U.S. maternal deaths – those during pregnancy or within 42 days of pregnancy – increased substantially by 33.3% after March 2020 corresponding to the COVID-19 pandemic onset, according to new research published in JAMA Network Open.
Data from the National Center for Health Statistics (NCHS) revealed this rise in maternal deaths was higher than the 22% overall excess death estimate associated with the pandemic in 2020.
Increases were highest for Hispanic and non-Hispanic Black women, exacerbating already high rates of disparity in comparison with White women, wrote Marie E. Thoma, PhD, an associate professor at the University of Maryland, College Park, and Eugene R. Declercq, PhD, a professor at Boston University.
The authors noted that this spike in maternal deaths might be caused either by conditions directly related to COVID-19, such as respiratory or viral infections, or by conditions worsened by pandemic-associated health care disruptions including those for diabetes or cardiovascular disease.
The precise causes, however, could not be discerned from the data, the authors noted.
The NCHS reported an 18.4% increase in U.S. maternal mortality from 2019 to 2020. The relative increase was 44.4% among Hispanic, 25.7% among non-Hispanic Black, and 6.1% among non-Hispanic White women.
“The rise in maternal mortality among Hispanic women was unprecedented,” Dr. Thoma said in an interview. Given a 16.8% increase in overall U.S. mortality in 2020, largely attributed to the COVID-19 pandemic, the authors examined the pandemic’s role in [the higher] maternal death rates for 2020.
“Prior to this report, the NCHS released an e-report that there had been a rise in maternal mortality in 2020, but questions remained about the role of the pandemic in this rise that their report hadn’t addressed,” Dr. Thoma said in an interview “So we decided to look at the data further to assess whether the rise coincided with the pandemic and how this differed by race/ethnicity, whether there were changes in the causes of maternal death, and how often COVID-19 was listed as a contributory factor in those deaths.”
A total of 1,588 maternal deaths (18.8 per 100,000 live births) occurred before the pandemic versus 684 deaths (25.1 per 100,000 live births) during the 2020 phase of the pandemic, for a relative increase of 33.3%.
Direct obstetrical causes of death included diabetes, hypertensive and liver disorders, pregnancy-related infections, and obstetrical hemorrhage and embolism. Indirect causes comprised, among others, nonobstetrical infections and diseases of the circulatory and respiratory systems as well as mental and nervous disorders.
Relative increases in direct causes (27.7%) were mostly associated with diabetes (95.9%), hypertensive disorders (39.0%), and other specified pregnancy-related conditions (48.0%).
COVID-19 was commonly listed as a lethal condition along with other viral diseases (16 of 16 deaths and diseases of the respiratory system (11 of 19 deaths).
Late maternal mortality – defined as more than 42 days but less than 1 year after pregnancy – increased by 41%. “This was surprising as we might anticipate risk being higher during pregnancy given that pregnant women may be more susceptible, but we see that this rise was also found among people in the later postpartum period,” Dr. Thoma said.
Absolute and relative changes were highest for Hispanic women (8.9 per 100,000 live births and 74.2%, respectively) and non-Hispanic Black women (16.8 per 100,000 live births and 40.2%). In contrast, non-Hispanic White women saw increases of just 2.9 per 100,000 live births and 17.2%.
“Overall, we found the rise in maternal mortality in 2020 was concentrated after the start of pandemic, particularly for non-Hispanic Black and Hispanic women, and we saw a dramatic rise in respiratory-related conditions,” Dr. Thoma said.
In a comment, Steven Woolf, MD, MPH, director emeritus of the Center on Society and Health at Virginia Commonwealth University, Richmond, said the findings are very consistent with his and others research showing dramatic increases in overall death rates from many causes during the pandemic, with these ranging from COVID-19 leading conditions such as diabetes, cardiovascular and Alzheimer’s disease to less-studied causes such as drug overdoses and alcoholism caused by the stresses of the pandemic. Again, deaths were likely caused by both COVID-19 infections and disruptions in diagnosis and care.
“So a rise in maternal mortality would unfortunately also be expected, and these researchers have shown that,” he said in an interview. In addition, they have confirmed “the pattern of stark health disparities in the Hispanic and Black populations relative to the White. Our group has shown marked decreases in the life expectancies of the Black and Hispanic populations relative to the White population.”
While he might take issue with the study’s research methodology, Dr. Woolf said, “The work is useful partly because we need to work out the best research methods to do this kind of analysis because we really need to understand the effects on maternal mortality.”
He said sorting out the best way to do this type of research will be important for looking at excess deaths and maternal mortality following other events, for example, in the wake of the Supreme Court’s recent decision to reverse Roe v. Wade.
The authors acknowledged certain study limitations, including the large percentage of COVID-19 cases with a nonspecific underlying cause. According to Dr. Thoma and Dr. Declercq, that reflects a maternal death coding problem that needs to be addressed, as well as a partitioning of data. The latter resulted in small numbers for some categories, with rates suppressed for fewer than 16 deaths because of reduced reliability.
“We found that more specific information is often available on death certificates but is lost in the process of coding,” said Dr. Thoma. “We were able to reclassify many of these causes to a more specific cause that we attributed to be the primary cause of death.”
The authors said future studies of maternal death should examine the contribution of the pandemic to racial and ethnic disparities and should identify specific causes of maternal deaths overall and associated with COVID-19.
In earlier research, the authors previously warned of possible misclassifications of maternal deaths.
They found evidence of both underreporting and overreporting of deaths, with possible overreporting predominant, whereas accurate data are essential for measuring the effectiveness of maternal mortality reduction programs.
Dr. Thoma’s group will continue to monitor mortality trends with the release of 2021 data. “We hope we will see improvements in 2021 given greater access to vaccines, treatments, and fewer health care disruptions,” Dr. Thoma said. “It will be important to continue to stress the importance of COVID-19 vaccines for pregnant and postpartum people.”
This study had no external funding. The authors disclosed no competing interests. Dr. Woolf declared no conflicts of interest.