Nitrous Oxide for External Cephalic Version: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Nitrous Oxide for External Cephalic Version (NCT03502915), a randomized phase 3 study. The primary outcome is Mean Pain Score Experienced During Version.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group Phase 3 trial with quadruple masking, sponsored by the University of North Carolina, Chapel Hill. The registry lists the study as completed. Its recorded start date is 2017-01-30, its primary completion date is 2019-02-14, and enrollment was 48. The registry record does not specify the roles covered by quadruple masking or provide further details of the randomization process.

The two arms were nitrous oxide and oxygen. Participants assigned to nitrous oxide received it during the version procedure; participants assigned to the comparator received placebo described as 100% oxygen. For the posted primary outcome, each arm had 24 participants. The registry’s primary endpoint was mean pain experienced during version, assessed over each procedure, described as lasting up to approximately 30 minutes on average. If there was more than one attempt, scores were averaged to form a single score for the procedure.

The registry lists four secondary outcomes: anxiety during version, post-procedure pain, post-procedure patient satisfaction, and provider-assessed difficulty. Results are posted for anxiety, post-procedure pain, and satisfaction. Provider-assessed difficulty is named as an outcome, but its results are not reported in the registry record. The registry also reports adverse-event counts for each arm.

Information about eligibility criteria, allocation concealment, the exact masking procedures, the planned analysis population, and the original sample-size rationale is not reported in the registry record. The number of participants enrolled describes the trial’s recorded enrollment; it does not by itself establish how many participants contributed to every outcome or analysis.

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

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 Mean Pain Score Experienced During Version, measured over During each version procedure, a total average of up to approximately 30 minutes.

The primary endpoint was mean pain experienced during each external cephalic version procedure. Pain scores were collected after each attempt using an 11-point scale running from 0, no pain at all, to 10, worst pain imaginable. Higher values indicate more pain and lower values indicate less pain. When a participant had more than one attempt, the registry says the attempt scores were averaged to produce one score for the procedure. The timeframe is during each procedure, described as a total average of up to approximately 30 minutes.

A mean is a measure of the average score across participants in an arm. The standard deviation describes how spread out individual scores are around that average. These measures help distinguish the average reported experience from variation between participants. The endpoint was used to compare pain associated with the two assigned gases during the version procedure, rather than pain at a later time.

The registry reports a mean of 5.486 with a standard deviation of 2.307 for the nitrous oxide arm and a mean of 5.433 with a standard deviation of 2.695 for the oxygen arm. Those values describe the posted outcome summaries; they do not by themselves establish whether the groups differ beyond sampling uncertainty. The registry’s own comparison and its confidence interval are relevant to that interpretation. Further details about endpoint handling, including missing or unusable scores, are not reported in the registry record.

7. Results Posted in the Registry

• Oxygen: mean 5.43 versus 5.49 on control; difference -0.05 (95% CI -1.47 to 1.37), standardized effect d = -0.02; z-test p = 0.942.

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.

Mean Anxiety Score Experienced During Version

• 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:
  • Oxygen: mean 4.25 versus 4.70 on control; difference -0.44 (95% CI -2.07 to 1.18), standardized effect d = -0.15; z-test p = 0.593.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Oxygen and the control. The interval gives the range of effects that remain compatible with the data.

Mean Post-procedure Pain Score

• 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:
  • Oxygen: mean 0.88 versus 1.00 on control; difference -0.12 (95% CI -0.84 to 0.60), standardized effect d = -0.09; z-test p = 0.745.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Oxygen and the control. The interval gives the range of effects that remain compatible with the data.

Mean Post-procedure Patient Satisfaction Score

• 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:
  • Oxygen: mean 6.92 versus 4.29 on control; difference 2.63 (95% CI 0.47 to 4.79), standardized effect d = 0.69; z-test p = 0.017.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Oxygen than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcome pattern is mixed rather than consistent across domains. The registry reports comparisons for anxiety during version, pain immediately after the procedure, and patient satisfaction. The posted results for anxiety and post-procedure pain do not show clear statistical evidence of differences in the registry’s two-sided tests. Satisfaction is the outcome with a registry-reported p-value below 0.05, and its group summaries favor the oxygen arm on the posted satisfaction score. Provider-assessed difficulty is listed among the secondary outcomes, but its results are not reported in the registry record.

This pattern does not alter the primary finding about pain during version. The primary endpoint addresses a distinct time and experience from post-procedure pain, and it was designated as the primary outcome. Secondary measures can add context, but they do not replace the primary comparison or resolve its uncertainty. A difference in one secondary measure alongside less clear results for others does not establish a consistent overall effect across all outcomes.

Unadjusted secondary results carry less weight when many outcomes are assessed. Without adjustment, a prespecified hierarchy, or other multiplicity strategy, the chance of at least one apparently unusual finding can rise as more comparisons are made. Whether such procedures were planned is not reported in the registry record. The satisfaction result should therefore be treated cautiously and in the context of the full set of outcomes, rather than as a standalone confirmatory finding.

The registry also does not report whether the secondary analyses were prespecified in their posted form, how missing observations were handled, or the analysis populations. These limitations make it difficult to judge the strength of the secondary evidence. The broad reading is that the posted secondary outcomes do not form a uniform pattern, and they provide limited additional certainty about the primary question.

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

• Mean Anxiety Score Experienced During Version
• Mean Post-procedure Pain Score
• Mean Post-procedure Patient Satisfaction Score
• Mean Post-procedure Provider Assessed Level of Difficulty Score

9. Statistical Interpretation

10. Statistical Methods Explained

Continuous outcomes and group summaries. The primary outcome and the posted secondary outcomes are continuous scores. A group mean represents the average score among participants included in that group’s summary. The standard deviation describes how much individual scores vary around the mean. A mean difference compares the two group averages, but its sign depends on the order in which the groups are subtracted. A positive or negative sign should therefore be interpreted using the stated group order, rather than as an inherent indication of benefit.

Confidence intervals. A confidence interval describes the range of values compatible with the observed data and the statistical model at the stated confidence level. A wider interval indicates less precision than a narrower interval. For the primary registry analysis, the 95% confidence interval runs from -1.402 to 1.507 for the reported mean difference. Because it includes zero, the analysis does not clearly distinguish a difference from no difference at the conventional two-sided significance threshold. The interval also permits effects in either direction, so the estimate should not be read as a precise determination of the underlying group contrast.

Hypothesis tests and p-values. The registry reports two-sided t-tests for the posted analyses. In general, a two-sided test evaluates evidence against a null hypothesis of no difference in either direction. A p-value is calculated under the assumptions of the test and describes how unusual the observed result, or a more extreme result, would be if the null hypothesis were true. It is not the probability that the null hypothesis is true, nor does it measure the size or practical importance of an effect. The primary registry p-value is 0.943. That result is consistent with substantial uncertainty about the group contrast; it does not prove that the groups are identical.

Registry analysis and simple recalculation. The registry posts its own two-sided t-test analyses. The separate unadjusted recalculations use posted group summary statistics to estimate group contrasts and their uncertainty; they are not a replacement for an analysis based on participant-level data. Small differences between a summary-based calculation and the registry’s reported test can arise from differences in methods or rounding. The primary recalculated contrast is expressed as oxygen minus nitrous oxide, while the registry reports its own group-order estimate. Direction and sign must be considered accordingly.

Equivalence terminology. The registry marks its analyses as equivalence analyses. In general, an equivalence analysis asks whether the plausible difference is contained within a prespecified range considered acceptably small. This requires a stated equivalence margin and an inferential approach designed for that question. The registry record does not report an equivalence margin or enough detail to establish how the equivalence designation was implemented. Consequently, the label alone cannot establish that equivalence was demonstrated. A conventional difference test that fails to find a statistically clear difference is not, by itself, an equivalence conclusion.

Randomization and masking. Random allocation is intended to reduce systematic differences between assigned groups, on average, while masking can limit the influence of expectations on care, reporting, or assessment. The registry identifies the trial as randomized and quadruple masked, with a parallel design. It does not explain which roles were masked, how allocation was concealed, or what procedures maintained masking. Those details cannot be inferred from the design labels alone.

Over-dispersion and score variability. In general, greater variation among participant scores makes a group mean less precise when the number of observations is held fixed. The posted standard deviations describe the observed spread of scores in each arm. For continuous outcomes, the t-test uses the group means, variability, and sample sizes to assess the mean contrast. In other settings, over-dispersion refers to variability greater than a statistical model expects; accounting for it matters because an overly restrictive variance assumption can make uncertainty look too small. The posted continuous-outcome analyses do not report a separate over-dispersion assessment.

Outcome timing and repeated attempts. The primary pain measure concerns the procedure itself, and the registry specifies averaging scores if there was more than one version attempt. This creates a single procedure-level score for the primary endpoint as described in the record. The registry does not report further details about the number of attempts, handling of incomplete score sets, or whether any other repeated-measure structure was included in the analysis. The secondary outcomes use their own stated timeframes, including immediate post-procedure assessments for pain and satisfaction.

Analysis populations and missingness. The registry does not describe whether the analyses followed an intention-to-treat approach, how participants were classified for analysis, or how missing outcomes were handled. These details affect interpretation because the number randomized, the number with available measurements, and the population included in a particular analysis need not be identical. The posted arm counts for the primary outcome are 24 in each group, but this alone does not explain how all randomized participants were handled across all outcomes.

Multiple outcomes. The registry lists several secondary outcomes and posts results for some of them. When many outcomes are tested, the chance of obtaining at least one apparently unusual result can increase. Adjustments or a prespecified hierarchy can address that issue, but the registry record does not report whether either was used. Secondary p-values should therefore be regarded as exploratory unless further analysis details establish otherwise.

11. Confidence Intervals and Precision

The registry-reported primary mean difference is 0.053, with a 95% confidence interval from -1.402 to 1.507. This interval is broad relative to the point estimate and includes zero. It also includes values on both sides of zero, meaning that the posted data are compatible with a range of possible directions and magnitudes for the group contrast. The estimate is therefore not precise enough to establish a narrowly defined difference from the registry summary alone.

A confidence interval is not a guarantee that the true value lies within its endpoints. In general, it is a procedure that, under its assumptions, produces intervals that cover the underlying parameter at the stated rate over repeated studies. Its practical value is that it displays the uncertainty around an estimate, rather than presenting a point estimate or p-value alone. The width depends on factors such as variability and the amount of information available.

The primary outcome summaries include 24 participants in each arm, with standard deviations of 2.307 for nitrous oxide and 2.695 for oxygen. Those spreads, alongside the arm sizes, help explain why the comparison retains uncertainty. The registry’s interval should be used when describing its own analysis. A summary-based recalculation has a different estimate and interval; it is not a substitute for the trial’s reported analysis.

The interval does not establish equivalence. To support equivalence, an acceptable range for the difference would need to be specified and the uncertainty assessed against that range. No such margin is reported in the registry record. Thus, the primary estimate is best described as showing no clear statistical difference in this trial, with an interval that leaves meaningful uncertainty about the contrast.

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 illustrations are planning calculations based on the observed effect, not the trial’s original design rationale. For these continuous outcomes, the figures are participants per arm, not total events. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The calculations consider detecting the stated fraction of the observed effect. The original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column gives the illustrated number of participants per arm for a design with an 80% chance of detecting the selected effect under the stated assumptions, if that effect were the true effect and the model assumptions held. The 90% power column gives the corresponding planning count for a 90% chance. These are prospective planning concepts, not probabilities that the trial’s observed result is correct. This is not observed power.

The rows that use 100%, 75%, and 50% of the observed effect ask what planning counts would look like if the target effect were progressively smaller than the observed estimate. Smaller target effects generally require more participants to distinguish them from random variation, all else being equal. The primary outcome’s planning figures are especially large because its observed group contrast is close to zero, so detecting a fraction of that observed contrast would require a very large trial under the listed assumptions.

The illustrations should not be treated as recommendations or as evidence about the actual design. Planning from an observed effect can be unstable, particularly when the estimate comes from a small trial, and the trial’s design may have used different assumptions. The curve is a visual representation of the planning relationship; it does not add an observed outcome or change the assumptions listed for the calculations.

13. How the Sample Size Is Derived

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

Formula

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

Worked calculation for this trial

1. Observed difference Δ = -0.053; pooled standard deviation σ = 2.509.
2. n = 2 × (1.96 + 0.842)² × 2.509² / -0.053² = 35,165.6, rounded up to 35,166 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 <- -0.0529999999999999                      # observed difference in means
sigma <- 2.50851290608599                     # 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(3502915)
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 = 35,166 (80% power), 47,077 (90% power)
 75% of observed effect: n per arm = 62,517 (80% power), 83,693 (90% power)
 50% of observed effect: n per arm = 140,663 (80% power), 188,308 (90% power)
power.t.test: 35166.6 
Simulated power at n = 35,166 per arm: 0.794 (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. Analyses Reported by the Sponsor

The registry’s own primary analysis is a two-sided t-test of the final mean values. It reports a mean difference of 0.053, a 95% confidence interval from -1.402 to 1.507, and p = 0.943. Those are the trial’s posted inferential results and should be the starting point when describing the sponsor-reported analysis.

The unadjusted recalculation uses the posted group summaries and expresses the contrast as oxygen minus nitrous oxide. It estimates -0.053, with a 95% confidence interval from -1.472 to 1.366 and p = 0.942. Its sign is reversed because the group order is reversed. The recalculation is a simple analysis of summary statistics, not a reanalysis of participant-level records. The registry does not flag a discrepancy between the calculations.

The estimates and p-values are close but not identical. The distinction matters: the sponsor’s analysis is the registry-reported result, while the recalculation is a separate unadjusted check based on posted summaries. Neither should be described as establishing equivalence, because the registry record does not state an equivalence margin. Both analyses indicate considerable uncertainty around the primary group contrast and do not provide clear evidence of a difference.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists several secondary outcomes in addition to the primary pain endpoint. Testing multiple outcomes creates a multiplicity issue: when many hypotheses are examined, at least one result may appear statistically unusual by chance even if none of the underlying effects is real. Statistical approaches such as a prespecified testing hierarchy or adjustment for multiple comparisons can control aspects of this problem. Whether any such approach was planned or used is not reported in the registry record.

The posted secondary p-values should therefore be read as unadjusted unless the registry states otherwise. In particular, a single secondary result with p = 0.025 should not be interpreted in isolation as definitive evidence. Its context includes the other outcomes, the primary result, and the absence of reported multiplicity procedures. The record does not report a prespecified order of testing or an overall confirmatory analysis plan.

Missing-data handling and analysis populations are also not reported in the registry record. It is not stated whether all randomized participants were analyzed in their assigned groups, whether analyses used only participants with observed outcomes, or how any missing scores were addressed. The posted primary summaries list 24 participants per arm, but this does not resolve how participants were treated across every endpoint or whether all randomized participants contributed data.

Other potentially important analysis details are not reported, including a detailed statistical analysis plan, the original sample-size justification, interim analyses, and the definition of the population used for each endpoint. The registry identifies the design as randomized and quadruple masked, but does not give the operational details of randomization or masking. These omissions limit independent assessment of whether outcome selection, analysis choices, or missing observations could influence the posted results. They do not prove that a particular bias occurred; they leave those questions unresolved in the registry record.

16. 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.

Most frequent non-serious adverse event terms

The registry reports zero serious adverse events in each arm among 24 participants at risk per arm. It also reports five other adverse events in the nitrous oxide arm and three in the oxygen arm, with 24 at risk in each arm. Deaths are reported as zero in both groups. The registry does not list top serious adverse-event terms.

These are counts from a small trial and should be read as descriptive safety information, not as precise estimates of comparative risk. A count of zero serious events in an arm means none were recorded in that arm in the registry results; it does not establish that serious events cannot occur. The record does not provide further definitions, event-level descriptions, severity details for the other events, or an explanation of how events were collected. It also does not state whether participants could contribute more than one other event, so the counts should not be interpreted as counts of unique participants.

The reported adverse-event figures do not establish a safety advantage for either arm. They describe the recorded events and participants at risk in this trial, with limited information about event types and ascertainment.

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

The main result is a comparison of average pain scores during external cephalic version among participants assigned to nitrous oxide or oxygen. The registry reports means of 5.486 and 5.433, respectively, and a mean difference of 0.053 with a 95% confidence interval from -1.402 to 1.507. The p-value is 0.943. The posted analysis does not provide clear statistical evidence of a difference in the primary outcome, but the confidence interval leaves uncertainty about the size and direction of the group contrast.

This does not mean that the two groups have been shown to be identical or equivalent. Equivalence requires a prespecified margin defining what range of differences would count as acceptably small, alongside an appropriate analysis. The margin and supporting details are not reported in the registry record. A nonsignificant test is not proof of no effect, and the p-value is not the probability that the null hypothesis is true.

The secondary outcomes give additional but mixed information. The registry-reported analyses do not show clear differences for anxiety during version or post-procedure pain, while satisfaction has a different pattern. Multiplicity handling is not reported, and results for provider-assessed difficulty are not reported in the registry record. The secondary findings therefore do not turn the primary result into a definitive conclusion about all aspects of the procedure.

The trial enrolled 48 participants, and the primary outcome summaries include 24 per arm. A trial of this size may leave substantial uncertainty for outcomes with considerable variation. The registry record also lacks information about missing-data handling, analysis populations, and the original sample-size rationale. The results should be understood as the reported evidence from this trial, not as a guarantee that the same pattern applies to all patients or settings. They do not, by themselves, establish a clinical recommendation.

18. Why This Trial Matters Statistically

This trial illustrates why a point estimate, a p-value, and a confidence interval answer different questions. The primary group means are close, and the registry-reported p-value does not indicate a statistically clear difference. The confidence interval adds essential context by showing that a range of contrasts remains compatible with the data. Interpreting only the p-value could obscure the uncertainty about the magnitude and direction of the difference.

It also demonstrates the distinction between failure to detect a difference and evidence of equivalence. The registry labels the posted analyses as equivalence analyses, but does not report an equivalence margin. Without that margin, readers cannot determine whether the full uncertainty interval lies within a clinically acceptable range. This is an important distinction in trial interpretation: a result that is not statistically significant is not automatically evidence that two interventions are alike.

The secondary results illustrate the importance of multiplicity and outcome hierarchy. Several outcomes were assessed, and one posted secondary comparison has a smaller p-value than the others. Without reported multiplicity procedures or a detailed analysis plan, that result needs cautious interpretation. The overall pattern matters more than selecting a single result after the fact.

Finally, the planning illustrations show how sample-size calculations depend on the effect size chosen for detection. A very small observed contrast leads to large planning counts when that contrast is used as the target. Such calculations are not evidence that the trial was adequately powered or underpowered according to its original plan, because the original design assumptions are not reported. The example reinforces the need to separate observed results, uncertainty, and prospective planning.

19. Limitations

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

Several limitations specific to this trial constrain interpretation. Enrollment was 48, with 24 participants in each arm for the posted primary outcome. This leaves limited information for estimating group differences, particularly when individual pain scores vary. The primary confidence interval is broad and includes contrasts in both directions. The registry result should not be treated as a precise estimate of a small effect.

The registry does not report the original sample-size justification or design assumptions. Consequently, the planning illustrations based on the observed effect cannot establish whether the study met its intended power or whether its enrollment target was chosen to answer the primary question. The record also does not describe interim analyses, if any.

The primary outcome is an average of scores across attempts when participants had more than one attempt. The registry does not provide the distribution of attempts, details about score completeness, or additional information about how the averaging rule was applied in the analysis. It also does not report the analysis population or the handling of missing outcomes. These omissions make it difficult to assess whether the posted summaries represent all randomized participants.

The equivalence label is not accompanied by a reported margin, limiting interpretation of the analysis as an equivalence assessment. The registry also does not report multiplicity handling for the secondary outcomes. The posted secondary results cover anxiety, post-procedure pain, and satisfaction, but results for provider-assessed difficulty are not reported in the registry record.

Finally, the record does not explain the mechanics of randomization, allocation concealment, or which roles were masked. The trial is identified as randomized and quadruple masked, but the labels alone do not allow assessment of implementation. Further eligibility, setting, and participant-characteristic details are not included in the record summarized for these outcomes, limiting assessment of generalizability.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Nitrous Oxide for External Cephalic Version — NCT03502915.
• Registry record last updated 2019-10-30; status last verified 2019-03.
• Results first posted 2019-10-30.
• 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.

23. Record Summary

NCT03502915 is a completed randomized, parallel-group Phase 3 trial of nitrous oxide versus oxygen during external cephalic version. It was sponsored by the University of North Carolina, Chapel Hill, and recorded enrollment was 48. The registry identifies the study as quadruple masked.

The primary endpoint was mean pain experienced during version. The registry reports means of 5.486 for nitrous oxide and 5.433 for oxygen, with a sponsor-reported mean difference of 0.053, a 95% confidence interval from -1.402 to 1.507, and p = 0.943. This does not show clear statistical evidence of a difference, but it also does not establish equivalence.

Secondary results are posted for anxiety during version, post-procedure pain, and satisfaction. Provider-assessed difficulty is listed but its results are not reported in the registry record. The record also reports no serious adverse events or deaths in either arm, and counts of other adverse events. Important details—including missing-data handling, analysis populations, multiplicity procedures, and the equivalence margin—are not reported in the registry record.