Unsettling truths about maternal mortality in the U.S.

Grave warnings of a crisis in maternal deaths are being slightly eroded by articles suggesting that advocates overreached and that the “crisis” was primarily a function of measurement error.

Both sides are partially right, because measuring maternal mortality is uniquely challenging. The question at the heart of identifying a maternal death — “Would this person have died if she hadn’t been pregnant?” — often can’t be answered simply from a medical record. In medical emergencies that occur during birth, such as unstoppable bleeding, the decision is straightforward. The answer in a death that occurs several months after giving birth, like was a stroke or opioid overdose related to the pregnancy, is a challenging judgment call and honest people can disagree.

The U.S. uses three different systems to identify maternal deaths; each has its particular strengths and limitations. The National Vital Statistics System (NVSS) has the most timely, if least precise, data and makes it publicly available for analysis. The Centers for Disease Control and Prevention’s Pregnancy Mortality Surveillance System (PMSS) is more reliable but slower, doesn’t provide state level estimates, and is available only to CDC researchers. State Maternal Mortality Review Committees can do the richest analyses and provide concrete, localized recommendations, but they’re also generally the slowest, are not consistent across states, and don’t provide national estimates.

The NVSS rates, because their data are public, have been the subject of most of the criticism, including from me. But the fact that each of these systems has limitations doesn’t mean the U.S. has no maternal mortality problem. Even using the most conservative measure, the U.S. fares poorly in global comparisons and racial disparities persist.

What’s needed is to draw on the respective strengths of each system and recognize that individuals and organizations developing these estimates — and they’ll always be estimates — need to collaborate to achieve greater accuracy, transparency, and consistency in making judgments on how pregnancy-related a death is.

U.S. maternal mortality isn’t going up…or down

For the past decade, virtually every article about U.S. maternal mortality begins with some form of the phrase “Maternal mortality is rising at an alarming rate in the U.S.” Start at the year 2000, however, and examine data from the most reliable system, the PMSS, and you’ll find that maternal mortality rates have changed little and have been essentially the same from 2009 (17.8 per 100,000) through 2019 (17.6) prior to the pandemic.

These plateaued rates have been on the CDC website for years, but the message they convey — unacceptably high, but flat, rates of maternal deaths — isn’t very compelling to reporters or policymakers. The fact that rates in the NVSS appeared to be rising (partially because of the measurement changes critics have attacked) was a much more effective message to get the public engaged. And it worked. Likewise, claiming now that there is no problem is news.

But shouldn’t having, even under the most conservative measure, the highest maternal death rate among industrialized countries be enough to motivate action?

Looking at the wrong timeframe

The World Health Organization’s standard measure of maternal mortality is “The annual number of female deaths from any cause related to or aggravated by pregnancy or its management (excluding accidental or incidental causes) during pregnancy and childbirth or within 42 days of termination of pregnancy….”

The CDC has for decades been using a broader timeframe, extending the analysis of deaths up to one year postpartum. In 2015, it also began reporting the timing of maternal deaths. The impact of these decisions has been profound, not only in adding a significant number of late postpartum pregnancy-related deaths to the totals, but in changing the way we think about maternal death. The CDC found that only one in nine deaths (11%) happen on the day of delivery. Another one-quarter occur during pregnancy, with more than half (63%) occurring in the year following delivery.

This longer timeframe changed the perception of maternal death from an exclusively medical issue for obstetricians and hospitals to a public health challenge to be addressed by communities. The U.S. has long relied on a system more focused on healthy babies rather than healthy mothers. Once the baby is safely born, commitment to the mother’s health fades. The result? It is now possible to document hundreds of preventable postpartum deaths each year.

Covid exacerbated maternal mortality disparities

As overall death rates surged during the pandemic, so did pregnancy-related deaths. Yet Covid’s impact made existing racial/ethnic disparities much worse. Overall Covid vaccination rates for non-Hispanic Blacks and Hispanics were significantly lower than for non-Hispanic whites. Two examples illustrate how deeply the pandemic intensified inter-racial differences:

Comparing pregnancy-related mortality rates from 2017 to 2019 (before the pandemic) to 2020 (when the pandemic first took hold), the rate for non-Hispanic whites rose 28%, while the rates in other groups rose at far greater paces — non-Hispanic Blacks, 40%; Hispanics, 95%; and American Indian Alaskan natives, 98% — increasing existing interracial/ethnic disparities.

Was Covid responsible for these differences? A study of 2020 data found that only 2% of deaths among non-Hispanic white mothers were Covid-related, versus 11% among non-Hispanic Blacks and a stunning 29% among Hispanics.

The U.S. public health system is meant to mitigate disparities, but there was a failure to effectively communicate the importance of vaccination, especially to people of color. Combined with a rise in misinformation and/or a failure to add and adapt resources in response to the crisis, existing disparities in maternal death got notably worse during the pandemic.

What lessons can be drawn from these uncomfortable truths? Start with humility. The U.S. will never have a perfect measurement for maternal morbidity and mortality, but all three existing maternal mortality assessment systems can help educate decision-makers and the public about what’s real, without relying on hyperbole to scare them into action or using claims of bad data to do nothing. The fact that U.S. maternal mortality has plateaued at its current dreadful rate should be sufficient justification to merit action.

Most importantly, maternal mortality must be understood as a public health challenge requiring concerted commitment not only because it can lead to healthier babies, but because we value pregnant people for who they are.

Eugene Declercq, Ph.D., is a professor of community health sciences at the Boston University School of Public Health, professor of obstetrics and gynecology at the Boston University School of Medicine whose research focuses primarily on maternal mortality and morbidity, founder of the website www.birthbythenumbers.org, and a member of the Massachusetts Maternal Mortality Review Committee.