Who can forget Mark Twain’s famous maxim: “There are three kinds of lies: lies, damned lies, and statistics”?
I’ve been brooding over this line while watching the pro-abortion left hard at work promoting the startling (and erroneous) claim that “an American woman is 14 times more likely to die” from childbirth than from abortion. They want to move this false claim into the popular imagination, to make the conversation solely about pregnant women’s safety and ignite an unreasonable fear of carrying a child to term. Because what they want — radical abortion license to the day of birth, for any reason or no reason, and paid for by you-the-taxpayer in every state of the nation — is way out of line with average Americans’ sentiments. They’d rather gin up panic over labor and delivery, which has never, in the history of mankind, been safer.
It behooves physicians like me to examine the studies critically and then to put the “lie” to that statistic — in the interests of truth and expectant mothers’ peace of mind.
The left, including the dissenting Supreme Court justices in Dobbs v. Jackson Women’s Health Organization, relies on a faulty 2012 study that used numbers that are dead wrong, the result of our country’s seriously flawed maternal mortality data collection methods — and equally flawed abortion data-gathering. Very basically, the researchers overcounted deaths from childbearing and undercounted deaths from abortion.
Undercounting and Overcounting
The American maternal mortality death rate used by these and other similar studies started to rise sharply in 2003, coinciding with a change in surveillance methods: the pregnancy checkbox introduced that year. State health departments added this checkbox to death certificates to indicate whether a woman had been pregnant within the last year. As more and more states implemented the checkbox, maternal death rates rose to stratospheric levels.
It’s easy to see why. The checkbox gathered data not only on women who had died of complications from pregnancy or birth, but also on women who died of homicide, suicide, and overdose. (In fact, these are the most common causes of death in pregnant and recently pregnant women in the United States.) It also happens to be the case that the checkbox is often checked erroneously for women who have not recently been pregnant. For instance, 187 deceased women aged 85 and older were identified as “maternal deaths” in the year 2013. That was a year in which a total of 679 women were counted. Almost a third (28 percent), then, we now know had not been pregnant in decades, if ever. Ah, statistics.
Inaccurate Reporting
The Centers for Disease Control and Prevention, to its credit, admits that coming up with accurate numbers is difficult. They refer to “issues associated with the reporting of maternal deaths on death certificates,” perhaps referring to the bizarre inclusion of elderly deceased women as “maternal deaths.”
However, even if the maternal death rate were accurate, comparing that statistic to that of abortion mortality remains a case of “garbage in, garbage out.” (A phrase we used in medical school to point out the importance of accurate data-gathering.)
To understand why, it is crucial to understand that abortion reporting is not required by federal law. It is done on a state-by-state basis. For one, California — where almost 1 in 8 Americans live — does not report abortions. In addition, 23 states do not require any type of abortion complication reporting, including deaths.
Even in states that do require reporting, most women leave abortion facilities within hours of the procedure and go to hospital emergency rooms when they experience complications. Those events are not reliably linked back to the induced abortion that caused them. Indirect abortion deaths, caused by the psychological disturbances that increase the likelihood of dying after abortion, are not captured at all. Therefore, neither the denominator (the number of abortions), nor the numerator (the number of resultant deaths) is accurate in the famously quoted 2012 study.
Better Data
You get a starkly different picture when using complete surveillance methods, with retrospective linking to medical records in countries with a single health care system. A Finnish study, for example, found that women are four times more likely to die in the year following abortion than women who give birth. In fact, women who had given birth had a lower death rate than women who had not been pregnant at all. Closer to home, a study using a complete data set recovered from California state insurance records showed twice as many women died in the two years after an abortion than after a birth (see table 1 of page 2). Causes of death in these post-abortive women included complications from the procedure — sepsis and hemorrhage, for example — though the most common causes were “indirect”: suicide, homicide, overdose, and other risk-taking behaviors.
So, when pro-abortion activists say, “childbirth is vastly more dangerous than abortion,” they don’t have the data to support it. This misuse of statistics serves many political purposes: to gin up opposition to health and safety regulations on abortion, to increase funding for abortion businesses such as Planned Parenthood, and even to discredit pregnancy resource centers that support women who choose birth for their children. But these fake claims also have real-life negative consequences for the very women about whom pro-abortion activists claim to care. They scare women away from the delight of motherhood. At the same time, their disingenuous talking point minimizes the serious medical risks that women experience in having an abortion.
Beyond all this, their claims distract from what should be our society’s unpoliticized public health focus: initiatives that address the real sources of pregnancy-related mortality, whether that pregnancy ended in abortion or in the birth of a child. Only then can we arm ourselves properly to fight every avoidable maternal death.