Throughput: Mifepristone Edition

I am not a lawyer.

I always note that anytime I put a toe into legal issues. While I am fascinated by the law, I have no training or experience in it¹. What little I know has been gleaned from wiser heads like Em and Burt. So it’s possible some of what follows will trip over legal issues.

But I did want to write about the recent decision to repeal the FDA’s approval of the abortion drug mifepristone. Because there are significant issues of science and medicine wrapped up in this debate. And I think it’s important to not only put the facts on the table but to look at how Judge Matthew Kacsmaryk’s decision has addressed these issues.

Just as I am not a lawyer, judges are not scientists. We have an ongoing crisis in forensic science because judges, lacking the knowledge to know if forensic techniques are valid or not, tend to defer to each other. So a judge not knowing the scientific issues involved in an issue as complex as chemical abortion is understandable.

However, after reading through the mifepristone decision, I believe that Kacsmaryk’s decision is, scientifically, a collection of claims from anti-abortion groups presented uncritically and without any consideration of opposing claims. As such, it tends to rely on specious claims and mischaracterizations about mifepristone, gestation and pregnancy. That is consistent with what actual legal experts claim is poor legal reasoning.

What I will not talk about

To put my cards on the table, I describe myself as “anti-abortion but reluctantly pro-choice”. That is, while I have personal oppositions to abortion, I think the issues are too personal, too difficult and too fraught for me to judge anyone’s decision, let alone for the government to do so. I’m not here to persuade anyone to that position. I respect both sides as I think both have valid arguments. And I disrespect both sides because I think both sides choose to put their worst arguments first. But that’s neither her nor there. I’m not here to talk about law or morality; I’m here to talk about science.

And that’s important, because science can’t answer the question of whether abortion should be legal or not:

How do we do the calculus on that? It depends on whether you think the fetus has rights or not. If you do, then abortion must be outlawed as it destroys a life. If you don’t, then it should be legal. This is not an issue that lends itself to the weight of evidence because people disagree on what that evidence means. Where do you draw the line on outlawing abortion: Birth? Viability? Fetal pain? Brain development? When the chance of spontaneous abortion (miscarriage) reduces to a certain threshold? Each of those points has a “rational” argument behind it. But which argument you accept depends on your values and how you value the rights of the unborn versus the rights of women.

Science has a role to play in the abortion issue of telling us what methods work, what the side effects are, how dangerous the methods are and what stage of development the fetus is in during a pregnancy. That is why I wanted to write his post: not to persuade anyone, but to set a better stage on which these issues can be debated. If we can agree on the facts, then we can debate the law and morality. When the facts are muddied, the debate suffers.

So let’s get to the facts.

What is Mifepristone?

Mifepristone is chemical that blocks the body from absorbing progesterone. It can be used to prevent ovulation, but is mostly used as the first stage in a chemical abortion. It blocks progesterone from sustaining the uterine lining and the embryo. It is followed shortly after by a dose of misoprostol, which causes the uterus to contract and expel the embryo. The latter is often used on its own to expel a fetus that has miscarried but it still inside the uterus.

Mifepristone, also known as RU-486, has been controversial since it was approved by the FDA in 2000. Those who believe that life begins at conception latch onto this pill essentially starving the fetus of nutrients and preventing further implantation. While Mifepristone does have other uses, abortion is it primary use and it is now the method of more than half the abortions in the United States. It should be noted, however, that this abortion method is only approved in the United States up to ten weeks of pregnancy. After ten weeks, the effectiveness of the method drops precipitously and there are growing safety concerns.

OK, so what does Kacsmaryk claim about mifepristone?

On page 7, while addressing the subject of standing, Kacsmaryk repeats their claim that “adverse events from chemical abortion drugs can overwhelm the medical system and place ‘enormous pressure and stress’ on doctors during emergencies and complications.” This claim is nonsense. The abortion pill has been in use for 23 years and has been used for millions of abortions. If it were going to overwhelm our system with emergencies, it would already have done so. I chatted with Em about this and she noted that, in some legal disputes, the judge doesn’t consider whether the claims of a party are true, but whether they would have an impact if they were true. But considering how at variance this claim is with reality, it is as though the plaintiffs claimed that approving the abortion pill would cause frogs to rain down from the sky. I mean … sure, if that were true, it would be something to consider. But it’s not only untrue, it’s a complete fantasy.

The rate of adverse effects of mifepristone is well known and well-studied. It is generally well-tolerated and the side effects are usually temporary. While Kacsmaryk and the pro-life plaintiffs make various hand-wavy arguments about how many women are dying of chemical abortion, the FDA’s data indicates that about five deaths per million are occurring, which is a death rate lower than many over-the-counter or prescription drugs. By comparison, carrying a pregnancy to term has a death rate of 330 per million and rising, if dying women is what you’re worried about. Putting aside any arguments about whether the fetus counts as a life — and therefore the death rate is a million per million — mifepristone, like most legal abortion methods, is safer than the alternative, at least for the woman.

And this is not a trivial point. A huge part of Kacsmaryk’s decision rests on the idea that women are going to have deleterious side effects of chemical abortion and that this will strain the system, particularly in “maternity care deserts” where there are few OB-GYN practitioners. But this arguments applies literally one hundred times stronger to pregnancy itself. Throughout the decision, the very real risks of pregnancy and childbirth are basically ignored. Kacsmaryk even pretends that the pro-life physicians have no conflict of interest in favoring continued pregnancies and the thousands of dollars of healthcare business generated by it.

Throughout the decision, the pro-life position is cast as genuine concern about the welfare of women while the pro-choice positions is cast as cynically unconcerned if women drop dead. Now maybe you see it that way; I’m not here to judge anyone’s abortion position. But a federal judge should at least consider both sides before issuing a national ban, no?

On page 8, Kacsmaryk claims that 14% of women getting the abortion pill received no information about potential side effects. This study is based on … surveys done on anti-abortion sites. So … not exactly objective analysis here.

On page 9, the judge makes two assertions. First, that “Women who have aborted a child — especially through chemical abortion drugs that necessitate the woman seeing her aborted child once it passes — often experience shame, regret, anxiety, depression, drug abuse, and suicidal thoughts because of the abortion.” However, an early-stage fetus does not generally resemble a human and is tiny. We’re used to close-up pictures that show a human-like appearance but those are generally less than a centimeter in size. Women who’ve had early miscarriages sometimes don’t even realize they were pregnant; they just think they had a heavy late period.²

Kacsmaryk further goes on to cite a study that claims that women who’ve had abortions have a 154% increase in suicide. This study is from David Reardon, an engineer who founded his own abortion research institute and makes various claims of the deleterious health effects of abortion. His claims are at variance with every other study of this issue. To be fair, this is a hard issue to study since there are many confounding factors³. But to be really fair, that’s why this issue is studied by experts in the field, none of whom Kacsmaryk deigns to cite.

On Page 12, he asserts that physicians don’t have enough information on the potential adverse side effects of mifepristone to advise their patients because the FDA doesn’t require these events to be reported. This is a half-truth. Adverse events are required to be reported to the manufacturer, who then reports them to the FDA. And the FDA’s website has pages and pages about the potential adverse effects of the abortion pill.

On page 14, he cites claims that 8% or 18% of women who take the abortion pill will require surgical intervention. I can’t find where these numbers come from, but they are wildly at variance with any reasonable study of the issue, which show far lower rates of incomplete or failed abortion (less than 5%).

On Page 29, Kacsmaryk again brings up the trauma of seeing the aborted fetus, this time citing a study from the UK, not the US, advising better informing of patients. He also raised a semi-reasonable point, that some studies have indicated that complications from a chemical abortion are sometimes miscoded as miscarriages. However, he ignores several critical parts of the study. For example, chemical abortions are less likely to end up in the ER than surgical ones. Additionally, the study suggests that the problem here is stigma, that some women are afraid of telling staff they had an abortion and therefore their symptoms are misdiagnosed. Three guesses as to whether that problem will get better or worse in the post-Dobbs era.

On Page 40, Kacsmaryk disputes the FDA invoking its power to legalize medications for serious and life-threatening conditions by claiming pregnancy is not a serious or life-threatening condition. Which would be news to the hundreds of thousands of women who experience healthcare complications during pregnancy and the hundreds of women who die of it. No, pregnancy is not an illness. We can debate whether it is “serious or life-threatening” enough to invoke Subpart H. But to pretend the FDA has no argument here is absurd.

On page 45, Kacsmaryk attacks the FDA’s claim that chemical abortion has a “therapeutic benefit” over surgical abortion — that is, that taking a pill is better than having an invasive medical procedure — by citing studies claiming a higher complication rate. Principally, he sites one from Finland claiming four times the complication rate from chemical abortion. But in doing so — picking one study as a beacon — he ignores massively larger and more thorough studies that show this is not the case. He cites the study on misdiagnoses miscarriages linked above to claim that chemical abortions are more likely to result in ER visits but this a post-hoc study, not a case-controlled study, and the authors point is that women not telling their doctors they had a medical abortion is a problem. He then goes on to site more web surveys.

Look, you can always find one study that will tell you whatever you want. Insight is obtained by looking at multiple studies or meta-studies. As I said earlier, I don’t judge people for their views on abortion — I can respect both sides. But in only citing the studies that claim chemical abortion is dangerous, Kacsmaryk is letting his pro-life bias dictate his reality. In chasing down his sources, I found every source was cited only by pro-life institutions and activists.

On page 49, he contrasts the safety precautions used in the trials of mifepristone to general use, noting that, when the drug was in trials, there were much stricter safety protocols. But this is true for every medication that is trialed because, during trials, we don’t know what the side effects will be. That’s … why we have clinical trials. Once the trials are done, we don’t monitor everyone taking them. See, for example, every drug that is ever taken. At that point, we rely on different reporting mechanisms to identify problems because monitoring every patient would be horrendously expensive and impractical.

He does have a semi-decent point in that women may get a chemical abortion without knowing that they have an ectopic pregnancy. He also has a semi-decent point about not being able to measure gestational age precisely (although, bizarrely, he illustrates this with a woman had a chemical abortion at 30-36 weeks, which is way outside the parameters for using the drug and while an ultrasound is more accurate at estimating gestational age than the last period generally, it’s not dramatically so, especially at less than ten weeks). But these cases suggest, at most, a return to the FDA’s earlier regulatory regime, not a complete ban on the medication. And again, the proof is in the pudding — the success rate of chemical abortion is 95%. He sites a few alarming-sounded numbers on deaths and complications without noting that this is in comparison to millions of medical abortions.

Most of the rest of the decision is on legal issues. If I may be forgiven for venturing outside my lane, but I’m actually somewhat sympathetic to his argument that the FDA erred or exceeded its authority in allowing these drugs to be dispensed by mail.⁴ And as scientist, I do have to admit that the health concerns raised in dispensing these drugs by mail, while not as ominous as Kacsmaryk makes it sounds, are not completely illegitimate. As I’ve noted above, it bears considering whether ectopic pregnancies will be missed, whether women will take the medication well beyond the 10-week limit and whether complications will be mistaken for side effects.

However, the entirety of this decision does not make me think that these issues are being seriously considered. Time after time, dangers are exaggerated, studies are cherry-picked and science is twisted into a pro-life shape. Again, I’m not here to persuade anyone on the abortion issue generally. But if you’re going to make the case for a judge to unilaterally ban a drug that has been safely used for 23 years … you’re going to need to do better than this.

1. Thankfully, since, if you have no training in the law, the easiest way to gain experience is to be targeted by it.
2. We have some experience in this. When my wife and I were trying to have our second, we had twins that miscarried at about ten weeks but failed to abort naturally. She had to use misoprostol after a failed D&C. There was a lot of pain and a lot of blood. But no miniature doll-like humans.
3. E.g., women with mental health issues are both more prone to engage in unsafe sex, more prone to seek abortion and more prone to further mental health problems.
4. The legal analysis does include perhaps the most ominous part: a detailed application of the Comstock Acts. This is the set of laws passed by President Grant that banned the mailing of obscenity, birth control, sex toys, abortifacients and even personal letters with sexual information. Just in case you are wondering all this might be headed.