Placing Emergency Contraception in the Hands of Women

For more than 50 years, since the successful synthesis of estrogens and progestins, safe and effective pregnancy prevention has been possible. Nonetheless, in the United States an estimated 3.5 million unwanted pregnancies occur annually, one third of which involve teenagers.¹ Among the many possible explanations for this paradox, barriers to health care figure prominently. For instance, lack of health insurance, an issue for more than 44 million Americans, creates a formidable barrier to access. Religious beliefs, concerns about contraceptive safety, and psychological barriers also contribute. Contraception may not be sought, especially by teens and other women for whom a physician visit and discussion of sexual behavior may prove embarrassing.

Women without access to effective contraception were enthusiastic about the discovery that certain formulations of contraceptive hormones or hormone receptor inhibitors could prevent implantation of a fertilized ovum. “Morning after” or “emergency contraception” (EC) provided these women a second chance to prevent an unwanted pregnancy. When taken within 72 hours of an episode of unprotected intercourse, the “Yuzpe” regimen (consisting of 2 doses of combined oral contraceptive pills, with the first taken within 72 hours of unprotected intercourse and the second, 12 hours later [a total of 100-120 mg of estrogen and 1.0-1.2 mg of progestin, depending on the formulation of the oral contraceptive pill]) is effective in preventing pregnancy in 75% of cases.² As with regular contraception, however, use of these agents has been less frequent than anticipated, and several other factors appear to restrict EC use. For example, patient knowledge of these agents is limited and some physicians are reluctant to prescribe them citing fears that their availability will lead to increased risk-taking behavior, especially among teens.

Some countries, including the United Kingdom, have removed some real and potential barriers to EC use by making these drugs available over-the-counter. This approach precludes the need for insurance, offers timely access, and eliminates the need for a physician visit for women who would find seeking EC embarrassing or otherwise difficult to accomplish. Indeed, studies have demonstrated that the rate of EC use is doubled when it is provided in advance of need.³ In the United States, 6 states (Washington and Maine, with pilot programs in Alaska, California, New Mexico, and Hawaii) have legislatively sanctioned pharmacy access to EC without needing a physician’s prescription,^{4 - 5} but the effects of this availability on clinical outcomes and rates of use have not yet been evaluated.

In this issue of JAMA, Raine and colleagues⁶ report important data on the effect of access to EC on clinical outcomes. In their study, 2117 young women 15 to 24 years of age were randomly assigned to pharmacy access without a prescription, advance provision, or usual care that required a clinic visit to obtain EC. The authors assessed the effects of increased access on pregnancy rates, acquisition of new sexually transmitted infections (STIs), contraceptive use, and sexual behaviors. Strengths of this study include the large number of women enrolled and the use of biomarkers, rather than self-report, for most of the outcome measures, including tests for diagnosis of Chlamydia trachomatis and herpes simplex virus type 2 and pregnancy tests.

After 6 months of follow-up, the authors documented a near doubling of EC use in the advance access group (37.4%) relative to usual care (21.0%) and comparable rates of use for pharmacy access and usual care (24.2% and 21.0%, respectively). Pregnancy rates were comparable in all groups as was the incidence of new STIs. Easier access to EC did not compromise regular contraceptive use or lead to an increase in risky sexual behaviors.

Limitations of the study include self-reported use of EC rather than performance of bioassays of hormone metabolites to verify use. Moreover, the women appear to have self-selected by presenting for care at clinics that provide family planning services. Although women requesting EC were excluded, the authors do not report whether contraception was the reason for their visit. That notwithstanding, the very fact these women sought health care suggests that they are more sophisticated than many young women in need of pregnancy protection. Accordingly, the findings of this study cannot be generalized to a nonselected group of young women or to youth in states with more restrictive provision of contraception to minors.

Another limitation of the study is the loss to follow-up of almost 10% of the inception cohort. This raises questions of differential outcomes in what is likely to have been a higher risk group of patients. Moreover, the short, 6-month follow-up may not have been sufficient to detect a reduction in pregnancies, a limitation acknowledged by the researchers. An unexplained result is less use of EC during the study than reported lifetime use, despite the facilitated access provided in the study.

While Raine et al have provided new and important information on the use of EC in this population of high-risk young women, it is unfortunate that the study population was not stratified by age and ethnic group. For example, matching on age, rather than analyzing age as a continuous variable, might have provided important information about the effect of the intervention on younger adolescents who are at highest risk of unprotected intercourse and unwanted pregnancy. Analyses that controlled for past use of EC also would have been of interest. It is possible that prior experience with EC may have influenced compliance as well as perceived adverse effects. Further, experience with adverse effects may have influenced use across the entire study, and exploration of this variable may have added important information to what is already known about the frequency of nausea and vomiting in patients using EC.⁷

Despite these limitations, the report by Raine et al makes an important contribution to current knowledge of EC use and should help dispel concerns that easier access to EC increases the risk of STIs or leads to abandonment of regular contraception. The finding that women who were provided EC at a clinic visit prior to the time of need were almost twice as likely to use EC than women in the clinic access control group is critically important, even though a reduction in pregnancy was not demonstrated. Several explanations for this finding, which could not be assessed within the context of the study, include timing of unprotected intercourse in relationship to ovulation, as well as the timing and accuracy of use of EC. The fact remains that women are more likely to use EC if it is readily available. It is similarly significant that no apparent downside of EC was demonstrated in the study. There was no evidence of a decrement in use of other methods of contraception such as oral contraceptives or condoms. Equally important, to counter the frequent argument that provision of EC will increase high-risk behavior, this study demonstrated no increase in STIs or reports of unprotected intercourse.

The task remains to improve access to EC by expanding the number and types of facilities where it can be obtained. Education about EC for both physicians and the public should be improved, especially now that other agents (including RU-486 [mifepristone]) have been shown to be effective in preventing unwanted pregnancy after an episode of unprotected intercourse.⁴ Importantly, physicians may wish to reconsider their position on EC in relation to other contraceptive methods; as this study suggests, regular contraception and EC are not in conflict. Sexually active women who do not desire to become pregnant should be counseled about regular methods of birth control, but they can be assured that EC is available if they are concerned their regular method may not have protected them from the possibility of an unwanted pregnancy.