A study published in JAMA Network Open on October 16 found that homicides of women and girls aged 15 to 44 in states with near-total abortion bans rose above expected levels after the Supreme Court’s 2022 Dobbs decision.
Researchers estimated 48 excess deaths in 2023 and 87 in 2024. However, the study’s design cannot prove that the bans caused these deaths.
Study Finds Excess Deaths in Abortion Ban States
The study, conducted by Liana Woskie of Tufts University and Brown University and Maeve Wallace of the University of Arizona, analyzed CDC death certificate data from January 2016 to December 2024.
It compared 11 states with sustained total or near-total abortion bans, including Texas, Tennessee, Louisiana, Alabama, and Oklahoma, with other states.
In these ban states, homicide rates were approximately 7% above expected levels in 2023 and 15% above in 2024, translating to about 0.34 extra deaths per 100,000 women in 2023 and 0.61 in 2024.
The findings serve as a warning signal but do not provide a definitive conclusion. The authors stress that their work cannot prove causation due to its ecological design, which examines groups of states over time rather than individual cases.
Limitations and Uncertainties in the Analysis
Death certificates lack critical details, such as pregnancy status, relationship to the perpetrator, and involvement of intimate partner violence. These omissions prevent the study from establishing a direct link between abortion bans and homicides.
The results also varied depending on the statistical models used. A more flexible model showed larger effects in 2023 and early 2024, but the late-2024 estimate relied heavily on data from a single month with unusually low homicide rates.
For women in states with abortion bans, the findings may reflect a harsh reality where limited access to reproductive healthcare intersects with heightened risks of violence. The inability to terminate unwanted pregnancies could trap some women in abusive relationships, increasing their vulnerability to harm.
The inclusion of Texas in the dataset complicates the analysis. Its abortion restrictions began in September 2021, before the Dobbs decision, and its data was less precise. Excluding Texas, the results for other ban states remained significant.
Firearm homicides constituted a substantial and growing portion of overall homicides in both ban and non-ban states. The study could not fully account for the potential influence of differing gun laws or other policies on these rates.
National Trends and Model Sensitivity
The study found no significant national change in homicide rates after the Dobbs decision.
The estimates had wide ranges. The 2024 projection of 87 excess deaths could range from 20 to 149. The month-by-month trend comparison only narrowly reached statistical significance, highlighting the sensitivity of the results to model choices.
Checks and Comparisons
The researchers tested their approach with “placebo” start dates before Dobbs, finding no significant effect. They also compared trends by age, noting that divergences in ban states were more pronounced for women aged 15 to 29 than for those aged 45 to 59, who are less affected by abortion policies.
Funded by the National Institutes of Health, the authors reported no conflicts of interest.
This research contributes to a growing body of work on abortion policy and women’s health. An April study in JAMA Network Open found no significant change in pregnancy-associated deaths in ban states, with authors noting early and imprecise results.
Readers can evaluate such studies by asking key questions: Does the design compare like with like? Are the estimate ranges narrow or wide? Do the results hold under different assumptions? Do the authors explain the potential mechanism of the effect?
Individuals with differing views on abortion may interpret these findings differently. The study’s limitations show the need for data on pregnancy status and victim-perpetrator relationships.
For those experiencing intimate partner violence, help is available 24/7 through the National Domestic Violence Hotline at 1-800-799-7233 or by texting START to 88788. In emergencies, call 911.
Additional data, state-level analyses, and studies focused on intimate partner homicide could clarify whether this pattern persists. Currently, the study raises serious questions but does not establish causation, and its estimates should be interpreted with caution.
