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 (NCT00391807), a non-randomized phase 3 study. The primary outcome is Pregnancy Rate (Expressed as Pearl Index) in Women Aged 18 to 35, MITT Population,.

2. Clinical Question

In practical statistical terms, the study asked how many pregnancies occurred relative to time on treatment in the defined population. A Pearl Index expresses pregnancy incidence per 100 women-years of treatment, as specified in the endpoint description. The registry characterizes the study as non-comparative; it was not designed as a direct estimate of the difference between this contraceptive and another treatment or a control. The record also names secondary questions about intracyclic bleeding or spotting, withdrawal bleeding, and safety and tolerability. Those additional outcomes do not change the central interpretive issue: without a posted numerical primary result, the record does not reveal the estimated pregnancy rate or its precision.

3. Trial Design

This was a Phase 3, open-label study with non-randomized allocation and a single-group model. The registry describes it as non-comparative. Participants were listed under the study drug, norethindrone/ethinyl estradiol; no control arm is recorded. The planned enrollment was 1683. The study status is completed, with a recorded start of 2006-11 and primary completion in 2008-08.

These design features determine what can be learned from the posted record. A single-group study can describe outcomes observed among participants receiving the study drug, but it does not directly compare those outcomes with a concurrent control group. Without random allocation or a comparator, any observed event rate cannot by itself establish how the rate would have differed under another treatment or no treatment. Open-label status means that masking was absent; the record does not describe additional procedures intended to limit awareness of treatment assignment.

The registry names pregnancy rate as the primary outcome and gives its timeframe as 13 cycles, 28 days each (1 year). The specified primary population is women aged 18 to 35 in the modified intent-to-treat population. Secondary outcomes concern bleeding and withdrawal bleeding at specified cycles, alongside safety and tolerability. Although the registry lists outcome measures, it reports no analyses. The study protocol’s details about planned statistical methods, analysis timing, or the reasons for the enrollment target are not reported in the registry record. Consequently, the stated design can be described, but its intended statistical decision rules cannot be reconstructed from this information.

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: Modified Intent to Treat Population (MITT), Women Aged 18-35, 27 women had no cycles evaluable for pregnancy because they used alternative contraceptive methods in all cycles - 1555 subjects evaluable in MITT population. Number of pregnancies per 100 women-years of treatment in MITT Population

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) in Women Aged 18 to 35, MITT Population,, measured over 13 cycles, 28 days each (1 year).

The primary endpoint is pregnancy rate, expressed as a Pearl Index, for women aged 18 to 35 in the modified intent-to-treat population. Its timeframe is 13 cycles, 28 days each (1 year). The endpoint description defines the measure as the number of pregnancies per 100 women-years of treatment in the modified intent-to-treat population.

As a general statistical concept, a rate relates the number of events to the amount of time people were observed and at risk. This differs from a simple proportion, which counts people with an event over a defined group without expressing the observation time in the denominator. A Pearl Index can therefore summarize pregnancy incidence in relation to exposure time, which is relevant when participants may contribute different amounts of observation.

The endpoint is used to describe pregnancy incidence during treatment, but the registry record does not report its numerical estimate, an uncertainty interval, or a test result. The endpoint definition alone does not establish how many pregnancies were observed or how precise the estimate was. Nor does a rate from a single-group study quantify a treatment contrast: there is no comparator rate in the record against which to estimate a difference.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Several statistical ideas help explain what the study design and endpoint can support. The key distinction is between defining an outcome and reporting an estimate of that outcome. The registry identifies a primary endpoint and its population and timeframe, but does not report a numerical result or analysis. General explanations below clarify how such measures are commonly interpreted; they do not establish which specific methods this study used.

What does a single-group design estimate?

In a single-group study, outcome observations come from participants in one treatment group. A summary such as a pregnancy rate can describe the experience recorded in that group, subject to the observation process and analysis population. A treatment contrast requires a second relevant outcome estimate or another design that supports a comparison. Here, the registry identifies no comparator, and its study description is explicitly non-comparative. Consequently, there is no between-group difference or ratio to interpret from the record.

Randomization is generally used to make treatment groups comparable in expectation with respect to measured and unmeasured characteristics. This study is recorded as non-randomized and single-group, so random allocation did not create a concurrent comparison. Even if the primary event rate had been posted, it would describe the observed group rather than directly estimate what would have happened to the same population under a different treatment. This is a design limitation, not a finding about the direction or size of any treatment effect.

What does the Pearl Index represent?

As a general concept, an event rate uses events relative to accumulated exposure time. A rate can be informative when participants contribute differing amounts of time under observation, whereas a proportion uses a specified group as its denominator. The endpoint description defines the Pearl Index as pregnancies per 100 women-years of treatment. It also gives a timeframe of 13 cycles, 28 days each (1 year). The registry does not provide the observed number of pregnancies, the accumulated exposure time used in an analysis, or the numerical rate. Those missing results prevent an independent assessment of the estimate.

A rate estimate is not complete evidence of precision by itself. A confidence interval, in general, describes a range of values compatible with the data under a specified statistical procedure and assumptions. A narrow interval usually indicates greater statistical precision than a wide interval, but an interval does not remove design limitations or establish causality. No confidence interval for the primary endpoint is reported in the registry record, so the precision of the pregnancy-rate estimate cannot be described.

What can the analysis population tell readers?

The primary population is called the modified intent-to-treat population and is restricted in the endpoint title to women aged 18 to 35. The record says 1555 subjects were evaluable in that population and notes that 27 women had no cycles evaluable for pregnancy because they used alternative contraceptive methods in all cycles. An analysis population defines who contributes to a particular summary; it is not interchangeable with the planned enrollment. The registry does not give a full participant flow, a detailed operational definition of the modified intent-to-treat population, or the handling of all other missing or incomplete observations.

In general, excluding observations can affect the population to which an estimate applies, especially if the reason for exclusion relates to exposure or outcome risk. The stated reason for the 27 women is clear, but the record does not report sensitivity analyses or other methods for evaluating the implications of those exclusions. It would therefore be inappropriate to infer how their inclusion or exclusion changes the endpoint estimate.

What is the role of statistical significance?

Hypothesis tests and p-values are commonly used to assess how compatible observed data are with a specified null hypothesis under a statistical model. They do not measure the probability that a treatment works, and statistical significance is not a substitute for an effect estimate and its precision. The registry reports no analysis results, p-value, confidence interval, or statistical plan for the primary outcome. No conclusion about statistical significance can be drawn from the listed endpoint definition or the study’s completed status.

The sample-size assumptions describe planning illustrations based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. These are not the study’s reported design justification. The trial itself is recorded as single-group, so the illustrative equal-allocation comparison does not describe its recorded design. The registry does not report the sample-size justification or clarify a trial-specific statistical plan. Planning illustrations are not observed power and cannot replace a reported analysis.

Why might over-dispersion matter?

As a general explanation, over-dispersion occurs when the variability in count data is greater than a chosen statistical model expects. A model that assumes too little variability can produce uncertainty estimates that are too narrow and tests that are too optimistic. Methods that account for over-dispersion can yield more realistic uncertainty when the data support that feature. The registry does not report a model for the pregnancy endpoint, dispersion estimates, or whether over-dispersion was assessed. It is therefore possible to explain the concept, but not to say whether it was relevant to the sponsor’s analysis.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 17 outcome measures and records no analyses. It identifies pregnancy rate as primary and names secondary outcomes involving intracyclic bleeding or spotting and withdrawal bleeding, including measures at specified cycles. The presence of multiple outcomes raises a general statistical issue: if many hypotheses are tested, the chance of at least one apparently unusual result can increase unless the analysis plan accounts for the number and structure of tests. Statistical approaches to multiplicity can include prespecifying a primary comparison, adjusting significance thresholds, or treating secondary outcomes as exploratory. The specific approach for this study is not reported in the registry record.

The absence of analyses means there are no posted p-values or confidence intervals to assess, adjusted or otherwise. It also means the record does not show whether secondary outcomes were interpreted as confirmatory, descriptive, or exploratory. No testing hierarchy, multiplicity adjustment, or interim-analysis plan is reported. These omissions prevent readers from reconstructing the intended control of false-positive findings or determining whether any statistical claims followed a prespecified sequence.

The primary population description notes 27 women with no cycles evaluable for pregnancy because they used alternative contraceptive methods in all cycles and identifies 1555 subjects as evaluable in the modified intent-to-treat population. Beyond this note, the registry does not report a full missing-data plan, the number of missing values for each outcome, or sensitivity analyses. Missing-data methods can affect estimates when participants do not contribute complete observations; the appropriate approach depends on why the information is missing and on the outcome structure. Those details cannot be inferred from the population label alone.

The record also does not state how incomplete cycle-level observations were handled for the secondary outcomes. It would be unwarranted to assume that missing observations were ignored, imputed, or treated in any particular way. The reported population note is useful, but it does not substitute for an analysis-population definition and outcome-specific missing-data account.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: appendicitis (0.2%), deep vein thrombosis (0.1%), hyperemesis gravidarum (0.1%), ovarian cyst (0.1%), venous thrombosis (ovarian) (0.1%).

The safety record lists one treated group with 1660 participants at risk for serious adverse events and 15 participants affected. For other adverse events, it lists 1660 at risk and 390 affected. These are counts in the study group, not comparative estimates. Since there is no control arm, the record cannot show whether the counts differ from what would have occurred in a comparable untreated or differently treated group.

The listed serious adverse-event terms and percentages are appendicitis, 0.2%; deep vein thrombosis, 0.1%; hyperemesis gravidarum, 0.1%; ovarian cyst, 0.1%; venous thrombosis (ovarian), 0.1%; and cholecystitis, 0.1%. These entries describe the registry’s listed terms and percentages. The record does not provide further case-level details, event timing, severity information beyond the serious-event classification, or an attribution assessment for these terms. The death count is not reported in the registry record.

Counts of participants affected should not be confused with counts of all events: the record identifies people affected, but does not provide a full event-level account. Nor do the entries alone establish that the study drug caused an event. Safety interpretation requires attention to the population at risk, ascertainment, follow-up, and comparison context, and those details are not all reported in the registry record.

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

The record establishes that a completed Phase 3, open-label, single-group study evaluated norethindrone acetate and ethinyl estradiol, with pregnancy rate expressed as a Pearl Index designated as the primary outcome. It specifies the primary population as women aged 18 to 35 in the modified intent-to-treat population and gives a timeframe of 13 cycles, 28 days each (1 year). It also lists selected bleeding outcomes and safety counts. These details describe the study question and recorded design.

The record does not report the numerical primary endpoint result, its confidence interval, or a statistical test. It therefore does not allow a reader to judge the observed pregnancy rate or the precision of that estimate. It also has no comparator group, so it cannot provide a direct treatment comparison. The absence of a posted result is not a zero rate, a null finding, or evidence of no effect; it is an absence of reported numerical outcome information in the registry record.

The safety counts show that participants in the study group were recorded as affected by serious and other adverse events. They do not establish whether those events were caused by treatment, nor can they be contrasted with event counts in a control group. The listed serious-event terms provide some descriptive detail, but the record does not provide the full information needed to assess event timing, attribution, or comparative risk.

More broadly, a registry outcome definition and a completed-study designation are not substitutes for a results analysis. In general, a useful interpretation of an estimated treatment effect considers its size, uncertainty, study design, and the population analyzed. Here, the numerical estimate and uncertainty are not reported, and the design is non-comparative. The record can therefore support a description of the trial and its listed safety data, but not a quantitative conclusion about comparative efficacy or the precision of a pregnancy-rate estimate.

13. Why This Trial Matters Statistically

This trial is statistically instructive because its primary endpoint is clearly named while its numerical result is absent from the registry record. That distinction is important for registry-based interpretation: an endpoint title explains what investigators intended to measure, but does not reveal the observed value, its uncertainty, or the analysis used to obtain it. Readers should not treat a defined outcome as though it were a reported result.

The study also illustrates how design constrains interpretation. A single-group, non-comparative study can describe outcomes in the enrolled group, but it cannot supply a randomized between-group contrast. Non-randomized allocation and the lack of masking are recorded design features; neither provides a basis for reconstructing a control outcome. Even a precise single-group rate would not answer a comparative question by itself.

The primary endpoint’s use of a rate makes the exposure-time definition relevant. The registry states the intended denominator in women-years and gives a cycle-based timeframe. This makes the endpoint conceptually distinct from a simple participant proportion. Yet without the observed rate and uncertainty interval, readers cannot assess the estimate’s precision or independently interpret the contribution of observation time.

Finally, the record distinguishes planned enrollment from the population described as evaluable for the primary endpoint and notes exclusions related to alternative contraceptive use. Such details matter because analysis populations determine which observations contribute to an estimate. The record does not supply the full missing-data or analysis plan, so the effect of those choices cannot be assessed. Together, these features make the record useful for understanding design and reporting limits, while underscoring why outcome definitions, analysis methods, and numerical results must be considered separately.

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 central limitation for interpretation is that the registry record does not report a numerical primary pregnancy-rate result, confidence interval, or statistical test. The study is described as completed, but completion does not supply the missing outcome estimate. The record therefore cannot support an assessment of the primary endpoint’s magnitude or statistical precision.

The study was non-comparative, non-randomized, open-label, and single-group. These recorded features mean there is no concurrent control outcome and no randomized treatment contrast to interpret. Any descriptive result from the treated group would not, by itself, establish what would have happened under another treatment. The absence of masking may also be relevant to outcomes involving participant reporting or assessment, but the registry does not describe outcome-ascertainment procedures in enough detail to assess that issue.

The primary population description reports 1555 evaluable subjects and 27 women who had no cycles evaluable for pregnancy because they used alternative contraceptive methods in all cycles. The record does not give a full participant flow, a complete modified intent-to-treat definition, or outcome-specific missing-data methods. It also does not report sensitivity analyses that could show how alternative analysis choices affect results.

Although 17 outcome measures are listed, no analyses are recorded. The statistical plan, multiplicity approach, interim-analysis procedures, and sample-size justification are not reported in the registry record. The safety entries provide counts and selected serious-event terms, but not enough detail to assess timing, attribution, or comparative risk. These limitations concern what can be concluded from the registry information; they should not be mistaken for evidence that an analysis or event was absent from the full study record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Study of Safety and Efficacy of an Oral Contraceptive — NCT00391807.
• Registry record last updated 2013-04-22; status last verified 2013-04.
• Results first posted 2010-12-22.
• 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

The registry describes a completed Phase 3, open-label, non-randomized, single-group study of norethindrone acetate and ethinyl estradiol for contraception. Enrollment was 1683, with a recorded start of 2006-11 and primary completion in 2008-08. The primary outcome was pregnancy rate expressed as a Pearl Index among women aged 18 to 35 in the modified intent-to-treat population, over 13 cycles, 28 days each (1 year).

The record identifies 1555 subjects as evaluable in that population and notes 27 women had no cycles evaluable for pregnancy because they used alternative contraceptive methods in all cycles. It lists 17 outcome measures and no analyses. Numerical primary endpoint results and their precision are not reported in the registry record. Safety entries include 15 participants affected by serious adverse events and 390 affected by other adverse events, each among 1660 at risk. With no comparator group, the design does not provide a direct treatment comparison.