Number of pregnancies per year in each treatment arm.

All Mirena participants were discontinued by end of Year 5; therefore, Years 6-10 are considered "not available" for that arm.

The Pearl Index (PI) is an estimate of the number of pregnancies per 100 women-years of contraceptive exposure. The Pearl Index is calculated as the number of "on treatment" pregnancies in the study divided by the total number of complete 28-day cycles of use in the study multiplied by 1,300 (13 cycles/yr x 100 years). Cycles where other birth control methods were used were excluded from this measure. The Pearl Index for MIRENA was not a pre-specified primary outcome measure and is therefore not included. There is only 8 years of data due to the early termination of the study.

The Pearl Index for MIRENA was not a pre-specified primary outcome measure and is therefore not included. The number of participants analyzed for this measure is the MITT population.

The safety and tolerability based on the report of adverse events and serious adverse events

All Mirena participants were discontinued by end of Year 5.

The incidence of diary reported bleeding and/or spotting by 90-day interval, as an assessment of IUS-induced amenorrhea, is summarized as the number of subjects reporting no bleeding and/or spotting.

All Mirena participants were discontinued by end of Year 5.

The rate of continuation of usage over time.

All Mirena participants were discontinued by the end of Year 5 per the study protocol. All LNG20 (36-45 Yr Old) participants were discontinued by the end of Year 8 per the study protocol. Month 6 is defined as 182 days of use, Year 1= 365 days, Year 2= 730 days, Year 3= 1095 days, Year 4= 1460 days, Year 5= 1825 days, Year 6= 2190 days, Year 7= 2555 days, Year 8= 2920 days, Year 9= 3285 days, and Year 10= 3650 days of product use.

Subjects who discontinued from the trial and who were not pregnant or did not start a hormonal contraceptive were followed to assess the return of menses after IUS removal. If after three months menses had not occurred, the subject was to be evaluated for the cause of their secondary amenorrhea and continued to be followed monthly until the diagnosis was established.

Only those subjects who completed this evaluation are included.

Subjects who discontinued from the trial and desired pregnancy were followed for up to 12 months to assess fertility after IUS removal.

Distribution of Pregnancies (Months Post-IUS Removal) [N(%)]

The Mean Plasma Pharmacokinetics of LNG in a Subset of Subjects through 60 Months of Use. "Non-obese" was defined as a BMI less than 30 kg/m^2. "Obese" was defined as a BMI of 30 kg/m^2 or higher.

Mean Plasma Levonorgestrel Concentrations (pg/mL) for All Subjects in the PK Substudy.

The Plasma LNG Levels in All Subjects With Continuing Use From Month 36 Through Month 120.

Plasma Levonorgestrel Concentrations for All Subjects Beginning at Month 36. All Mirena participants were discontinued by the end of Year 5.

The Changes in Endometrial Thickness Based on Transvaginal Ultrasonography at One Year, Five Years and Ten Years.

Endometrial thickness (ET) was evaluated in 52 LILETTA Efficacy Group subjects in the LNG20 (16-35 Yr Olds) arm only, with data both at Baseline and Month 12 (ET population) as part of a substudy conducted within the main protocol.

Analysis of residual LNG content from an appropriate sampling of IUSs that were removed or expelled during the study was to be summarized and plotted as a function of time of removal/expulsion. Residual drug content analysis was performed on a sampling of removed or expelled samples to determine the average in vivo drug release rate during the time the samples were implanted. All available samples were sorted and grouped by duration of use in 90-day intervals. The drug content and time of exposure of the samples analyzed were fit by an exponential regression to calculate the initial and average in vivo release rate over the duration of the study.

NOTE: Year = 360 day intervals

In this substudy, a sampling of removed or expelled LNG20 IUSs were evaluated for residual drug content from Day 1 through Year 8. The total number of IUSs analyzed was 130 units from a total of 130 participants. All IUSs came from participants in the LNG20 (16-35 Yr Olds) arm. IUSs from participants in the Mirena arm were not evaluated in this substudy.

From data based on the daily diary log, number of IUS subjects experiencing some bleeding and/or spotting.

Any incidences of bleeding and/or spotting were self-reported via daily diary log by remaining participants at regular intervals throughout the length of the study. All Mirena participants were discontinued by the end of Year 5.

Life table methods were used to estimate the cumulative pregnancy rate after each year. Confirmed pregnancy was the failure outcome and the corresponding time to pregnancy was the failure time. Time to pregnancy was calculated as the date a confirmed pregnancy was initially reported minus the date of IUS placement plus one.

Note: a year is considered 364 days (13 women-months). Only the efficacy population (LNG20 16-35 years old) was evaluated, per protocol.

The expulsion rate includes the number of subjects with reported expulsions or removal of the IUS, regardless of reason.

All Mirena participants were discontinued by end of Year 5 per the study protocol. All LNG20 (36-45 Yr Old) participants were discontinued by end of Year 8 per the study protocol.

Measuring the change in hemoglobin results from the screening visit and following 12 months, 60 months and 120 months of product use.

No standard deviation since there was only one Mirena participant measured at month 120

Time to event was calculated as the date of continuation/expulsion/removal minus the date of IUS placement plus one. Subjects without a continuation/expulsion/removal were censored at the last contact date and failure time was calculated as the last contact date minus the date of IUS placement plus one.

All Mirena participants were discontinued by end of Year 5 per the study protocol. All LNG20 (36-45 Yr Old) participants were discontinued by Year 8 per the study protocol.