Originally published on Health Affairs Blog.

The U.S. abortion landscape is changing rapidly. Large swaths of the country are enacting ever more extreme abortion restrictions, while a number of states are racing to protect or even expand access. In 2020, the conservative-leaning U.S. Supreme Court will consider its first major abortion rights case since Justices Neil Gorsuch and Brett Kavanaugh were confirmed, and additional cases are at the Court’s doorstep. And all the while, the U.S. abortion rate continues to decline: According to a September report from the Guttmacher Institute, the abortion rate has reached a record low, with concurrent declines in birthrates suggesting that fewer people are becoming pregnant in the first place.

Against this backdrop, another set of questions keeps coming up in media reports and policy debates: What do we know about the extent to which people are self-managing abortion? Is it happening commonly enough to offset a significant portion of the reported decline in abortion provided in clinical settings? And will its role expand, in part as a response to changes in abortion policy and access?

What Is Self-Managed Abortion?

Self-managed abortion refers to the practice of ending a pregnancy without the formal supervision of a health care professional. While self-managed abortion is not new, medication abortion has changed how we think about it by offering a method that has proven to be simple, safe and effective. As a result, self-managed abortion using abortion pills outside of a clinical setting is increasingly discussed by advocates as an option that enhances reproductive autonomy—a core principle that should guide all discussions about reproductive health policy—while also potentially serving as a work-around to restrictive policies and gaps in access.

One option for self-managed medication abortion is misoprostol, one of two drugs used in the Food and Drug Administration (FDA)-approved regimen for medication abortion. The drug is up to 85% effective at ending a pregnancy when used on its own. Because it was designed and approved for purposes other than abortion, it is available by prescription in the United States. And, in some countries, including Mexico, the drug can be purchased without a prescription.

A second option is the FDA-approved two-drug protocol, involving both misoprostol and mifepristone. Despite an extensive safety record, that second drug, mifepristone, is heavily restricted in the United States, making misoprostol on its own the more readily available option. However, new websites offering the two-drug protocol are changing that situation and may be expanding informal access to self-managed medication abortion in the United States.  

Measuring Self-Managed Abortion

It is difficult to accurately gauge the extent to which people are self-managing abortion, precisely because it occurs outside of the formal health care system. The Guttmacher Institute’s own state and national abortion data, updated every three years, only capture abortions that occur in clinical settings—hospitals, clinics and physicians’ offices. Nonetheless, research on this practice in the United States extends back several decades.

Most of the studies in this area measure lifetime incidence of self-managed abortion—in other words, whether the women participating in the studies had ever attempted to self-manage at any point over the course of their lives. For example:

  • A study conducted in 1999 found that 37% of patients at obstetrics-gynecology clinics in New York City were familiar with misoprostol as an abortifacient, and 5% reported personal use of the drug for this purpose.
  • Researchers collecting information from patients at 1,425 obstetrics-gynecology facilities in three U.S. cities in 2008–2009 found that 4% had ever attempted to self-manage. Notably, methods of self-management were not limited to misoprostol, and interviews with 30 women who said they had attempted to self-manage their abortion found that a majority reported using less effective methods, such as vitamins and herbs, illicit drugs and alcohol.
  • A similar pattern was seen among individuals accessing clinical abortion care in 2014, where 1.3% reported having ever used misoprostol to end a pregnancy on their own, and 0.9% reported using something else.
  • In a study conducted in 2014–2015, researchers estimated that 1.7% of Texas women had ever attempted to end a pregnancy on their own, the equivalent of 100,000 residents. While the survey was based on a sample of 779 Texas women, the estimate of 100,000 was based on responses from approximately 13 respondents, suggesting some amount of imprecision.
  • Most recently, a national survey of 7,022 U.S. women conducted in 2017 found that 1.4% reported ever attempting to end a pregnancy on their own. The majority of these individuals reported using drugs or substances other than misoprostol, and only 28% successfully ended the pregnancy.

The available evidence, limited as it is, clearly shows that self-managed abortion does occur in the United States, but not yet on a large scale and not always using the most effective methods. Lifetime incidence in the published studies was between 1% and 5%. For these reasons, it is unlikely that self-managed abortion was a primary driver of the national abortion declines, even in Guttmacher’s most recent survey period from 2014 to 2017. 

In addition, the extent to which abortion restrictions contribute to interest in or incidence of self-managed abortion requires further study. The Texas study received considerable media attention because the state had recently passed abortion restrictions that resulted in the closure of half of the clinics in the state, and the study findings were frequently interpreted as fallout from the law. As with the rest of these studies, the findings were about lifetime incidence, and there was no evidence that most, or even a substantial minority, of these attempted abortions had occurred in response to the recent restrictions.

Indirect Measures

In addition to studies looking directly at the incidence of self-managed abortion, there are several studies suggesting substantial interest in this approach. One informal, though widely cited, investigation found that Google searches for terms related to self-managed abortion increased from 119,000 in 2011 to 700,000 in 2015. Similarly, researchers conducting an online study of individuals using Google to obtain information on self-managed abortion found that there were 210,000 relevant searches over a 32-day period in 2017.

Most recently, a study using data from Women on Web, a site that provides information and instructions on self-managed abortion (but  does not ship medication to the U.S.), documented 6,022 specific requests for medication abortion over a 10-month period between 2017 and 2018. Moreover, they found that residents in states with restrictive abortion policies were most likely to make these requests.

However, interest does not necessarily translate into practice: Yet another study where researchers conducted interviews with 32 individuals who actively sought out abortion drugs online found that none purchased them due to issues such as lack of information on the websites, concerns about fraud and cumbersome billing processes. Instead most study participants accessed clinical abortion care; two pregnancies were expected to be carried to term.

While Guttmacher’s flagship studies on abortion incidence do not capture the number of self-managed abortions, our most recent two studies have included a related measure: In 2017, 18% of nonhospital facilities treated at least one patient who had attempted to end a pregnancy on their own, up from 12% in 2014. Most of these facilities reported only seeing one or two such patients in either year, but the increase in facilities treating patients after self-managed abortion lends support to the idea that more people in the United States are attempting to self-manage. It is important to remember that this measure reflects the subset of people who seek follow-up care from an abortion provider after attempting to self-manage an abortion. It does not reflect people who turn up for follow-up care elsewhere, nor does it include anyone who self-manages an abortion and does not need or want any follow up care.

Looking Ahead

Recent developments suggest that self-managed abortion will be of increasing importance in the United States. A 2018 study found that the highly effective combination of mifepristone and misoprostol—the same regimen that accounts for about four in 10 abortions provided in clinic settings—can now be purchased online for $110 to $360.

Then in March 2018, a physician licensed in Austria launched the website Aid Access. Individuals in the United States who visit the website provide information about their pregnancy that is screened by a physician to assess whether they are eligible for medication abortion. If so, the physician prescribes the medication and refers the person to a pharmacy in India that will provide the pills. The combination of mifepristone and misoprostol is mailed to individuals for a suggested donation of $90. Users also receive instructions on how to use the drugs, how to recognize complications and how to access follow-up care if needed without revealing the attempt to end a pregnancy on their own. Legal documents report that Aid Access provided 11,100 consultations for medication abortion in the 10 months it was operating in 2018. Notably, they also reported that only 2,600 prescriptions were filled—less than a quarter of the number of consultations and a figure equivalent to 0.3% of all the abortions that Guttmacher estimated to have occurred in clinical settings in 2017.

Still, it is likely that Aid Access and potentially other services along these lines will be increasingly important resources for individuals who want or need to obtain abortion pills online. For many, this may be the case because medication abortion—or any abortion care—is not readily available or easily accessible in their communities, especially with many states piling on more abortion restrictions by the year and with abortion rights once again before the U.S. Supreme Court. For other individuals seeking to end a pregnancy, online access may simply be a better option for reasons of privacy or other personal preferences. Either way, as we look ahead at what the future holds for abortion access in the United States, it will be critical to put in place policies and practices that ensure that people self-managing their care have the information and support they need.