Cervical Pessary to Prevent Preterm Birth in Singleton Pregnancies With a Sonographically Measured Short Cervix: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Cervical Pessary to Prevent Preterm Birth in Singleton Pregnancies With a Sonographically Measured Short Cervix (NCT02357394), a randomized na study. The primary outcome is Preterm Birth Before 37 Weeks.

2. Clinical Question

The trial title and summary describe the intention to learn whether adding an Arabin pessary can reduce preterm birth in singleton pregnancies with a short cervix. The posted results do not establish an effect in either direction: the groups are small, and the registry record does not report an effect estimate, confidence interval, or test of the difference. The question addressed by the record is therefore narrower than a definitive assessment of the intervention. It documents the randomized comparison and its observed event counts, but it does not provide a statistical estimate of how those counts should be compared or how uncertain any group contrast is.

3. Trial Design

The study is registered as an open-label randomized controlled trial with parallel groups. Randomization means that assignment to the pessary or standard-of-care group was intended to be determined by chance rather than chosen solely by participants or clinicians. The registry record does not report the randomization sequence, allocation concealment, stratification, or the method used to implement assignment. Masking is listed as none, so the design is described as open-label.

The study enrolled 7 participants. The posted outcome groups contain 4.0 participants assigned to the Arabin Pessary group and 3.0 assigned to Standard of Care. The primary outcome counts are 3.0 participants with preterm birth in the pessary group and 1.0 in the standard-of-care group. These are observed counts, not estimates adjusted for baseline characteristics or other factors.

The pessary group was assigned to receive the device. The comparison group was assigned standard care, described as cervical-length surveillance, vaginal progesterone, and emergency cerclage. The record does not describe permitted co-interventions in the pessary group, adherence, treatment crossovers, or whether the standard-care components were applied uniformly.

The study status is terminated. The start is recorded as 2015-05 and primary completion as 2017-06-06. The reason for termination, whether recruitment stopped as planned, and whether the planned enrollment was reached are not reported in the registry record. The sponsor is Gene Lee, MD. A sample-size justification for the trial’s own design is not reported in the registry record, and the registry does not describe an interim analysis plan.

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 Preterm Birth Before 37 Weeks, measured over Up to 37 weeks 0 days.

The primary endpoint is the number of participants who delivered before 37 weeks, with the timeframe stated as up to 37 weeks 0 days. It is a binary outcome at the participant level: each participant either had a delivery before that threshold or did not. The registry describes the result as a count of participants, rather than a time-to-event measure or a continuous measure of gestational age at delivery.

Preterm birth is used as the primary outcome because the study’s stated question concerns whether the pessary can reduce delivery before the specified gestational threshold. A threshold outcome is straightforward to interpret as an event count in each randomized group. However, it does not show how early or late the delivery occurred within the preterm period, nor does it describe outcomes among pregnancies that continued to the threshold. The registry lists preterm birth before 34 weeks as a secondary outcome, but a result for that outcome is not included in the registry record of the posted record.

For the primary endpoint, the posted counts are 3.0 participants in the pessary group and 1.0 participant in the standard-of-care group. Because no inferential analysis is reported, these counts alone do not quantify the uncertainty around a group contrast. The endpoint definition also does not specify, in the registry record, how missing delivery information or withdrawals were handled.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does randomization contribute?

General statistical explanation: Randomization assigns participants to groups by chance and is intended to reduce systematic differences in prognostic factors between groups. It supports a fair comparison on average across repeated trials, but it cannot guarantee balance in a particular small trial. In this record, allocation is listed as randomized, while sequence generation, concealment, and any stratification are not reported in the registry record. Those details matter for judging how assignment was implemented, but their absence does not justify guessing the procedure.

The design is open-label, with masking listed as none. In general, lack of masking can influence care, participant behavior, or outcome assessment when those processes are susceptible to knowledge of assignment. The primary outcome is delivery before a gestational threshold, but the registry does not explain how the outcome was adjudicated or whether assessment procedures differed by group. No assumption about the effect of open-label conduct can be made from the record alone.

How should a binary endpoint be analyzed?

General statistical explanation: A binary endpoint records whether an event occurred for each participant. Group counts can be described directly, while a comparative analysis may estimate a difference in event probability, a relative measure, or another prespecified contrast. An interval estimate is useful because it describes the range of effect sizes reasonably compatible with the data and model; it is not a guarantee that the true effect lies within that range. A hypothesis test addresses compatibility with a specified null model, but a p-value does not measure the size or clinical importance of an effect.

The registry record gives only the primary event counts and group sizes in the results at hand. It reports that a two-group analysis is not estimable or that required values are missing for the binary endpoint, and it lists no analyses. Therefore, there is no sponsor-reported comparative estimate or uncertainty interval to interpret. The descriptive counts should not be mistaken for a formal test or an adjusted treatment effect. No method for handling sparse data is reported in the registry record.

What do confidence intervals and sparse data imply?

General statistical explanation: Confidence intervals communicate estimation uncertainty. When the number of observations and events is small, an estimate can be highly sensitive to the outcome of an individual participant, and interval estimates can be wide. Standard large-sample approximations may also perform poorly with sparse binary data. Alternative methods exist, but the choice depends on the prespecified analysis plan and design; neither the planned method nor an alternative method is reported in the registry record.

These considerations are especially relevant to the small groups in this trial. Even if a simple contrast were calculated from the counts, a single participant’s outcome would have substantial influence. Such a recalculation would be an unadjusted analysis of posted summary statistics, not a replacement for a prespecified sponsor analysis. No recalculated analysis is included in the record, and the discrepancy indicator is false.

What is the role of sample-size planning?

General statistical explanation: A sample-size calculation is a planning exercise that asks how many participants would be needed under specified assumptions to detect a chosen effect with a chosen error rate and power. The record describes planning illustrations based on a fraction of the observed effect, a two-sided significance level of 0.05, equal allocation between the two groups, and a normal approximation. It also distinguishes calculations for time-to-event outcomes from calculations reported as participants per arm. These assumptions define the illustration; they do not establish the trial’s original design assumptions.

The planning illustrations are not observed power. Observed power is not a useful substitute for an effect estimate and its uncertainty after results are known. In particular, a calculation anchored to the observed effect is not evidence that the trial had a specified probability of detecting an effect before enrollment. The trial’s own sample-size justification and planned target are not reported in the registry record. Specific planning estimates are not stated in the record details described for this analysis.

What is not available for interpretation?

The registry lists no analyses, so there is no reported covariate-adjusted result, subgroup estimate, interim analysis result, or sensitivity analysis to assess. The record does not report the prespecified statistical analysis plan, the analysis population, how participants with missing outcome data were treated, or whether the analysis followed assignment regardless of treatment received. These absences limit assessment of the analytic choices and prevent reconstruction of a formal comparison from the reported counts alone.

In general, an intention-to-treat analysis compares participants according to randomized assignment and preserves the benefits of randomization, while a per-protocol analysis may exclude or reclassify participants based on treatment received or adherence. Neither analysis population is specified in the registry record. The reported group counts can be read as descriptive results for the posted groups, but they should not be labelled as an intention-to-treat or per-protocol estimate without that information.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Device: Arabin Pessary", "Standard of Care")
n      <- c(4, 3)
events <- c(3, 1)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Device: Arabin Pessary  4  3  75.0  19.4  99.4  30.1  95.4  28.4  94.7
ARM  Standard of Care  3  1  33.3  0.8  90.6  6.1  79.2  6.8  80.6

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 17 outcome measures and 16 secondary outcomes, in addition to the primary endpoint. Secondary outcomes include preterm birth before 34 weeks and neonatal morbidity, neonatal length of stay, neonatal intensive care admission and duration, ventilator support, retinopathy of prematurity requiring treatment, and birthweight < 1500 Grams. The registry record does not report a multiplicity adjustment, a hierarchy of testing, or a strategy for distinguishing confirmatory outcomes from exploratory ones. Without such information, multiple outcome results, if available, would need to be interpreted with care because examining many outcomes can increase the chance of apparently notable findings arising by chance.

Missing-data handling is not reported in the registry record. The primary outcome is described as a count of participants who delivered before the threshold, but the record does not explain whether all randomized participants had delivery information, how withdrawals were handled, or whether any outcomes were imputed. The reported group sizes and event counts do not establish that no data were missing.

The analysis population is also not reported. It is not possible to tell whether the primary results include every randomized participant according to assigned group, or whether any participants were excluded or analyzed differently. No protocol deviations, adherence information, or crossover data are described. These details matter because exclusions or reclassification after randomization can weaken the comparability that random assignment is intended to produce.

The record lists no analyses for the primary endpoint. It therefore offers no reported sensitivity analysis to show whether conclusions would change under alternative assumptions about missing data, and no adjusted or subgroup analysis to evaluate robustness. These gaps do not prove that data were missing or that analyses were performed; the corresponding information is simply not reported in the registry record.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

The registry’s adverse-event figures report 0 serious adverse events among 4 participants at risk in the Arabin Pessary group and 0 among 3 participants at risk in Standard of Care. Other adverse events are also recorded as 0 in each group. Deaths are recorded as 0 in both groups. These are event counts for the listed at-risk groups, not estimates of the probability of an event.

Zero recorded events in a small group do not establish that an event is impossible or that the intervention has no safety risks. With only 4 participants and 3 participants at risk in the respective groups, the registry figures offer limited information about uncommon outcomes. They also do not establish whether every potential event was captured or whether follow-up was complete; those details are not reported in the registry record.

The adverse-event fields should be read separately from the primary preterm-birth outcome. The primary result counts preterm births, whereas the adverse-event figures describe recorded serious and other adverse events and deaths. The registry lists no top serious-event terms. The record does not provide further detail about event definitions, observation duration, severity grading beyond the serious-event category, or causal attribution. The counts therefore describe what was posted, without supporting broader conclusions about safety.

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

The result establishes that the registry reports 3.0 participants with preterm birth before 37 weeks 0 days among 4.0 assigned to the Arabin Pessary group, and 1.0 participant among 3.0 assigned to Standard of Care. Those are the observed primary outcome counts for the enrolled sample. The registry does not report a statistical comparison, uncertainty interval, or p-value, so it does not quantify how precisely the underlying group contrast has been estimated.

The counts do not show that the pessary caused a higher or lower risk of preterm birth. Randomization supports comparative interpretation in principle, but a small sample can produce unstable results, and the record lacks the analysis details needed to assess the planned comparison. The raw count difference should not be treated as a definitive estimate of treatment effect. Nor should the absence of a reported statistical analysis be interpreted as evidence that the groups are equivalent.

The results also do not establish what happened for the listed secondary neonatal outcomes. Although those outcomes are registered, their results are not reported in the record details used for this analysis. Likewise, the adverse-event counts do not establish the absence of safety risks. They document zero recorded events in the listed categories for the small groups at risk.

The study is marked terminated, but the reason for termination and the planned enrollment are not reported in the registry record. As a result, the status cannot be used to infer why recruitment ended, whether the study stopped because of outcomes, or whether it fulfilled its intended design. The record’s contribution is a description of a randomized comparison and the observed counts, not a conclusive judgment about efficacy, safety, or applicability to other pregnancies.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of the gap between a randomized design and an adequately precise estimate. Random assignment is a strong framework for comparing groups, but it does not remove uncertainty when only a small number of participants are enrolled. Here, the registry reports 7 participants overall, with 4.0 in the pessary group and 3.0 in the standard-of-care group. The primary event counts are therefore important to read as descriptive observations, not as a stable estimate of the effect that might be expected in a larger population.

The record also illustrates why an endpoint definition and an analysis result are different things. The primary endpoint is clearly expressed as a participant count before a specified gestational threshold. Yet no two-group inferential analysis is reported. Without an estimate and uncertainty interval, readers cannot distinguish a potentially meaningful contrast from variation that could readily arise in a small sample.

Further, the record makes the distinction between a trial’s own design and retrospective planning illustrations especially relevant. Calculations based on an observed effect can illustrate sample-size requirements under stated assumptions, but they do not reveal the original rationale for the study or the probability that the trial would detect an effect as planned. The registry does not report the original sample-size justification.

Finally, registered secondary outcomes and adverse-event counts demonstrate the breadth of questions a trial may consider, while also highlighting the need for prespecified analyses, transparent missing-data handling, and careful interpretation when many outcomes are considered. With analyses and key procedural details absent, the record is useful for understanding what was registered and what was observed, but its statistical contribution remains limited.

15. 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 limitations constrain interpretation. Enrollment was 7, with 4.0 participants in the pessary group and 3.0 in the standard-of-care group. Such a small sample means the primary counts provide little precision about group differences. No confidence interval or inferential result is reported, and the registry indicates that a two-group comparison of the binary endpoint is not estimable or has missing required values.

The study is open-label, with no masking. The registry record does not describe blinded outcome assessment or safeguards against differences in care or reporting between groups. The primary outcome is a delivery threshold, but the method for confirming and recording the endpoint is not reported. Details about eligibility criteria, including the precise short-cervix definition, are also not reported in the registry record.

The reason for termination, planned enrollment, and the trial’s own sample-size justification are not reported. Nor are recruitment flow, participant withdrawals, adherence, crossovers, or protocol deviations. These omissions prevent assessment of whether the analyzed groups fully reflect the randomized participants and whether the study reached its intended recruitment target.

Finally, the record does not report the statistical analysis plan, analysis population, handling of missing outcomes, multiplicity strategy, or results for the registered secondary outcomes. The posted adverse-event figures have small at-risk groups and lack detail about ascertainment and follow-up. Collectively, these limitations leave the outcome counts interpretable as observations in the registry, but not as a precise or complete comparative assessment.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Cervical Pessary to Prevent Preterm Birth in Singleton Pregnancies With a Sonographically Measured Short Cervix — NCT02357394.
• Registry record last updated 2019-10-31; status last verified 2019-10.
• Results first posted 2019-10-31.
• 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.

19. Record Summary

The registry record describes an open-label, randomized, parallel-group trial of an Arabin pessary versus standard of care for singleton pregnancies with a sonographically measured short cervix. The study is marked terminated. Enrollment was 7; the primary outcome was preterm birth before 37 weeks 0 days.

The posted primary counts are 3.0 participants with preterm birth among 4.0 in the pessary group and 1.0 among 3.0 in standard care. No two-group analysis is reported, and the registry indicates that values are missing or not estimable for the binary comparison. The record lists 17 outcome measures, including 16 secondary outcomes, but their results are not stated in the registry record described for this analysis.

Serious adverse events, other adverse events, and deaths are each recorded as 0 in both groups. These small-group counts are descriptive and do not establish the absence of risk. The termination reason, original sample-size rationale, analysis population, missing-data approach, and statistical analysis plan are not reported in the registry record.