Three Different Cardiopulmonary Resuscitation (CPR) Training Methods: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Three Different Cardiopulmonary Resuscitation (CPR) Training Methods (NCT01361919), a randomized na study. The primary outcome is CPR Depth.

2. Clinical Question

3. Trial Design

The registry record describes a randomized controlled study with single masking and a factorial model. It lists three CPR training procedures: feedback during both training and testing, feedback during training but not testing, and no feedback during either. The study’s brief description frames the population as medical and nursing students learning basic life-saving skills. Enrollment is recorded as 298. The sponsor is Unity Health Toronto, and the study status is completed.

The listed study start is 2008-09 and primary completion is 2009-09. The record describes a training phase, immediate testing, and 12 week retention testing in the arm descriptions. However, the posted primary outcome and the ventilation secondary outcome are both labelled baseline. Their numerical results therefore describe the registry’s baseline-labelled measurements; they should not be treated as estimates of immediate post-training performance or retained performance.

Although the design is labelled factorial, the registry record does not report the factor structure, the allocation ratio, details of sequence generation, or how masking was maintained. It also does not explain which participants or assessors were masked. The arm descriptions identify the types of devices and whether feedback was available, but some descriptions end before providing complete procedural detail. There are no posted reported analyses. The record does not report the original analysis plan, participant flow, or reasons for any difference between enrollment and outcome-specific numbers. Those details cannot be inferred from the design labels or outcome summaries.

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: some participants did not record some of the baseline data elements prior to being assigned

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.

Population: All enrolled participants (n=298) completed baseline measures, however, due to technical difficulties - outcomes were not able to be collected for 32 participants as outlined above.

6. Primary Endpoint

The registry lists the primary outcome as CPR Depth, measured over baseline.

The primary endpoint is CPR depth: the mean depth of chest compressions, measured in millimeters. Compression depth is a direct quantitative measure of how far the manikin’s chest is compressed during CPR practice. The study used a simulation manikin with an accelerometer pad to collect CPR performance data, as described in the arm information. The registry lists the endpoint’s timeframe as baseline and gives a mean and standard deviation for each group.

As a continuous measure, depth preserves more information than a simple count of participants who do or do not meet a threshold. Comparing group means describes the average level of compression depth in each group, while the standard deviation describes variability among the observations. These are descriptive summaries; they do not by themselves establish that group differences were caused by the training approach.

The study summary also identifies compression speed and other CPR skills as areas of interest, but the posted primary outcome is depth. No target threshold or clinically important difference for depth is reported in the registry record. The baseline label is especially important for interpretation: the results cannot be read as evidence about how much students learned, how performance changed after instruction, or whether any difference persisted at 12 week testing.

7. Results Posted in the Registry

• Feedback During CPR Training Not Testing: mean 42.20 versus 43.70 on control; difference -1.50 (95% CI -3.33 to 0.33), standardized effect d = -0.24; z-test p = 0.108.
• No Feedback Group: mean 35.30 versus 43.70 on control; difference -8.40 (95% CI -10.39 to -6.41), standardized effect d = -1.24; z-test p = <0.001.

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.

Ventilations

• 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:
  • Feedback During CPR Training Not Testing: mean 18.00 versus 20.00 on control; difference -2.00 (95% CI -2.75 to -1.25), standardized effect d = -0.79; z-test p = <0.001.
  • No Feedback Group: mean 19.00 versus 20.00 on control; difference -1.00 (95% CI -1.75 to -0.25), standardized effect d = -0.39; z-test p = 0.009.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Feedback During CPR Training Not Testing than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for No Feedback Group than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary outcome is ventilation rate, measured as number per minute and labelled baseline. Its group means show a pattern that is not identical to the primary endpoint: the feedback-during-training-and-testing group has the highest reported mean, while the other two group means are lower. The magnitude and uncertainty of each contrast are available in the posted results, but these secondary summaries should be interpreted as baseline descriptions rather than evidence about later training performance.

At a broad level, the direction of the ventilation pattern is consistent with the primary baseline pattern in that the feedback-during-training-and-testing group has the higher group mean in both measures. The relative ordering of the other groups differs between the measures. This variation is a reminder that chest-compression depth and ventilation rate are distinct aspects of CPR performance; one cannot stand in for the other. The secondary outcome neither confirms nor overturns the primary result, and the registry does not report a combined measure of overall CPR skill.

The secondary comparisons are unadjusted. When multiple outcomes or group contrasts are examined, the chance of finding at least one apparently unusual result can increase. A single unadjusted p-value does not account for that broader set of questions. The registry record does not report a multiplicity procedure or identify a prespecified hierarchy for interpreting secondary outcomes. For that reason, the secondary results carry less confirmatory weight than an outcome and analysis plan designated in advance, and the registry’s reported-analyses list is empty.

The record also does not provide secondary outcome results for immediate testing or 12 week retention. Conclusions about learning, transfer to a device without feedback, or retention therefore cannot be drawn from these baseline ventilation summaries. The useful reading is limited to how the posted group means compare at the stated timeframe and how those descriptions sit alongside the primary measure.

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

• Ventilations

9. Statistical Interpretation

10. Statistical Methods Explained

Both posted outcomes are continuous measures reported as means and standard deviations. A mean gives the arithmetic average for a group. A standard deviation describes how spread out individual measurements are around that average: a larger value indicates more variability, while a smaller value indicates that measurements are more clustered. The standard deviation is not the same as the uncertainty in the estimated mean or in a difference between group means.

What does a mean difference represent?

A mean difference subtracts one group mean from another. In the posted comparisons, the reference is feedback during CPR training and testing. The negative signs indicate that the mean for the named comparison group is lower than the reference group’s mean. For CPR depth, the contrast with feedback during training but not testing is -1.5 millimeters, and the contrast with no feedback is -8.4 millimeters. These are unadjusted contrasts between group summaries. They do not account for covariates, and no adjusted analyses are reported in the registry record.

In a randomized trial, a difference between groups can be used to compare assigned approaches when the analysis and outcome timing match the trial question. Here the registry’s endpoint timeframe is baseline. That label limits interpretation: the contrast is not a reported change from before to after training. Nor does it estimate performance at immediate testing or 12 week retention testing. The registry does not report a model linking baseline measurements to later outcomes.

What does a confidence interval add?

A confidence interval gives a range of values compatible with the observed estimate under the statistical method used. It conveys uncertainty around the estimated mean difference, rather than describing the spread of individual participants’ measurements. For example, the CPR-depth interval for the comparison with feedback during training but not testing runs from -3.327 to 0.327. Because it includes zero, it does not rule out no average difference under this unadjusted analysis. The interval for no feedback runs from -10.39 to -6.411 and does not include zero. Neither interval addresses possible bias from outcome timing, missing data, or analysis choices not described in the record.

A p-value measures how unusual data at least as inconsistent with a specified null hypothesis would be if that null hypothesis and the statistical assumptions held. It is not the probability that the null hypothesis is true, nor does it measure the size or importance of a difference. The posted p-values are unadjusted: 0.108 for feedback during training but not testing versus feedback during training and testing, and <0.001 for no feedback versus feedback during training and testing. The registry does not report multiplicity adjustment or a prespecified decision threshold for these comparisons.

Why does the outcome scale matter?

CPR depth is measured in millimeters, while ventilation is measured in number per minute. A mean difference remains in the original outcome’s units, which makes it possible to understand the direction and scale of a contrast. The standard deviation is also expressed in the outcome’s units. The posted effect-size summaries are standardized comparisons, but they do not replace the original-scale differences. Standardized quantities can help describe a difference relative to variability, while the clinical meaning of any value depends on context and on a meaningful threshold. Such a threshold is not reported in the registry record.

These summaries rely on group-level means, standard deviations, and sample counts. They do not show the full distribution of measurements or whether assumptions such as an approximately suitable sampling distribution were checked. The record describes the trial as randomized but does not report the analysis population, the handling of missing measurements, or a statistical model. Accordingly, the posted unadjusted calculations are useful for reading the numerical contrast, but they are not a substitute for a complete trial analysis.

What can and cannot be inferred from the design label?

The trial is labelled randomized, single-masked, and factorial. These are design descriptors, not enough detail to reconstruct the randomization procedure, masking, or factorial contrasts. The record does not explain the factors or report interaction analyses. The posted results give two comparisons with the feedback-during-training-and-testing group, not a full account of all possible design effects. The comparison results should therefore be interpreted as the particular pairwise contrasts reported, without assuming a model or analysis plan that is not described.

11. Confidence Intervals and Precision

Precision concerns how tightly the data estimate a group difference. For CPR depth, the contrast between feedback during training but not testing and feedback during both is -1.5 millimeters. Its 95% confidence interval extends from -3.327 to 0.327. The interval spans zero, so the estimate is compatible with a small difference in either direction as well as no difference. The p-value is 0.108. Together, these results do not establish a clear difference in the unadjusted baseline means, but they also do not demonstrate equivalence.

For no feedback compared with feedback during both training and testing, the mean difference is -8.4 millimeters, and the 95% confidence interval extends from -10.39 to -6.411. This interval is entirely below zero. It indicates a more precisely located negative contrast than the first comparison, while still reflecting sampling uncertainty around the group-level estimate. The p-value is <0.001. Confidence intervals do not account for uncertainties that are absent from the reported analysis, such as how missing outcomes were handled or whether comparisons were adjusted for multiple testing.

The standard deviations describe variation within the groups; they are not confidence intervals. The standard deviations for CPR depth are 5.8, 6.6, and 7.6 millimeters across the three groups. The reported confidence intervals instead describe uncertainty around the mean differences. Neither type of summary reveals the full distribution of individual performance.

Most importantly, statistical precision cannot repair a mismatch between the recorded timeframe and the research question. The endpoint is labelled baseline, whereas the study aims describe immediate and retention testing. The intervals quantify uncertainty in the posted baseline contrasts; they do not quantify uncertainty in an unreported post-training or retention effect. No confidence intervals for later performance are reported in the registry record.

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 figures are participants per arm, not total events. They are illustrative calculations based on the observed effect, not the trial’s original design assumptions. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set at the stated fraction of the observed effect. The trial’s own original sample-size assumptions are not reported in the registry record.

The 80% power column gives the planned number of participants per arm for a design intended to have an 80% chance of detecting the specified effect under those assumptions, if that effect and the statistical model are correct. The 90% power column describes the corresponding planning target for a 90% chance. These are prospective planning illustrations, not observed power, and they do not tell whether the completed trial did or did not detect an effect.

Rows planning for 100%, 75%, and 50% of the observed effect illustrate the impact of choosing a smaller target difference. Smaller effects are harder to distinguish from random variation, so they require more participants per arm for the same nominal power. For example, the CPR-depth comparison between feedback during training but not testing and feedback during both has planning values of 269 participants per arm for 80% power and 360 for 90% power at 100% of the observed effect; the smaller-effect rows require larger planning numbers. The numbers are tied to the observed estimate and should not be read as a recommendation for a future study without considering its own question, design, and assumptions.

The registry enrollment and the outcome-specific group counts describe the completed record; the planning illustrations answer a different question. No sample-size justification from the original protocol is reported.

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 = -1.500; pooled standard deviation σ = 6.208.
2. n = 2 × (1.96 + 0.842)² × 6.208² / -1.500² = 268.9, rounded up to 269 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 <- -1.5                      # observed difference in means
sigma <- 6.20835214558874                     # 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(1361919)
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 = 269 (80% power), 360 (90% power)
 75% of observed effect: n per arm = 479 (80% power), 640 (90% power)
 50% of observed effect: n per arm = 1,076 (80% power), 1,440 (90% power)
power.t.test: 269.9 
Simulated power at n = 269 per arm: 0.801 (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 record does not report a statistical analysis plan or a multiplicity adjustment. The primary outcome has two posted comparisons against the feedback-during-training-and-testing group, and the secondary outcome also has two such comparisons. Considering several comparisons can increase the chance that at least one result appears statistically unusual by chance. Unadjusted p-values do not control that family-wide risk. The record does not say whether the primary comparison was designated in advance, how secondary findings were to be interpreted, or whether any other outcomes were included in the original testing plan.

No reported analyses are listed. The numerical comparisons are unadjusted recalculations from the group summaries, rather than results of an analysis described by the study investigators. The registry does not report whether models adjusted for baseline characteristics, how the factorial design was represented statistically, or whether interaction terms were tested. It would be inappropriate to infer any of these methods from the design label alone.

Missing-data handling is also not reported in the registry record. The outcome summaries include group-specific sample counts, but the record does not explain whether all randomized participants had measurements, why the outcome counts differ from the overall enrollment, or how incomplete observations were treated. It does not identify an intention-to-treat, per-protocol, or other analysis population. The posted counts should therefore be read as the counts attached to the reported group summaries, not as proof of a particular population definition.

The record does not describe imputation, sensitivity analyses, exclusions, or rules for handling implausible measurements. It also does not report whether analyses were conducted according to randomized assignment. Without those details, the degree to which the posted estimates may be affected by missingness or analysis choices cannot be assessed. These omissions are limitations of the available registry information; they do not establish that a particular procedure was or was not used outside the record.

15. Safety Results

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

Other (non-serious) adverse events

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

The registry lists serious adverse events and other adverse events separately. For the feedback-during-training-and-testing group, the record lists 0 serious events among 100 at risk and 0 other events among 100 at risk. For feedback during training but not testing, it lists 0 serious events among 100 at risk and 0 other events among 100 at risk. For the no-feedback group, it lists 0 serious events among 98 at risk and 0 other events among 98 at risk.

These figures mean that no events were recorded in the listed categories for the listed groups. They do not establish that the true risk is zero, and the registry record does not describe the event-collection procedures or the observation period for safety reporting. The safety denominators differ from the outcome-specific participant counts, and the record does not explain why. Death counts are not reported in the registry record. No serious adverse event terms are listed.

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

The posted results describe group means and unadjusted comparisons for baseline-labelled CPR depth and ventilation. For CPR depth, the no-feedback group’s recorded mean is lower than the mean in the group receiving feedback during training and testing, while the contrast between feedback during training only and feedback during both includes zero in its confidence interval. Those are statistical descriptions of the posted measurements, not proof that one teaching approach produces better later performance.

The study’s stated aims include comparing teaching methods and assessing skills at immediate and 12 week testing. However, the primary and secondary results available in the registry are labelled baseline. No numerical results for immediate testing or retention are posted in the record. Consequently, the available figures do not show whether students improved after training, whether training effects transferred to testing without feedback, or whether skills were retained.

A statistically small p-value does not establish practical importance, and an interval that includes zero does not establish that two approaches are equivalent. The record does not report a threshold for a meaningful difference in compression depth or ventilation, so the practical importance of the observed contrasts cannot be determined from these statistics alone. The measures also do not summarize every skill named in the study description.

The randomized design is relevant to comparing assigned approaches, but interpretation still depends on the timing and analysis of outcomes. Because the posted endpoint timeframe is baseline and analysis details are missing, the estimates should not be presented as causal post-training effects. The adverse-event listings record no events in the listed categories, but they do not establish the absence of all possible harms. Overall, the registry supports a narrow reading of baseline group summaries and their uncertainty, not a broad conclusion about which training method is superior.

17. Why This Trial Matters Statistically

This study raises a useful statistical distinction between a trial’s stated educational question and the outcome timing represented in its posted results. The study description focuses on comparing methods of teaching CPR skills, including whether device feedback is available during training and testing. Yet the primary outcome is labelled baseline. Recognizing that distinction prevents a group contrast from being mistaken for evidence of learning or retention.

The three approaches also illustrate that a comparator is not always a simple active-versus-control contrast. One group has feedback in both training and testing, one has feedback in training but not testing, and one has no feedback. These arrangements address different aspects of the intervention, such as continued feedback and assessment without it. The registry label “factorial” does not provide enough detail to determine the intended factor structure or the analysis for those questions, so the posted pairwise contrasts need to be read on their own terms.

Statistically, the record makes clear why estimates, intervals, and outcome context should be considered together. The confidence interval for one primary comparison includes zero, while the interval for the other lies below zero. Neither finding alone answers whether a difference is educationally meaningful, and neither supplies missing post-training data. The unadjusted nature of the comparisons and the absence of a reported multiplicity plan further limit confirmatory interpretation.

Finally, this record illustrates why sample-size planning illustrations should not be confused with the original trial rationale or with the power of a completed study. Planning values based on observed effects are useful for exploring how target effect size affects future sample requirements, but they do not explain how this trial was designed. The registry does not report the original sample-size justification or a complete analysis plan.

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.

A key limitation is that both posted outcomes are labelled baseline even though the study description emphasizes training, immediate testing, and 12 week retention. The registry record does not report numerical results for those later timepoints. This prevents the posted outcome summaries from answering the central learning and retention questions described in the study.

The record does not provide a detailed protocol, eligibility criteria, participant characteristics, allocation procedure, masking procedures, or a complete description of the factorial structure. It does not report the analysis population, missing-data approach, covariate adjustment, or a multiplicity strategy. These omissions make it difficult to assess how the analyses relate to the original design and whether the posted contrasts were specified in advance.

The arm descriptions are incomplete in the registry record, and the reasons for the safety denominators differing from the outcome counts are not explained. The overall enrollment is 298, while the primary outcome summaries list group counts of 90, 88, and 88. The record does not explain the relationship between these figures, so no reason for the difference can be assigned. Nor does it report participant flow or exclusions.

The reported comparisons are unadjusted recalculations from group summary statistics; no investigator-reported analyses are listed. Group means and standard deviations do not reveal the full distribution of performance or the presence of unusual observations. No meaningful-difference threshold is reported for the outcomes. Finally, the safety listings report no events in the specified categories, but event ascertainment and duration are not described, and death counts are not reported in the registry record.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Three Different Cardiopulmonary Resuscitation (CPR) Training Methods — NCT01361919.
• Registry record last updated 2019-07-15; status last verified 2019-05.
• Results first posted 2019-07-15.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

NCT01361919 is a completed randomized controlled study sponsored by Unity Health Toronto, with enrollment of 298. It is labelled single-masked and factorial. The study compared feedback during training and testing, feedback during training but not testing, and no feedback during either. The listed start is 2008-09 and primary completion is 2009-09.

The primary outcome is mean CPR depth in millimeters, labelled baseline. Ventilation rate is the posted secondary outcome and is also labelled baseline. The registry record does not list reported analyses or numerical results for immediate testing or 12 week retention. The pairwise comparisons are unadjusted recalculations from the group summaries. The record does not report the original sample-size assumptions, analysis plan, missing-data methods, or multiplicity procedures.