In the weeks since the Supreme Court’s momentous ruling in Dobbs v. Jackson Women’s Health Organization, much of the news coverage has focused on the relative safety of abortion versus pregnancy. A New Yorker article announced [1] that “pregnancy is more than thirty times more dangerous than abortion.” The Huffington Post published the prediction that “in 2022 alone, the Supreme Court’s decision will directly cause the deaths of hundreds of people as their bodies are used by the state against their will. Abortion is significantly safer than pregnancy—period.” On Twitter [2], the actor Halle Berry declared that “[t]he treatment for an ectopic pregnancy is an abortion. The treatment for a septic uterus is abortion. The treatment for a miscarriage that your body won’t release is abortion. If you can’t get those abortions, you die. You. Die.”

Some of these claims may be disputable or misleading, but they’re all rooted in a concern for maternal mortality. The Dobbs decision has sparked a long-overdue reckoning with the abysmal state of maternal health in the United States. It’s no secret that we have the highest maternal-mortality rate among developed nations. According to the Pew Trust [3], “[p]regnancy-related deaths among American women have risen markedly over the past 30 years, despite an overall downward trend worldwide.” Between 2000 and 2017, UNICEF [4] reported that the United States averaged roughly nineteen maternal deaths per 100,000 live births. During the same time period, Chile, Ireland, and Poland—all countries where elective abortion was illegal at the time—averaged, respectively, thirteen, six, and two deaths per 100,000. Canada, where there are fewer restrictions on abortion than in the United States, averaged ten deaths per 100,000, while the United Kingdom averaged eight, and Australia averaged six.

More recent data suggests that the high maternal-mortality rate in the United States has not declined. According to the CDC’s Pregnancy Mortality Surveillance System [5], the maternal-mortality rate rose 140 percent between 1987 and 2018—from 7.2 to 17.3. The agency’s National Vital Statistics System [6] reports that in 2020 the maternal mortality in the United States climbed to 23.8 deaths per 100,000. Meanwhile, the rate in other industrialized countries has either remained stable or declined.

Given this context, it is understandable that many people have expressed concern about how Dobbs could affect women’s health. Will a shift in medical practice endanger pregnant mothers and further widen the maternal-mortality gap between the United States and other developed nations? Will outcomes for white women and women of color continue to diverge? In recent weeks, stories about delayed care [7] of pregnant women facing medical emergencies and a lack of access to life-saving [8] intervention have been all over the press, and there is increasing anxiety about what may now happen to women who experience miscarriages, ectopic pregnancies, or other serious medical complications. Many insist these undeniably agonizing cases are direct evidence that unrestricted access to elective abortion is essential for safeguarding women’s health. But this poses a false dichotomy: either maintain one of the most permissive abortion regimes [9] in the world or condemn mothers to die from medical complications of pregnancy. This argument distracts from what is otherwise a critical conversation about maternal health, and from legitimate concerns about how new abortion restrictions are being implemented, and how they are being interpreted or misinterpreted by doctors, hospitals, and pharmacies in an already overwhelmed health-care system.

Such conversations must start with important distinctions in terminology. Some diversity of opinion exists among Catholic bioethicists and medical professionals about how to talk about abortions and other medical interventions that are performed to treat conditions like septic uterus, ruptured membranes, and ectopic pregnancies. In a clinical context the term “abortion” refers to any pregnancy loss that occurs before twenty weeks. Within this broad category, we can identify three distinct “types” of pregnancy loss. The first is a “spontaneous abortion,” or the death of a fetus in utero before twenty weeks’ gestation, more commonly called a miscarriage (a death after that is called a stillbirth). The second is what bioethicists have long referred to as an “indirect abortion”: any procedure that ends a pregnancy but does not have as its aim the death of the unborn child. This includes all procedures intended to preserve the life of the mother—procedures that ought to have been performed in nearly all the cases that have recently made headlines. Finally, there is “elective abortion,” which directly intends the death of a living fetus or embryo. Most of the estimated 50 to 66 million abortions that have been performed in the United States since 1973 have been elective abortions. And it is this third type of abortion that new laws are intended to restrict.

Obfuscation of these essential distinctions is evident on both sides of this issue. On the one hand, many abortion-rights advocates seek to expand public funding of elective abortion through all nine months of pregnancy, claiming that anything less puts women’s lives in jeopardy. Never mind that every law currently in effect, including Texas’s “Heartbeat law” (see section 170A.002), makes it clear that physicians are not only allowed but expected to intervene to save the life of the mother even if this intervention requires the termination of her pregnancy by means of an indirect abortion. On the other hand, a vocal minority of “abortion abolitionists,” such as Scott Herndon [10], a Republican candidate for the Idaho senate, support the elimination of all exceptions [11] to abortion bans, even those that would save the life of the mother, as well as the criminal prosecution of women who procure abortions. These abortion abolitionists refuse to make any distinction between direct, elective abortion and indirect abortion, and their legislative proposals have provoked serious concern that, after Dobbs, we are on our way to total bans of the medical interventions necessary to save the lives of pregnant women. Pro-lifers should reject all legislation that fails to make an explicit and feasible exception for such indirect abortions. There can be no ambiguity about this.

That brings us back to the relationship between elective abortion and maternal health. Pro-choice activists insist that new restrictions on elective abortion will inevitably result in the deaths of thousands of women. [12] Last month, a piece in the New Yorker [13] cited a study [14] claiming that a “hypothetical total abortion ban” would lead to a 21 percent rise in pregnancy-related deaths. In an L.A. Times [15] op-ed published shortly after the draft leak, three social scientists argued that “losing abortion rights puts women’s lives at stake.” During a panel discussion hosted by the New York Times, Laura MacIsaac, an obstetrician-gynecologist at Mount Sinai Hospital, claimed that “maternal mortality without the availably of abortion will absolutely go up. We’ve seen it since the beginning of time and it will continue,” without providing any evidence to back up this claim. Assertions like these are so common, and so confidently presented, that the average layperson does not dare to question them.

But is abortion really safer than pregnancy? This turns out to be a hard question to answer because of two factors: the difficulty of measuring pregnancy mortality and the difficulty of collecting accurate abortion-related data. The first difficulty is fairly straightforward: it is hard to accurately assess the rate of maternal death without a uniform definition of “maternal death.” Different organizations and reporting bodies use different definitions, depending on different criteria and covering different postnatal periods.

There are other statistical challenges. Pregnancy mortality is measured per live birth, not per pregnancy. This means that the pregnancies of women who have early miscarriages usually go uncounted. The Cleveland Clinic [16] estimates that a third to half of all pregnancies end in miscarriage. Eighty percent of these miscarriages happen early in pregnancy, many before a woman even knows she’s pregnant. All of these pregnancies are excluded from the data because they do not result in live births. They show up in the statistics only if a woman dies.

This creates serious problems for accurately assessing the actual risk of pregnancy. The Elliot Institute’s Amicus Brief in the Dobbs case points out that this method of statistical accounting reduces the relevant baseline population “by excluding cases of pregnancy losses (no live birth), yet the total number of deaths still includes those maternal deaths resulting from these very same excluded—uncounted—pregnancies.” This might suggest that our maternal-mortality rate is lower than currently thought (though still higher than that of other wealthy countries that use the same standards of measurement). And that would be very good news. The fact remains, however, that the current statistical parameters make it almost impossible to offer definitive comparisons between the safety of pregnancy and the safety of abortion. Indeed, in 2004 the director of the CDC, Dr. Julie Louise Gerberding, wrote that maternal-mortality rates and abortion-mortality rates “are conceptually different and are used by the CDC for different public health purposes.” This alone should keep us from making sweeping claims about the relative safety of abortion.

The second challenge of collecting accurate abortion-related data is significantly more complex. It is really a set of challenges, rather than just one. The United States lacks universal mandatory reporting for abortions and also for medical complications and deaths related to abortion. Because of this, determining the relative safety of abortion is nearly impossible. CDC data is based on voluntary state reporting, and it is not always consistent with that of other reporting institutions. For example, the Guttmacher Institute often reports significantly higher numbers of abortion than the CDC, even though their data is also based on voluntary reporting. According to Guttmacher [17], several key states—including California, Maryland, and New Hampshire—don’t report abortion data at all. In other states, like New Jersey, the state health department does report abortion data, but its data is based on voluntary reporting by physicians. It is worth noting that California, Maryland, and New Jersey are among the states with the highest annual abortion rates, yet their data is not reflected in CDC reporting.

If the standards for reporting abortion itself are uneven, the standards for reporting on the health risks of abortion are downright abysmal. Only twenty-seven states and the District of Columbia require the reporting of complications from abortion. And even the states that require reporting lack enforceable penalties for abortion providers who fail to comply. This means that we do not have a reliable measure of abortion-related complications, and, without that, there can be no useful comparison between the risks of abortion and those of pregnancy.

This challenge is compounded by the fact that abortion-related complications and deaths are often reported as pregnancy-related complications or deaths. In 2017 an otherwise healthy twenty-three-year-old Black woman named Keisha Atkins died of a pulmonary thromboembolism (a blood clot in the lungs) during a late-term abortion: she was twenty-four weeks pregnant at the time. (Ninety-two percent of abortions take place before the thirteenth week of gestation.) According to the medical examiner [18], Atkins had begun the abortion process and was at a clinic preparing for the final stage. While there, she began to experience symptoms of distress—cramping (normal during an abortion), but also shortness of breath and low blood oxygen levels. She was transferred to a hospital, where she continued to have cramping, an elevated heart rate, and low blood-oxygen levels. Further testing revealed fluid in her lungs and reduced heart function. The medical examiner goes on to explain:

Due to rapid decomposition in her clinical status (requiring the placement of a breathing tube) and the concern for a significant infection, she was taken emergently to the operating room to complete the abortion procedure. During the operation, she sustained a cardiac arrest. Extensive resuscitation efforts were ultimately unsuccessful.

The medical examiner notes that Atkins had a septic uterus due to the abortion procedure itself, and the autopsy “revealed a well-developed, well-nourished young woman with extensive medical intervention.” Her family went on to sue the abortion clinic and hospital, which settled [19] for $1.26 million in May 2022. Still, Atkins’s death certificate reports her cause of death as “pulmonary thromboembolism due to pregnancy.” The CDC would not say whether her case made it into the national statistical data “because all states do not report to the CDC.” While New Mexico, Atkins’ home state, does report abortion data to the CDC, it does not report complications. Thus, even though she died during an abortion procedure, her death would be categorized as a pregnancy-related mortality, not an abortion-related mortality.

Another challenge in assessing abortion-related risk is that statistics often lump together abortions at every stage of gestation. Pregnancy is not a static physiological phenomenon but one that dramatically changes over the course of forty weeks. Dr. Monique Chireau Wubbenhorst, an ob-gyn and former professor of obstetrics and gynecology at Duke University School of Medicine, notes that the risk [20] of abortion mortality increases with gestational age, just like the risk of pregnancy-related mortality. A study led by Dr. Linda Bartlett that was published in the journal, Obstetrics & Gynecology, shows that death from abortion increases exponentially by 38 percent for each additional week of gestation. The same study shows that abortion mortality for women of color is three times that for white women. Other data also suggest that after eighteen weeks’ gestation the mortality rate for vaginal childbirth (3.6 deaths per 100,000) is less than half the mortality rate for abortions performed during the same period (7.4 deaths per 100,000). This directly contradicts the undifferentiated claim that abortion is safer than pregnancy.

Some abortion-rights advocates have suggested that the increased mortality rate associated with late-term abortions is due to the fact that they are much more likely to be the result of medical emergencies that threaten a pregnant woman’s life. Appearing on Face the Nation [21] in the wake of the Planned Parenthood fetal-tissue scandal in 2015, Hillary Clinton said, “I think that the kind of late-term abortions that take place are because of medical necessity. And, therefore, I would hate to see the government interfering with that decision.” The Annenberg Public Policy Center [21] fact-checked this statement: “A spokesman for Clinton’s campaign told us that she meant that many late-term abortions—not all or even most—are because of medical reasons. But that’s not what she said. Her statement left the impression that the majority, if not all, late-term abortions are medically necessary. The available evidence does not support that assertion.” Most available data suggest that late-term abortions, such as the one Atkins underwent, are not medically necessary. In 1997, Dr. Ron Fitzsimmons, the then-executive director of the National Coalition of Abortion Providers, told the New York Times [22] that “the procedure is performed far more often than his colleagues have acknowledged, and on healthy women bearing healthy fetuses.” We can thus conclude that, after a certain point in gestation, a healthy pregnant woman is safer giving birth to a child than aborting it. Yet abortion-rights advocates often claim that even late-term abortions are safer than live childbirth. The L.A. Times op-ed made this very claim even though the article they cited in support of it includes no maternal-mortality statistics.

In our current debates, economically vulnerable women and women of color [23] are routinely held up as the reason we need to maintain unrestricted access to elective abortion. The previously cited L.A. Times op-ed claims that “scientific and medical research consistently shows that childbirth is far riskier than terminating a pregnancy, particularly for poor and minority women.” Women of color, especially Black women, are three times [24] more likely to die during pregnancy or childbirth than white women, regardless of income or education. This is not evidence of the efficacy of abortion, but an indictment of our failure to address inequities in maternal health. Setting aside the challenges of assessing abortion-related data, we might ask ourselves why abortion is so often presented as the silver bullet for addressing poor maternal-health outcomes for poor women and women of color despite the fact that the poor and people of color are more likely [25] to oppose abortion than wealthy and white people. Evidence [26] from other industrialized nations also suggests that our failure to address maternal mortality among women of color is not related to abortion access, but to our failure to provide adequate health care.

The Dobbs ruling is no doubt polarizing, but it may also give us a rare opportunity to build a bipartisan consensus around ways to address maternal mortality, particularly for the most vulnerable women. A recent study from the Commonwealth Fund [27], a non-partisan independent research organization, concluded that “high maternal mortality in the U.S. is not the result of any single factor, and reducing it will require an integrated effort involving policy and practice changes to improve hospital and community care for all women while advancing racial equity.” The report goes on to note “the shortage of maternity care providers (both ob-gyns and midwives) relative to births,” adding that “in most other countries, midwives outnumber ob-gyns by severalfold, and primary care plays a central role in the health system. Although a large share of [U.S.] maternal deaths occur post-birth, the U.S. is the only country not to guarantee access to provider home visits or paid parental leave in the postpartum period.” It is worth noting that the researchers do not include abortion access in their list of policy recommendations, but focus on bolstering health care, insurance coverage, postpartum care, and parental leave.

Addressing our scandalous level of maternal mortality will require that all states—especially those like Texas and Mississippi that have passed strict restrictions on elective abortion and where maternal-health outcomes are among the worst in the nation—increase funding for programs that provide mothers and their children with the support they need. Any state that invokes the sanctity of life needs to start from a principle formulated by the former Archbishop of Canterbury, Rowan Williams: “the poor deserve the best.” This would require that all women have access to quality health care—before, during, and after birth, when they are still vulnerable to life-threatening infections and other complications. A focus on reducing maternal mortality would mean prioritizing underserved populations and addressing health inequities between white women and women of color. First, it is vital to ensure that women receive quality health care before they become pregnant, as many pregnancy-related deaths stem from underlying health conditions. Second, women must have access to quality prenatal and postnatal care, and should be encouraged to use it. Third, it is critical to ensure paid parental leave for all employees, especially those who work entry-level or hourly positions. Fourth, states ought to invest in robust midwife-led care, which is proven [26] to reduce maternal and neonatal mortality. Such measures will require major investments in vulnerable communities and in the education of nurses and midwives.

Pro-life activists should not imagine that one can build an authentic culture of life without greater public investment in health care and social services. But pro-choice activists, politicians, and journalists should not pretend that the only way to protect women from life-threatening complications of pregnancy is to ensure that they have unrestricted access to elective abortion, or that we know more about the relative risks of abortion and pregnancy than we actually do. The sooner we come to terms with these realities and let the needs of vulnerable mothers and babies dictate our policies, the sooner we can do the important work of making pregnancy as safe in the United States as it is elsewhere in the world. Liberal abortion laws did not keep us from having the highest maternal-mortality rates among industrialized nations, nor will new abortion restrictions alone be adequate to meet the needs of pregnant mothers and lower the maternal-mortality rate. Maternal mortality is not mainly a function of abortion access; it is a function of how well or poorly we provide for expectant mothers.