Comparison of Oral Contraceptive Therapies

Berlin, Germany—A database analysis of insured women found that those who took 84 days of levonorgestrel/ethinyl estradiol plus ethinyl estradiol 0.01 mg for 7 days were significantly less likely to become pregnant compared with those who took 84 days of levonorgestrel/ethinyl estradiol plus placebo for 7 days.

The results were presented during a poster session at the ISPOR European Congress. The poster was titled Comparison of Unintended Pregnancy among Women who Initiated Oral Contraceptive Therapy with an 84/7 Estradiol or an 84/7 Placebo.

The authors noted that 28% of women who use contraception in the United States use oral contraceptives primarily. Many take 84 days of levonorgestrel/ethinyl estradiol followed by 7 days of ethinyl estradiol or placebo because they prefer to control the timing of their menstrual cycles and menstruating less than once per month, according to the authors. They cited a study that indicated 25% of women believed menstruation had a negative impact on important parts of their lives. In another study, those who took ethinyl estradiol experienced fewer days of scheduled and unscheduled bleeding compared with a placebo group.

In this examination, the authors obtained information from the de-identified US i3 Invision^™ database, which includes patient-level demographics as well as inpatient, outpatient, and prescription drug claims data. The authors collected data from May 1, 2009, through December 31, 2011.

Patients were included if they were 15 to 40 years of age, had continuous insurance coverage, and filled a prescription for 84 days of levonorgestrel/ethinyl estradiol plus ethinyl estradiol 0.01 mg for 7 days or 84 days of levonorgestrel/ethinyl estradiol plus placebo for 7 days. The authors matched the 0.01-mg ethinyl estradiol group in a 2:1 ratio with the placebo group based on age, region of residence, insurance, business type, insurance product type, and year of the index date.

The mean age of patients was approximately 27 years, the most common region was the South (where 55% of patients lived), and the most common insurance type was a point of service plan (81% of patients).

There was an 8% increase in the medication possession ratio in the 0.01-mg ethinyl estradiol group compared with the placebo group, which represented a statistically significant difference (P<.0001). The researchers hypothesized that a reduction in bleeding in the 0.01-mg ethinyl estradiol group may have been associated with an improvement in adherence to therapy.

One year after initiating therapy, 3.0% of patients who took 0.01 mg of ethinyl estradiol became pregnant compared with 4.5% of patients in the placebo group (P<.0001).

The authors cited a few limitations to the study. They analyzed information from a claims database and only examined patients with medical and outpatient prescription benefit coverage, so the results may not be consistent in other populations. They also questioned the reliability of identifying patients based on diagnostic codes instead of clinical assessments. Further, they indicated the database could not distinguish between unplanned and planned pregnancies, and they could not determine the factors that predicted pregnancy because it was a descriptive study.

In the future, the authors said they plan to analyze other factors associated with pregnancy and compare side effects of oral contraceptives.