Hormonal Contraception and Vaginal Health: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Hormonal Contraception and Vaginal Health (NCT00612508), a randomized na study. The primary outcome is Thickness of the Vaginal Epithelium (in mm)With Means and Standard Deviations Reported..

2. Clinical Question

3. Trial Design

The registry record describes a randomized, parallel-group study with no masking. The study is marked completed, and its phase is listed as not applicable. Enrollment is recorded as 14. The stated study purpose was to examine whether the route of hormonal contraception is associated with different changes in the vaginal lining.

The two interventions were oral Desogen and intravaginal NuvaRing. The Desogen description specifies a daily tablet containing desogestrel and ethinyl estradiol, with inactive pills every 28 days. The NuvaRing description specifies ethinyl estradiol and etonogestrel, with the ring used for 3 weeks and removed for one week before a new ring is used. Both arm descriptions refer to baseline vaginal biopsy and follow-up biopsies after 3 and 6 cycles.

The registry record says 7 R and 7 P subjects were randomized and had an initial biopsy. It does not explain the R and P abbreviations in that population description. At the 84-day assessment, the record states that 11 subjects returned, with 6 R and 5 P. For the 168-day assessment, it states that 6 subjects returned, with 3 R and 4 P. Those arm counts do not agree with the stated total at that assessment; the record does not resolve the inconsistency.

The study dates are recorded as starting in 2007-05 and reaching primary completion in 2008-05. The registry record does not report further details about recruitment procedures, allocation concealment, or masking beyond identifying the study as randomized and unmasked.

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: A total of 14 subjects (7 R, 7 P) were randomized and had an initial biopsy; 11 (6 R, 5 P) returned for a biopsy at 3rd cycle (84 days), and 6 (3 R, 4 P) 6th cycle (168 days). The analysis used a paired T test comparing the baseline mean to the mean as the end of the 3rd cylce (e.g. 84 days) and the 6th cycle (168 days)

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 Thickness of the Vaginal Epithelium (in mm)With Means and Standard Deviations Reported., measured over baseline, 84 days, 168 days.

The primary endpoint is thickness of the vaginal epithelium, measured in millimeters. The registry description says vaginal tissue sections were evaluated histologically and their absolute thickness recorded. A biopsy-based measure gives a tissue-level outcome that can be compared across the assessment times specified in the record.

The registry lists measurements at baseline, 84 days, and 168 days. It describes comparisons between baseline and the assessment after 3 cycles, and between baseline and the assessment after 6 cycles. These are within-person comparisons when paired biopsy measurements are available for the same participant. Such comparisons focus on change from an individual’s own baseline rather than treating observations from different people as if they were paired.

The endpoint is reported as a mean with standard deviation. A mean describes the average thickness in the analyzed observations, while a standard deviation describes the spread of individual measurements around that average. Neither measure alone establishes the size or precision of a treatment contrast. The registry record does not provide the mean values, standard deviations, estimated changes, confidence intervals, or test results for this endpoint. It therefore does not allow the magnitude or precision of the observed thickness changes to be assessed.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: epithelial thickness is recorded as a measurement in millimeters. The registry says that results are expressed as means and standard deviations. A mean gives a central summary of observed measurements. A standard deviation describes their variability, but it does not describe the precision of the estimated mean by itself. Precision depends on the amount and structure of the available data, as well as the variability.

Why use paired comparisons?

The record describes paired t tests comparing baseline with follow-up. Pairing is appropriate when two measurements belong to the same participant. Instead of analyzing the baseline and follow-up values as unrelated groups, a paired analysis evaluates each participant’s within-person difference. This can account for stable differences among participants and can make the comparison more informative when repeated measurements are correlated.

A paired analysis requires usable measurements at both time points for a participant. Someone with a baseline biopsy but no biopsy at a particular follow-up cannot contribute a complete pair to that specific comparison under a conventional paired t test. The registry gives counts of participants who returned for biopsies, but does not report how many complete pairs entered each test, the amount of missing measurement data, or the handling of incomplete pairs.

What does the t test assess?

In general, a paired t test evaluates whether the average within-participant difference is distinguishable from zero under the assumptions of the test. Its estimate is the mean of the paired differences. A p-value, if reported, describes how compatible the observed test statistic is with a specified null hypothesis under the model; it is not a measure of the size or clinical importance of a change. The registry record does not report a p-value or an estimate for either stated comparison.

The usual t-test framework also relies on assumptions about the distribution of the paired differences and the independence of differences across participants. The registry record does not report diagnostic checks, transformations, robust alternatives, or details of these assumptions. With a small number of observations, unusual values or departures from assumptions can have substantial influence, but the available record does not permit assessment of whether this occurred.

Within-arm change versus a comparison of treatments

Comparing baseline with follow-up within each arm addresses whether the measurements changed over time among participants with paired observations in that arm. It is not the same as directly comparing the amount of change between randomized groups. To assess whether the route of contraception is associated with different changes, an analysis would ordinarily need to estimate a between-arm contrast in change, or use another specified model that addresses the randomized comparison. The registry description does not report such an estimate or clearly explain whether the paired tests were conducted by arm.

Randomization is intended to make treatment groups comparable at allocation, on average, but it does not ensure that the groups remain comparable among only those with follow-up measurements. When participants do not return for biopsy, complete-case results can be affected if follow-up availability relates to the outcome or treatment. The record gives follow-up counts but does not describe missing-data assumptions or methods.

How to read uncertainty and statistical significance

In general, a confidence interval expresses uncertainty about an estimated quantity under the analysis assumptions. It helps readers judge both the range of effects compatible with the data and the estimate’s precision. A p-value addresses compatibility with a null hypothesis, but does not replace an effect estimate and interval. Neither confidence intervals nor p-values for the primary outcome are reported in the registry record.

Standard deviations and confidence intervals have different meanings. The standard deviation describes variation among measurements, whereas a confidence interval describes uncertainty in an estimated quantity. A reported mean and standard deviation would help characterize the observed data, but without paired differences or an appropriate treatment contrast, they would not fully establish how much the groups differed in change.

What the available record permits

The registry describes the outcome, time points, and paired testing approach, but lists no analysis entries and supplies no numerical primary-outcome results. Consequently, the methods can be explained, but the result of the planned comparison cannot be reconstructed. There is no basis in the record for calculating an effect estimate, confidence interval, or p-value. Any such calculation would require numerical outcome data that are not reported.

The record also does not report a statistical analysis plan, sample-size justification, interim analyses, or multiplicity adjustment. Their absence means those features cannot be characterized; it does not establish what procedures were or were not used outside the registry record. General statistical considerations should therefore be kept separate from claims about the study’s actual conduct.

10. Multiplicity, Missing Data and Analysis Populations

The primary outcome is assessed at baseline, 84 days, and 168 days, and the registry describes baseline comparisons with each follow-up assessment. Considering more than one assessment can raise a multiplicity question because repeated testing may increase the chance of at least one false-positive finding when all null hypotheses are true. The registry record does not report whether a multiplicity adjustment was planned or used, nor does it describe a hierarchy among the comparisons. No primary-outcome test results are reported.

The record identifies adverse events as a secondary outcome, but gives no further analysis plan for that outcome. It does not report whether additional outcomes or subgroup analyses were prespecified. A comprehensive account of the family of statistical tests is therefore not available in the registry record.

Missing follow-up is relevant to interpretation of the paired outcome. The record says 14 subjects had an initial biopsy, with fewer returning at follow-up. It reports 11 returns at 84 days, listed as 6 R and 5 P. At 168 days it reports 6 returns, but gives arm counts of 3 R and 4 P, which are inconsistent with that total. Reasons for missing biopsies, characteristics of those not returning, and methods for handling missing data are not reported in the registry record.

The record does not define the analysis population beyond the enrollment and biopsy counts, and does not state whether analyses followed randomized assignment regardless of treatment received. It does not describe imputation, sensitivity analyses, or other missing-data procedures. Accordingly, it is not possible to determine whether the intended analyses included all randomized participants or only those with complete paired measurements.

11. 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 1% reporting threshold.

Most frequent non-serious adverse event terms

The registry lists serious adverse events for the Desogen and NuvaRing groups. Each group has 7 subjects recorded as at risk for serious events, and 0 affected subjects in each group. For other adverse events, 7 subjects are recorded as at risk in each group; 1 affected subject is listed for Desogen and 0 for NuvaRing.

These counts describe what is entered in the registry record. The record does not provide the event term or clinical details for the other adverse event, and the list of top serious-event terms is empty. Death counts are not reported in the registry record. The figures should not be read as a complete comparison of safety, since the record does not provide event timing, exposure-adjusted rates, statistical comparisons, or details about ascertainment.

The counts are small and descriptive. A zero in the affected-subject field means no affected subjects are listed for that category in the record; it does not establish that an event is impossible or that the treatments have equivalent safety. Likewise, the single other event listed for Desogen does not establish that the groups differ in risk. The registry record does not provide the detail needed to make those broader judgments.

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

This record is useful for identifying the study question, interventions, endpoint, and intended follow-up comparisons. It describes a randomized parallel-group comparison of oral and intravaginal hormonal contraception and a tissue-based measure of vaginal epithelial thickness. It also indicates that biopsies were planned at baseline and after treatment cycles.

What it does not provide is the numerical primary-outcome result. There are no reported means, standard deviations, paired changes, between-arm estimates, confidence intervals, or test results. The direction and magnitude of any change, and the precision with which it was estimated, are therefore not available from the registry record. Lack of a posted estimate is not evidence that the treatments produced the same outcome, nor evidence that either treatment caused a particular change.

The registry describes paired comparisons of baseline with follow-up, but does not clearly state how those analyses relate to the randomized treatment contrast. A within-group change over time does not by itself show that routes differ. A direct comparison of change between randomized arms would be needed to address whether the changes differ, with appropriate attention to follow-up availability and uncertainty. No such result is reported.

The biopsy return counts also require caution. The record gives fewer returning subjects at follow-up than at baseline and includes an inconsistency between the stated total and arm counts at 168 days. It does not explain the discrepancy or missing follow-up. The available information thus does not establish how much the observed follow-up sample represents the randomized groups.

Finally, the safety counts are descriptive and sparse, with no detailed event descriptions or inferential analysis. The study record can inform readers about what was measured and how the study was organized, but the posted information is insufficient to quantify the primary treatment comparison or to draw a broad comparative safety conclusion.

13. Why This Trial Matters Statistically

This trial raises a design and analysis question that is common in clinical research: how to distinguish changes over time within a treatment group from differences in change between randomized groups. The registry describes repeated biopsy measurements and paired testing, making the relationship between within-person change and the randomized contrast especially important to interpretation.

Repeated measures can be valuable because each participant serves as a reference for their own baseline. At the same time, a paired analysis depends on follow-up observations being available for the same people. When participants do not return, the analyzed pairs may not represent everyone randomized. Transparent reporting of the paired sample, missing-data reasons, and analysis population helps readers understand what the estimated change applies to.

The record also illustrates why an outcome definition and an analysis description are not substitutes for results. It specifies a histologic measure and assessment times, but does not report the numerical findings or uncertainty. Without estimates and intervals, readers cannot assess effect size, precision, or compatibility with a range of possible changes. A stated test method alone cannot answer the clinical comparison.

For small studies, clarity about the estimand, missing follow-up, and uncertainty is particularly important. The record’s inconsistency in the 168-day follow-up counts further underscores the value of internally coherent participant accounting. These are statistical reporting issues, not evidence for or against a treatment effect. The trial matters as an example of the questions that must be answered before repeated-measure results can be interpreted as a comparative effect.

14. 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 primary limitation for statistical interpretation is that numerical results for the primary endpoint are not reported in the registry record. The endpoint title identifies means and standard deviations, but the record does not provide the values, paired differences, treatment contrast, confidence intervals, or test results. Consequently, neither effect size nor precision can be evaluated.

The record describes paired comparisons with baseline, but does not make clear whether the tests were run separately by arm or whether a direct comparison between arms was performed. The distinction matters because within-arm change does not establish a difference between treatments. The analysis population, treatment of participants with incomplete follow-up, and missing-data methods are also not reported.

Follow-up is incomplete relative to the baseline biopsy count, and the record gives no reasons for non-return. In addition, the stated 168-day total does not align with the arm counts stated for that assessment. The registry record does not explain or correct this discrepancy. Without clarification, the follow-up accounting and the number of usable paired observations at that assessment are uncertain.

The record does not report the statistical analysis plan, sample-size justification, multiplicity procedures, or assumption checks for the paired t tests. These omissions prevent assessment of whether the reported approach was prespecified or how multiple comparisons and model assumptions were handled. The record also does not describe the population’s eligibility characteristics in detail or report a threshold for a meaningful change in epithelial thickness.

Safety information is limited to affected-subject counts, without event descriptions for the other adverse event, timing, or a comparative analysis. Death counts are not reported in the registry record. These limitations leave both the primary comparison and the safety profile incompletely characterized.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Hormonal Contraception and Vaginal Health — NCT00612508.
• Registry record last updated 2012-11-08; status last verified 2012-10.
• Results first posted 2012-11-08.
• 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.

18. Record Summary

The registry record describes a completed, randomized, unmasked, parallel-group study of oral Desogen versus intravaginal NuvaRing. Enrollment is recorded as 14, and the primary outcome is histologic vaginal epithelial thickness in millimeters, assessed at baseline, 84 days, and 168 days.

The stated primary analysis uses paired t tests for baseline-to-follow-up comparisons. Numerical primary-outcome results and analysis entries are not reported. Follow-up biopsy counts are given, but the 168-day total conflicts with its stated arm counts. The safety record lists 0 subjects affected by serious adverse events in each group, and one other adverse event for Desogen versus none for NuvaRing.