The Unequal Impacts of Abortion Bans

Since the Supreme Court’s 2022 Dobbs decision overturned Roe v. Wade, some states have imposed severe restrictions on access to abortion, effectively forcing pregnant people to continue unwanted or unsafe pregnancies to term. 

The result, according to new research from associate professors in Population, Family and Reproductive Health Alison Gemmill and Suzanne Bell, was an increase in live births—but also infant deaths. The impacts were worse among certain populations: Black infants, for example, died at a rate 11% higher than would have been expected in the absence of bans. 

In the February 24 episode of Public Health On Call, Gemmill, PhD, MPH, MA, and Bell, PhD ’18, MPH, spoke with host Lindsay Smith Rogers about their study and how abortion bans unequally impact certain subgroups, and discuss the negative physical, mental, and economic implications of being unable to obtain an abortion. This Q&A is adapted from their discussion.

Can you give us an overview of the previous work you did on the effects of abortion bans?

Alison Gemmill: Our original work looked at the impact of a specific Texas policy known as SB8, which was one of the first major abortion bans in the country. Because SB8 was implemented nearly a year before Dobbs, we were able to investigate the impacts of that policy first. We did that by conducting two separate analyses using birth and death certificate data.

The first analysis examined the change in live births after the policy was implemented, and the other looked at what happened to infant mortality. We found that S8 was associated with an almost additional 10,000 births in the state of Texas through December 2022, and an unexpected increase of about 13% in infant deaths. The rest of the U.S. didn’t experience that increase. There was also evidence to suggest that the increase in infant mortality was partly driven by an increase in congenital birth defects.

Why was that significant?

AG: Before the abortion ban was in place, if a fetal anomaly was detected in an anatomy scan, then that couple had the option to terminate the pregnancy. With the ban in place, the option to terminate is off the table. As a consequence, we think we’re seeing people carrying essentially doomed pregnancies to term, and they’re delivering babies with severe defects that are incompatible with life, resulting in infant deaths.

Did your new research answer any questions your previous work did not? 

Suzanne Bell: Our earlier research didn’t answer how specific populations may be differentially impacted by abortion bans. Our two recently published studies provide the broadest evidence to date of the unequal impact of abortion bans across and within states and subgroups. 

We again used birth certificate and death certificate data, as well as census data from 2012 through 2023 for all 50 states and D.C. To analyze these data, we used statistical modeling to estimate how abortion restrictions affected the rate of live births and infant mortality in states with complete or six-week abortion bans. This method allowed us to estimate how many live births and infant deaths there would have been if these restrictions had not been in place, and then to compare that to what we actually observed. We also estimated effects by subgroups. 

What did you learn about live births?

SB: We estimated that there was one additional live birth above expectation per 1,000 reproductive-age females following the adoption of these abortion bans. This is a 1.7% increase—equivalent to more than 22,000 excess births by the end of 2023—and this indicates that approximately 10% of abortions that we think would have occurred in these states during this time period ultimately resulted in a live birth. Impacts were largest among racially minoritized individuals—non-Hispanic Black, Hispanic, and other racial/ethnic groups; those without a college degree; Medicaid beneficiaries; unmarried individuals; younger individuals less than 35 years of age; and those residing in southern states.

When we looked across states, the estimated changes in fertility rates [live births] above expectation range from 0.3% above expectation to 2.3%, with the largest estimated changes in Texas at 2.3%, Kentucky at 1.4%, and Mississippi at 1.4%

Do we know why this is happening?

SB: The assumption is that these individuals, and in particular those experiencing the greatest structural disadvantages, are unable to overcome the barriers imposed by these abortion bans and are forced to continue an unwanted or unsafe pregnancy to term, resulting in higher fertility in these states that have imposed these bans.

What did this new work tell you about infant mortality?

AG: In our earlier work in Texas, we found an increase in infant mortality. With this work, we wanted to know whether we would see a similar increase in other states following the Dobbs decision. Our paper shows a 6% increase in infant mortality among states with bans, corresponding to an additional 478 infant deaths above what we would have expected in the absence of these bans.

We found similar patterns to what we found in our initial Texas paper, but in this paper, we also examined specific states that were impacted. We found that the largest increase in infant mortality—9%—occurred in Texas and Kentucky. In general, we found that these increases were concentrated in Southern states. We also looked at infant mortality among racial/ethnic groups, and found that there was an 11% increase in the infant mortality rate among non-Hispanic Black infants. This increase in Black infant mortality was consistent across states with bans.

Our infant mortality results suggest that abortion restrictions may be stopping or even reversing improvements in infant mortality that have been made in recent decades.

What does this tell us about the legislation itself?

SB: These findings confirm hypotheses in some of our prior work in Texas that indicate many pregnant people were ultimately unable to overcome barriers to access abortion services, and instead were forced to continue unwanted or unsafe pregnancies to term, including doomed pregnancies diagnosed with fatal anomalies.

Even the Herculean efforts that many abortion groups have taken and continue to take to support pregnant people accessing needed abortion care out of state, has not fully offset the demand or need for abortion among people residing in states that have imposed these abortion bans.

We find the effects of these abortion bans are not uniformly felt, with the largest estimated impacts among populations experiencing the greatest structural disadvantages and in states with among the worst maternal and child health outcomes. Many of these populations already have worse pregnancy outcomes to begin with.

Our infant mortality results suggest that abortion restrictions may be stopping or even reversing improvements in infant mortality that have been made in recent decades, with the greatest harm and undue suffering experienced by Black infants and infants born in the South.

Even before the Dobbs decision, the states that imposed abortion bans had much worse maternal and infant health outcomes, with many counties in these states considered maternity care deserts—a situation that is only worsening in recent years. Our findings suggest these disparities across states worsen when restrictions are imposed on reproductive autonomy.

Isn’t more live births the intended outcome of an abortion ban?

SB: This isn’t just more babies being born. It is populations that are already at greater disadvantage that are being forced to continue pregnancies that, in the absence of these bans, might have otherwise been terminated. Thinking about these additional births in the context of research like the Turnaway Study, we see there are harms done to families that are struggling with an unplanned birth—not only for the birthing person but also for the existing children of these families, as most people having abortions already have children—and these extend to the mental and economic health of the family as well.

What other impacts do you want to dig into?

AG: There’s been a lot of discussion around what’s happening with pregnancy-related care and potential negative consequences for maternal mortality and maternal morbidity, and we’re looking at both of those outcomes right now.

For the maternal mortality research, we’re using death certificate data; and for maternal morbidity, we are using a variety of sources, including hospitalization data, which is close to the gold standard for measuring severe complications of pregnancy. It’s important to look at those complications, because we estimate that for every maternal death, there are about 100 cases of severe maternal complications known as severe maternal morbidity.

 One thing I want to highlight: With our infant mortality findings, we found an increase in the infant mortality rate due to congenital anomalies. But what struck me was that we also found an increase in deaths due to noncongenital anomalies, caused by things like maternal complications, but also potentially infant accidents and infant suffocation. We were a bit restricted in the way that we could analyze those other causes of death, because it’s such a rare outcome. 

It’s not just a story about congenital anomalies or defects. There’s a larger story there about negative spillover effects happening to both mothers and their babies—that’s something that we hope to disentangle in the next year.

This article was edited for length and clarity by Morgan Coulson.

RELATED

• Two New Studies Provide Broadest Evidence to Date of Unequal Impacts of Abortion Bans
• Analysis Suggests 2021 Texas Abortion Ban Resulted in Increase in Infant Deaths in State in Year After Law Went into Effect
• A Spike in Births and Other Potential Impacts of Texas’ Abortion Restrictions
• A Brief History of Abortion in the U.S.