Mild Versus Moderate Therapeutic Hypothermia in Out-of-hospital Cardiac Arrest Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Mild Versus Moderate Therapeutic Hypothermia in Out-of-hospital Cardiac Arrest Patients (NCT02011568), a randomized na study. The primary outcome is Number of Participants With Death or Poor Neurologic Outcome at 6 Months.

2. Clinical Question

The primary outcome combines death with poor neurologic outcome. For participants assessed neurologically, the registry defines poor outcome as a Disability Ratings Scale score >5, assessed by a specialist in rehabilitation medicine. The comparison therefore concerns the occurrence of either component by the six-month assessment, rather than neurologic function alone.

The result should be read as a comparison between the randomized groups, not as a comparison with no temperature management. Both groups received therapeutic hypothermia, at different target temperatures. The registry’s primary outcome counts participants with the combined event; it does not report separate primary-outcome counts for death and poor neurologic outcome. The record also does not report the detailed procedures used to achieve or maintain the target temperatures.

3. Trial Design

The registry describes a randomized, double-blind, parallel-group trial. It records an enrollment of 389 and a completed status. The listed study dates are 2013-08 for start and 2020-03 for primary completion. The sponsor is Ottawa Heart Institute Research Corporation, and the official title identifies the study as CAPITAL CHILL. The phase is listed as NA.

The trial compared moderate hypothermia at 31 degrees Celsius with mild hypothermia at 34 degrees Celsius after out-of-hospital cardiac arrest. The registry describes the intended participants as comatose survivors with return of spontaneous circulation. Its summary calls the study an investigator-initiated prospective clinical trial at the University of Ottawa Heart Institute and says the plan was to expand it as a multicenter project. The registry record does not report whether that expansion occurred, so the stated single-center description is the basis for describing the trial setting.

The primary endpoint was assessed at six months and combined death or poor neurologic outcome. The posted arm counts for that endpoint are 89 participants among 184 in the moderate group and 83 among 183 in the mild group. These outcome denominators are not the same as the overall enrollment figure; the registry record does not explain the difference. It does not report the trial’s original sample-size justification, the detailed randomization procedure, or the specific masking procedures.

The registry lists secondary outcomes including mortality, stroke, bleeding, length of stay, cardiogenic shock, repeat circulatory arrest requiring cardiopulmonary resuscitation, and seizures. Some secondary outcome results are posted, while other listed outcomes do not have results in the registry record. The record reports zero formal analyses. Consequently, the estimates discussed in the statistical interpretation are unadjusted recalculations from the posted summary counts, rather than reported sponsor analyses.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

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 With Death or Poor Neurologic Outcome at 6 Months, measured over Six months.

The primary endpoint is the number of participants who experienced death or poor neurologic outcome at six months. The registry defines poor neurologic outcome using the Disability Ratings Scale: a score >5 qualifies as poor, with assessment by a specialist in rehabilitation medicine. Death and neurologic disability are combined into one endpoint, so the outcome reflects either event and does not isolate the contribution of each component.

A combined endpoint can capture more than one important unfavorable outcome in a single comparison. That can make the result relevant to a broad question about survival and neurologic status after cardiac arrest. At the same time, the combined count does not show whether any difference arises mainly from deaths, neurologic outcomes among survivors, or both. Separate primary endpoint component counts are not reported in the registry record.

The six-month timeframe identifies when the primary outcome is assessed. The posted counts are 89 of 184 participants in the moderate group and 83 of 183 in the mild group. These counts describe the recorded endpoint in each group; they do not, on their own, establish that the groups differ beyond what could plausibly arise through sampling variation. That uncertainty is central to interpreting the comparative estimate and its confidence interval.

7. Results Posted in the Registry

• Mild Hypothermia: 45.4% (83/183) versus 48.4% (89/184) on control; risk difference -3.0 percentage points (95% CI -13.2 to 7.2), risk ratio 0.94 (0.75–1.17); unadjusted two-proportion z-test p = 0.563.

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.

Number of Mortality

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Mild Hypothermia: 41.0% (75/183) versus 43.5% (80/184) on control; risk difference -2.5 percentage points (95% CI -12.6 to 7.6), risk ratio 0.94 (0.74–1.20); unadjusted two-proportion z-test p = 0.629.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Mild Hypothermia and the control. The interval gives the range of effects that remain compatible with the data.

Number of Participants With Bleeding

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Mild Hypothermia: 19.7% (36/183) versus 23.4% (43/184) on control; risk difference -3.7 percentage points (95% CI -12.1 to 4.7), risk ratio 0.84 (0.57–1.25); unadjusted two-proportion z-test p = 0.389.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Mild Hypothermia and the control. The interval gives the range of effects that remain compatible with the data.

Number of Participants With Cardiogenic Shock

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Mild Hypothermia: 33.3% (61/183) versus 38.6% (71/184) on control; risk difference -5.3 percentage points (95% CI -15.1 to 4.6), risk ratio 0.86 (0.66–1.14); unadjusted two-proportion z-test p = 0.294.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Mild Hypothermia and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists secondary outcomes spanning mortality, stroke, bleeding, length of stay, cardiogenic shock, repeat circulatory arrest requiring cardiopulmonary resuscitation, and seizures. Numerical comparisons are posted for mortality, bleeding, and cardiogenic shock. The available unadjusted estimates point in the same direction as the primary point estimate, but their confidence intervals span zero. This broad pattern does not make the primary finding more certain: each secondary comparison has its own uncertainty, and the outcomes are not substitutes for the primary endpoint.

The secondary outcomes address different aspects of the clinical course. Mortality is part of the primary composite, while bleeding and cardiogenic shock capture other events during the hospitalization. Their timeframes and meanings differ, so concordance in estimate direction does not establish a common treatment effect. The secondary comparisons cannot explain whether the primary composite’s pattern is attributable to death, neurologic outcome, or both; separate primary-component results are not reported in the registry record.

Unadjusted secondary results carry less weight because several outcomes are listed and the registry record does not report a multiplicity adjustment or a detailed statistical analysis plan. When many comparisons are considered, some results can appear notable by chance alone. Secondary endpoints can help describe the overall pattern, but they should not be treated as independently confirmatory without information about prespecification and multiplicity handling. Results for the other listed secondary outcomes are not reported in the registry record.

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

• Number of Mortality
• Number of Participants With Stroke
• Number of Participants With Bleeding
• Length of Stay in the Unit
• Length of Stay in the Hospital
• Number of Participants With Cardiogenic Shock
• Number of Participants With Repeat Circulatory Arrest Requiring Cardiopulmonary Resuscitation (CPR)
• Number of Participants With Seizures

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is binary: each participant either had the combined event of death or poor neurologic outcome by six months, or did not. For a binary endpoint, the event count and the number assessed in each group describe the observed data. Here, the posted counts are 89 of 184 in the moderate group and 83 of 183 in the mild group. The statistical comparison uses those group summaries.

What does a risk difference describe?

A risk difference compares the event proportion in one group with the event proportion in the other. The stated comparison is mild hypothermia relative to moderate hypothermia, so a negative estimate indicates a lower observed event proportion in the mild group. Unlike a relative measure, a risk difference expresses an absolute contrast in proportions. It is useful for seeing the size and direction of the observed separation, but the estimate alone does not show how uncertain that separation is.

How should the confidence interval be read?

A confidence interval gives a range of values compatible with the observed data under the method’s assumptions. A narrower interval indicates greater precision; a wider interval indicates more uncertainty about the magnitude and direction of the underlying difference. For the primary comparison, the reported interval extends from a negative value to a positive value and therefore includes zero, the value corresponding to no risk difference. That does not prove no difference. It means the data are compatible with no difference as well as with differences in either direction within the interval.

What does the p-value contribute?

The unadjusted p-value is 0.563 for the primary comparison. A p-value evaluates how unusual data at least as discordant with a null comparison would be if that null model and the test assumptions held. It is not the probability that the null is true, and it does not measure the probability that the result is due to chance. A p-value also does not measure clinical importance or provide the range of plausible effect sizes. The estimate and interval are needed to understand direction and precision.

What is meant by an unadjusted recalculation?

The comparisons are simple unadjusted recalculations from the posted counts. They do not account for covariates or other analysis features that might be part of a trial’s full statistical plan. The registry record reports zero formal analyses and contains no sponsor-reported analysis for these outcomes. The record does not report whether models, subgroup analyses, interim analyses, or other analytic procedures were planned or performed. Their absence from the registry record should not be interpreted as evidence that they did or did not occur.

Why do secondary comparisons need caution?

Several secondary outcomes are listed, and multiple comparisons raise the possibility of chance findings when many questions are examined. An unadjusted p-value assesses one comparison without accounting for the set of outcomes. Without a reported multiplicity strategy, a nominal result from one secondary comparison would carry less weight than a result supported by a prespecified analysis that accounts for the broader testing context. The posted secondary comparisons are therefore best read as descriptive and exploratory, particularly when their confidence intervals are broad.

What can randomization and masking contribute?

Random allocation is intended to make the groups comparable on average, including factors that may not be measured. Double masking is recorded, but the registry record does not give details about who was masked or how masking was maintained. These design features can support a fair comparison, but they do not remove sampling uncertainty, missing-data concerns, or the need to understand how outcomes were analyzed. Detailed information about the allocation process and analysis populations is not reported in the registry record.

11. Confidence Intervals and Precision

The primary risk-difference confidence interval runs from -0.132 to 0.072. Because it crosses zero, the data are compatible with no difference as well as with a lower event proportion in either group. The point estimate lies in the negative direction, but the interval is wide enough to include a positive difference as well. The estimate therefore has limited precision about both the direction and magnitude of the underlying group contrast.

The confidence interval should be considered alongside the observed counts: 89 of 184 participants in the moderate group and 83 of 183 in the mild group. Counts alone can make a comparison look more definite than it is; the interval communicates sampling uncertainty. The p-value of 0.563 is consistent with this uncertainty, but it does not replace the interval or establish that the groups are equivalent.

Intervals for the posted secondary risk-difference comparisons also cross zero. This indicates that those estimates likewise do not pin down a clear direction of difference. Precision depends on the amount of information in the data and the variability of the outcome, among other features. The registry record does not report a prespecified clinically important difference, so it is not possible to judge the interval against such a threshold from the registry information alone.

These intervals accompany unadjusted recalculations from summary counts. They do not incorporate any unreported adjustment or analysis decisions. Their interpretation is conditional on the calculation method and the posted data being appropriate for the comparison. The registry record does not provide enough detail to assess other sources of uncertainty, such as missing outcome data or departures from the intended analysis population.

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 estimates are participants per arm for the binary outcomes. They are not total event counts. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning scenario uses a target effect expressed as a stated fraction of the observed effect. The original trial’s own sample-size justification and design assumptions are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm for a planned study to have an 80% probability of detecting the assumed effect under the stated planning conditions, if that effect and the assumptions were correct. The 90% power column gives the corresponding planning target for a 90% probability. These are prospective design calculations, not descriptions of the probability that the completed trial found a true effect.

For the primary endpoint, planning for 100% of the observed effect gives 4,301 participants per arm for 80% power and 5,758 for 90% power. Planning for 75% of the observed effect gives 7,654 participants per arm for 80% power and 10,246 for 90% power. Planning for 50% of the observed effect gives 17,236 participants per arm for 80% power and 23,074 for 90% power.

The pattern illustrates that smaller effects require more participants to detect under otherwise unchanged assumptions. The estimates use the observed effect as their starting point, so they are planning illustrations rather than independent confirmation of an effect. This is not observed power.

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 = 0.484; treatment proportion p₁ = 0.454; average p̄ = 0.469.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7057 = 1.3832.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.7054 = 0.5937.
4. n = (1.3832 + 0.5937)² / (-0.0301)² = 4,300.8, rounded up to 4,301 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 <- 0.483695652173913                          # control proportion
p1 <- 0.453551912568306                          # 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(2011568)
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 = 4,301 (80% power), 5,758 (90% power)
 75% of observed effect: n per arm = 7,654 (80% power), 10,246 (90% power)
 50% of observed effect: n per arm = 17,236 (80% power), 23,074 (90% power)
power.prop.test: 4300.8 
Simulated power at n = 4,301 per arm: 0.805 (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 lists numerous secondary outcomes, including several distinct clinical events and measures of hospital stay. The record reports no formal analyses and does not describe a multiplicity strategy. It is therefore not reported in the registry record whether the trial used a hierarchy of outcomes, adjusted significance thresholds, or another method to account for repeated testing. This matters because examining multiple outcomes can increase the chance of finding an apparently unusual result even when no real group difference exists.

The primary endpoint is identified in the record, but the registry does not provide a full statistical analysis plan. The planned model, any covariate adjustment, interim analyses, subgroup analyses, and rules for handling multiple looks at the data are not reported in the registry record. The presented comparisons are unadjusted recalculations from posted summary counts, not a replacement for a protocol-defined sponsor analysis.

The record also does not report the analysis population, such as whether the primary comparison followed an intention-to-treat approach, or how participants with unavailable six-month assessments were handled. The posted outcome denominators are 184 and 183, while overall enrollment is 389; the record does not explain the relationship between those figures. It is not possible to infer from the counts whether participants were excluded, had missing outcomes, or were handled in another way.

Missing-data methods can affect estimates if outcome availability differs between groups or is related to participants’ outcomes. Without information on missingness and analysis populations, the extent of that concern cannot be assessed from the registry record. These omissions limit appraisal of the analysis, but they do not by themselves establish that the reported counts are biased. They identify information needed to understand how the posted results were generated.

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

Serious events with the highest arm-level rate: deep vein thrombosis (11.4%), inferior vena cava thrombosis (7.7%).

The registry records serious adverse events among 28 of 184 participants in the moderate group and 34 of 183 participants in the mild group. It records deaths of 80 and 75, respectively, in the adverse-event group information. These counts should be read in the context of the registry’s adverse-event reporting, not as a separate adjusted comparison. The registry record does not provide a detailed description of event attribution or the criteria used to classify serious events.

The most frequent serious-event terms listed are Deep Vein Thrombosis at 11.4% and Inferior Vena cava thrombosis at 7.7%. The record does not identify these term percentages by treatment group in the listed summary, so they should not be assigned to one group. Other adverse events are recorded as 0 in both groups. Event counts and terms are descriptive; they do not by themselves establish whether an event was caused by treatment or whether group differences are statistically reliable.

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

The posted primary outcome counts show that death or poor neurologic outcome at six months occurred in 89 participants in the moderate group and 83 in the mild group. The unadjusted risk-difference estimate is in the direction of a lower observed event proportion with mild hypothermia, but its confidence interval includes zero and extends in both directions. The result therefore remains uncertain about the direction and size of any underlying difference.

This comparison does not establish that the two temperature targets are equivalent. Establishing equivalence or noninferiority requires an appropriate design, prespecified margins, and analysis; those details are not reported in the registry record. Nor does the result establish a clinically important advantage for either target. The registry record does not report a threshold for clinical importance, and statistical uncertainty is not the same as evidence that an effect is absent.

The endpoint combines death and poor neurologic outcome, so the overall result does not show which component contributes to the observed counts. The listed secondary outcomes add context but are also uncertain and are not reported with a multiplicity strategy. The safety summaries provide event counts and terms, not proof of causation or a complete comparative safety analysis.

The trial record supports a careful description of what was posted: randomized groups, a six-month composite endpoint, group counts, and an imprecise unadjusted comparison. It does not support treatment advice or a definitive conclusion about the relative clinical merits of the two target temperatures. Additional information about the analysis plan, missing outcomes, and component endpoints would be needed for a fuller interpretation.

17. Why This Trial Matters Statistically

This trial presents a useful statistical distinction between the direction of an estimate and the certainty of that estimate. The primary point estimate favors a lower observed event proportion in the mild group, but the interval spans zero and the unadjusted p-value is 0.563. Reporting the estimate together with its uncertainty prevents the direction of the point estimate from being mistaken for a conclusive group difference.

The endpoint also illustrates a trade-off in composite outcomes. Combining death with poor neurologic outcome gives a broad measure of an unfavorable six-month outcome, while leaving unanswered how much each component contributes. That distinction matters when the outcome has different components with different meanings. The registry record does not report separate primary component counts.

The planning illustrations emphasize how sample-size requirements depend on the effect a study is designed to detect. Smaller assumed effects require more participants per arm under the stated planning assumptions. Because these calculations use the observed effect and are not the trial’s original design justification, they should be read as planning aids rather than as a judgment of the completed trial’s power.

Finally, the registry record demonstrates why statistical interpretation depends on reporting beyond event counts. Information about analysis populations, missing data, multiplicity, and the sponsor’s statistical plan would help readers assess how the results were obtained. Where that information is not reported, transparent uncertainty is more informative than a categorical conclusion.

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 main limitation for interpretation is the limited analytical detail in the registry record. It does not report a sponsor analysis for the primary or posted secondary outcomes, a statistical analysis plan, the original sample-size justification, or whether covariate adjustment was planned. The comparisons discussed are unadjusted recalculations from posted summary counts.

The primary outcome is a composite, and separate counts for death and poor neurologic outcome are not reported. The record also does not explain the difference between overall enrollment of 389 and the primary outcome denominators of 184 and 183. Missing outcome counts, reasons for unavailable assessments, and the analysis population are not reported, so the impact of missingness cannot be evaluated.

The trial is described as initiated at a single center, while the summary mentions a plan for multicenter expansion. Whether expansion occurred is not reported in the registry record. The record also does not give detailed information on randomization procedures, masking implementation, treatment delivery, or outcome-assessment completeness. These details would help readers judge how well the comparison reflects the intended design.

Numerous secondary outcomes are listed, but the record does not report a multiplicity approach and numerical results are not available for every listed outcome. Serious adverse event summaries lack detailed attribution information. These limitations do not invalidate the posted counts, but they constrain the conclusions that can be drawn from them.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Mild Versus Moderate Therapeutic Hypothermia in Out-of-hospital Cardiac Arrest Patients — NCT02011568.
• Registry record last updated 2024-12-06; status last verified 2023-10.
• Results first posted 2024-12-06.
• 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

NCT02011568 is a completed randomized, double-blind, parallel-group trial comparing therapeutic hypothermia at 31 degrees Celsius with therapeutic hypothermia at 34 degrees Celsius after out-of-hospital cardiac arrest. The registry records enrollment of 389, a start date of 2013-08, and primary completion in 2020-03.

The primary outcome was death or poor neurologic outcome at six months. The posted counts were 89 of 184 in the moderate group and 83 of 183 in the mild group. The unadjusted recalculation has a risk-difference confidence interval that crosses zero; the registry record does not report a sponsor analysis. Secondary outcomes and adverse-event summaries provide additional descriptive information, but multiplicity and missing-data methods are not reported in the registry record.