The War on Mifepristone: How Junk Science and False Narratives Threaten US Abortion Access

Kelly Baden, Guttmacher Institute, Joerg Dreweke, Guttmacher Institute and Rachel K. Jones, Guttmacher Institute

False information and faulty science regarding the safety of the drug mifepristone, which is used in the vast majority of medication abortions in the United States, is at the center of a narrative that may further reduce abortion access. Mifepristone, and medication abortion more broadly, is safe, effective and widely used in the United States and globally. Despite its demonstrated safety, relentless policy and legal attacks aim to restrict or even ban access to this method of abortion.

This analysis will explore and counter several pernicious aspects of such attacks. These include the misrepresentation of normal signs a medication abortion is working as intended—for example, cramping or bleeding—as serious medical complications; and the conflation of routine or precautionary care-seeking by medication abortion patients with emergency treatment for serious adverse events. 

While false narratives about abortion have long been a premise of various attempts to restrict reproductive rights through litigation and legislation, their potential to cause devastating harm is greater than ever before. One major reason is the degree to which medication abortion has become the cornerstone of US abortion access, accounting for at least 63% of clinician-provided abortions in US states without total bans as of 2023. Another reason is the arrival of the second Trump administration, with anti-abortion officials now in key leadership roles at federal departments and agencies that could feature prominently in attempts to restrict mifepristone, including the Food and Drug Administration and the Department of Justice. Increasingly aggressive legal tactics by anti-abortion policymakers and activists further compound these dangers.

Mifepristone's Exemplary Safety Record

Mifepristone is backed by decades of data showing very low complication and mortality rates, supported by leading US and global medical authorities, widely used in the United States and globally, and proven safe when accessed through telehealth or when self-managing an abortion without the involvement of a provider.

Mifepristone was approved by the US Food and Drug Administration (FDA) in 2000 for use in combination with misoprostol to end early pregnancies. (Studies show misoprostol is also a safe and effective method of abortion when used on its own.) At the time, mifepristone had already been approved and was being used in other nations, and the US approval was preceded by a comprehensive review of studies on the drug’s safety and efficacy.

The scientific consensus regarding mifepristone’s safety is grounded in a number of factors. Over the past 25 years, medication abortion using mifepristone has been:

Opposition to Mifepristone Is Not Grounded in Science

The science on mifepristone’s safety and efficacy is settled. Still, as part of its long-term goal of banning all abortions, the anti-abortion movement—after failing to block the initial US approval of mifepristone—has worked in myriad ways to restrict access. One prong of this strategy is an ongoing effort to lend a veneer of scientific respectability to false narratives about the safety of medication abortion.

At the core of these narratives is a distorted depiction of people seeking care following medication abortion that falsely suggests they are experiencing serious adverse events. A successful medication abortion using the mifepristone–misoprostol regimen involves a range of symptoms and side effects. This can include heavy bleeding, cramping, pain, nausea, headache, dizziness and more. But people who experience more bleeding or clotting, or who have more pain or discomfort than they expected may seek care during or after the process to confirm that their symptoms are normal. People may also seek care after a medication abortion to confirm that they are no longer pregnant because pregnancy tests can continue to show false positives for several weeks after a successful abortion.

Well-conducted studies illustrate this reality. For example, an early study of 1,130 telehealth medication abortion patients found that 17% made an unplanned in-person health care visit related to the abortion, most commonly for bleeding, pain, or concerns about pregnancy test results. About 4% of participants ultimately received a procedural abortion, and findings from an earlier paper from this study found a hospitalization rate of less than 1%.

In contrast, several recent papers released by anti-abortion organizations claim to have found much higher rates of serious adverse events following medication abortion than had been previously established. At first glance, these papers paint an alarming picture about medication abortion, but two were subsequently retracted for serious methodological flaws by the journal that published them. A subsequent paper self-published by the anti-abortion Ethics and Public Policy Center (EPPC) has garnered widespread attention. Among other impacts, the EPPC paper has prompted the FDA, in line with demands from anti-abortion policymakers, to review evidence for the safety of mifepristone. Such a review is now underway, according to Department of Health and Human Services Secretary Robert F. Kennedy Jr., but the FDA has not provided a timeframe nor clinical standards for the review.

Science is iterative, and findings that go against the established scientific consensus are not necessarily wrong. But when a paper is a clear outlier in the literature, its findings, data sources, methodology, interpretation and conclusion require close scrutiny. This is where the papers questioning the safety of mifepristone fall far short. One egregious error, particularly in the EPPC study, is the repeated reference to the outcomes examined as “serious adverse events.” However, the FDA criteria for designation as a serious adverse event require one of seven outcomes: hospitalization, disability, permanent impairment or intervention to prevent permanent impairment, congenital anomaly/birth defect, death or some other life-threatening or serious event. The majority of outcomes in the EPCC paper do not meet these qualifications, as further detailed below. All of this underscores why these papers must not be relied upon for use in litigation, drug safety reviews or other policymaking processes. Major concerns about these papers include:

Categorizing ER visits during the medication abortion process as serious adverse events: Several anti-abortion papers, among them the self-published EPPC paper, erroneously characterize people seeking care at an emergency room or other health care setting during or after a medication abortion as a serious adverse event. However, an ER visit alone is not evidence of a complication. There are many reasons that someone who has had a medication abortion might present at an ER, including to confirm that they are no longer pregnant or for reassurance that the bleeding they are experiencing is normal. For many individuals without insurance or poor access to community care, the emergency department serves as their primary point of entry when seeking care for any of their health conditions or concerns. The conflation of ER visits with serious adverse events has become a signifier for shoddy science and a methodological flaw that also featured prominently in one of the anti-abortion papers retracted in 2024.

Presenting complications occurring at expected rates as new information: Because the safety record of mifepristone is well established, we also have a nuanced understanding of the serious adverse events that some patients may experience in very rare circumstances. The literature shows that serious complications—like sepsis, blood loss requiring a transfusion, or other complications requiring hospitalization—occur in well below 1% of all medication abortions using mifepristone. Deaths associated with medication abortion are not only exceedingly rare but reported figures cannot be causally linked to its use, and nearly half of the reported cases were evidently entirely unrelated (e.g., cases of homicide, drug overdose or suicide that occurred within a given period after medication abortion but were not caused by it). Still, the anti-abortion movement has routinely portrayed the rare serious adverse events that have been documented within mifepristone’s overall exemplary safety record as evidence that medication abortion is unsafe.

Papers questioning mifepristone’s safety have also adopted this tactic, framing, for instance, the long-established (and low) rate of patients needing a blood transfusion after medication abortion as new information that supposedly undermines mifepristone’s safety record. That record clearly shows that patient safety is paramount for abortion providers, but by selectively framing long-established data as new and alarming results, these papers contribute to a culture of fear and stigma around abortion at a time when reproductive rights are under attack.

Characterizing incomplete abortion as a serious adverse event: Medication abortion is highly effective, but like most medical treatments, it does not work 100% of the time. It is normal and expected that for some people who have a medication abortion the pregnancy may not fully pass or may take longer than desired by the patient. Additional rounds of misoprostol or a procedural abortion are both safe and effective ways to complete the process. Treatment for incomplete abortion, which the scientific literature suggests is needed in about 3–5% of medication abortions, is part of standard medical practice and does not indicate a serious adverse event—just a different path to the same safe outcome. However, patients seeking and obtaining care for an incomplete abortion at rates long established by scientific literature has been falsely portrayed as a pattern of serious adverse events by some anti-abortion papers. Instead, such data is further evidence that medication abortion is highly effective and results in a complete abortion in upward of 95% of cases.

Including unrelated conditions as serious adverse events: This common flaw in anti-abortion papers inflates the number of serious adverse events by including medical conditions or events not related to the use of mifepristone. For instance, the self-published EPPC paper lists ectopic pregnancy as a serious adverse event resulting from medication abortion, even though mifepristone use does not cause an ectopic pregnancy and does not cause harm if taken by someone with an ectopic pregnancy. The same paper also lists health issues such as cardiac and pulmonary events as resulting from the use of mifepristone, without any indication that they were causally related to having a medication abortion. One study, which has since been retracted, asserted that any ER visit for any reason within 30 days of an abortion was to treat an abortion complication.

Obscuring research methods to hinder reproducibility: In some cases, the authors of papers questioning the safety of mifepristone have made it difficult to fully evaluate the validity of their findings by withholding specifics about their methodology. In the case of the self-published EPPC paper, this includes failing to name the dataset they purchased and not detailing the medical diagnosis and treatment codes used to conduct the analysis—information that is commonly disclosed in studies so that other researchers can attempt to replicate and build on the findings. But neither the methodology section of the EPPC paper itself nor separate fact sheet and frequently asked questions documents included these codes. Such opacity about research methods would likely have been flagged by peer reviewers and could have resulted in the paper being rejected by a high-quality scientific journal.

Reaching conclusions not supported by the findings: A common feature of poorly constructed papers is that the authors’ policy recommendations do not align with their findings. For instance, the policy implications section of the EPPC paper states that mifepristone should be provided only by physicians, but the paper does not present information indicating that outcomes differed by provider type (e.g., physicians versus other qualified providers, like advanced practice clinicians). Likewise, the EPPC authors recommend that mifepristone be provided only in person, but the paper does not distinguish how individuals obtained mifepristone or whether outcomes differed for in-person versus telehealth provision. Indeed, mifepristone was for the most part only available via in-person office visits until 2020, and the study did not examine changes in serious adverse events after that point. Additionally, the authors state that mifepristone should be provided only for pregnancies up to seven weeks, but the paper does not examine outcomes by gestational duration.

Selectively emphasizing “real-world” data over clinical trials: Another dubious tactic employed in anti-abortion papers is their insistence that claims of higher mifepristone complication rates are based on “real-world” (and therefore superior) data, as opposed to the more controlled conditions of clinical trials. However, this tactic ignores the body of real-world evidence confirming the safety and effectiveness of mifepristone—with results that mirror those of clinical trials. For instance, a peer-reviewed study using real-world claims data for 54,911 abortions reimbursed through California’s Medicaid program found rates of major complications for medication abortion comparable to those of clinical trials (0.31%). Another study using linked claims data from Ontario Canada’s single-payer health system found fewer than six adverse events among 25,744 medication abortions and a serious complication rate of 0.76%; 5% of patients had a subsequent uterine evacuation to address an incomplete abortion.

At the same time, using real-world data is not a guarantor of sound science. One anti-abortion study that used such data was retracted, while another by EPPC has serious methodological flaws that render its findings unreliable. Therefore, rather than privileging one type of study over another, research evaluation should focus on whether a study’s methodology meets the standards of rigorous scientific evidence.

Science Over Stigma: The Case for Patient-Centered Abortion Policy

Seeking care after a medication abortion—even in the absence of serious complications—is a normal and expected part of the process for some patients. Unfortunately, faulty science and abortion stigma have distorted this reality, framing routine care seeking by some patients as evidence of risk and using it to justify further restrictions—or even outright bans.

While most individuals can determine on their own whether their abortion is complete or if complications are present, abortion stigma has contributed to a perception that abortion poses acute medical risks. For example, bleeding is a necessary outcome of medication abortion and a sign that it is working effectively, but can nevertheless be alarming for some patients, especially those who have been exposed to misinformation about the safety of abortion. It is possible that expanding patient education and providing accurate information about what to expect during a medication abortion could alleviate some of these concerns. But just as importantly, medication abortion patients need access to appropriate and supportive sources of care before, during and after the process. These needs may be particularly acute for residents of ban states who accessed medication abortion by mail after a virtual visit with a health care provider or returned home after accessing in-person care in a state where abortion is less restricted. Fears of potential criminalization are justified for these and other patients, since reports by health care providers are the most common way that people accused of self-managed abortion come to the attention of law enforcement.

As more people access abortion via telehealth—which by the end of 2024 accounted for one in four abortions in the United States—and many others travel long distances for care, clear guidance on what to expect becomes even more important. Everyone accessing medication abortion—regardless of how or where they receive care—deserves clear, accurate information and evidence-based care, free from stigma and bias. Patients should be able to safely access abortion-related care at any point in the process, for any reason, without fear of criminalization or of having their concerns weaponized to stigmatize and restrict abortion access.

To push back against false narratives and distortions, it is important to highlight the overwhelming scientific consensus: medication abortion is safe, effective, and commonly used around the world. Efforts to conflate normal patient behavior with medical emergencies are not rooted in evidence but in ideology. Reclaiming the narrative means establishing once and for all that seeking care—whether for reassurance, information, or support—is not a sign of danger, but a reflection of patient agency and responsible health care. Countering faulty science in this context is not only about correcting the record; it is about protecting access to essential care and defending the integrity of evidence-based medicine.

Acknowledgments

The authors thank Kirsten Moore of the EMAA (Expanding Medication Abortion Access) Project and Ushma Upadhyay of ANSIRH (Advancing New Standards in Reproductive Health) at the University of California, San Francisco, for reviewing a draft of this analysis. The views expressed here are those of the authors and the Guttmacher Institute. This analysis was edited by Ian Lague.

Source URL: https://www.guttmacher.org/2025/10/war-mifepristone-how-junk-science-and-false-narratives-threaten-us-abortion-access