Background
Previous emergency contraception studies have excluded women who report >1 episode of unprotected or underprotected intercourse. Thus, clinical recommendations are based on exposure to a single episode of underprotected intercourse.
Objective
We sought to assess the prevalence and timing of underprotected intercourse episodes among women requesting emergency contraception and to examine the probability of pregnancy following an emergency contraception regimen including placement of either a copper intrauterine device or a levonorgestrel intrauterine device with simultaneous administration of an oral levonorgestrel pill in women reporting multiple underprotected intercourse episodes, including episodes beyond the Food and Drug Administration–approved emergency contraception time frame (6-14 days).
Study Design
Women seeking emergency contraception who had a negative pregnancy test and desired either a copper intrauterine device or levonorgestrel emergency contraception regimen enrolled in this prospective observational study. At enrollment, participants reported the number and timing of underprotected intercourse episodes in the previous 14 days. Two weeks later, participants reported the results of a self-administered home pregnancy test.
Results
Of the 176 women who presented for emergency contraception and received a same-day intrauterine device, 43% (n = 76) reported multiple underprotected intercourse episodes in the 14 days prior to presenting for emergency contraception. Women with multiple underprotected intercourse episodes reported a median of 3 events (range 2-20). Two-week pregnancy data were available for 172 (98%) participants. Only 1 participant had a positive pregnancy test. Pregnancy occurred in 0 of 97 (0%; 95% confidence interval, 0–3.7%) women with a single underprotected intercourse episode and 1 of 75 (1.3%; 95% confidence interval, 0–7.2%) women reporting multiple underprotected intercourse episodes; this includes 1 of 40 (2.5%; 95% confidence interval, 0–13.2%) women reporting underprotected intercourse 6-14 days prior to intrauterine device insertion.
Conclusion
Women seeking emergency contraception from clinics commonly reported multiple recent underprotected intercourse episodes, including episodes occurring beyond the Food and Drug Administration–approved emergency contraception time frame. However, the probability of pregnancy was low following same-day intrauterine device placement.
Introduction
Previous emergency contraception (EC) studies have excluded women who report >1 recent episode of unprotected or underprotected intercourse (UPI). However, women may experience multiple UPI episodes before presenting for EC, including episodes outside of the Food and Drug Administration–approved time frame for efficacy of EC options. The risk of luteal phase pregnancies may increase when UPI episodes occur >5 days prior to EC use or contraceptive initiation ; thus, patients’ report of having UPI >5 days beforehand may cause clinicians to delay contraception. This delay can place women at increased risk of unintended pregnancy.
Three forms of EC are currently available in the United States: oral levonorgestrel (LNG), oral ulipristal acetate, and the copper intrauterine device (IUD). All of these methods of EC have demonstrated efficacy within 120 hours of UPI, with failure rates ranging from 0.09% for the copper IUD to 2.6% for oral LNG. Predictors of oral EC failure include higher body mass index, sex during the fertile window (from 5 days before ovulation to the day of ovulation), and repeat UPI after using EC. Although repeated use of EC pills as a pericoital method of contraception has been investigated, there are minimal data on single use of EC following multiple UPI episodes in the same menstrual cycle to inform clinical recommendations for this population.
Our primary aim was to determine the prevalence of multiple UPI episodes within the 14 days prior to presenting to a family planning clinic for EC. Our secondary aim was to provide preliminary data on pregnancy rates following EC regimens that included IUD placement for women who reported multiple UPI episodes, including UPI between 6-14 days prior to receiving an IUD EC regimen.
Materials and Methods
We conducted a prospective observational study of women presenting for EC at a single family planning clinic in Salt Lake City, UT. The original study aimed to evaluate EC user interest in and the efficacy of oral LNG plus a LNG IUD for EC. Eligible women were 18-35 years old, had at least 1 UPI episode in the past 120 hours, had a negative urine pregnancy test, were fluent in English or Spanish, and agreed to comply with study guidelines. Women who reported multiple UPI episodes and UPI >120 hours were included in the study population. Women with a positive urine pregnancy test; who were currently breast-feeding; who had vaginal bleeding of unknown etiology or current intrauterine infection; who had untreated Neisseria gonorrhea or Chlamydia trachomatis infection; or who had known abnormalities of the uterine cavity such as bicornuate uterus, uterine didelphys, or a uterine leiomyoma impinging on the uterine cavity were excluded from participation. We obtained data on oral EC use after the last sexual encounter prior to study enrollment. Further details regarding the original study have been previously published. Trained study staff obtained informed consent from all participants, and skilled clinicians provided either the copper IUD or the LNG + EC regimen as requested by the study participant.
At the time of enrollment, we asked participants, “How many times in the last 2 weeks have you had sex when you did not use a method of birth control or were worried that the method you used did not work?” Participants then used a calendar to identify on which of the last 14 days they had either: (a) unprotected intercourse, defined as sex “with no method of birth control,” or (b) UPI, defined as sex in which they had “used a method where you were worried you might get pregnant (for example: broken condom, missed birth control pills, etc.).” We defined unprotected intercourse/UPI episodes as the sum of both categories. Prior to leaving the enrollment clinic, participants were given a high-sensitivity qualitative urine pregnancy test (Osom card pregnancy test, Sekisui Diagnostics LLC, Lexington, MA) (negative result with human chorionic gonadotropin ≤20 IU/L) and instructed on how to complete home urine pregnancy testing. Two weeks after presenting for EC and receiving a same-day IUD, participants completed a home pregnancy test and reported results via telephone or World Wide Web–based survey tools.
The primary outcomes of this study include: (1) the number of UPI episodes in the 14 days before IUD placement; and (2) the pregnancy rate following IUD EC regimens for women reporting multiple UPI events prior to obtaining EC, including those reporting UPI 6-14 days prior to obtaining EC. The primary analytic cohort included women who had a successful IUD insertion and who reported timing of intercourse. We compared the characteristics of women who reported a single UPI episode and those who reported multiple UPI episodes using χ 2 tests. We calculated point estimates and 95% confidence intervals (CI) for pregnancy rates using Clopper-Pearson binomial exact CI. We conducted all data analysis with statistical software (Stata 13; StataCorp LP, College Station, TX). The sample size was determined by the primary goal of the main study, to provide a point estimate for the risk of EC failure with 2-week outcome data from 100 evaluable LNG users. The University of Utah Institutional Review Board approved this investigational protocol ( ClinicalTrials.gov identifier: NCT01963962 ).
Materials and Methods
We conducted a prospective observational study of women presenting for EC at a single family planning clinic in Salt Lake City, UT. The original study aimed to evaluate EC user interest in and the efficacy of oral LNG plus a LNG IUD for EC. Eligible women were 18-35 years old, had at least 1 UPI episode in the past 120 hours, had a negative urine pregnancy test, were fluent in English or Spanish, and agreed to comply with study guidelines. Women who reported multiple UPI episodes and UPI >120 hours were included in the study population. Women with a positive urine pregnancy test; who were currently breast-feeding; who had vaginal bleeding of unknown etiology or current intrauterine infection; who had untreated Neisseria gonorrhea or Chlamydia trachomatis infection; or who had known abnormalities of the uterine cavity such as bicornuate uterus, uterine didelphys, or a uterine leiomyoma impinging on the uterine cavity were excluded from participation. We obtained data on oral EC use after the last sexual encounter prior to study enrollment. Further details regarding the original study have been previously published. Trained study staff obtained informed consent from all participants, and skilled clinicians provided either the copper IUD or the LNG + EC regimen as requested by the study participant.
At the time of enrollment, we asked participants, “How many times in the last 2 weeks have you had sex when you did not use a method of birth control or were worried that the method you used did not work?” Participants then used a calendar to identify on which of the last 14 days they had either: (a) unprotected intercourse, defined as sex “with no method of birth control,” or (b) UPI, defined as sex in which they had “used a method where you were worried you might get pregnant (for example: broken condom, missed birth control pills, etc.).” We defined unprotected intercourse/UPI episodes as the sum of both categories. Prior to leaving the enrollment clinic, participants were given a high-sensitivity qualitative urine pregnancy test (Osom card pregnancy test, Sekisui Diagnostics LLC, Lexington, MA) (negative result with human chorionic gonadotropin ≤20 IU/L) and instructed on how to complete home urine pregnancy testing. Two weeks after presenting for EC and receiving a same-day IUD, participants completed a home pregnancy test and reported results via telephone or World Wide Web–based survey tools.
The primary outcomes of this study include: (1) the number of UPI episodes in the 14 days before IUD placement; and (2) the pregnancy rate following IUD EC regimens for women reporting multiple UPI events prior to obtaining EC, including those reporting UPI 6-14 days prior to obtaining EC. The primary analytic cohort included women who had a successful IUD insertion and who reported timing of intercourse. We compared the characteristics of women who reported a single UPI episode and those who reported multiple UPI episodes using χ 2 tests. We calculated point estimates and 95% confidence intervals (CI) for pregnancy rates using Clopper-Pearson binomial exact CI. We conducted all data analysis with statistical software (Stata 13; StataCorp LP, College Station, TX). The sample size was determined by the primary goal of the main study, to provide a point estimate for the risk of EC failure with 2-week outcome data from 100 evaluable LNG users. The University of Utah Institutional Review Board approved this investigational protocol ( ClinicalTrials.gov identifier: NCT01963962 ).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
