Safety & Effectiveness of Mifepristone for Medication Abortion in a National Claims Cohort, 2017-2023
- 1Department of Obstetrics, Gynecology, and Reproductive Sciences, University of California, San Francisco, San Francisco, California, USA
Abstract
BACKGROUNDExtensive research has documented medication abortion safety. Large-scale insurance claims data offer an important complement to clinical studies by assessing real-world outcomes in routine practice. In claims data, researchers must infer mifepristone use, clinical outcomes, and event severity from diagnosis, treatment, and utilization codes. Understanding how these analytic choices influence safety and effectiveness estimates is important for interpreting evidence used to inform care delivery and medication policy.
METHODSWe conducted a retrospective cohort study of mifepristone use with national multipayer claims data, both medical and pharmacy, from 2017 through 2023. We identified both a broad cohort and a more specific medication-abortion cohort after excluding evidence of other indications. We classified potential safety events, effectiveness outcomes, and health care utilization within 45 days of mifepristone use. Serious adverse events required claims-observable consequences consistent with FDA seriousness criteria.
RESULTSThe broad cohort included 869,409 mifepristone uses among 698,780 unique patients for medication abortion or other indications. After restriction to uses with additional evidence of medication abortion, 775,931 uses among 626,443 unique patients remained. In the medication-abortion cohort, 3,306 uses were followed by at least one serious adverse event, resulting in an absolute 45-day risk of 0.43%. Serious bleeding or hemorrhage occurred after 0.24% of uses, serious infection or sepsis after 0.05%, serious hypersensitivity or anaphylaxis after 0.01%, and serious ectopic pregnancy outcomes after 0.05%.
CONCLUSIONSSerious adverse events after mifepristone use were rare in this national study. Applying established, clinically grounded criteria markedly reduced the number of events classified as serious.
Related
- Mifepristone Patient Agreement Form and Medication Abortion Knowledge: A Randomized Clinical Trial.
- E-Cervix Assessment of Cervical Softening During Medication Abortion: A Single-Center Observational Study.
- Telehealth vs In-Clinic Medication Abortion Provision
- Effectiveness of a pharmacy-led personalised intervention program aimed at improving adherence to oral diabetes and/or hypertension medication in people with type 2 diabetes
- Women's experiences of emergency post-abortion care at Kawempe National Referral Hospital, Uganda: a qualitative descriptive study
BibTeX
@misc{woskie2026safety,
title = {{Safety & Effectiveness of Mifepristone for Medication Abortion in a National Claims Cohort, 2017-2023}},
author = {L. R. Woskie and J. Strasser and R. Griffin and Q. Luo and M. I. Rodriguez and U. Upadhyay},
howpublished = {medRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.02.26364611},
url = {https://www.medrxiv.org/content/10.64898/2026.10.02.26364611v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Safety & Effectiveness of Mifepristone for Medication Abortion in a National Claims Cohort, 2017-2023
AU - L. R. Woskie
AU - J. Strasser
AU - R. Griffin
AU - Q. Luo
AU - M. I. Rodriguez
AU - U. Upadhyay
PY - 2026
DA - 2026/10/02
PB - medRxiv
DO - 10.64898/2026.10.02.26364611
UR - https://www.medrxiv.org/content/10.64898/2026.10.02.26364611v1
AB - BACKGROUNDExtensive research has documented medication abortion safety. Large-scale insurance claims data offer an important complement to clinical studies by assessing real-world outcomes in routine practice. In claims data, researchers must infer mifepristone use, clinical outcomes, and event severity from diagnosis, treatment, and utilization codes. Understanding how these analytic choices influence safety and effectiveness estimates is important for interpreting evidence used to inform care delivery and medication policy. METHODSWe conducted a retrospective cohort study of mifepristone use with national multipayer claims data, both medical and pharmacy, from 2017 through 2023. We identified both a broad cohort and a more specific medication-abortion cohort after excluding evidence of other indications. We classified potential safety events, effectiveness outcomes, and health care utilization within 45 days of mifepristone use. Serious adverse events required claims-observable consequences consistent with FDA seriousness criteria. RESULTSThe broad cohort included 869,409 mifepristone uses among 698,780 unique patients for medication abortion or other indications. After restriction to uses with additional evidence of medication abortion, 775,931 uses among 626,443 unique patients remained. In the medication-abortion cohort, 3,306 uses were followed by at least one serious adverse event, resulting in an absolute 45-day risk of 0.43%. Serious bleeding or hemorrhage occurred after 0.24% of uses, serious infection or sepsis after 0.05%, serious hypersensitivity or anaphylaxis after 0.01%, and serious ectopic pregnancy outcomes after 0.05%. CONCLUSIONSSerious adverse events after mifepristone use were rare in this national study. Applying established, clinically grounded criteria markedly reduced the number of events classified as serious.
ER -