For the past year, author and former Google data scientist Seth Stephens-Davidowitz has been making the media rounds. His research—and his book, Everybody Lies—uses big data to uncover behaviors and attitudes that Americans wouldn’t normally admit to. Some of the findings are fun (including tips for how to get a second date) and some feel depressingly self-evident (Americans are pretty racist), but none have caused the level of progressive panic as Stephens-Davidowitz’s research on abortion. Using data tracked from Google searches, he concludes that abortion restrictions have led to “a hidden demand for self-induced abortion reminiscent of the era before Roe v. Wade.”
This inference is alarming and it has garnered significant press, including articles in Vox and The New York Times. However, there’s a basic misunderstanding at the core of the research that could harm women’s access to comprehensive reproductive health care—particularly affordable and safe abortions.
Get TalkPoverty In Your Inbox
Stephens-Davidowitz’s research mistakenly conflates “self-induced abortion” with “illegal abortion,” though the two terms apply to two very different procedures. A self-induced abortion is simply an abortion that can be conducted within the comfort of one’s home. That includes medical abortions, also referred to as the “abortion pill,” which can be used to end early-term pregnancies. An illegal abortion, on the other hand, is often what we think of as a “coat hanger abortion”—it’s one of the risky procedures women undergo when other options (such as self-induced medication abortions) are not available.
Self-induced abortions are safe and fairly common: They accounted for 31 percent of all nonhospital abortions in 2014. “People choose to self-induce for a variety of reasons,” said Jill Adams, founding executive director of the Center on Reproductive Rights and Justice at Berkeley Law. “The flexibility of conducting the procedure at home on one’s own timeline is paramount, and self-induced abortion can be significantly cheaper than surgical abortion.”
But misinformation about self-induced abortion, namely that using the abortion pill is a dangerous practice, could ultimately make it harder to access. Anti-choice advocates and legislators have seized on this type of misinformation in the past, most notably through TRAP lawsTargeted Regulation of Abortion Providers, or TRAP laws, are specific legal requirements for abortion providers that are different (and more difficult to comply with) than the requirements for other medical practices. Examples include specifying specific hallway widths, staffing requirements, or admitting privileges. that require medically unnecessary updates to clinics that provide abortions as an indirect way to reduce abortions.
If that happens, it will hit women with few other options the hardest. Medical abortion is particularly crucial for people who would otherwise struggle to access reproductive health care, including people living in rural areas and women of color. Rural patients face clear physical barriers: 31 percent of women living in rural areas traveled more than 100 miles to access abortion services, and an additional 43 percent traveled between 50 and 100 miles. For women of color, who often suffer from a variety of barriers to abortion—such as financial instability, limited access to a broad range of providers, and distance from clinics—medication abortion can be the most cost-effective and low-risk abortion procedure.
There’s a chance that the searches Stephens-Davidowitz reports could simply represent an increase in medically accurate information. Telemedicine has revolutionized abortion care for the aforementioned groups. In 2006, Planned Parenthood of the Heartland in Iowa launched a telemedicine service that provided medical abortion care at rural clinics. The initiative was wildly successful: Research showed that telemedicine availability increased access to abortion care for people living in remote parts of the state. Moreover, the study showed that telemedicine availability increased access for women seeking abortion services at earlier gestational stages, for which medical abortion could be a silver bullet against the cost, distance, and stigma of an in-clinic abortion. In short, Google results for “self-induced abortion” may have ticked up because more women are simply aware that it exists.
While it seems intuitive that restrictive abortion laws would increase the incidence of illegal abortions, the inference that Stephens-Davidowitz draws about self-induced abortions is not necessarily backed up by the Google search data. Clear, detailed terminology is critical in discussing abortion, especially when the consequences can result in devastating outcomes for people seeking health care. Mistaken inferences, even when they have good intentions, have harmful consequences when placed in the nefarious hands of anti-choice activists—and can result in even more limitations on women’s health care.