as 0 in both groups. These counts describe the posted safety summary; they do not establish whether an event was caused by study medication.

The registry lists car accident at 5.9% and hernia at 3.0% among the top serious adverse-event terms. The record does not explain the denominator or event-level details for those listed percentages, so they should not be converted into participant counts or interpreted as treatment-related rates. The “other” adverse-event figures list 21 affected participants among 33 at risk for saracatinib and 8 among 17 at risk for placebo.

These summaries do not provide event timing, severity beyond the serious-event classification, relatedness assessments, or detailed descriptions of the other events. They also do not support a comparative safety conclusion on their own. The figures are best read as registry-reported counts and terms, with the limitations of the record’s level of detail.

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

The primary result is an estimated difference between groups in change in drinks consumed during a laboratory session. Its point estimate is 0.48 drinks, and its confidence interval extends from -1.954 to 2.914. The result is therefore uncertain in both direction and magnitude. The unadjusted p-value of 0.699 does not provide strong evidence against a null between-arm difference, but it also does not prove that the groups are equivalent or that no difference exists.

The outcome is specific to the study setting and measurement: participants received an initial drink and could choose additional drinks during a three-hour laboratory period. The result does not by itself describe drinking in everyday life, longer-term behavior, or effects outside the assessment conditions. The study summary mentions a hypothesis about craving and drinking, but the posted primary endpoint is a change in the number of drinks consumed, not a direct measure of craving.

The secondary outcomes describe stimulation and sedation response scales during the Day 8 session. Their differing statistical patterns should be considered separately from the primary outcome and cautiously because the p-values are unadjusted and the registry does not report a multiplicity plan. They do not override the primary comparison or turn the laboratory measure into evidence about broader outcomes.

General statistical explanation: a non-small p-value is not proof of no effect, especially when the interval remains wide enough to include differences in either direction. Conversely, an interval excluding zero on a secondary measure does not, by itself, establish a reliable or clinically meaningful effect. Such conclusions depend on the outcome’s interpretation, prespecification, analysis quality, and broader evidence, which are not established by these registry summaries.

The registry record supports a cautious account of the posted estimates, their uncertainty, and the limits of the available reporting. It does not support medical advice or a general conclusion about saracatinib beyond these reported laboratory outcomes.

17. Why This Trial Matters Statistically

This trial is statistically informative because it illustrates the distinction between an observed point estimate and the precision of the estimate. The primary mean difference is modest in relation to a confidence interval that spans negative and positive values. Reading the interval alongside the p-value makes clear that the result is not a precise estimate of a single treatment contrast.

It also illustrates why outcome definition matters. The primary endpoint is a change score from baseline to Day 8 in a controlled drinking session. The secondary outcomes are response-scale areas under the curve. Although all relate to the alcohol session, they represent different constructs and use different scales. Their estimates cannot be combined or interpreted as if they were repeated versions of the same endpoint.

The difference in reported participant counts across outcome summaries is another practical statistical concern. The primary and secondary estimates do not use the same listed group sizes, and the registry does not explain why. Without information on missingness and analysis populations, readers cannot determine whether the difference reflects available measurements, analysis choices, or another reason. That uncertainty limits the ability to assess comparability across outcomes.

Finally, the planning illustrations show how strongly sample-size requirements depend on the effect targeted. Smaller assumed effects lead to larger planning counts, and these calculations should not be confused with observed power or with the original design rationale. The original sample-size assumptions are not reported in the registry record. Together, these issues make the trial a useful example of interpreting effect estimates, uncertainty, outcome-specific populations, and planning calculations without overstating what a registry summary can establish.

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.

Several limitations are specific to what is posted for this trial. First, the primary comparison is based on group counts of 26 and 15, while the secondary and safety summaries list 33 and 17. The reason for this difference is not reported in the registry record. Without analysis-population definitions and missing-data information, it is not possible to determine how the populations differ or whether selection into the outcome summaries could affect interpretation.

Second, the primary estimate is imprecise: its confidence interval ranges from -1.954 to 2.914. The registry does not provide enough analysis detail to evaluate the model behind the least-squares means, including any covariate adjustment or assumptions. The posted summaries therefore do not permit a full assessment of robustness to alternative modeling choices.

Third, the secondary comparisons are unadjusted, and the registry does not state whether they were confirmatory or how multiplicity was handled. A positive result on one response measure should not be treated as a definitive finding without that context. The registry also does not list reported analyses, leaving the relationship between the posted outcomes and a prespecified analysis plan unclear.

Finally, the outcomes are measured in a laboratory drinking session after an 8-day treatment period. The registry does not provide information that would establish how these results relate to behavior outside the study setting or over longer periods. Participant eligibility and characteristics are not reported in the summary details, limiting assessment of how broadly the enrolled group represents other populations. These limitations concern the available record and do not establish that the trial itself was conducted improperly.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Saracatinib and Alcohol Drinking — NCT02955186.
• Registry record last updated 2020-11-13; status last verified 2020-10.
• Results first posted 2020-11-13.
• 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

NCT02955186 is a completed, randomized, triple-masked, parallel phase 2 trial sponsored by Yale University. It enrolled 50 participants and compared 125 mg saracatinib daily for 8 days with placebo daily for 8 days in a laboratory drinking study. The primary endpoint was change in drinks consumed from baseline to Day 8.

The posted primary mean difference was 0.48 drinks, with a 95% confidence interval from -1.954 to 2.914 and an unadjusted p-value of 0.699. The registry also posts stimulation and sedation response outcomes, unadjusted comparisons, and safety summaries. Analysis populations, missing-data methods, multiplicity procedures, and the original sample-size justification are not reported in the registry record.