Transdermal Contraceptive Delivery System (TCDS), Also Known as AG200-15 Patch Arm: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Transdermal Contraceptive Delivery System (TCDS), Also Known as AG200-15 Patch Arm (NCT01236768), a randomized phase 3 study. The primary outcome is Pregnancy Reported as Pearl Index.

2. Clinical Question

3. Trial Design

This was a randomized, open-label, parallel-group Phase 3 study of the AG200-15 transdermal contraceptive delivery system and Levora, an oral contraceptive. The official title describes the study as a comparison of contraceptive efficacy and safety. The registry lists contraception as the condition and Agile Therapeutics as the sponsor. The study status is completed.

The registry reports enrollment of 407 participants. Its primary endpoint description specifies the intent-to-treat population ages 18-35 and a six-month timeframe. The primary endpoint results list 147 participants for AG200-15 and 163 for Levora. The record does not report why the endpoint counts are smaller than total enrollment, how participants were distributed among analysis populations, or whether all enrolled participants contributed endpoint data.

Randomization assigns participants to study groups according to a chance-based process, which is intended to reduce systematic differences between groups at assignment. A parallel design means participants are assigned to one group rather than switching between the treatments. In this trial, masking was none: participants and relevant study personnel were not masked to treatment assignment according to the registry description. Open-label conduct can matter for outcomes involving judgment or participant reporting. The primary outcome concerns pregnancies, but the registry does not describe procedures for confirming or adjudicating pregnancy reports.

The posted primary outcome is pregnancy reported as a Pearl Index, with a 95% confidence interval for each arm. Safety is listed as a secondary outcome. The record lists six outcome measures and one analysis. It does not report the full analysis plan, the allocation ratio, the randomization method, the intended enrollment rationale, or the detailed follow-up and participant disposition. These design details should not be inferred from the trial title or outcome numbers.

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

Analysis population for the primary outcome, as defined in the registry: Intent-to-treat population ages 18-35.

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.

Population: Safety Population (number of subjects that received drug)

6. Primary Endpoint

The registry lists the primary outcome as Pregnancy Reported as Pearl Index, measured over 6 months.

The primary endpoint, pregnancy reported as Pearl Index, is a rate-based measure of contraceptive efficacy during product use. The registry defines it as the number of on-therapy pregnancies times 1300, divided by the number of 28-day on-therapy cycles. It is described as an estimate of the number of pregnancies per 100 woman-years of product use. The reported timeframe is six months, and the specified population is the intent-to-treat population ages 18-35.

A rate expresses events in relation to time or exposure rather than simply counting events. This can help describe pregnancy occurrence while participants are using a product, accounting for the amount of on-therapy cycle time represented in the data. It also makes clear that the endpoint is not merely the number of pregnancies among all people enrolled: the definition is based on on-therapy pregnancies and on-therapy cycles. The registry does not provide the underlying count of pregnancies or the number of cycles used to calculate the posted arm estimates.

The registry reports a Pearl Index of 8.19 for AG200-15, with a 95% confidence interval from 0.19 to 16.19. For Levora, it reports 6.8, with a 95% confidence interval from 0.15 to 13.44. These are arm-level estimates. A direct estimate of the difference or ratio between the treatments, and a confidence interval or test for that comparison, are not reported in the registry record.

Pearl Index is useful as an exposure-based summary, but it does not by itself describe every dimension of contraceptive use or every pattern of risk over time. Its interpretation depends on how pregnancies and on-therapy cycles are defined and recorded. Those operational details are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary result is a Pearl Index, a rate measure rather than a raw event count. As defined in the registry, it uses on-therapy pregnancies and 28-day on-therapy cycles, with the calculation expressed as pregnancies per 100 woman-years of product use. This scaling allows the rate to be interpreted in relation to exposure time. The underlying pregnancy counts and cycle totals are not reported, so the posted rate cannot be independently reconstructed from the registry information.

What does a confidence interval describe?

Each arm’s estimate is accompanied by a 95% confidence interval: 0.19 to 16.19 for AG200-15 and 0.15 to 13.44 for Levora. In general, a confidence interval describes the uncertainty associated with an estimate under the statistical method used. A narrower interval indicates greater precision, while a wider interval indicates less precision. These intervals are wide relative to their point estimates, so the arm estimates leave considerable uncertainty about the underlying Pearl Index values. A confidence interval is not a guarantee that the true value lies within its endpoints, and its interpretation depends on the method and assumptions used to calculate it. The registry does not identify the interval-calculation method.

Why is an arm-level interval not a treatment comparison?

The posted confidence intervals describe each arm separately. The scientific question in a randomized comparison is usually framed in terms of a contrast between treatments. That contrast could be expressed in a suitable scale as a difference or ratio, but the registry does not post such a comparative estimate. A valid uncertainty interval or test for the contrast must account for the comparison itself. It cannot be recovered simply by reading the two arm estimates as if one were a direct test of the other. The record also does not report a comparative p-value or a prespecified decision threshold.

In general, a p-value evaluates how compatible observed data are with a specified statistical null model, subject to the analysis assumptions. It does not measure the probability that a treatment is effective, nor does it describe the size or practical importance of an effect. No p-value for the primary arm-to-arm comparison is reported in the registry record. Accordingly, there is no basis in these posted results for labeling the contrast statistically significant or nonsignificant.

Why can event rates be uncertain?

Rate estimates depend on both the number of events and the amount of exposure contributing to the denominator. When event counts are limited, estimates may be unstable, and their confidence intervals can be broad. The registry does not provide the pregnancy counts or cycle totals, so the contribution of event frequency and exposure duration cannot be assessed separately. General statistical practice treats a rate estimate as a summary of observed data, not as a guarantee of the future rate for a different population or period.

In count outcomes, variability can sometimes exceed what a simple Poisson model expects. This is called over-dispersion. In general, over-dispersion can make standard errors too small if the model assumes less variability than the data contain; methods that account for extra variability may produce more appropriate uncertainty. The registry does not report whether a count model was used, whether over-dispersion was assessed, or how the confidence intervals were constructed. This is a general consideration, not a claim about the analysis in this trial.

What does randomization contribute?

Randomization is intended to make treatment groups comparable, on average, with respect to both measured and unmeasured characteristics at assignment. It supports a causal comparison when allocation is maintained, follow-up and outcome ascertainment are adequate, and the analysis is appropriate. It does not eliminate uncertainty, guarantee balance in a particular trial, or correct problems that arise after assignment. The registry describes the study as randomized but does not report the method used to generate or conceal assignments.

The trial was open-label, with no masking. In general, awareness of assignment can affect participant behavior, reporting, or clinical decisions. The direction and extent of any such influence depend on the outcome and study procedures. Pregnancy is a comparatively concrete endpoint, but the registry does not describe confirmation, adjudication, or ascertainment procedures. Thus, the open-label design is relevant context without demonstrating that outcome assessment was biased.

How should planning calculations be distinguished from trial results?

Planning illustrations based on an observed effect answer a hypothetical design question: how many participants per arm might be required under stated assumptions to detect a chosen fraction of that effect. They are not observed outcomes and are not the trial’s original sample-size justification. The registry record does not report the original sample-size rationale. A planning calculation based on the observed effect also depends on assumptions about significance level, allocation, and the statistical approximation. It should not be interpreted as observed power, nor as evidence that the trial was adequately powered for its original objective.

The record states assumptions for planning illustrations, including a two-sided significance level of 0.05, equal allocation, and a normal approximation. It also distinguishes participant-per-arm calculations from total-event calculations for time-to-event outcomes. The primary endpoint in this study is reported as a Pearl Index; no time-to-event primary analysis is described. These planning conventions do not establish the method used for the trial’s primary analysis.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures and one analysis, with safety identified as a secondary outcome. It does not provide enough detail to establish the full family of planned statistical tests, how outcomes were prioritized, or whether any adjustment for multiple testing was planned or performed. Multiplicity arises when several hypotheses or outcomes are assessed: in general, more testing can increase the chance of at least one apparently unusual result. Adjustment methods and hierarchical testing plans can address that issue, but the registry does not report such procedures. The number of listed outcomes alone does not establish which results were inferentially tested.

The registry identifies the primary endpoint population as intent-to-treat ages 18-35. It does not explain the exact analysis set, how participants with no on-therapy cycles or missing pregnancy information were handled, or whether participants were analyzed according to their randomized assignment. The distinction matters because different analysis populations can answer different questions and can yield different estimates. No imputation procedure, censoring rule, or sensitivity analysis for missing endpoint data is reported in the registry record.

Overall enrollment is 407, while primary endpoint results list 147 participants for AG200-15 and 163 for Levora. Safety results use denominators of 177 and 188. These different counts may reflect different endpoint populations or reporting conventions, but the registry does not explain their relationship. It would be speculative to attribute the differences to withdrawals, exclusions, missing observations, or specific analysis rules.

General statistical interpretation of missing data depends on why information is absent and how the analysis addresses it. The registry does not report the amount or pattern of missing primary endpoint data, so the potential effect on the estimates cannot be evaluated from this record. The posted confidence intervals quantify uncertainty according to their calculation method, but do not by themselves account for unreported sources of bias from missing data or outcome ascertainment.

11. 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 4% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: murine typhus (0.5%), convulsion (0.5%).

The registry reports serious adverse events in 0 of 177 participants in the AG200-15 group and 2 of 188 in the Levora group. It reports other adverse events in 42 of 177 and 45 of 188, respectively. Deaths are reported as 0 in both groups. These figures are descriptive counts with the stated group denominators. They do not, by themselves, establish that treatment assignment caused an event or that one treatment is safer than the other.

The registry also lists Murine Typhus and Convulsion among the top serious adverse event terms, each at 0.5%. The record does not map these terms to an arm in the information, so they should not be assigned to either group. The broader serious-event counts and the named terms are different forms of reporting; the terms do not replace the group-level counts.

Safety comparisons can depend on the definitions of an adverse event, seriousness, time at risk, follow-up, and the population included. Although group denominators are reported, the registry does not describe the safety analysis population or exposure duration in enough detail to assess those features. Differences in raw counts should therefore be read as posted observations, not as proof of a treatment effect. The absence of reported deaths does not establish the absence of other risks or events outside the registry’s stated reporting.

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

The posted efficacy results describe Pearl Index estimates for AG200-15 and Levora in the intent-to-treat population ages 18-35 over six months. The estimates are 8.19 and 6.8, with 95% confidence intervals of 0.19 to 16.19 and 0.15 to 13.44, respectively. Those intervals convey substantial uncertainty around each arm-level rate. The results do not include a direct contrast, comparative confidence interval, or comparative p-value in the registry record.

These findings therefore describe the values reported for the two groups; they do not establish whether the treatments differ, are equivalent, or meet a noninferiority criterion. A numerical difference between point estimates is not itself evidence that a true treatment difference exists. Equivalence and noninferiority conclusions require a suitable prespecified framework, including an analysis of the contrast and, for noninferiority, a margin. Such details are not reported in the registry record.

The endpoint is a rate of on-therapy pregnancies relative to product-use exposure, not a direct prediction of an individual participant’s experience. The estimate is specific to the study’s population, timeframe, endpoint definition, and observed follow-up. The record does not provide underlying pregnancy counts, on-therapy cycle totals, detailed confirmation procedures, or the reasons for the differing participant counts across endpoint and safety results.

Safety results are also descriptive. The listed serious adverse event counts are 0 of 177 for AG200-15 and 2 of 188 for Levora; other adverse events are 42 of 177 and 45 of 188; and deaths are 0 in both groups. These observations do not establish causation, and the registry does not provide enough detail to evaluate differences in exposure or outcome ascertainment. The named serious adverse event terms are reported without an arm assignment in the available record.

In general, randomized trials can support comparative conclusions when the design, analysis, outcome ascertainment, and reporting provide an interpretable estimate of the treatment contrast. Here, randomization is reported, but the posted results do not include that contrast or key analysis details. The most defensible reading is limited to the arm-specific estimates, their reported uncertainty, and the descriptive safety counts. Broader claims about comparative performance or safety go beyond what the registry record reports.

13. Why This Trial Matters Statistically

This trial illustrates why outcome definitions and denominators matter in clinical statistics. A Pearl Index is not simply a count of pregnancies: it relates on-therapy pregnancies to on-therapy cycles and expresses the result as an estimate per 100 woman-years of use. That makes exposure part of the endpoint’s meaning. Without the underlying event and exposure counts, readers can interpret the reported rate but cannot independently reconstruct it or examine how the numerator and denominator contributed to its uncertainty.

It also illustrates the distinction between an estimate within each treatment group and a comparison between treatments. Separate arm-level estimates and confidence intervals are useful descriptions, but the scientific comparison requires a contrast and uncertainty around that contrast. Without that result, the numerical values should not be treated as a formal test of difference. This distinction is central to interpreting randomized clinical trials across many types of outcomes.

The record also highlights how enrollment and analysis denominators may differ. The total enrollment is 407, while the primary efficacy and safety results use different group counts. Those differences do not necessarily indicate an error, but they require explanation to understand which participants contributed to each result. The registry does not report that explanation or the handling of missing endpoint information.

Finally, the open-label design and the limited statistical detail are important context. Randomization can help make treatment groups comparable at assignment, but it does not substitute for transparent outcome definitions, analysis populations, or missing-data methods. The posted results are most informative when read as estimates with uncertainty and as descriptive safety counts, rather than as a complete comparative analysis.

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 trial-specific details needed for a fuller assessment are not reported in the registry record. The primary outcome results do not include a direct comparison between AG200-15 and Levora, such as a contrast estimate with a confidence interval or a comparative test. The record also does not report a prespecified hypothesis framework, a noninferiority margin, or an equivalence margin. Therefore, the arm-level values cannot establish those types of comparative conclusions.

The primary endpoint is reported as a Pearl Index, but the registry does not provide the underlying number of on-therapy pregnancies or 28-day on-therapy cycles. It does not describe the interval-calculation method or details of pregnancy confirmation and outcome adjudication. These omissions limit assessment of how the estimates were constructed and how uncertainty was quantified.

The registry reports 407 enrolled participants, but the primary endpoint and safety results use different denominators. It does not explain the flow of participants through follow-up, the relationship between the reported groups, the number with missing endpoint information, or the analysis rules for participants who discontinued or did not contribute on-therapy cycles. No missing-data strategy or sensitivity analysis is reported. The record also does not give the original sample-size justification, so planning illustrations based on an observed effect should not be mistaken for the original design rationale.

The trial was open-label, and the registry does not detail the procedures used to ascertain or verify pregnancies. That absence does not demonstrate biased assessment, but it limits evaluation of the potential influence of unmasked treatment assignment. Safety reporting is also limited: the record gives group counts and selected serious event terms but does not describe event definitions, exposure time, or the methods used to collect and summarize events. These limitations constrain interpretation without changing the values that are posted.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Transdermal Contraceptive Delivery System (TCDS), Also Known as AG200-15 Patch Arm — NCT01236768.
• Registry record last updated 2017-09-15; status last verified 2017-08.
• Results first posted 2017-09-15.
• 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

NCT01236768 is a completed, randomized, open-label, parallel-group Phase 3 study sponsored by Agile Therapeutics. It compared the AG200-15 transdermal contraceptive delivery system with Levora, an oral contraceptive containing 150 mcg of levonorgestrel and 30 mcg of ethinyl estradiol. The registry reports enrollment of 407 participants.

The primary outcome was pregnancy reported as Pearl Index over six months in the intent-to-treat population ages 18-35. The reported estimates are 8.19 for AG200-15, with a 95% confidence interval from 0.19 to 16.19, and 6.8 for Levora, with a 95% confidence interval from 0.15 to 13.44. A direct comparative analysis is not reported in the registry record.

Safety is listed as a secondary outcome. Serious adverse events are reported for 0 of 177 participants in the AG200-15 group and 2 of 188 in the Levora group; deaths are 0 in both groups. The registry does not explain the differences among enrollment, endpoint, and safety denominators or describe the missing-data and analysis procedures.