Objectives: To estimate the incidence of ectopic pregnancy (EP) associated with current prescription contraceptive use Methods: We performed a retrospective cohort study of women aged 15-44 years enrolled at Kaiser Permanente Northern and Southern California from 2009 to 2019. We identified EPs and prescription contraceptive use based on validated algorithms using diagnosis, procedural, and medication codes abstracted from electronic health records. Contraceptive use was updated as women started, stopped, or changed methods. Contraceptive pills, patches, and rings were combined in one category. EPs that were identified at least 14 days after starting, during, or within 42 days of stopping a method were attributed to that method. EP Incidence rates and 95% confidence intervals (CI) were estimated per 10,000 woman-years overall and for each contraceptive category or when no prescription method was used. Results: There were 12,472 EPs among 3,380,648 women for an overall EP incidence rate of 9.7 per 10,000 woman-years. The vast majority of EPs (n= 10,435, 83.9%) occurred when women used no prescription contraceptive for a rate of 11.3 (CI 11.1-11.5). The incidence rate was lower with any current contraceptive use except for the progesterone-only pill (17.1 (CI 14.9-19.6)). It was lowest with depot medroxyprogesterone acetate (2.2 (CI 1.6-3.0)) and levonorgestrel Intrauterine devices (IUD) (3.9 (CI 3.5-4.4)). The incidence was 8.5 (CI 7.4-9.8) for the copper IUD and 8.4 (CI 7.0-10.0) for IUDs of unknown type. Rates were similar for pill, patch, or ring (5.2[4.9-5.5] and contraceptive implants (5.1 (CI 4.2-6.2)). Conclusions: The absolute risk of EP is low. Prescription contraceptive use is protective of EP, but the risk is higher for the progesterone-only pill.
