A Comparative Evaluation of the Performance and Fit of Pediatric Face Masks: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Comparative Evaluation of the Performance and Fit of Pediatric Face Masks (NCT02224612), a randomized phase 4 study. The primary outcome is Change From Baseline in Carbon Dioxide Levels.

2. Clinical Question

The study concerns a short-term comparison of mask fit and measured respiratory-related outcomes in healthy children, rather than a comparison of clinical treatment strategies. The primary endpoint is a continuous measurement in mmHg. The registry also lists physical fit as a secondary outcome. It does not report a detailed definition of physical fit or the method used to score it.

The posted result compares the mean change associated with each mask. The Medline arm’s reported mean change is -0.07 mmHg, while the KC Childs arm’s reported mean change is -0.79 mmHg. The reported mean difference for KC Childs relative to Medline is -0.72 mmHg. These figures describe the recorded endpoint during a brief mask assessment; they do not by themselves establish what the difference means for longer use or for children outside the enrolled healthy population.

3. Trial Design

The registry describes a randomized, open-label crossover study. In a crossover design, participants receive or experience more than one study condition, so comparisons can be made across conditions within the same participant. This feature can be useful when outcomes vary substantially between individuals, because each participant can contribute measurements under both conditions. The registry states that children wore both masks, with each worn for 2-3 minutes.

The two devices were the Medline Pediatric Face Mask and the KC Childs Mask. Both are described as single-use disposable masks. The record divides the arm descriptions into three age groups: children 4-5 years, children 6-9 years, and children 10-12 years. It reports total enrollment of 96. The posted outcome summaries list 96 observations for each mask, but do not provide a separate participant count for each age group.

The study is marked completed, and the phase is listed as phase 4. The registry reports a start date of 2014-01 and a primary completion date of 2014-01. Medline Industries is named as the sponsor. These details describe the registry entry; they do not supply additional information about recruitment procedures or how the crossover sessions were conducted.

Although the study is described as randomized, the sequence-generation method, allocation concealment, order of mask use, and any washout period are not reported in the registry record. Masking is listed as none. The record does not explain whether this refers to participants, assessors, or all study personnel. It also does not report the original statistical analysis plan or the assumptions used to choose the study’s enrollment target.

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: Statistical analysis projected ninety-nine (99) Subjects were needed to assess mask equivalence, however only 96 subjects were enrolled at the facility.

Overall Study

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Carbon Dioxide Levels, measured over Baseline and 2 minutes.

The primary endpoint is change from baseline in carbon dioxide levels, measured in mmHg. The registry identifies the measure as a mean and gives the assessment times as baseline and 2 minutes. It states that ETCO2 was assessed by a capnographer while each pediatric face mask was worn. The reported comparison is the difference between masks in change from baseline, rather than a comparison of the absolute carbon dioxide level at a single time.

In general, comparing change from baseline is one way to describe how a measurement shifts during an assessment relative to its starting value. Using the same endpoint for both masks allows the posted result to express the contrast in a common unit. Here the unit is mmHg, and the reported mean difference is -0.72 mmHg for KC Childs relative to Medline.

The endpoint is used as a short-term measured outcome during mask wear. The registry also describes respiration assessments and lists physical fit as a secondary outcome, but it does not give detailed measurement procedures or results for those assessments in the posted outcome information. The record does not state why this particular endpoint was selected as primary or provide a clinical threshold for interpreting a difference in mmHg.

7. Results Posted in the Registry

• KC Childs Mask: mean -0.79 versus -0.07 on control; difference -0.72 (95% CI -1.44 to 0.00), standardized effect d = -0.28; z-test p = 0.051.

8. Statistical Interpretation

9. Statistical Methods Explained

A continuous outcome can be summarized by its mean and standard deviation. The mean describes the average observed value in a group; the standard deviation describes how spread out individual observations are around that average. For this study, the endpoint is change from baseline in carbon dioxide levels, measured in mmHg. The posted arm means are -0.07 for Medline and -0.79 for KC Childs, with standard deviations of 3.56 and 0.65, respectively.

What does the mean difference describe?

The registry reports the comparison as a mean difference of -0.72 mmHg for KC Childs relative to Medline. A difference is formed by expressing one group’s mean relative to the other’s, so the sign depends on the stated comparison order. In this record, the negative estimate corresponds to a lower reported mean change for KC Childs. It is an estimate of the size and direction of the contrast, not a statement about the importance of that contrast.

The registry labels the comparison metric as a mean difference. Its design is crossover, meaning each participant was exposed to both masks according to the arm descriptions. In general, crossover analyses can use the within-participant relationship between measurements. That relationship affects the uncertainty of a paired contrast. The registry does not describe the exact model, whether the analysis used paired measurements, or how sequence and period effects were handled.

What does the confidence interval add?

A confidence interval gives a range of values compatible with the observed data under the method’s assumptions. It helps readers judge the uncertainty around a point estimate. The posted interval for the mean difference runs from -1.444 mmHg to an upper bound just above zero. Since zero lies within that interval, the result does not sharply distinguish the observed contrast from no difference under the interval method. The interval also indicates that the estimate is not precise enough to pin down a narrow range wholly on one side of zero.

The interval depends on assumptions and on the analysis used to estimate its standard error. A standard error describes the expected variability of an estimate across repeated samples under a statistical model. The registry gives the comparison estimate and interval, but does not report the full model specification or participant-level data needed to examine those assumptions. The trial’s original statistical plan is not reported in the registry record.

How should the p-value be read?

The unadjusted p-value is 0.051. In general, a p-value measures how compatible the observed data are with a specified null model, under the statistical assumptions used. It does not measure the size of the effect, the probability that the null model is true, or the likelihood that the result will repeat. Here, the reported p-value is close to 0.05, and the confidence interval also reaches across zero. The two summaries provide complementary ways to understand uncertainty around the contrast.

The label unadjusted means the reported value is not presented as adjusted for multiple comparisons. The registry lists one secondary outcome and three outcome measures, but does not provide analysis details or say how any multiplicity was handled. It is not reported in the registry record whether the primary comparison was designated as the sole confirmatory test or whether other analyses were planned.

Why can over-dispersion matter?

In general, over-dispersion refers to more variability in the observed data than a statistical model expects. For continuous outcomes, unexpectedly large spread can make standard errors too small if the analysis does not account for it, leading to intervals that are too narrow and p-values that are too optimistic. This is a general statistical consideration, not a claim that over-dispersion occurred in this trial. The posted standard deviations differ between mask summaries, but the registry does not report an over-dispersion assessment or the model used to estimate the comparison.

The crossover structure is another reason not to infer the analysis method solely from the summary statistics. Within-child correlation can increase or decrease the precision of a paired comparison relative to treating the two mask measurements as independent. The summary record does not provide the covariance, sequence details, or period-specific outcomes. Consequently, an independent reanalysis from the displayed means and standard deviations would not reproduce all design-specific information.

The record includes no posted analyses in the analysis count, while the primary outcome summaries and one secondary outcome are described. It does not report the analysis population, missing-data method, handling of protocol deviations, or whether all randomized participants contributed to the primary comparison. These are important parts of interpreting a clinical trial analysis and should not be inferred from the enrollment count alone.

10. Confidence Intervals and Precision

The point estimate for the mean difference is -0.72 mmHg, and its confidence interval extends from -1.444 mmHg to an upper limit just above zero. The interval’s inclusion of zero means that the posted data do not tightly rule out no average difference between masks. At the same time, values below zero remain compatible with the data across a range reaching to -1.444 mmHg. The interval is not a measure of the range of individual children’s responses.

Precision concerns how narrowly an estimate is determined. A narrow interval would indicate less uncertainty about the estimated mean contrast, while a wider interval would indicate more. The reported interval includes both a range of negative differences and values close to zero, so the estimate is not sharply determined in relation to zero. The unadjusted p-value of 0.051 conveys a related feature of the same comparison, but does not replace the interval or describe the magnitude of possible effects.

The registry reports 96 observations for each mask. However, the study is crossover, and the record does not give the paired correlation or the standard error method. Those details influence precision. The arm standard deviations are 3.56 mmHg for Medline and 0.65 mmHg for KC Childs, but these summaries alone do not show how individual children’s changes under the two masks were related.

The upper confidence bound is very close to zero, while the lower bound is farther from zero. This asymmetry is part of the posted interval and should be read as reported, rather than replaced by a symmetric range around the point estimate. The registry does not explain the interval construction or provide enough detail to reproduce it from the summary information.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total event counts. They are illustrations based on the observed effect, not a report of the trial’s original enrollment calculation. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning row uses a stated fraction of the observed effect as the effect to detect.

For planning at 100% of the observed effect, the figures are 199 participants per arm for 80% power and 266 participants per arm for 90% power. At 75% of the observed effect, they are 353 participants per arm for 80% power and 472 participants per arm for 90% power. At 50% of the observed effect, they are 794 participants per arm for 80% power and 1,062 participants per arm for 90% power.

In plain language, the 80% power column describes the planned number per arm under which the assumed effect would be detected by the specified test in 80% of repeated studies, if the assumptions held. The 90% column uses a higher probability of detecting that assumed effect, so it requires more participants in each listed scenario. Smaller effects are harder to distinguish from random variation and therefore require more participants than larger effects under the same planning framework.

These calculations are planning illustrations, not observed power for the completed study. The registry does not report the trial’s own original sample-size justification or design assumptions. The crossover structure and any within-participant correlation are not described in the planning assumptions, so the figures should not be treated as a reconstruction of the trial’s original design.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -0.720; pooled standard deviation σ = 2.559.
2. n = 2 × (1.96 + 0.842)² × 2.559² / -0.720² = 198.3, rounded up to 199 participants per arm for 80% power.

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

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -0.72                      # observed difference in means
sigma <- 2.55891578603126                     # 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(2224612)
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 = 199 (80% power), 266 (90% power)
 75% of observed effect: n per arm = 353 (80% power), 472 (90% power)
 50% of observed effect: n per arm = 794 (80% power), 1,062 (90% power)
power.t.test: 199.2 
Simulated power at n = 199 per arm: 0.804 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary outcome of change from baseline in carbon dioxide levels and a secondary outcome of physical fit of pediatric face masks. It also reports three outcome measures, but the detailed set of measures and any hierarchy among them are not fully described in the posted information. No analyses are listed in the analysis count. The record does not report a prespecified multiplicity strategy, such as how interpretation would account for testing more than one outcome or analysis.

Multiplicity matters because examining several outcomes or comparisons can increase the chance of at least one apparently notable result even when none reflects a genuine underlying difference. This is a general statistical issue. The reported p-value is explicitly unadjusted, and the registry does not say whether it was intended as a confirmatory test, whether other comparisons were planned, or whether any adjustment was made outside the posted record.

Missing-data handling is not reported in the registry record. The record does not state how many participants completed both mask assessments, whether any measurements were unavailable, how incomplete crossover sequences were treated, or whether imputation or another method was used. Enrollment of 96 and outcome summaries with 96 observations per mask do not, by themselves, answer these questions about completeness or analysis populations.

The analysis population is also not reported in the registry record. It is not stated whether the posted comparison includes all randomized participants, only those who completed both assessments, or another defined group. The registry does not describe exclusions after randomization, protocol deviations, or the handling of any participants who did not follow the assigned sequence. These details can affect how closely the analyzed results reflect the randomized study population.

For a crossover study, the record would ideally clarify how measurements from the same child were modeled and whether sequence or period effects were considered. Those details are not reported. As a result, the summary statistics support a description of the posted estimate, but do not allow a full assessment of analysis choices, missingness, or multiplicity.

14. 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 reports zero serious adverse events for each mask group. For the Medline Pediatric Face Mask, 0 of 96 participants at risk had a serious adverse event; for the KC Childs Face Mask, 0 of 96 participants at risk had a serious adverse event. The reported death count is zero for each group. The registry also reports zero other adverse events among 96 participants at risk in each group.

These counts describe events recorded for the study and the listed at-risk groups. A count of zero means that no events of that category were posted for the stated group; it does not establish that the possibility of such events is absent. Small or brief studies may not characterize uncommon events, and the registry does not provide further detail about adverse-event collection, definitions, or follow-up duration.

The assessments involved wearing each mask for 2-3 minutes. The record does not provide a longer safety follow-up period. The adverse-event figures should therefore be interpreted within the scope of the recorded study and not as a broad estimate of safety in other settings or with longer use.

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

The posted result describes a short-term contrast in mean change from baseline in carbon dioxide levels while children wore two pediatric face masks. The estimated mean difference is -0.72 mmHg for KC Childs relative to Medline. The confidence interval extends from -1.444 mmHg to an upper limit just above zero, and the unadjusted p-value is 0.051. Together, these statistics show an estimate that is negative but uncertain in relation to zero.

The result does not establish that one mask is better overall. The record does not report a clinical threshold that would indicate whether a difference of this size is important, nor does it provide detailed results for the physical-fit secondary outcome. The carbon dioxide measure was assessed during a brief period of mask wear, so the result does not directly describe longer-duration use, repeated use, or outcomes beyond the recorded assessment.

The population was described as healthy children, with age-group descriptions covering 4-5, 6-9, and 10-12 years. The registry does not provide separate estimates by age group or detailed participant characteristics. The overall estimate therefore cannot show whether the numerical contrast differed across those age ranges or among children with different characteristics.

The crossover design means that participants experienced both masks, but the registry does not report the sequence, period effects, or detailed paired analysis. The posted summaries cannot establish whether those design features influenced the estimate. The record also does not report missing-data methods, analysis populations, or a multiplicity plan. These limits do not invalidate the displayed numbers, but they constrain the conclusions that can be drawn from them.

In statistical terms, the result is an estimate with a confidence interval that includes values close to no difference. It is not a precise demonstration of an exact effect, and the p-value should not be treated as a probability that the masks are equivalent or different. The appropriate reading is limited to the posted short-term comparison and its reported uncertainty.

16. Why This Trial Matters Statistically

This study is statistically useful as an example of a randomized crossover comparison of two devices in a pediatric population. In a crossover design, measurements from the same participant under both conditions can help separate treatment-condition differences from stable differences between participants. That advantage depends on appropriate design and analysis details, including order and within-person association. The registry does not report those details, making the design label important but not sufficient to reconstruct the analysis.

The endpoint is continuous, so the mean difference expresses the contrast in the outcome’s original unit, mmHg. This makes the estimate directly interpretable as a difference in average change, while the confidence interval conveys its uncertainty. The posted interval spans zero and the unadjusted p-value is 0.051, illustrating why a point estimate alone should not be treated as a complete summary of evidence.

The arm standard deviations are notably different in the posted summaries: 3.56 mmHg for Medline and 0.65 mmHg for KC Childs. Differences in spread can matter when choosing and evaluating statistical models. In this crossover setting, the paired relationship between measurements would also influence uncertainty. The registry does not provide enough detail to determine how the reported comparison accounts for these features.

The record also illustrates the distinction between enrollment and analysis planning. The trial enrolled 96 participants, while planning illustrations based on the observed effect show participant counts per arm under stated assumptions. Those illustrations are not the original design justification and should not be read as observed power. Separating design assumptions, observed estimates, and precision is central to sound interpretation.

17. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

Several trial-specific details needed to assess the result are not reported in the registry record. These include the sequence-generation method, allocation concealment, mask-use order, any washout period, and the handling of period or sequence effects. The record identifies the study as randomized and crossover, but these additional details would help assess whether the comparison and its uncertainty were analyzed in a way that reflects the design.

The primary endpoint is described as a difference between masks in change from baseline, with measurement at baseline and 2 minutes. The record does not provide a detailed protocol for measurement timing, calibration, or the handling of repeated readings. It also does not provide a clinical threshold for interpreting changes in mmHg. This limits assessment of practical importance as distinct from statistical uncertainty.

The record does not report participant characteristics beyond healthy status and age-group labels, or give outcome estimates within those age groups. It does not explain the substantial difference between the posted arm standard deviations. Participant-level data and the within-child correlation are not available in the registry entry, so the comparison cannot be independently reconstructed from the displayed summaries alone.

Missing-data handling, analysis populations, protocol deviations, multiplicity procedures, and the original sample-size justification are not reported in the registry record. The physical-fit outcome is listed as secondary, but detailed results and scoring procedures are not reported in the posted information. Finally, the assessments were brief, and the record does not give longer follow-up. These limitations restrict interpretation to the outcomes and assessment conditions described in the registry.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: A Comparative Evaluation of the Performance and Fit of Pediatric Face Masks — NCT02224612.
• Registry record last updated 2019-04-23; status last verified 2017-08.
• Results first posted 2019-04-23.
• 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.

21. Record Summary

The registry identifies NCT02224612 as a completed, randomized, open-label crossover study comparing Medline Pediatric Face Masks with KC Childs Mask in healthy children. Enrollment was 96, and the age-group descriptions are 4-5 years, 6-9 years, and 10-12 years. Both masks were worn for 2-3 minutes for carbon dioxide, respiration, and physical-fit assessments.

The primary outcome was change from baseline in carbon dioxide levels, measured in mmHg at baseline and 2 minutes. The posted mean difference for KC Childs relative to Medline is -0.72 mmHg, with a confidence interval extending from -1.444 mmHg to an upper bound just above zero and an unadjusted p-value of 0.051. The registry lists physical fit as a secondary outcome and reports no serious adverse events or deaths in either group.

Medline Industries is named as sponsor. The listed start and primary completion dates are both 2014-01. Details about the original design assumptions, analysis population, missing-data methods, multiplicity, and full crossover analysis are not reported in the registry record.