A new study led by researchers at the University of Oxford, published today in JAMA Pediatrics, offers fresh insight into trends in maternal mortality in the United States. For the first time, the study disentangles genuine changes in health outcomes from shifts caused by how deaths are recorded. Nevertheless, the study confirms the devastating impact of the COVID-19 pandemic on maternal death rates for women of all racial and ethnic groups.
The study, based on data from 2000 to 2023, investigated how the introduction of a ‘pregnancy checkbox’ on death certificates—implemented at different times across US states—affected reported maternal death rates. The team, which included researchers from Brown University and Columbia University, compared states where the checkbox had been implemented with those who had yet to introduce the policy – an approach known as ‘difference-in-differences analysis.’
According to the results, introducing the checkbox accounted for two thirds (66%) of the rise in maternal deaths between 2000-2019, adding an additional 6.8 deaths per 100,000 live births each year. Once adjusted, maternal mortality rates remained largely stable for much of the pre-pandemic period.
However, a sharp increase was recorded in 2021 during the COVID-19 pandemic, with adjusted maternal mortality rates peaking at 18.9 deaths per 100,000 live births—nearly double the rate of previous years (between 6.8 and 10.2 deaths per 100,000 live births). This sharp increase was seen for women of all age groups (except 15-19 years), and across all racial and ethnic groups. Rates returned to pre-pandemic levels by 2022.
Senior author Associate Professor Seth Flaxman (Department of Computer Science, University of Oxford) said: “We started this study by asking what we assumed was a simple question: has the number of pregnant women who die each year in the United States gone up or down since the year 2000? As is so often the case in public health, the numbers do not speak for themselves. At the height of the COVID-19 pandemic, pregnant women died at significantly higher rates than they did before the pandemic. But when we took a careful look at the two decades before the pandemic, we found remarkably little change: in 2019 and 2023, pregnant women died at roughly the same rates as they did a generation earlier.”
The study also highlighted persistent and striking disparities among different demographic groups. Maternal mortality among non-Hispanic Black women was consistently around three to four times higher than that of non-Hispanic White women throughout the study period. For instance, before the COVID-19 pandemic, maternal mortality for non-Hispanic Black women was 20.7 per 100,000 live births compared with 5.0 for non-Hispanic White women.
Whilst all demographic groups were impacted by the COVID-19 pandemic, Native American or Alaska Native women saw the largest increase during the pandemic period, with mortality rates nearly tripling from 10.7 to 27.5 deaths per 100,000 live births.
By contrast, fetal and infant mortality rates showed a general decline from 2000 through 2020. Infant mortality dropped from 6.9 to 5.4 deaths per 1,000 live births during this period, though a small increase was observed from 2021 onward.
Lead author Robin Y. Park (PhD student in Engineering Science, University of Oxford) said: “By accounting for changes in reporting, our study clarifies the significant impact of public health emergencies on maternal outcomes. Any future initiatives on pandemic preparedness should consider provisions for childbirth and postpartum care to limit preventable maternal deaths.”
Co-author Assistant Professor Robbie M. Parks (Columbia University) said: “The public should be informed about how important accurately tracking death rates are for public health action. Our analysis here is an example of how important it is to take into account changes in data collection, which could lead in some cases to opposite conclusions”.
Co-author Assistant Professor Alyssa Bilinski (Brown University) added: “Amidst declining funding for public health surveillance in the US, our results underscore the importance of investing in systems that systematically collect and rigorously analyse maternal and infant health data to inform policy. They also suggest additional questions, particularly how best to reduce significant racial disparities in poor outcomes and how to best understand the comparability of US metrics to peer countries.”
The data for the study was provided via the National Vital Statistics Service (NVSS), a repository of publicly available data managed by the Centres of Disease Control (CDC).
