Abstract
It is estimated that the risk of pregnancy after a single act of unprotected intercourse may be reduced by 75% with the use of emergency contraception (EC). Studies have demonstrated that the rate of EC use doubles when it is provided ahead of need, and perhaps half of unintended pregnancies each year might be averted if EC were easily accessible. 1
Concerns that easier access would reduce regular contraceptive use and/or increase sexually transmitted infections (STIs) allowed the Food and Drug Administration to reject an application to switch levonorgestrel EC to over-the-counter status. However, to date there has been little or no evidence to challenge or support such fears.
Raine et al. conducted a single-blind, six-month trial at four clinics in one California city on sexually active women, aged 15 to 24 years, at risk of unintended pregnancy (n = 2117). Primary outcomes were EC use, rate of pregnancy, and STIs at six months. Secondary outcomes were changes in contraceptive use and sexual behaviours. 1 California is one of six states which allow access to EC through pharmacies. In Canada, British Columbia, Saskatchewan, and Quebec permit pharmacists to prescribe EC.
Patients were randomized to either advance provision of three packets of EC (n = 884), pharmacy access to 13 pharmacies to obtain EC at no cost when needed (n = 889), or the control group, clinic access to EC when needed (n = 344). No matter the group, EC offered to all patients consisted of two doses of 0.75 mg of levonorgestrel, 12 hours apart within 72 hours.
Baseline assessment included a pregnancy test and a questionnaire about contraceptive use, sexual behaviours, and STI testing. Women were excluded from the study if they used long-acting contraception such as intrauterine devices, had recently sought EC, had had unprotected intercourse within 72 hours, or were pregnant. Mean age of the patients was 19.9 years; half of the study participants were teenagers. While half of the sample used oral contraceptives, half used condoms as birth control.
Follow-up at six months, using chart reviews and questionnaires, assessed EC use, pregnancy rate, sexual behaviours, and STIs. Participants were tested for pregnancy and STIs. About 10% in each group were lost to follow-up. Analysis was by modified intent-to-treat and accounted for clinic site. Mean follow-up was 6.9 months.
Overall, 29% of the women used EC. Compared with the control group, women in the pharmacy access group were no more likely to use EC (24.2% vs 21 %; p = 0.25). Women in the advance provision group were significantly more likely to use EC at least once, as were women in the clinic access group (37.4% vs 21 %; p <0.001), even though the frequency of unprotected intercourse was similar (39.8% vs 41 %; p = 0.46).
During the study, 37.5% reported having unprotected intercourse, not significantly different among the three groups; 46.7% of participants who had unprotected intercourse used EC and 7.7% became pregnant. Compared with the control group, the pharmacy access and advance provision groups did not have a significant reduction in pregnancy rate (pharmacy access group, p = 0.93; advance provision group, p = 0.71).
There were no differences in patterns of contraceptive use, sexual risk behaviours (numbers of partners, sexual frequency), and rate of STIs (12% acquired an STI during the trial).
Study limitations include crossover of treatment groups and loss of patients to follow-up. Relatively high EC use in the clinic access group, as compared with other studies, may explain the lack of difference between the control and pharmacy groups. Strengths include the study size and not relying on self-reporting alone for results.
An accompanying editorial recommends that, despite trial limitations, access to EC be improved and awareness of it increased. 2
In conclusion, women given easy access to EC use it more than women with restricted access. However, EC availability did not change sexual behaviours or use of other contraceptive methods.