As the Supreme Court decides the future of abortion laws in the U.S., a key question to be considered is whether access to the procedure has positive or negative consequences for the people who get an abortion, and for society in general.
Dobbs v. Jackson Women’s Health Organization concerns the constitutionality of a new Mississippi law that would ban abortions after 15 weeks of pregnancy. The case challenges the Supreme Court’s 1973 Roe v. Wade decision, a precedent that protects abortion access before fetal viability—a point at around 24 weeks of gestation, when a fetus is considered able to survive outside the uterus.
Antiabortion activists often contend that abortion harms women physically and psychologically. Another argument, made by Mississippi Attorney General Lynn Fitch, is that there are now many laws that protect equal economic opportunity—suggesting that abortion access is no longer necessary to help women pursue financial independence.
Examining the validity of these claims has been notoriously difficult. The most scientific way to measure how getting an abortion impacts a pregnant person would be through a randomized controlled trial—which would require the obviously impossible scenario of people who seek an abortion being designated by researchers to either receive one or not. “The challenge is: we cannot feasibly or ethically randomly assign abortion access to people to measure what happens to their lives,” says Caitlin Myers, a professor of economics at Middlebury College.
Thus, until recently, the few studies that tried to measure the impacts of abortion access had an important limitation. They compared groups that were too different to begin with—for example, people who got abortions and people who chose to give birth.
But one study has made great progress in measuring the impact of abortion access. To address the methodological limitations in previous studies, researcher Diana Greene Foster, a professor in the department of obstetrics, gynecology and reproductive sciences at the University of California, San Francisco, designed a novel approach. Her team recruited women at abortion clinics, and compared outcomes among those who were just over the gestational limit and were denied an abortion with those who were just under the limit and had the procedure. Foster called the investigation the Turnaway Study, a reference to clinics turning some people away because they are too far along in their pregnancy. Its results have been described in 50 scientific papers, almost all of which were published in peer-reviewed journals from 2012 to 2020. And to date, the study is one of the most comprehensive in the field.
“The real innovation of the Turnaway Study,” Foster says, “was to take people who received an abortion and compare them to [those who had] the only other outcome that is available to somebody who is pregnant and doesn’t want to be—which is to carry that pregnancy to term.”
From 2008 to 2010 the study recruited nearly 1,000 women seeking abortions at 30 facilities in 21 states. Afterward the participants were interviewed by phone every six months for a span of five years. (While transgender men and nonbinary people also experience pregnancy and seek abortions, the Turnaway Study focused specifically on pregnant women.) The study found that, compared with women who received an abortion, those who wanted the procedure but were denied it fared worse in numerous aspects of their life, including financial situation, education, and physical and mental health.
Because of how it was able to isolate the effect of abortion access in a natural experiment, the Turnaway Study is recognized by the scientific community as an important contribution to the field, generating data that many scientists say should be considered by policy makers. “The rigid cutoff generates a situation where you’ve got women immediately on either side of the cutoff being extremely similar in characteristics and circumstances—but one can have an abortion and the other one can’t,” says Phillip Levine, a Wellesley College professor of economics, who was not involved with the study. “The Turnaway Study has the advantage of providing information that would otherwise be very difficult to ascertain regarding the impact of an abortion on women.”
“It’s impossible to overstate how scholarly the design of the study is,” says Amanda Stevenson, an assistant professor of sociology at the University of Colorado Boulder, who was also not involved with the investigation. She says that the researchers cleverly drew from their knowledge of how abortion care delivery happens in the U.S. to find a natural experiment that answered their questions.
Poverty and Unemployment
When it comes to the socioeconomic implications of abortion access, the Turnaway Study found that women who were denied the procedure and carried the pregnancy to term were more likely to live in poverty. Six months after they sought an abortion, 61 percent of them were below the U.S. federal poverty level, compared with 45 percent among the group who received an abortion just under the limit. And at the same point, people in the former group were also more likely to be unemployed (51 percent versus 37 percent among women who had an abortion) and to receive financial benefits from the Temporary Assistance for Needy Families (TANF) program (more than 15 percent versus less than 8 percent). Between one and five years after seeking an abortion, women who were denied the procedure and ended up giving birth were more likely to report not having enough money to cover living expenses, a more subjective poverty measure.
Myers, who was not involved with the work and whose research focuses on the effects of reproductive policies, says she was initially a bit skeptical about the study’s design. “I thought that there still could be something different about the people who showed up a little too late—that they might unobservably have other factors in their lives that were causing them to show up a little bit too late that were also causing them to have different outcomes,” she says.
For Myers, what put those concerns to rest was seeing a subsequent analysis, also a part of the Turnaway Study, in which researchers linked the study’s participants to their credit scores. “What I found so powerful was that they [the authors] don’t just match them at the time they seek the abortion. They match them for an extensive time period prior to experiencing the unintended pregnancy,” Myers says. “And they show that their financial circumstances are trending very similarly right up until the moment that they go to the abortion providers and experience different outcomes.”
The credit score analysis was led by Sarah Miller, an assistant professor of business economics and public policy at the University of Michigan, who had the idea after seeing the Turnaway Study’s initial socioeconomic findings. Her work shows that women who were denied an abortion suffered an increase in financial distress that lasted several years. And these women’s debt increased by 78 percent, compared with their average prebirth debt. Their number of negative public records, such as bankruptcy and eviction, increased by 81 percent, also compared with their prebirth average.
“What this study has done is that it has very convincingly shown that being denied an abortion has these economic effects,” says Jason Lindo, a professor of economics at Texas A&M University, who provided feedback for the credit score paper but was not directly involved in the work. He notes that the findings are consistent with the broader scientific literature, which indicates that having children in general can lead to negative economic consequences.
Health and Emotions
One of the key questions that Foster was looking to answer when she designed the Turnaway Study was: Does abortion hurt women? The investigation found that, in the long term, more women who gave birth (27 percent) reported fair or poor physical health, compared with women who had an abortion (20 percent for a first-trimester abortion and 21 percent for second-trimester one). “The biggest differences we saw, besides the socioeconomic differences, are in physical health,” Foster says. “That’s consistent with the medical literature that shows that carrying a pregnancy to term—the many months of continued pregnancy and childbirth—are associated with much greater risk than having an abortion, even a later abortion.”
Two women in the Turnaway Study died from childbirth; there were no abortion-related deaths. Stevenson notes that, while the investigation was not designed to study mortality, two pregnancy-related deaths are more than what would be normally expected for a population of that size. “What that tells us is that it is quite possible that people who are seeking abortion care, right on the edge of when abortion is currently available in most places, are at elevated risk of pregnancy-related mortality,” she says. “This is not conclusive evidence because that was not the intention of the study. But it is a very surprising finding.”
When it comes to the participants’ mental health, the study found no evidence of harm from having an abortion. “There were no emergent cases of depression, anxiety or suicidality,” Foster says. What she and her colleagues did find was a short-term decline in mental health among those who were denied an abortion, with increased self-reported anxiety symptoms and lower self-esteem and life satisfaction.
Prior studies suggest that women seeking an abortion tend to have more existing mental health disorders than women who choose to carry their pregnancy to term. “It explains why we see a correlation or association between abortion and subsequent mental health in some data,” says Julia Steinberg, an associate professor at the department of family science at the University of Maryland School of Public Health.
Correlation is not necessarily causation, as statisticians often note. Epidemiologist Chelsea Polis, a researcher in the department of epidemiology at the Johns Hopkins Bloomberg School of Public Health, and her colleagues did a systematic review of studies looking into long-term mental health outcomes potentially related to abortion. They found that higher-quality studies reported no significant differences in long-term mental health between women who choose to abort a pregnancy and those who do not. “We also found that the opposite holds true for studies with the weakest research methodologies: those studies consistently found negative mental effects from abortion,” Polis says.
The Turnaway Study found no evidence of emerging negative emotions five years after an abortion among the participants who had access to the procedure. In fact, relief was the predominant emotion reported at all times during the study. Additionally, at each data collection point throughout the five years, 95 percent of these women reported that having an abortion was the right decision.
What the data show, Foster says, is that women who seek an abortion understand the consequences of carrying a pregnancy to term and make a decision that reflects the outcomes they are likely to experience if they are unable to end the pregnancy. “They say that they don’t have the money to support a child, and we see that their economic outcomes suffer. They say that their relationship isn’t strong enough to raise a child, and we see that their relationship deteriorates. They say that they want to pursue other life outcomes, and we see that those life outcomes are strongly affected by whether they are able to get their abortion.”
Foster says she is now setting up a new study to document the potential impact of the end of Roe in the U.S. “It seems very likely that our Supreme Court will allow 15-week bans or allow states to ban abortion entirely,” she says. “Either way, I am designing a study to recruit people served just before the law was implemented and those who were turned away just after.” Foster adds that she is also working with scientists in Nepal to study the effect of abortion receipt and denial in a context with high maternal mortality, widespread unsafe abortions and high malnutrition among children.
Consequences of Stricter Abortion Laws
Scientists caution that the Turnaway Study looked at a specific group of women: those who sought an abortion relatively late in their pregnancy. More than 90 percent of women who seek the procedure in the U.S. do so within the first 13 weeks of gestation. “My guess, based on the evidence, is that the results would likely be the same with an earlier cutoff if there were such experiments,” Levine says. “But it’s also important to recognize that we don’t necessarily know that.”
Even if a direct extrapolation is not possible, the Turnaway findings hint at what could happen if the Supreme Court allows states to ban or further restrict abortion, according to Liza Fuentes, a senior research scientist at the Guttmacher Institute. “It gives us insight into the types of negative outcomes that we could expect for people—if they are unable to travel, to obtain an abortion—living in a state that makes abortion unavailable,” says Fuentes, who was not involved with the study.
Myers notes that one of the pillars of Mississippi’s argument for overturning Roe is that access to abortion has made little difference in people’s lives. “That is patently false,” she says. “We have solid scientific evidence showing that Roe mattered to people’s lives, and we know that abortion access matters to people’s lives now. And that is where I think it’s very important for science to come in and say, ‘We have answers to these questions of fact.’”