Sophia Piral moved to Columbia, Missouri, from Guatemala in 2019 to study education policy. Her husband and three-month-old son came with her.
“It was difficult in the sense that we were first-time parents and, I mean, there was a different culture, different food, different language,” Piral said. “We didn’t know exactly what to do, or who to ask for help.”
Piral struggled to navigate the U.S. healthcare system. Shortly after she arrived in the U.S., the COVID-19 pandemic hit, and she lost her mother.
Throughout this big life transition, Piral noticed a change in herself and suspected she was either struggling to adjust to the U.S. or experiencing postpartum depression, which is a risk factor in pregnancy-related deaths. A pregnancy-related death is when a person dies during pregnancy or within a year after the pregnancy ends due to complications or pre-existing conditions.
“I remember not having any resources, like I mean, at all,” Piral said.
Health care organizations use pregnancy-related death statistics, or maternal mortality data, to identify where resources – like counseling and prenatal care – are needed. That's a problem in Missouri, where information about maternal mortality rates among Hispanic women is not reliable because the population is statistically too small.
In some Midwest states, maternal mortality rates of all people are higher than the national aggregate rate of 23.5 deaths per 100,000 live births from 2018-2021.
According to the Centers for Disease Control and Prevention, Hispanic maternal mortality rates in the U.S. are rising.
The most recent data published in 2021 shows the rate at 28 maternal deaths per 100,000 live births, compared to 11.8 in 2018. Although Missouri ranks in the middle of other Midwestern states in terms of maternal mortality, its statistics on Hispanic pregnancy-related deaths can’t be taken at face value.
Missing in Missouri
Data from Missouri’s Pregnancy-Associated Mortality Review Board, which functions within the Missouri Department of Health and Senior Services, shows Hispanic women’s death rates sit higher than white women and lower than Black women.
According to the Missouri DHSS, between 2017 and 2019, white women in the state had a maternal mortality rate of 18.4 deaths per 100,000 live births. For the same period, Black women had the highest rate at 61.6 deaths per 100,000 live births.
The state of Missouri doesn’t have clear numbers on how many Hispanic women have died due to pregnancy or childbirth, though.
From 2017 to 2019, the state reported 23.2 Hispanic maternal deaths for every 100,000 live births. But there’s an asterisk next to that number; it says the data should be “interpreted with caution.”
According to senior researcher Daniel Quay, who analyzes the statistics for PAMR’s annual report, the asterisk essentially means the data is what researchers call an “unstable ratio.” He said that, for now, it’s not possible to accurately calculate the rate of maternal deaths among Hispanic people in Missouri because the population isn’t big enough.
“[Hispanic data has] not quite met the threshold of being able to be a stable ratio. We don’t want to erase these women’s deaths so to speak. It’s trying to honor the fact that this was still a person that died. I cannot mathematically use this information at this time because my numbers are too low,” Quay said.
This essentially means the scientists can’t guarantee the ratio they end up with will accurately represent the extent of maternal mortality in the Hispanic population - especially when compared to larger groups in Missouri.
For instance: If one out of the nearly five million white people in Missouri died as a result of pregnancy, it wouldn’t significantly impact the ratio of pregnancy-related deaths because the population is so large. But for the roughly 300,000 Hispanic or Latino people in Missouri, it’s a different story.
“A single death has a disproportionately large impact in such small populations,” Quay said.
Because of this, the ratio may appear larger or smaller than the actual ratio of pregnancy-related deaths occurring - hence the asterisk.
Self-reporting in Nebraska
According to the Nebraska Department of Health and Human Services, during the years 2014-2018, the Hispanic maternal mortality rate was 28.3 per every 100,000 population births. This data was based on patients’ previous self-reporting and birth certificates.
Scientists largely use self-identification data from health care patients and information from coroner’s reports to determine the racial background of someone who has died. Both of those data collection methods have accuracy issues.
While self-identification is the preferred race and ethnicity data collection method, according to the U.S. Office of Minority Health, discrimination is still present in health care.
Piral said immigrants like her don’t always disclose their ethnicity, especially in medical records, because of fear of discrimination. And, coroners who identify the race of someone who has died don’t always get it right.
“We have a lot of concerns about the ‘self-reported’ race data in health care environments across Nebraska and frankly across the United States,” said Ann Anderson Berry, a professor of pediatrics with the University of Nebraska Medical Center.
She said she and many of her colleagues don’t necessarily trust the accuracy of self-reported data because race and ethnicity can also be assumed by medical providers, who then record it based on their assumptions into a patient’s health record.
Nebraska also has an asterisk next to its maternal mortality data. It’s by the word “ethnicity” and the little star indicates Nebraska bases its ethnic data solely on self-reported numbers.
Anderson Berry is also the medical director of the Nebraska Perinatal Quality Improvement Collaborative — a group of medical professionals and perinatal stakeholders in the state who work on improving access to maternal and infant health. The Collaborative makes decisions based on data, including the disaggregated data on race and ethnicity, from the CDC for its reports and to develop future initiatives.
“Without data, we do not understand where we should target our work. And we can’t measure the impact of our work to move forward to better outcomes for moms and babies and families in Nebraska,” she said.
“We have a lot of concerns about the ‘self-reported’ race data in health care environments across Nebraska and frankly across the United States."Ann Anderson Berry, a professor of pediatrics with the University of Nebraska Medical Center
Anderson Berry said not only does the data from the state’s Department of Health and Human Services rely on self-reporting, but the information is also from 2018 and outdated. She added NPQIC cannot rely on data from the state because it is not delivered in a timely manner.
“Data transparency in the state of Nebraska is an ongoing concern for NPQIC and not just for NPQIC, but for other public health initiatives across the state,” she said. “The key pieces of data that we need are not available from the state. So we are working with older data than what the citizens of Nebraska deserve to do our work.”
Anderson Berry holds South Dakota’s Pregnancy-Associated and Infant Deaths report as a good example for other states and organizations seeking to address problems with data presentation. The report defines every term, disaggregates data based on race and ethnicity and outlines how and why it uses numbers.
Piral said untrustworthy data makes it harder to advocate, demand accountability and improve programs for people who need them.
“Even though numbers don't represent the whole reality, right? There's so many things that are hidden. And there's so many things that are not tracked,” she said.
Because records don’t always tell the whole story, Missouri’s PAMR team has an investigator, called an abstractor, who is assigned to help gather information about a person who has died to help determine their background and the factors that may have influenced their death around pregnancy.
The abstractor combs through the deceased person’s digital footprint and any available medical and public records. Sometimes, this includes discovering more about the person’s race and ethnicity.
The human side
Larger studies over multiple years could help scientists capture a more accurate picture of Hispanic maternal health. But in the short term, the unstable ratios are an unavoidable issue.
For Missouri, until the Hispanic population — which is currently at about 5% in the state — grows significantly larger, scientists can’t draw a specific conclusion about Hispanic maternal mortality rates. The population is increasing, though. From 2010 to 2020, it grew by 42.6%. And yet, even if the number of maternal deaths increases in the state, there will be no red flag to alert researchers and know reliable data from previous years.
“A lot of statistical agencies wouldn't report unstable numbers, because you don't know if that number is random, or if it's indicative of the health of the population,” said sociologist and demographer Juanita Chinn, a program director with the Eunice Kennedy Shriver National Institute of Child Health and Human Development.
“But then there's the human side. And people want to know how they're represented,” she said. “They want to know how they’re being impacted by this public health crisis that is going on nationwide, but also within the state of Missouri.”
Other Midwest states with higher percentages of Hispanic populations haven’t reported problems with unstable statistics. Missouri has one of the lowest Hispanic populations in the region. But that doesn’t mean there aren’t concerns about the numbers from Missouri’s neighbors.
From 2016-2018, Kansas did not release its pregnancy-related death rate but reported that Hispanic people made up 23.1% of pregnancy-related deaths.
In 2021, Iowa reported 13% of pregnancy-related deaths were self-reported as Hispanic but did not include race information.
Chinn pointed to a National Institute of Health-funded study that shows racial and ethnic disparities in maternal mortality in the U.S. may be larger than reported in previous years.
“It’s that kind of work that keeps us informed, that allows the data to be presented in these disaggregated ways and that are accurate and reliable. So that we know where we can go in and intervene,” she said.
And in Missouri, some health officials recognize the data may not be telling the whole story.
“One incorrect assumption or analysis would be that there's not a problem,” PAMR board member Alison Williams said. “Due to small sample size, that might lead you to believe that there's not enough deaths with the Hispanic population, so it's probably not that big of a problem - but that's not the case at all. We know that Hispanic populations are often marginalized, often have decreased access to care.”
Williams, also vice president of clinical quality improvement with the Missouri Hospital Association, said the state is set to pilot a new discrimination assessment tool to determine the role discrimination plays in healthcare access and equity. What that tool looks like is unclear right now.
The lack of data is an obstacle for organizations that need funding to create programming geared toward certain groups, though.
Allocating resources
Generate Health in St. Louis is a non-profit organization that primarily provides health services for Black families. Executive Director Kendra Copanas uses PAMR data to identify community needs, create programming and advocate for racially equitable health policies.
“We actively use those reports and have conversations about what each organization is doing or can do to address the issues that are showing up in the report,” Copanas said.
Generate Health and similar non-profit organizations often rely on grant funding. Local health departments are usually in the same boat.
One of the largest sources of federal funding for maternal-child health comes from Title V, a federal block grant program. Title V provides funding to all 50 states for the improvement of public health services and systems for mothers, children and families.
Title V is different from other grants. It’s a “block” of money from the U.S. Health Resources and Services Administration that is distributed among states. State Title V offices allocate their portion of the funding.
Under Title V, subcontractors are the organizations and departments that use the block grant funding to create systems of health programs and services within a state. States also combine the federal funds from HRSA with local funds. Each state Title V office allocates the funding to its counties differently.
As part of the block grant agreement, all subcontractors list any health gaps in the community and what they are doing to address them. All states also must list their health priorities each year.
This year, Nebraska, Iowa, Kansas and Missouri, did not specifically state addressing racial or ethnic health disparities as a priority for pregnant women and maternal health. And according to records obtained by the Midwest Newsroom, about 15 of 111 counties in Missouri provided specific support to Hispanic people with Title V funding since 2017. That support included hiring bilingual service providers and translating some informational materials. The remaining four counties in Missouri don’t subcontract with Title V.
Each state is also required to show what percentage of its target population is being served by the Title V program. Missouri has one of the lowest percentages of pregnant people served, at 52.6% in 2021. This is in stark contrast to Nebraska and Iowa – and many other states– where the states report they served 100% of pregnant people.
As a result of this reporting, the Missouri PAMR report has now been translated into Spanish.
“I need to keep going”
In St. Louis, Lakisha Reddit is pushing forward to serve pregnant people in her community.
Reddit runs doula training classes where students learn how to guide and support others through the birthing process.
Reddit recently received funding to translate all of her doula-training materials into Spanish, making them among the few non-English resources provided in Missouri. The materials will also be translated into French, Arabic and Somali.
“I need to keep going. The funding that I currently have is two years and that’s it after this,” Reddit said. “I don’t know what’s to come next.”
Reddit said cultural relevance is important to her, and she wants all future doulas – especially those born outside the U.S. – to know that they should share their culture and traditions.
She has one Spanish speaker in the class of 19: Mayelin Escobar, originally from Cuba.
Escobar said one of her goals is to make sure Hispanic women feel seen by their healthcare providers.
“I work with all the Latino moms and they have issues with the hospitals like Black moms and Black women, and they have a problem with the translation,” Escobar said. “I want to help my moms and the hospitals, too.”
As a certified doula, Escobar said she hopes to help Hispanic moms who encounter healthcare inequities. She graduated from the program in 2023.
Piral couldn’t find any Spanish-language mental health services in Columbia, Missouri, or any resources culturally relevant to a family from Guatemala.
In her home country, Piral would have been able to walk down to the local tienda to pick up a food supplement called Incaparina for her son. It’s a popular powdered drink many people in Guatemala feed their babies to prevent malnutrition. But in Missouri, she struggled to find the comforts and resources of home.
Like Reddit, Piral decided to take matters into her own hands, starting a support group for Latina moms in the area.
“We barely have medical access. And we don't have psychological support,” she said. “In our culture, we don’t talk about mental health. So here in the U.S., we are not going to have access to that. And we don’t have the resources to pay for that. So, I was like, ‘Okay, we can just talk. And that’s gonna help us release all the things we’re going through to help us, right?’”
Piral is now working on her Ph.D. in the Educational Leadership and Policy Analysis program at the University of Missouri-Columbia. Her focus is understanding the experiences of Latin American immigrant families and teachers in U.S. schools. She hopes to help develop policies that allow children to have school careers free of stigma, discrimination or any type of violence.
Her son will be a kindergartner this August.
“Something that I want to do with this is to recognize that every Latina mom has knowledge, has a certain experience,” she said. “And yeah, we may need support from others, but still we have an expertise that has to be valued.”
This story comes from the Midwest Newsroom, an investigative journalism collaboration including IPR, KCUR 89.3, Nebraska Public Media News, St. Louis Public Radio and NPR.
