Study of Safety and Efficacy of an Oral Contraceptive: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Safety and Efficacy of an Oral Contraceptive (NCT00932321), a randomized phase 3 study. The primary outcome is Pregnancy Rate (Expressed as Pearl Index) for Women 18 to 45 Years Old, MITT Population.

2. Clinical Question

3. Trial Design

This was a randomized, open-label, comparator-controlled, parallel-group phase 3 study of two schedules of norethindrone acetate and ethinyl estradiol. The record gives the overall enrollment as 938 and lists the study status as completed. The recorded start was 2004-01, and primary completion was 2004-11. The sponsor was Warner Chilcott.

Both groups received the same stated doses: norethindrone acetate 1 mg and ethinyl estradiol 20 mcg. The schedules differed in the number of days of medication in each 28 day cycle: 24 days in one arm and 21 days in the other. The study’s primary objective was to assess efficacy in preventing pregnancy. Secondary objectives concerned intracyclic bleeding and safety and tolerability.

Random allocation means participants were assigned to study groups by a random process, which is intended to support a fair comparison at the outset. The registry record does not report the allocation procedure or concealment details. Open-label status means masking was none; participants and study personnel were not recorded as blinded to the assigned schedule. The record does not report whether outcome assessment was masked.

The primary outcome analysis was described as modified intention to treat. The registry defines that population as a subset of all treated participants who were evaluated for pregnancy at least once after beginning medication. The primary outcome group sizes were 705 for 24 Day NA/EE and 181 for 21 Day NA/EE. The record does not explain the difference between these endpoint group sizes and overall enrollment, or provide a participant flow account explaining exclusions, discontinuations, or missing outcome data.

The primary endpoint timeframe was 5.6 months (6 - 28 day cycles). The record also lists mean intracyclic bleeding or spotting days in cycles 2-6 for the modified intention-to-treat population. It does not report the full protocol, detailed follow-up procedures, or a rationale for the chosen enrollment.

4. Participant Flow and Accounting

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

Before assignment: MITT Population defined as subset of All Treated Subjects who were evaluated for pregnancy, either positive or negative, at least once after beginning the study medication. Completed subjects - subset of MITT subjects completing at least 161 days of treatment based on returned drug or diary.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Modified intention to treat (MITT) subset of all treat subjects - evaluated for pregnancy at least once after beginning study medication.

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 Pregnancy Rate (Expressed as Pearl Index) for Women 18 to 45 Years Old, MITT Population, measured over 5.6 months (6 - 28 day cycles).

The primary endpoint is the pregnancy rate expressed as a Pearl Index for women 18 to 45 years old in the modified intention-to-treat population. The registry defines the calculation as 1300 * number of pregnancies/number of women-cycles of treatment. It records the timeframe as 5.6 months (6 - 28 day cycles). In other words, the endpoint relates the number of pregnancies to the amount of treatment exposure measured in women-cycles, using the stated scaling factor.

A rate rather than a simple count is useful when participants contribute different amounts of time at risk or different numbers of treatment cycles. Relating events to exposure allows the result to reflect the observed treatment time, rather than treating every participant as though they had identical follow-up. This is a general statistical explanation; the registry record does not provide participant-level exposure or pregnancy counts from which to assess the calculation independently.

The posted Pearl Index is 1.823 for 24 Day NA/EE, based on 705 participants, and 2.978 for 21 Day NA/EE, based on 181 participants. These are point estimates. The registry record does not report a confidence interval or a sponsor-reported statistical test for the primary endpoint. Without an interval or another reported measure of uncertainty, the precision of these estimates and the uncertainty around their between-arm contrast cannot be assessed from the posted primary results alone.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies a randomized, parallel-group comparison and posts an arm-specific Pearl Index for each schedule. The primary endpoint is a rate based on pregnancies and women-cycles of treatment, multiplied by the stated factor of 1300. This form of rate accounts for treatment exposure in its denominator. It is different from a simple proportion of participants with an event, because it uses women-cycles rather than only the number of participants.

The primary analysis population is described as modified intention to treat: treated participants evaluated for pregnancy at least once after beginning study medication. An analysis population determines which participants contribute to an estimate. Modified intention-to-treat definitions vary across studies, so the operational definition matters. The registry supplies a definition, but does not report a participant flow account or explain how participants without an evaluation after treatment initiation were handled. It also does not provide the event and exposure totals needed to reproduce the posted rates or estimate their sampling uncertainty.

What does a confidence interval communicate?

In general, a confidence interval expresses the uncertainty around an estimated effect under a specified statistical method and its assumptions. A narrower interval indicates greater precision than a wider interval, while the interval’s location describes which effect values are compatible with the observed data under that method. It is not a guarantee that the true effect lies within the interval. No confidence interval is posted for the primary Pearl Index results, so their comparative precision cannot be evaluated from the primary outcome record.

The secondary continuous outcome is the mean number of intracyclic bleeding or spotting days in cycles 2-6. The posted means are 6.31 days in the 24-day arm and 7.31 days in the 21-day arm, with standard deviations of 9.16 and 10.51. A standard deviation describes the spread of individual observations around a group mean; it is not the uncertainty of the mean itself. The recalculated unadjusted mean difference is 1.0 days, with a confidence interval from -0.674 to 2.674 and p = 0.242. This recalculation uses the posted group summaries.

A mean difference is interpretable in the outcome’s original units, here days. Its confidence interval describes uncertainty around that estimated contrast under the recalculation’s assumptions. The interval includes values on either side of zero, so the recalculated estimate does not provide clear evidence of a difference under a conventional null-hypothesis framework. This statement concerns only the unadjusted recalculation; it does not establish equivalence or non-inferiority.

How should the reported secondary analysis be read?

The registry labels the reported secondary analysis as Cochran-Mantel-Haenszel and as a non-inferiority or equivalence analysis. It gives p = 0.311 but does not report the test parameter, an effect estimate, a confidence interval, or the margin used for the non-inferiority or equivalence assessment. Without the margin and an estimate with its uncertainty, the p-value alone cannot establish whether a non-inferiority or equivalence criterion was met. The reported method label also does not give enough detail to reconstruct the test or resolve how it was applied to the continuous endpoint.

The recalculated comparison and the reported analysis are distinct. The recalculation is an unadjusted comparison derived from the posted means, standard deviations, and group sizes. It does not replace the analysis described by the registry, and it does not incorporate any unreported adjustment or prespecified testing procedure. The registry record reports no discrepancy flag; in any case, the reported analysis and the recalculation should be distinguished by their methods and available details.

What do the planning illustrations mean?

The planning illustrations are based on the observed secondary mean difference, with the stated assumptions of a two-sided significance level of 0.05, equal allocation, and a normal approximation. They give participants per arm for 80% power and participants per arm for 90% power under scenarios targeting 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. For example, planning for 100% of the observed effect gives 1,403 participants per arm for 80% power and 1,877 participants per arm for 90% power. These are hypothetical planning quantities, not observed power and not a record of the trial’s own design assumptions. The registry record does not report the original sample-size justification or its assumptions.

Over-dispersion is a general concern when count outcomes vary more than a simple model assumes. If present, it can affect standard errors and the reliability of tests that do not account for it. The registry classifies the primary endpoint as a count-type outcome, but does not report a dispersion estimate or model details for that endpoint. No conclusion about over-dispersion or its handling can be drawn from the posted primary results.

10. Multiplicity, Missing Data and Analysis Populations

The registry record does not report how the study handled multiplicity: whether any adjustment was planned, whether outcomes followed a testing hierarchy, or how the primary and secondary analyses were related for decision-making. Multiplicity is a general concern when several outcomes or comparisons are tested, because repeated testing can increase the chance of at least one apparently positive finding even when no underlying effect exists. The record lists one secondary endpoint, but that alone does not establish the full number of analyses conducted or the prespecified testing strategy.

The record also does not report whether interim analyses were planned or conducted. No interim analysis schedule, stopping boundary, or monitoring procedure is described. These details are important when interpreting a trial’s reported significance tests, but their absence does not justify assuming that any particular procedure was used.

The primary endpoint population is defined as a modified intention-to-treat subset of treated participants evaluated for pregnancy at least once after starting study medication. The registry does not report the number of participants excluded from that population, reasons for exclusion, follow-up completion, missing pregnancy assessments, or methods for handling missing data. It also does not report sensitivity analyses addressing missing outcomes. These omissions limit assessment of whether the analyzed groups fully represent the randomized participants.

The secondary endpoint is described for the modified intention-to-treat population, and the posted group sizes are stated. The record does not explain the handling of missing bleeding diaries or observations, whether all participants contributed data in cycles 2-6, or how incomplete follow-up affected the means. The posted summaries therefore cannot establish the extent or impact of missing secondary outcome data.

These reporting gaps do not show that missingness or multiplicity affected the results; they mean that the registry record does not report enough information to assess those issues. The statistical plan, missing-data strategy, and interim-monitoring details are not reported in the registry record.

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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: partial thyroidectomy (0.1%), thyroid cancer (0.1%), back injury (0.1%).

The registry posts serious adverse event counts for each safety group: 3 among 743 at risk in the 24 Day NA/EE group and 0 among 186 at risk in the 21 Day NA/EE group. It also lists Partial Thyroidectomy, Thyroid Cancer, and Back Injury among the top serious terms, each with a recorded value of 0.1%. These figures are descriptive counts and listings; the record does not provide a comparative statistical analysis or confidence intervals for serious adverse events.

Other adverse events are recorded for 457 among 743 at risk in the 24 Day NA/EE group and 99 among 186 at risk in the 21 Day NA/EE group. The registry does not define the “other” category in the information posted, so its scope and relationship to individual event terms are unclear. These counts should not be treated as directly comparable measures of severity or attributed to a specific event without further definitions.

The record gives no death counts for either group. It also does not report event timing, duration, recurrence, treatment relatedness, exposure-adjusted rates, or a detailed safety analysis plan. Rare-event counts can be imprecise, particularly when group sizes and event numbers are limited; this is a general statistical consideration, not an additional result from this study. The posted safety information supports a description of the recorded counts, but not a broad conclusion about comparative safety.

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

The results report arm-specific Pearl Index estimates for the two schedules and a secondary measure of intracyclic bleeding or spotting days. The primary point estimates are 1.823 for the 24-day schedule and 2.978 for the 21-day schedule, based on the posted modified intention-to-treat groups. These values describe the recorded estimates over the stated timeframe; the record does not provide primary confidence intervals or a sponsor-reported comparison test.

Without reported uncertainty for the primary endpoint, the numerical ordering of the point estimates should not be treated as definitive evidence of a difference between schedules. Nor does the absence of a reported test establish that the schedules are equivalent. Equivalence and non-inferiority require a defined margin and an appropriate estimate with uncertainty, neither of which is reported for the primary outcome. The non-inferiority or equivalence label belongs to the reported secondary analysis, for which the margin and effect estimate are also not reported.

The secondary summaries concern bleeding or spotting, not pregnancy prevention. The recalculated unadjusted result is not the same analysis as the registry-reported secondary result, and neither supplies the missing precision information for the primary endpoint. Safety counts likewise describe recorded events and do not, without fuller definitions and analysis, establish a comparative safety conclusion.

The record is therefore useful for identifying the study design, endpoint definitions, group summaries, and what analyses were posted. It is not sufficient to reconstruct the primary analysis, judge its precision, or assess the full impact of missing outcomes. This is a limitation of the available registry reporting, not proof that the trial did not collect additional information.

No treatment recommendation follows from these statistics. They describe the recorded trial results and their reporting limitations. Judging how the findings apply beyond the study would require information not reported in the registry record, including fuller methods, participant flow, exposure data, and uncertainty estimates for the primary comparison.

13. Why This Trial Matters Statistically

This trial illustrates why an endpoint definition and an estimate are not enough to evaluate a randomized comparison. The Pearl Index is explicitly defined in terms of pregnancies and women-cycles, and the registry posts a point estimate for each schedule. Yet the record does not provide confidence intervals, event and exposure totals, or a reported primary between-arm test. Without those elements, readers can identify the numerical estimates but cannot judge their precision from the primary results.

It also illustrates the importance of distinguishing outcomes and analyses. The primary endpoint concerns pregnancy, while the secondary endpoint concerns bleeding or spotting days. A result for the secondary outcome cannot stand in for an analysis of the primary outcome. In addition, the registry records a non-inferiority or equivalence label for the secondary analysis but omits the margin and effect estimate needed to interpret that label. Method names and p-values alone may not communicate the actual decision criterion.

The record provides a useful example of why modified intention-to-treat definitions should be read closely. Here, the definition specifies treated participants evaluated for pregnancy at least once after starting medication, but does not include a participant flow account or missing-data handling. The definition narrows what can be inferred about who contributed to the primary analysis.

Finally, planning illustrations based on an observed effect are not the same as the trial’s original sample-size rationale and are not observed power. They can show how planning quantities depend on assumptions about the effect to detect, but they cannot establish what the investigators planned or whether the trial was adequately powered for its primary question. Together, these features make transparent reporting of methods, denominators, uncertainty, and analysis populations central to sound statistical interpretation.

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.

Several limitations specific to this record affect interpretation. The primary endpoint is reported as a Pearl Index, but no confidence interval, standard error, or sponsor-reported comparison is posted. The record does not provide the pregnancy counts or women-cycle totals behind the arm-specific values, so the calculations cannot be independently checked from the posted summaries.

The endpoint group sizes differ from overall enrollment, and the record does not explain the participant flow or the reasons participants may not have contributed to the primary analysis. Although the modified intention-to-treat definition is given, the handling of participants without pregnancy evaluation, incomplete follow-up, and missing outcomes is not reported. This makes it difficult to assess potential selection or missing-data effects.

The study was open label, with masking recorded as none. The record does not say whether pregnancy assessment or outcome analysis was masked. It also does not report allocation concealment, protocol details, a statistical analysis plan, sample-size justification, interim analyses, or a multiplicity strategy. These details are not reported in the registry record and should not be inferred.

For the secondary endpoint, the registry reports a method label and p-value but no estimate, confidence interval, or non-inferiority or equivalence margin. The unadjusted recalculation is based on aggregate summaries and does not supply those missing protocol details. Safety reporting is similarly limited: serious event counts and selected terms are posted, but the record lacks fuller event definitions, timing, relatedness, and comparative uncertainty.

These limitations concern the information available in the registry record. They do not establish that additional analyses or documentation were never collected; such information is not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Study of Safety and Efficacy of an Oral Contraceptive — NCT00932321.
• Registry record last updated 2013-04-22; status last verified 2013-04.
• Results first posted 2011-03-22.
• 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

NCT00932321 is a completed, randomized, open-label, parallel-group phase 3 study sponsored by Warner Chilcott. Overall enrollment was 938. It compared norethindrone acetate 1 mg and ethinyl estradiol 20 mcg taken for 24 days of each 28 day cycle with the same doses taken for 21 days of each 28 day cycle.

The primary endpoint was the Pearl Index for pregnancy among women 18 to 45 years old in a modified intention-to-treat population, over 5.6 months (6 - 28 day cycles). The posted estimates were 1.823 for the 24-day arm and 2.978 for the 21-day arm. The registry record does not report a confidence interval or sponsor-reported comparison test for this endpoint.

The secondary outcome was mean intracyclic bleeding or spotting days in cycles 2-6. The record also posts limited serious and other adverse event counts. Missing-data handling, the primary analysis plan, sample-size justification, and multiplicity procedures are not reported in the registry record.