A new study published in the same medical journal tamps down the alarm, offering different data that reduce the number of deaths by nearly two-thirds.
Researchers with the Texas Department of State Health Services set out to evaluate the original findings published in the journal of Obstetrics & Gynecology, which, they say, showed 147 maternal deaths in their state in 2012, an unprecedented increase from 2010, 72 deaths. The 2012 number was significant enough to inspire legislative efforts, including the establishment and extension of a maternal mortality and morbidity task force.
What researchers found upon further analysis, which included examinations of medical and autopsy records, was that there were actually 56 — not 147 — maternal deaths in 2012. Human error while using an electronic death registration system appears to have been the biggest culprit in the discrepancy.
Electronic filing of death certificates by doctors, justices of the peace and medical examiners grew 44% in Texas from 2010 to 2012, researchers found. Those who filed the certificates most likely clicked on the wrong dropdown menu option when entering pregnancy status, thereby suggesting that more than twice the number of deceased women were pregnant upon their deaths when they were not.
“Other academic research has shown issues with the quality of death data, nationally,” said the new study’s lead author, Sonia Baeva, a maternal mortality and morbidity epidemiologist with the Texas Department of State Health Services, in a statement. “Death data can be a rich source of information, but our work shows that identification of rare events, such as maternal deaths, should be supported with additional evidence.”
The new study takes some heat off Texas, which state Sen. Lois Kolkhorst applauded in a statement.
“It is clear that Texas does not have the worst maternal mortality in the developed world and that previous reports were grossly inaccurate,” said Kolkhorst, who authored State Bill 17 in 2017 to extend the maternal mortality and morbidity task force. “Instead of overreacting to sensationalized reports in the media, we directed better data collection in order to develop targeted, evidence-based strategies to more effectively improve the lives and health of mothers in Texas.”
The lead author of the 2016 study is all for accurate data, but she clarifies that a followup study should have never been necessary in the first place.
Marian MacDorman of the Maryland Population Research Center explained that the data she and her co-authors worked with were drawn directly from the Vital Statistics Division of the Texas Department of Health, which — as in all 50 states — passed its information along to the National Center for Health Statistics.
“The conflict is not really between data from ‘my’ study and ‘their’ study,” she said. “The conflict is that the Vital Statistics section of the Texas Department of Health has different numbers from the Texas maternal mortality review task force.
“It’s appropriate that they did the study, given that the data was so bad,” she said, “but basically they shouldn’t have to do this kind of study.”
Time spent correcting counts in a data system takes away from maternal health work itself, MacDorman said. Her hope is that Texas can now focus on the women who need health services.