Why Are So Many Mothers Dying in Texas?

4 minute read

The number of women who are dying during pregnancy, or soon after giving birth, has risen nationwide, and a new report shows an alarming 87% spike in deaths in Texas. As of 2015, the maternal mortality rate in Texas is 32.5 per 100,000 live births, and the increase accounts for around 50 to 60 additional deaths per year in Texas.

“We think there is a very significant problem in Texas,” says Marian MacDorman, a research professor at the Maryland Population Research Center at the University of Maryland, College Park, and the lead author of the new study, published in the journal Birth.

MacDorman and her colleagues published a 2016 study showing a more than 20% increase in maternal mortality rates across the U.S. Since maternal mortality rates are declining worldwide, the increase in the U.S. is cause for alarm. But the researchers also noticed a “strange pattern” in Texas, where there was an especially notable spike in cases: an 87% increase in maternal deaths in the state, when comparing data from 2006-2010 to 2011-2015.

The researchers say that the findings highlight two key problems: a true increase in the number of women dying during or after pregnancy, and a lack of quality data collection that makes understanding the trend more difficult. The researchers found that in Texas, the highest maternal mortality rates were among women over age 40. From 2011 to 2015, the maternal mortality rate for women over 40 was 27 times the rate for women under 40.

There are a few possible reasons for the high rate in deaths among women over 40. Women of that age group are often considered higher risk pregnancies, and more women are having babies at that age than in the past. However, the actual deaths among women over 40 likely aren’t as numerous as the findings suggest, MacDorman says, and the discrepancy may be because of a reporting problem.

A new question was added to death certificates in the U.S. in 2003, which was meant to determine whether women who died were pregnant at their time of death or if they had already given birth. Texas made the addition in 2006. However, studies have shown that many errors are made in answering this question and that many cases are misreported. In several cases, women who were not pregnant or were not postpartum within one year of death are reported as being so.

MORE: Why U.S. Women Still Die During Childbirth

Yet even with weaknesses in the data reporting, MacDorman discovered what she says is a true spike in the maternal mortality rate in the state. When the researchers looked just at women under 40, who are less likely to be misclassified, there was still a 36% increase in maternal mortality from the two time periods the researchers studied. There are likely many reasons for the increase, including higher instances of chronic disease than in the past, access to care issues for women without insurance, a higher prevalence of cesarean sections and more.

In Texas, MacDorman noticed that there’s a very high number of deaths that are classified as from “nonspecific causes,” which means that people looking at the upward trend in maternal deaths are not getting very much information about what caused them. Without that information, places like hospitals cannot design interventions to prevent common causes of death. “You have to know what went wrong,” says MacDorman. “With [large numbers of] deaths going to nonspecific causes, it hampers the efforts to prevent the deaths.”

MacDorman says she would like to see an evaluation of the data by the National Center for Health Statistics—the government agency responsible for the National Vital Statistics System (NVSS), the source of official maternal mortality statistics—so that improvements can be made. When errors or anomalies arise, the group has the ability to investigate and make corrections—a process that MacDorman says is needed. “When prevention efforts are scattered and unfocused, more women die,” she says. “That’s what’s going on.”

More Must-Reads from TIME

Contact us at letters@time.com