A report from the U.S. Commission on Civil Rights last week claimed that women are 50% more likely to die from childbirth today than they were a generation ago.
Gail Heriot, who sits on the commission but regularly dissents to its extreme suggestions, published a rebuttal to the claims that women were dying as a result of childbirth at much higher rates than they were decades ago.
“Alas, like many of our Commission reports, this one is a disappointment. We’re unlikely to save lives if we don’t make a greater effort to sort out fact from fiction. I don’t see nearly enough of that effort in this report,” Heriot wrote in her rebuttal.
“[P]erhaps the most serious is its repeated allegation—usually in the form of unexamined quotes from supposed experts—that ‘racism’ is what’s causing racial disparities in maternal mortality. This allegation (which lately I have also been seeing on publicly funded posters and billboards here in San Diego) will not help us reduce maternal deaths. Instead, it will encourage racial minority mothers to view medical professionals as hostile or even malicious. That is much more likely to make things worse than better,” she added.
Heriot explained in her rebuttal that if the rate of maternal mortality had increased 50%, it would be “alarming”; however, the increase appears to be the result of “changes in how the United States keeps track of maternal mortality.”
Here’s how Heriot explained the change in record keeping:
In 2003, a pregnancy question was added to the revision of the U.S. Standard Certificate of Death. This question includes a series of checkboxes designed to elicit whether the decedent was pregnant at the time of her death or whether she had been pregnant in the last year. These checkboxes were added for a reason: Researchers feared that pregnancy-related deaths were being under-reported and hoped the checkboxes would improve the likelihood that a pregnancy-related death would be reported as such. The checkboxes made it more likely that a physician preparing a death certificate would inquire into the decedent’s pregnancy status. If it turned out she was or had recently been pregnant, the physician could be more attentive to the possibility that pregnancy increased the likelihood of the death. Such cases could, for example, be referred to medical professionals with expertise in making a judgment about pregnancy relatedness.
The federal government, however, does not directly control the form of death certificates. Individual states do. Not all states were quick to adopt the checkbox recommendation (and some had made efforts even before 2003 to improve the likelihood that a pregnancy-related death would be reported as such). As each state eventually fell into line, the reported rate of maternal mortality ticked up in that state—not because more pregnancy-related deaths were occurring, but rather because more deaths were being classified as pregnancy-related. This is exactly what those who recommended the checkboxes had hoped for. We shouldn’t be surprised that what they intended is what actually happened.
As Heriot explained, comparing numbers before and after a change to reporting was made is unreliable. She also wrote that the entire notion of a pregnancy-related death could be misinterpreted and that a clear definition is needed, as it could mean anything from a death that would not have occurred were it not for the pregnancy to someone dying while pregnant or recently pregnant and the pregnancy possibly increasing the likelihood of death.
Of course, media outlets have already jumped to report the commission’s original claim, and trying to correct the record will be difficult since media outlets tend to focus on stories that purport hardships for women.