In Person and mHealth Coping Skills Training for Symptom Management and Steps in Stem Cell Transplant Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for In Person and mHealth Coping Skills Training for Symptom Management and Steps in Stem Cell Transplant Patients (NCT03859765), a randomized na study. The primary outcome is Number of Participants Who Adhere to at Least 80% of the Study Protocol.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group trial with two arms and no masking. The registry record lists the study as completed and reports enrollment of 40 participants. Each arm included 20 participants for the primary adherence outcome. The study was registered as phase NA, and the condition listed was stem cell transplant.

The intervention arm, HCT Symptoms and Steps, combined coping-skills training with activity coaching. Its planned delivery included 3 in-person sessions and 4 video-conferencing sessions, along with a wireless activity tracker. The comparison arm, HCT Education, received 1 brief in-clinic session before discharge, a wireless activity tracker, and 6 brief phone calls after returning home. The intervention descriptions identify the general components, but further details about session content and delivery are not reported in the registry record.

The registry record gives a study start of 2019-07-23 and a primary completion date of 2021-12-20. The primary outcome timeframe is approximately 20-24 weeks. The three posted secondary outcome results use changes from pre-treatment baseline to post-treatment, approximately 7-12 weeks. These timeframes are not interchangeable: the primary measure concerns adherence over the protocol period, while the secondary results concern measured changes over the shorter stated interval.

Randomization is reported, but the method used to generate the allocation sequence, concealment procedures, and any stratification are not reported in the registry record. The record also does not report a masking plan beyond identifying masking as none. These details matter for judging how participants entered groups and how outcomes were assessed, particularly for behavioral interventions where participants are aware of the program they receive.

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

Analysis population for the primary outcome, as defined in the registry: Total number screened

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 Participants Who Adhere to at Least 80% of the Study Protocol, measured over approximately 20-24 weeks.

The primary endpoint counts participants who adhered to at least 80% of the study protocol. The registry record identifies this as a measure of feasibility and gives an approximate assessment period of 20-24 weeks. It reports 17 participants meeting the threshold in HCT Symptoms and Steps and 18 in HCT Education, with 20 participants in each group.

A feasibility endpoint asks whether a study process or intervention can be carried out and whether participants engage with it. For a multi-session program delivered partly after discharge, adherence can indicate whether participants complete a substantial share of the planned protocol. It does not directly measure pain, fatigue, disability, psychological distress, or physical activity. Those topics appear among the study's secondary outcomes, but adherence should not be treated as evidence that any symptom or activity outcome changed.

The threshold simplifies a more detailed pattern of participation into a count: participants either met the stated adherence criterion or did not. The registry record does not report how adherence across separate sessions or program elements was counted, how incomplete records were handled, or whether participants who did not meet the threshold had different amounts or types of participation. Such details would help clarify what the count represents in practice. The reported result therefore supports a narrow description of how many participants met the protocol adherence threshold, rather than a complete account of engagement or of the intervention's clinical effects.

7. Results Posted in the Registry

• HCT Education: 90.0% (18/20) versus 85.0% (17/20) on control; risk difference 5.0 percentage points (95% CI -15.4 to 25.4), risk ratio 1.06 (0.84–1.34); unadjusted two-proportion z-test p = 0.633.

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.

Change in Physical Disability as Measured by the the Functional Assessment of Cancer Therapy (FACT)

• 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:
  • HCT Education: mean 6.94 versus 4.78 on control; difference 2.16 (95% CI -1.66 to 5.98), standardized effect d = 0.37; z-test p = 0.268.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between HCT Education and the control. The interval gives the range of effects that remain compatible with the data.

Change in Pain Severity

• 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:
  • HCT Education: mean 0.02 versus -0.79 on control; difference 0.81 (95% CI -0.43 to 2.05), standardized effect d = 0.43; z-test p = 0.199.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between HCT Education and the control. The interval gives the range of effects that remain compatible with the data.

Change in Fatigue

• 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:
  • HCT Education: mean -4.72 versus -2.62 on control; difference -2.10 (95% CI -7.26 to 3.06), standardized effect d = -0.27; z-test p = 0.425.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between HCT Education and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists secondary outcomes covering physical disability, pain severity, fatigue, physical activity, depression, anxiety, and self-efficacy. However, posted numerical results are available for only three of these outcomes: physical disability, pain severity, and fatigue. Results for physical activity, depression, anxiety, and self-efficacy are not reported in the registry record. The record also does not provide outcome-specific analysis descriptions for the secondary measures.

Across the three posted results, the estimated mean differences do not point uniformly in one direction: the estimates for disability and pain are positive, while the fatigue estimate is negative. The intervals for all three comparisons include zero. Without the score direction for each measure, numerical signs alone do not establish whether a change is favorable or unfavorable. Together, these summaries do not provide a clear, precise pattern of differences across the symptom-related outcomes.

The secondary measures relate to the primary result by addressing different questions. The primary outcome concerns adherence and feasibility; the secondary outcomes concern measured changes in disability and symptoms, with other planned domains not numerically reported. Similar or different adherence counts cannot establish that symptoms changed, and symptom measurements do not determine whether participants completed the protocol. Each outcome must be interpreted on its own terms.

Unadjusted secondary results carry less weight because the registry lists several outcome domains and does not report a multiplicity strategy or adjusted analyses. When many outcomes are examined, chance findings become more likely if each is treated as a standalone test. This general concern does not mean every observed contrast is false; it means secondary comparisons need cautious interpretation and ideally a prespecified hierarchy or adjustment plan. Such a plan is not reported in the registry record. The incomplete numerical reporting across the listed secondary outcomes also limits any assessment of the overall pattern.

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

• Change in Physical Disability as Measured by the the Functional Assessment of Cancer Therapy (FACT)
• Change in Pain Severity
• Change in Fatigue
• Change in Physical Activity
• Change in Depression
• Change in Anxiety.
• Change in Self-Efficacy

9. Statistical Interpretation

10. Statistical Methods Explained

The registry record identifies a randomized parallel-group design. Randomization is intended, in general, to create groups that are comparable on average at the time of assignment. It does not guarantee that all characteristics will be balanced in a small trial. The specific randomization procedure and allocation concealment are not reported in the registry record, so the adequacy of those procedures cannot be assessed from the available description.

What does the primary comparison estimate?

The primary outcome is binary at the participant level: each person either met the threshold of adherence to at least 80% of the protocol or did not. The registry reports the count meeting the threshold in each group and a risk-difference estimate. A risk difference expresses the contrast between the proportions in the two groups using the comparison's stated ordering. A value of zero would represent equal observed proportions; a positive or negative value indicates which group has the higher observed proportion under that ordering. Here the estimate is 0.05, and its interval runs from -0.154 to 0.254.

The reported unadjusted p value is 0.633. A p value describes how compatible the observed data are with a specified null model of no group difference, under the assumptions of the test. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of a difference. Because the result is unadjusted, it should not be read as accounting for multiple outcomes or other covariates. The registry does not report an adjusted analysis for this endpoint.

How should the continuous outcomes be read?

For the posted secondary results, each group has a mean change and a standard deviation. A mean summarizes the average change among the participants included in that result. A standard deviation describes variability among individual observations around the mean; it is not a confidence interval and does not indicate the uncertainty in the mean by itself. The registry comparison gives a mean difference for each outcome, together with an interval and an unadjusted p value.

The mean difference is a contrast between the reported group means, with the ordering indicated by the registry comparison. Zero corresponds to no difference between those means. For the disability, pain, and fatigue outcomes, the direction of a favorable change cannot be inferred from the numerical signs alone unless the scoring direction and interpretation of each scale are known. Those details are not reported in the registry record. The results should therefore be described numerically rather than labeled improvement or worsening.

The comparisons also include a standardized effect estimate for the continuous outcomes. In general, standardization expresses a mean contrast relative to a measure of variability, allowing the scale-specific contrast to be described in a unitless way. It does not make scales clinically interchangeable, and it does not establish that a difference matters to patients. Clinical interpretation would need information about the instruments, score direction, and meaningful-change criteria, which are not reported in the registry record.

What do confidence intervals contribute?

A confidence interval gives a range of values compatible with the observed data under the method used. It communicates precision as well as direction: a broad interval means the estimate is uncertain over a wider range of possible differences. For the primary risk difference, the interval includes zero and extends to values on both sides of no difference. The same general observation applies to the intervals for the three posted secondary mean differences. The registry record does not state the method used for its original analyses, and no reported analysis details are listed.

What assumptions underlie the planning illustrations?

The planning illustrations are based on the observed effect, not on the trial's own design. Their stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each row uses the stated fraction of the observed effect as the effect to detect. These are planning assumptions for interpreting hypothetical future sample sizes, not evidence about the original protocol. The original sample-size assumptions and justification are not reported in the registry record.

In general, normal approximations simplify the calculation by using an approximate distribution for the estimate. Their suitability depends on the outcome and sample size. The planning displays should be read as illustrations under their stated assumptions, rather than as a substitute for a design tailored to a future trial's outcome definition, analysis model, expected attrition, and allocation procedures.

11. Confidence Intervals and Precision

The primary risk-difference estimate is 0.05, with a confidence interval from -0.154 to 0.254. The interval includes zero, so the posted data are compatible with no difference in adherence as well as with differences in either direction. The range is broad relative to the point estimate, indicating substantial uncertainty about the comparative adherence rate in a larger population. The unadjusted p value is 0.633; it should be read alongside the interval, not as a standalone measure of importance or certainty.

For the three secondary outcomes with posted comparisons, the intervals for mean differences are -1.662 to 5.982 for physical disability, -0.427 to 2.047 for pain severity, and -7.264 to 3.064 for fatigue. Each interval includes zero. Their widths indicate that the point estimates are not precise enough to isolate a narrow range of plausible group differences. The registry record does not give clinical thresholds for interpreting these scales, so the intervals cannot be translated into a statement about clinically meaningful change.

Confidence intervals are not guarantees that the true effect lies inside the reported range, and they do not assign probabilities to individual values after the data are observed. In general, they show the uncertainty that would arise from sampling under the analysis method. Precision is limited by the number of participants with posted outcome data: the primary groups each have 20 participants, while the secondary groups each have 18. The record does not explain why the secondary outcome counts differ from the primary counts or how missing measurements were handled.

These estimates are useful for describing the pilot and for illustrating uncertainty, but they should not be treated as stable estimates of effects in a broader HCT population. The interval widths, rather than the point estimates alone, are important when considering what the study can establish.

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 sample-size planning figures are participants per arm for 80% power and participants per arm for 90% power. They are not total events: the posted outcomes are participant counts or continuous measurements, and the record contains no time-to-event outcomes for which a total-event calculation would apply. Equal allocation is assumed between the arms.

The planning assumptions are a two-sided significance level of 0.05, equal allocation, and a normal approximation. The effect to detect is set in turn to the stated fraction of the observed effect. Read each planning line as a hypothetical: if a future study aimed to detect that assumed effect under those conditions, the estimate gives the participants needed in each arm. The planning figures are not the actual enrollment or analysis sample sizes for this trial.

The 80% power column represents a design intended to have an 80% chance of detecting the assumed effect if that effect is real and the planning assumptions hold. The 90% column represents a higher target chance of detection under those same conditions and therefore generally calls for more participants. In general, smaller effects are harder to distinguish from random variation and require more participants than larger effects, all else equal. The planning figures reflect this relationship as the assumed effect is reduced.

These calculations are planning illustrations based on the observed effect, not the trial's original design assumptions. The registry record does not report the original sample-size justification or design assumptions. They are not observed power: observed power is not a useful substitute for interpreting the estimates and their confidence intervals after a trial has been completed.

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

Worked calculation for this trial

1. Control proportion p₀ = 0.850; treatment proportion p₁ = 0.900; average p̄ = 0.875.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4677 = 0.9167.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4664 = 0.3925.
4. n = (0.9167 + 0.3925)² / (0.0500)² = 685.6, rounded up to 686 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 <- 0.85                          # control proportion
p1 <- 0.9                          # 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(3859765)
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 = 686 (80% power), 918 (90% power)
 75% of observed effect: n per arm = 1,272 (80% power), 1,702 (90% power)
 50% of observed effect: n per arm = 2,978 (80% power), 3,986 (90% power)
power.prop.test: 685.6 
Simulated power at n = 686 per arm: 0.802 (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.

14. Multiplicity, Missing Data and Analysis Populations

The registry record lists seven secondary outcome domains, but numerical results are posted for only physical disability, pain severity, and fatigue. Results for physical activity, depression, anxiety, and self-efficacy are not reported in the registry record. It lists no reported analyses. The reported outcome summaries and unadjusted comparisons therefore do not amount to a fully documented analysis plan for all measured outcomes.

Multiplicity is the statistical issue that arises when several outcomes or comparisons are tested. In general, the more comparisons examined, the greater the chance that at least one result may appear noteworthy by chance if each is considered separately. Researchers may address this through prespecified primary and secondary outcome roles, hierarchical testing, or statistical adjustment. The registry record does not report a multiplicity strategy, so the unadjusted secondary p values should be treated as exploratory rather than as a set of confirmatory tests.

Missing-data methods and analysis populations are not reported in the registry record. The primary outcome includes 20 participants in each group, while the three secondary outcomes include 18 in each group. The record does not state whether the difference reflects missing measurements, exclusions, outcome-specific availability, or another reason. It also does not identify whether the analyses followed an intention-to-treat approach, a per-protocol approach, or another population definition. Those omissions affect how representative the reported summaries are of all randomized participants.

The registry identifies the trial as randomized and unmasked, but it does not report allocation generation or concealment details, outcome-assessor procedures, or a statistical analysis plan. It also does not report interim analyses or a sample-size justification. These details should not be inferred from the pilot design or the posted results. Without them, the results can be described as reported, but the risk of bias, handling of incomplete outcome data, and prespecification of comparisons cannot be fully evaluated.

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 registry lists zero participants affected by serious adverse events in HCT Symptoms and Steps and zero in HCT Education. It also lists zero participants affected by other adverse events in each group. There were 20 participants at risk in each group, and the registry reports zero deaths in both groups.

These figures are counts recorded for the trial, not proof that no adverse event could occur or that the groups have identical safety profiles. With 20 participants at risk per group, the data provide limited information about uncommon events. The record does not list serious adverse-event terms, and it does not provide additional detail about event definitions, how events were collected, or the observation period used for safety reporting. The primary endpoint timeframe is approximately 20-24 weeks, but the registry does not state a separate safety follow-up period.

Accordingly, the safety entries should be read narrowly as the event counts recorded in this study. They do not support broad conclusions about safety beyond the trial participants and the reporting information in the registry record.

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

The posted primary result describes protocol adherence in a randomized pilot: 17 participants in HCT Symptoms and Steps and 18 in HCT Education met the threshold of at least 80% adherence, out of 20 participants in each group. The estimated risk difference is 0.05, and the confidence interval ranges from -0.154 to 0.254. This interval includes no difference and allows uncertainty in either direction. The result is therefore a limited estimate of feasibility rather than a precise comparison.

The outcome does not measure whether participants experienced less pain, fatigue, distress, or disability, or whether physical activity increased. The trial did list those kinds of outcomes, but only three secondary results are numerically posted. Their intervals include zero, and the registry does not report the scoring direction or a clinically meaningful threshold for the scales. It is not possible to use these summaries alone to make claims about the clinical value of the program.

Nor does the primary result establish equivalence between the approaches. The trial does not report an equivalence margin or an analysis designed to test equivalence. An inconclusive comparison is different from evidence that two programs have the same effect. The uncertainty in the interval and the limited number of participants remain central to interpretation.

The registry record describes an intervention tailored to coping and activity during HCT recovery, delivered through in-person and video sessions, and compares it with education, a tracker, and phone follow-up. The recorded data concern this particular comparison and the participants enrolled. They do not establish effects in populations or care settings not represented by the trial. The appropriate conclusion is descriptive: the registry reports similar adherence counts in the two groups, with substantial uncertainty about the comparative difference, and offers limited secondary outcome information.

17. Why This Trial Matters Statistically

This trial illustrates why a pilot's primary question may be feasibility rather than efficacy. The intervention involved multiple sessions and contact after discharge, so adherence was a direct way to describe whether participants completed a substantial portion of the protocol. A feasibility measure can inform whether a program is deliverable and worth studying further, but it does not substitute for a sufficiently precise estimate of clinical outcomes.

The primary comparison also shows why effect estimates should be read with their intervals. The point estimate alone is 0.05, but its interval extends from -0.154 to 0.254. That uncertainty matters more than treating a single p value as a verdict. The posted result does not identify a precise difference between the groups, nor does it demonstrate equivalence.

The trial offers a useful example of an active comparator. HCT Education included an in-clinic session, a wireless tracker, and phone calls, while HCT Symptoms and Steps involved more extensive coping-skills and activity-coaching sessions plus a tracker. The question is therefore about the contrast between two packages of support, not about intervention versus no support. Careful definition of the comparator is essential when interpreting any observed difference.

Finally, the record highlights the value of transparent outcome and analysis reporting. Several secondary domains are listed, yet numerical results are posted for only some of them, and analysis details are absent. A clear future report could specify scoring, missing-data handling, analysis populations, multiplicity procedures, and the rationale for sample size. Those elements help readers distinguish feasibility evidence from estimates intended to support conclusions about treatment effects.

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 primary comparison is based on 20 participants per group, and the confidence interval is broad. That limits precision for the adherence contrast and leaves a wide range of plausible differences. The registry record identifies the study as a pilot, but its original sample-size justification and design assumptions are not reported.

The record does not explain how protocol adherence was calculated across the intervention components or how incomplete participation was classified. It also does not report why the three posted secondary outcomes have 18 participants per group while the primary result has 20. Missing-data handling and analysis-population definitions are not reported, so the effect of incomplete outcome data cannot be assessed.

Numerical results are not reported for four listed secondary outcomes: physical activity, depression, anxiety, and self-efficacy. The posted continuous outcomes lack reported scale-direction details and clinically meaningful thresholds. This makes it difficult to interpret the signs and practical importance of the changes. The registry also provides no reported analysis plan, multiplicity approach, or adjusted analyses, and the entries for reported analyses are empty.

Other design details are absent, including the method of random sequence generation, allocation concealment, assessor procedures, and any interim analysis plan. The intervention and education descriptions are abbreviated, so the full content and fidelity procedures cannot be evaluated from the registry record. Finally, the record provides no separate description of the safety observation window or adverse-event collection methods. These limitations restrict interpretation of both comparative effects and generalizability beyond the enrolled participants.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: In Person and mHealth Coping Skills Training for Symptom Management and Steps in Stem Cell Transplant Patients — NCT03859765.
• Registry record last updated 2024-03-22; status last verified 2022-09.
• Results first posted 2024-03-22.
• 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

NCT03859765 is a completed randomized, unmasked, parallel-group pilot trial sponsored by Duke University. It enrolled 40 participants and compared HCT Symptoms and Steps with HCT Education for stem cell transplant patients. The primary endpoint was the number of participants adhering to at least 80% of the protocol over approximately 20-24 weeks.

The registry reports 17 of 20 participants meeting the threshold in HCT Symptoms and Steps and 18 of 20 in HCT Education. The risk-difference estimate is 0.05, with a confidence interval from -0.154 to 0.254. Three secondary outcomes have posted numerical results; results for four other listed secondary outcomes are not reported in the registry record. The record also lists zero serious adverse events, zero other adverse events, and zero deaths in each group. Analysis-plan, missing-data, and multiplicity details are not reported.