The Use of Progesterone to Reduce Preterm Birth: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Use of Progesterone to Reduce Preterm Birth (NCT00830765), a randomized na study. The primary outcome is Weeks Gestation at Birth Among Patients Receiving the Active Drug..

2. Clinical Question

3. Trial Design

The registry record describes a randomized, parallel-group trial with quadruple masking. It lists 144 participants as enrolled: 71 in the progesterone group and 73 in the placebo group. The trial status is completed. Its recorded start was 2004-06, and its primary completion was 2010-01. The sponsor was University of Mississippi Medical Center.

Participants assigned to the active group received weekly injections of 100mg of OHP17 from enrollment until 34 weeks' gestation or delivery, whichever occurred first. Those assigned to placebo followed the same stated injection schedule. The record does not identify which groups of people were included in the quadruple masking, nor does it describe the randomization sequence, allocation concealment, treatment adherence, or changes in assignment.

The trial description frames the target population broadly as people at risk of preterm delivery. It discusses prior preterm birth and additional risk groups, including cerclage, multifetal pregnancy, preterm rupture of membranes, and preterm labor that had been stopped. The registry record does not give eligibility rules or participant counts for these groups. It therefore does not establish how broadly the enrolled participants represented all of the groups named in the description.

The posted primary analysis population is intention to treat in both groups. One outcome measure is listed, and no secondary outcomes are posted. The record includes group means and standard deviations for gestational age at birth, along with a mean difference and an unadjusted p-value. It does not report a detailed statistical analysis plan, an original sample-size justification, or additional adjusted analyses.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intention to treat analysis in both groups

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 Weeks Gestation at Birth Among Patients Receiving the Active Drug., measured over Through delivery, until discharge up to 40 weeks gestation.

The posted primary measure is weeks of gestation at birth. It is a continuous outcome: each participant's gestational age at birth is expressed in weeks, and the group result is summarized as a mean with a standard deviation. A mean gives the central average for the group, while the standard deviation describes how spread out individual measurements are around that average. The posted mean was 31.9 weeks in the progesterone group and 30.2 weeks in the placebo group.

Gestational age at birth is used as a direct measure of when delivery occurred. In a trial concerned with preterm birth, comparing this measure between randomized groups can describe whether the observed timing of birth differed in the trial population. It does not, by itself, explain why individual births occurred when they did, or summarize all aspects of infant or maternal health.

The outcome description in the registry record also mentions the interval to delivery or neonatal morbidity. It does not clarify whether those phrases refer to additional components of the primary outcome, alternative outcomes, or descriptive wording. The posted numeric results are for weeks of gestation at birth; separate results for interval to delivery or neonatal morbidity are not reported in the registry record. The primary outcome timeframe is described as through delivery, until discharge up to 40 weeks gestation.

7. Results Posted in the Registry

• Progesterone Group: mean 31.90 versus 30.20 on control; difference 1.70 (95% CI 0.62 to 2.78), standardized effect d = 0.51; z-test p = 0.002.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous because it records gestational age in weeks. A common way to summarize a continuous outcome in each group is the mean and standard deviation. The mean is the arithmetic average of the observed measurements. The standard deviation describes how much those measurements tend to vary around their group mean. In the posted results, the progesterone group has a mean of 31.9 weeks and a standard deviation of 3.5 weeks; the placebo group has a mean of 30.2 weeks and a standard deviation of 3.1 weeks. A standard deviation is not a confidence interval and does not indicate how precisely the mean difference has been estimated.

The reported effect measure is a difference in means. Its direction depends on the order of subtraction: the registry comparison uses the progesterone group relative to placebo, yielding a reported mean difference of 1.7 weeks. A positive value in this direction means the posted mean is higher in the progesterone group. The estimate describes the difference between the group averages, not a change measured within each participant.

A confidence interval expresses uncertainty around an estimated effect. In general, the interval procedure produces ranges that would cover the target parameter at a stated frequency across repeated applications under its assumptions. The posted interval for the mean difference is 0.619 to 2.781 weeks. Its confidence level is not reported in the registry record, so it should not be assumed from the endpoint values alone. The interval is useful alongside the point estimate because it conveys a range of effect sizes compatible with the calculation, rather than presenting a single value as exact.

The p-value is reported as an unadjusted p-value of 0.002. In general, a p-value evaluates how compatible the observed data are with a specified null model, under the statistical assumptions used. It is not an effect size, a measure of precision, or the probability that a treatment has no effect. “Unadjusted” indicates that the reported value is not presented as corrected for multiple testing. The registry record does not state whether other tests were performed but not posted or whether a multiplicity strategy was specified.

For continuous outcomes, standard methods often rely on assumptions about the sampling distribution of the group summaries and the variability of observations. In general, unusual distributions, influential observations, or differences in spread can affect how well a method describes uncertainty. The registry record does not identify the specific inferential procedure used for the posted comparison, report diagnostics, or describe any transformation or robust analysis. The posted summary statistics alone cannot establish whether those assumptions were checked.

General statistical considerations also distinguish a group-level average from an individual outcome. Even when an estimated mean difference is precise, there can still be substantial variation among participants. Here, the group standard deviations describe that spread, while the interval for the mean difference describes uncertainty in the comparison. These quantities answer different questions and should not be substituted for one another.

The record states that the primary analysis population was intention to treat in both groups. In general, an intention-to-treat analysis aims to compare participants according to their randomized assignment, regardless of treatment received after assignment. The registry record does not explain its operational implementation: it does not state how participants without an outcome measurement were handled, whether all randomized participants were included, or how deviations from assigned treatment were treated.

The posted p-value and interval are not identified as adjusted for covariates or for the different risk groups described in the trial summary. Whether the comparison used a particular adjustment method is not reported in the registry record. This matters because an unadjusted group comparison and a covariate-adjusted analysis can answer related but not identical questions and can yield different estimates or uncertainty. No adjusted result should be inferred from the posted summaries.

Over-dispersion is a general concern for count outcomes when variability exceeds what a simple count model expects; it can make standard errors too small if ignored. The primary posted outcome is a continuous measure summarized by means and standard deviations, not a count outcome. The record does not report count-based primary analyses, so over-dispersion is not an identified feature of the posted primary comparison.

Finally, statistical significance and clinical importance are distinct concepts. A p-value summarizes compatibility with a statistical model, whereas the estimated difference and its interval describe the size and uncertainty of the contrast. Judging the importance of gestational age differences would require clinical context and relevant outcome information beyond the numerical comparison. The record's mention of neonatal morbidity does not include corresponding posted results.

10. Confidence Intervals and Precision

The mean difference is reported as 1.7 weeks, with a confidence interval from 0.619 to 2.781 weeks. The estimate is therefore accompanied by a range rather than being known with exactness. The interval lies on the positive side of zero, which is the no-difference value for a contrast between means. The confidence level associated with this interval is not reported in the registry record, so the range should be interpreted without assigning it an unreported confidence percentage.

The interval gives information about uncertainty in the estimated group contrast; it does not describe the range of gestational ages experienced by individual participants. The standard deviations, 3.5 weeks in the progesterone group and 3.1 weeks in the placebo group, describe the spread of participant measurements within each group. Individual gestational ages can vary even when the estimated group average is relatively well characterized.

The posted means are based on 71 participants in the progesterone group and 73 in the placebo group. The registry record does not report missing outcome counts, the number included in the comparison, or details of the interval calculation. It also does not report an adjusted confidence interval or subgroup-specific precision. These absences make it harder to evaluate whether the stated uncertainty accounts for all features of the trial analysis.

The unadjusted p-value is 0.002, but that value does not replace the interval as a description of effect size and precision. Nor does either statistic resolve uncertainty about how well the enrolled participants represent the risk groups described in the trial summary. The interval provides a range for the posted comparison under its analysis assumptions, not a guarantee about the effect in other populations.

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 sample-size estimates are planning illustrations based on fractions of the observed mean difference, not the original trial's design calculations. The displayed participant counts are per arm, not totals across both arms and not total event counts. For planning at 100% of the observed effect, the estimates are 60 participants per arm for 80% power and 80 participants per arm for 90% power. For planning at 75% of the observed effect, they are 106 participants per arm for 80% power and 142 participants per arm for 90% power. For planning at 50% of the observed effect, they are 238 participants per arm for 80% power and 318 participants per arm for 90% power.

Power is the probability that a study would meet its stated statistical significance criterion if the effect size used for planning were the true effect and the planning assumptions held. The 80% power column gives the estimated participants per arm for that planning target; the 90% power column gives the estimated participants per arm for the higher target. Higher power generally requires more participants when other assumptions are held constant.

The listed planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect. The smaller the assumed effect, the more participants are needed to distinguish it from no difference under these assumptions. These figures are not observed power. The trial's own original design assumptions and sample-size justification are not reported in the registry record, and the illustrations should not be read as evidence about how the trial was originally planned.

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 = 1.700; pooled standard deviation σ = 3.303.
2. n = 2 × (1.96 + 0.842)² × 3.303² / 1.700² = 59.3, rounded up to 60 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 <- 1.7                      # observed difference in means
sigma <- 3.30324211203239                     # 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(830765)
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 = 60 (80% power), 80 (90% power)
 75% of observed effect: n per arm = 106 (80% power), 142 (90% power)
 50% of observed effect: n per arm = 238 (80% power), 318 (90% power)
power.t.test: 60.2 
Simulated power at n = 60 per arm: 0.796 (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 record lists one outcome measure and no posted secondary outcomes. It includes an unadjusted p-value for the mean difference, but does not report a multiplicity plan, the number of analyses conducted, or whether additional outcomes or comparisons were considered during the trial. Without that information, it is not possible to determine whether the reported test was the only planned test or how selection among possible analyses was handled.

The primary analysis population is described as intention to treat in both groups. In general, this approach aims to retain the comparison created by random assignment. However, the registry record does not explain how the analysis dealt with participants who did not receive assigned treatment, withdrew, or lacked a recorded gestational age at birth. The number and reasons for missing primary outcome measurements are not reported in the registry record. No imputation method, complete-case approach, or sensitivity analysis is described.

The record does not provide a participant flow account or specify whether the outcome summaries include all randomized participants. Although the group counts are posted as 71 and 73, those counts alone do not establish how missing observations were handled or whether they are the exact analysis denominators. Details about protocol deviations, exclusions after randomization, and adherence are also not reported.

The trial description includes multiple categories of risk for preterm delivery, but no subgroup results or subgroup analysis plan are posted. The record does not state whether analyses were adjusted for baseline characteristics or stratified by risk group. Consequently, the overall estimate cannot establish whether the comparison was consistent across those groups. These reporting gaps constrain assessment of multiplicity, missing-data bias, and the precise implementation of the intention-to-treat analysis.

14. 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 serious adverse event figures list 0 affected participants among 73 at risk in the placebo group and 0 among 71 at risk in the progesterone group. These entries mean that no serious adverse events are recorded as affecting participants in either group in the posted data. They do not establish that no events occurred outside the record's reporting scope, or that the groups had identical safety profiles in all respects.

For non-serious adverse events, the registry record lists 0 participants at risk and 0 affected in each group. With no reported at-risk denominator, these entries do not provide an interpretable comparison of non-serious adverse events. Death counts are not reported in the registry record. No serious adverse event terms are listed.

Safety interpretation is limited by the absence of event descriptions, ascertainment details, follow-up completeness, and information on how adverse events were collected. The posted figures should be read as registry reporting, not as proof that a treatment has no adverse effects.

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

The posted results describe a higher mean gestational age at birth in the progesterone group than in the placebo group. The mean difference is 1.7 weeks, with a confidence interval from 0.619 to 2.781 weeks and an unadjusted p-value of 0.002. The comparison concerns the trial participants and the posted analysis of the continuous gestational-age outcome.

This result does not mean that every participant in the progesterone group delivered later than every participant in the placebo group. The standard deviations, 3.5 weeks and 3.1 weeks, indicate variation in gestational age within both groups. A difference between averages cannot describe each participant's experience or determine the outcome for a particular pregnancy.

The result also does not settle the broader claims raised in the trial summary. Separate results for interval to delivery and neonatal morbidity are not reported in the registry record, and outcomes by risk group are not posted. Therefore, the overall mean comparison does not show whether results differed for participants with a prior preterm birth, cerclage, multifetal gestation, preterm rupture of membranes, or preterm labor stopped during the current pregnancy.

The registry record does not report the original sample-size justification, the detailed analysis plan, missing outcome handling, or a confidence level for the interval. Those gaps limit assessment of the design and analysis. The serious adverse event entries show no affected participants recorded in either group, but non-serious event denominators and death counts are not available in an interpretable form. These results are a statistical account of the posted trial record; they do not on their own establish treatment recommendations or capture every outcome relevant to participants.

16. Why This Trial Matters Statistically

This trial illustrates how a randomized comparison can be reported through a continuous clinical outcome. Gestational age at birth is summarized with a mean and standard deviation in each group, while the mean difference expresses the contrast between groups. Keeping these measures distinct helps readers separate the average outcome, participant-level variability, and uncertainty about the comparison.

The reported estimate is accompanied by a confidence interval and an unadjusted p-value. Reading all three together is more informative than treating a significance threshold as the whole result: the estimate gives the direction and size of the contrast, the interval conveys uncertainty, and the p-value describes compatibility with a specified statistical null model under assumptions. None alone addresses clinical importance.

The record also demonstrates why outcome definition and reporting detail matter. The primary outcome title refers to weeks of gestation at birth, while its description also mentions interval to delivery or neonatal morbidity. Only gestational-age summaries are posted. The distinction between an outcome's label, its operational definition, and the results available for analysis is important when interpreting trial evidence.

Finally, the stated intention-to-treat population and the randomized, masked design are useful context, but they do not replace reporting on participant flow, missing data, analysis methods, or subgroup results. Without those details, readers cannot fully assess how the statistical comparison was implemented or whether it applies consistently across the varied risk groups in the trial description.

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.

The registry record leaves important trial-specific details unresolved. It does not report the eligibility criteria or how many enrolled participants belonged to each of the risk groups discussed in the summary. As a result, the composition of the trial population and the relevance of its overall estimate to each risk group cannot be assessed from the posted record.

The primary outcome is described using more than one concept: gestational age at birth, interval to delivery, and neonatal morbidity. Numeric results are posted for gestational age at birth only, and the record does not clarify whether the other terms are components, alternatives, or background description. No separate results for neonatal morbidity or interval to delivery are reported.

The record does not give a detailed statistical analysis plan, the original sample-size justification, the randomization method, or details of masking implementation. It also does not describe outcome missingness, participant flow, treatment adherence, protocol deviations, or sensitivity analyses. These omissions prevent a full appraisal of how the posted intention-to-treat comparison was conducted.

Only one outcome measure is listed, and no secondary outcomes or subgroup estimates are posted. The serious adverse event entries report zero affected in each group, but the record lacks event details and gives no evaluable non-serious adverse event denominators; death counts are not reported. The posted evidence therefore covers a limited set of outcomes and does not provide a comprehensive account of safety or the trial's broader clinical outcomes.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: The Use of Progesterone to Reduce Preterm Birth — NCT00830765.
• Registry record last updated 2013-02-20; status last verified 2013-01.
• Results first posted 2013-02-20.
• 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 record describes a completed randomized, quadruple-masked, parallel-group trial sponsored by University of Mississippi Medical Center. It lists 144 participants enrolled, with 71 assigned to weekly progesterone injections and 73 to weekly placebo injections. Treatment was scheduled from enrollment until 34 weeks' gestation or delivery, whichever occurred first.

The posted primary outcome is mean weeks of gestation at birth. The group means were 31.9 weeks for progesterone and 30.2 weeks for placebo, with a reported mean difference of 1.7 weeks, a confidence interval from 0.619 to 2.781 weeks, and an unadjusted p-value of 0.002. The registry specifies an intention-to-treat analysis but does not detail the analysis implementation or missing-data handling.

One outcome measure is listed and no secondary outcomes are posted. Serious adverse events are recorded as affecting 0 participants in either group. Non-serious event denominators and death counts are not reported in an interpretable way. The registry record does not provide the original sample-size justification or a full statistical analysis plan.