In an FAQ page about abortion pills—namely mifepristone, brand-named Mifeprex—the Food and Drug Administration is now including the number of deaths reported to the agency after people took the pills. The only problem is that there's no evidence the 24 deaths were caused by either Mifeprex or misoprostol, the two drugs that are used together to induce abortions in pregnancies that are 10 weeks along or less. The "24" figure has been touted by at least one anti-abortion legal advocacy group, but mainstream gynecologists say it's not a number they'd report to their patients when talking about the risks of medical abortion, because it doesn't convey just how unlikely it is for the pills to cause serious injury and death.
"It does not make sense to present the information like that," Daniel Grossman, director of Advancing New Standards in Reproductive Health, a research institute at the University of California–San Francisco, writes in an email.
The FDA's Office of Media Affairs referred Pacific Standard to the FAQ, as well as a few other pages, in response to requests for comment on the movement to get it to relax its rules around mifepristone. The way the United States regulates mifepristone—allowing only certified hospitals and doctors to dispense the pills—is out of step with international guidelines, unnecessarily limiting Americans' access to medical abortions, advocates argue.
As a justification for additional FDA regulation, the page doesn't make sense. It says that 24 deaths have been reported among Americans who took mifepristone and misoprostol, the abortion pills, but also notes that the FDA can't say whether the pills caused those deaths and other non-fatal, but serious, side effects.
In the FDA's full report on the 24 deaths, linked in the FAQ, many of the deaths seem obviously unrelated to medical abortion, including two homicides, one natural death from pulmonary emphysema, and one overdose on methadone, a prescription opioid. In a brief published in April, ANSIRH, Grossman's group, estimated that 13 of those deaths were "probably or possibly related to the abortion." The rest "appear to be unrelated to the abortion."
As of press time, the FDA Office of Media Affairs didn't answer questions about the agency's choice to report the 24 deaths in its FAQ.
It's no surprise that the FDA collects these numbers. Regulations say that when a doctor or patient sends a drug company a report that someone has died after taking the company's products, the company must forward that information to the FDA, no matter the cause of death. But it's misleading to potential patients to report the 24 number in a consumer-facing FAQ page, clinicians say. And it appears the decision to do so was recent, after a general overhaul of FDA.gov websites earlier this year.
Besides the fact that not all of those deaths are due to abortion, there's another problem with putting that number, and that number only, in an FAQ. "It's critical that any information about risk be presented as a ratio that includes a denominator," Grossman writes. The 24 deaths are out of an estimated 3.7 million doses of abortion pills that Americans have taken since the FDA approved Mifeprex in 2000. That means that, even including the unrelated fatalities, medical abortion's mortality rate is less than one in 100,000, a lower reported rate than penicillin's (two in 100,000) and Viagra's (four in 100,000), as ANSIRH's brief points out.
It's possible that doctors and nurses are under-reporting deaths. Surveys suggest clinicians often don't send in death reports—due to any drug—to the FDA's system, which is voluntary. Clinicians say they don't report some fatalities because they couldn't definitively attribute a death to a drug, or they just didn't have time. But this caveat would apply to any drug, not just mifepristone and misoprostol.
Gynecologists have their own standards for what information to talk about with patients, which don't align with the FAQ. A spokeswoman for the American College of Obstetricians and Gynecologists, which represents about 90 percent of practicing OB-GYNs in America, says the college can't comment on the FDA's communications decisions. Instead, she sent me the group's practice guidelines—which are written by experts who search existing studies and evaluate them using a standard method that's widely accepted by clinicians—for doctors who oversee medical abortions. The guidelines recommend ACOG members tell patients that bleeding and cramping are an expected part of the procedure; that they might also experience side effects such as nausea, vomiting, and diarrhea; and that if they bleed through two maxi pads per hour, two hours in a row, then they should seek medical help. Large studies suggest that less than 1 percent of patients will need emergency surgery because of excessive bleeding, the guideline says, and that this rate may be even lower among patients with especially early pregnancies, 49 days' gestation, or less. The guideline doesn't say anything about death.
"What I tell patients, when they talk to me, is that an abortion, as far your risk of death is concerned, is about 15 to 20 times safer than having a child," says Alice Mark, medical director of the National Abortion Federation, a professional and advocacy group for clinicians who perform abortions. In 2015, the rate of pregnancy- and birth-related deaths among Americans was 17.2 per 100,000 live births, according to the Centers for Disease Control and Prevention. American Indian and black women suffer from far higher rates than that.
Mark had never seen the FDA's Mifeprex FAQ before, despite her career in abortion advocacy and care. So it may not be reaching many patients. I had seen the FDA's fuller report on the 24 deaths once previously, in a press release that a spokeswoman for the anti-abortion legal advocacy group Liberty Counsel sent me.
Whatever their purpose, FDA-written FAQ pages about abortion pills have existed for a long time. The earliest Internet Archive record of one dates to 2009, according to Aaron Lemelin, an analyst at the Sunlight Foundation's Web Integrity Project, which monitors government webpages. That page says it was originally published in 2007. Several previous versions of the FAQ, which Lemelin identified for me, didn't mention the overall number of deaths reported to the FDA after someone had undergone a medical abortion.
Versions of the page dating to 2007 and 2009 enumerated deaths due to severe, septic infections. At the time, there was some question about whether medical abortion could cause sepsis, Mark says, but it turned out that there was a rash of similar deaths then, triggered by the same germs, among people who had had gynecological procedures and natural miscarriages, or just given birth. So something else was going on at the time that put people undergoing reproductive care at risk for these infections, Mark says. The exact problem was never determined.
A 2016 version of the page doesn't say the number of septic deaths.
The FDA's latest version of its abortion-pills FAQ seems to have been posted this April, around the time that the agency overhauled its website and gave thousands of webpages new URLs, including this one, Lemelin found.
Watchdog groups have found other agencies under the Trump administration to have edited their webpages, seemingly for political reasons. For example, at the beginning of 2018, volunteer advocates found that the feds had taken down hundreds of government pages about climate change. President Donald Trump has repeatedly stated his disbelief in climate change, and stocked his cabinet with climate deniers.
The Web Integrity Project hasn't found similarly troubling edits to FDA webpages, although it hasn't finished combing through the domain yet, Sarah John, the project's director, writes in an email. If the 24 deaths were added to the medical abortion FAQ for political reasons, it would fall in line with the Trump administration's platform. As president, Trump has talked frequently about being pro-life, as has Alex Azar, the head of the Department of Health and Human Services, the larger agency that FDA is a part of.