Maternal Mortality and the Data we don’t have

As part of my politics of childbirth series, this week we’re going to look at data, or more accurately, a lack of data.

Throughout my dissertation I thought it was strange that we lumped all Asian and Hispanic/Latina people together without dis-aggregating categories. But in my own qualitative research, my sample set was so small that I did the same thing. The lack of data by subgroup was a little annoyance that lived in the back of my head while the more pressing needs of dissertation coding, writing, and defending took precedence.

Until I met some seriously motivated students at Boston University.

Racial Disparities

Racial disparities in maternal health are well documented among Black women (and often by Black women who did a majority of the early work calling attention to it). Black women are 3-4 times more likely to die than white women in childbirth, in some places more likely. There are any number of factors here including medical racism, the “weathering” effect that chronic exposure to racism generally has on Black women’s health, and lack of prenatal and postpartum support, among other things.

Serena Williams and Beyonce, both of whom are more physically fit and wealthier than most of us put together will ever be, had “near misses” with maternal mortality. But as William said, she tried to get help, but “no one was listening.”  If these women with their physical health and economic privilege have a hard time getting doctors to listen, the rest of us are in trouble. 

As NPR reports, these near misses are common: “for every woman who dies in childbirth 70 come close.” Great, right?

There are some great hashtags out there showing the ways in which we don’t need more research (#FundTheResults). But we might need more research, or to look at the research we have in a different way. Because, it turns out, when you break apart racial classifications in Latinas and Asians, you get some very different stories.

Data Gaps

For my Politics of Childbirth class at BU, students chose their own final paper topic. After our class discussions on maternal mortality statistics and racial disparities, several of my students ran with those topics. Particularly, one chose to look at dis-aggregating the statistics on Asians and another on Hispanic/Latinas.

The student who was looking for statistics that dis-aggregated Southeast Asians from Asians came up to me furious and devastated. He said, “I thought I wouldn’t find a lot, I didn’t think I’d find nothing.”  Fortunately (for his paper and for public health) there is some recent research getting into ethnic differences and dis-aggregating the statistics on Asian and Southeast Asian infant and maternal mortality numbers, but he was right that it’s slim pickings. We are in desperate need of more information there.

Another student studied the data gap in Hispanic and Latina communities. She found research suggesting that while ethnicity is important in understanding different rates of infant and maternal mortality than just broad categories there’s still not a lot of information out there. In 2022, there was a surge in Hispanic/Latina maternal mortality…maybe? The numbers aren’t clear.

Like any area that tries to categorize social frames, like race or gender or class, it is difficult to know where to draw the lines in these categories. But there is evidence out there that we don’t have enough evidence to know how important dis-aggregating racial categories would be. I’m happy to see more research coming out, especially in the last five years, that casts a wider (or actually more specific) net but if you’re looking for maternal health research to support, go find some scholars working in racial disparities and help them out.

Of course, given the years of research into Black maternal mortality and limited amount of action in response, this might just be all academic anyway. I want to be clear that I want public health policies to keep pregnant and postnatal people safer, I’m just not entirely convinced data alone will be enough to do that. Still, it couldn’t hurt.

Maternal Mortality Crisis?

With new research suggesting that the maternal mortality crisis might not actually be a crisis, it is more important than ever to get this good data. I’m still assessing and adjusting to this new research, so my thoughts here might be wildly different in a few months, but at the moment, there are a few reasons I’m not sure the maternal mortality crisis can be chalked up to a matter of bad statistics.

One benefit of the new research is that it tries to bring US statistics in line with those globally: we count maternal mortality for an entire year, many other countries count it at three to six months postpartum. Our time frame of what counts as maternal mortality is longer, meaning the statistics could be skewed. However, what we do know is that a majority of maternal deaths in the US do occur in the first few months (the CDC says six weeks). This still puts us in the rough time frame where we can compare ourselves to other countries. What we also know is that there are disparities by race that will persist even if we alter the statistics.

With half of maternal mortality happening once patients leave the hospital, it does becomes harder to tell when a death is pregnancy/childbirth related and when it’s not. The new research rightly points out the problem with the “check box” many coroners use that indicates whether someone gave birth within a year of their death but does not always include whether their death was birth/postpartum related. That is something that needs to be assessed more (not a small task).

It’s going to be important to get that data to really know what’s happening.

Why I’m worried: claiming that the maternal mortality crisis was all a matter of bad data is going to obfuscate that there are real problems in our public health system. There are demonstrable racial disparities across medical fields. If the maternal mortality crisis broadly is less national crisis and more specific crisis, then we need to express it that way rather than saying it doesn’t exist.

If we’re going to talk about how we include “incidental” or loosely related maternal deaths, we need to be clear about what that means. Blood clots, heart attacks, all very clearly related. What about intimate partner violence or suicide? I say that’s likely related. Others won’t. Either way you have to assess that on a case-by-case basis which is really flipping difficult.

These are important conversations to have, and we should have them and go where the data leads. But, as it relates to today’s topic, it is also possible that calling the maternal mortality crisis overblown will pull resources away from those studying maternal mortality and racial disparities, meaning that when we need to actually get more into the data people might start pivoting to other research.

All in all, I’m suspiciously pessimistic about the developments in the field, but hopefully it will all work out.