Mifepristone and the rule of law, part II

The district court's decision is wrong. It must be reversed.

On April 7, 2023, the U.S. District Court for the Northern District of Texas issued an order overturning the FDA’s approval of mifepristone. The order will take effect within seven days unless it is stayed or reversed by a higher court.

In my prior post addressing this case, I opined that it was unfair to criticize the presiding judge for his views on abortion and his prior advocacy. I adhere to that view. I believe that a judge’s personal beliefs and professional background are irrelevant to his ability to be a good judge. There are many excellent judges who are personally pro-life and many excellent judges who are personally pro-choice. The focus should be on the judge’s legal reasoning, not his personal characteristics. In Judge Duncan’s Wall Street Journal editorial following the Stanford incident, he explained that one of “the basic concepts of legal discourse” is that “one must meet reason with reason.” I agree.

And there is plenty to say about the judge’s legal reasoning. The decision is indefensible. In this post, I will walk through the decision and explain just how bad it is.

¹

Standing

The plaintiffs are organizations of pro-life doctors seeking to overturn the FDA’s approval of mifepristone. As I explained in my prior post, they lack standing. Under Clapper v. Amnesty International USA, the plaintiffs bear the burden of proving a “certainly impending” injury from the FDA’s approval of mifepristone. Neither the organizations, nor their doctor-members, can meet that standard. The plaintiffs’ philosophical disagreement with abortion does not give them standing to seek a federal court order banning all doctors nationwide from prescribing mifepristone to their patients.

The district court nonetheless finds that the plaintiffs have both associational standing—i.e., standing based on injuries suffered by their doctor-members—and organizational standing—i.e., standing based on injuries to the organizations themselves.

The court starts with associational standing. It begins by reciting assertions from the plaintiffs’ filings and declaring, without additional analysis, that they establish standing. For example, the court quotes the plaintiffs’ assertions that “chemical abortion drugs can overwhelm the medical system” and “consume crucial limited resources” such as “blood for transfusions.” (p. 7).

These allegations do not establish standing. To establish standing, the doctor-members must show that they personally face an imminent risk of concrete and particularized injury. Vague speculation that someday, somewhere, some unspecified doctor will be “overwhelmed” by an onslaught of patients coming to the ER after taking mifepristone does not establish that these particular doctors face an imminent risk of a concrete and particularized injury.

Also, these allegations are disconnected from reality. Mifepristone has been on the market for the past 23 years. Patients suffering complications from mifepristone abortions have not “overwhelmed the medical system” or rendered blood unavailable for transfusions. The agency record shows that the rates of transfusions are 0-0.1 percent and hospitalization is 0-0.7 percent. The judge was required to defer to these factual findings if substantial evidence supports them, which it does: the FDA reached these conclusions by compiling eleven studies containing data on over 30,000 patients.

How does the court deal with these facts? The answer is that it doesn’t. The court appears completely uninterested in conducting any kind of neutral evaluation of whether the plaintiffs’ allegations are true. It simply quotes them and declares victory for the plaintiffs.

The court also credits the plaintiffs’ allegation that the doctors may be forced to be made “complicit” in an “elective chemical abortion” by forcing them into a situation in which they “need[] to remove a baby with a beating heart or pregnancy tissue as the only means to save the life of the woman or girl.” (pp. 7-8.). It doesn’t appear that any of the doctor-members have ever had to involuntarily “remove a baby with a beating heart” as a result of an unsuccessful mifepristone abortion; it’s unclear whether this has happened to any doctor, ever. But even if it has, it’s absurd to suggest that this is a “certainly impending” outcome for these doctors. Think of what these doctors are saying: they are claiming standing to ban all women from obtaining mifepristone from all doctors in all 50 states, and force all of these women to obtain surgical abortions if they want to terminate their pregnancies, in order to ward off against the hypothetical possibility that some unspecified woman, somewhere, might someday take mifepristone, have some extremely unusual reaction, enter a hospital, and randomly encounter one of these particular doctors, who personally will have to conduct an extremely rare surgical procedure to save the woman’s life, which will lead to the doctor experiencing emotional harm from being “complicit” with abortion. Really?

Next, the court says: “Plaintiffs also argue the challenged action ‘prevent Plaintiff doctors from practicing evidence-based medicine’ and have caused Plaintiffs to face increased exposure to allegations of malpractice and potential liability, along with higher insurance costs.” (p. 8). Each of these allegations is weaker than the last. The plaintiff doctors don’t prescribe mifepristone, so the fact that other doctors choose to do so does not prevent them from practicing “evidence-based medicine.” It is a joke to say that the plaintiff doctors face a “certainly impending” injury on the theory that the FDA’s approval of mifepristone 23 years ago might lead to some unspecified hypothetical lawsuit at some point in the future. And it is totally speculative to suggest that “insurance costs” will decrease if the plaintiffs prevail.

It gets worse. Even though literally millions of women have taken mifepristone, and there are many studies exhaustively studying the health effects of mifepristone, the district judge does not feel that there are enough studies, and therefore feels that “women and girls are prevented from giving informed consent to providers.” (p. 8). Of course this does not establish that the plaintiffs, who are pro-life doctors who do not prescribe mifepristone to their patients, have standing to block other doctors from prescribing mifepristone.

But let’s set that point aside for a moment and see how the the court justifies its conclusion regarding informed consent. The court rests its analysis on a study:

Women also perceive the harm to the informed-consent aspect of the physician-patient relationship. In one study, fourteen percent of women and girls reported having received insufficient information about (1) side effects, (2) the intensity of the cramping and bleeding, (3) the next steps after expelling the aborted human, and (4) potential negative emotional reactions like fear, uncertainty, sadness, regret, and pain. See Katherine A. Rafferty & Tessa Longbons, #AbortionChangesYou: A Case Study to Understand the Communicative Tensions in Women’s Medication Abortion Narratives, 36 HEALTH COMMC’N. 1485, 1485–94 (2021). Plaintiff physicians’ lack of pertinent information on chemical abortion harms their physician-patient relationships because they cannot receive informed consent from the women and girls they treat in their clinics.

What is this study? It’s a study consisting of an analysis of anonymous blog posts on a pro-life website. A group called the “Institute of Reproductive Grief Care” maintains a website called “Abortion Changes You” that, among other things, includes anonymous blog posts of women who regretted having abortions. In the study, the authors try to draw conclusions from these blog posts. Here’s a quote from the abstract:

Our contrapuntal analysis rendered four sites of dialectical tension found across women’s blog posts: only choice vs. other alternatives, unprepared vs. knowledgeable, relief vs. regret, and silence vs. openness. Each site of struggle characterized a different noteworthy moment within a woman’s medication abortion experience: the decision, the medication abortion process, identity after abortion, and managing the stigmatizing silence before and after the abortion.

OK, so, to recap, the judge asserts that “fourteen percent of women and girls report having received insufficient information,” without revealing that the entire sample consists of anonymous blog posts on a pro-life website designed for women who regret their abortions. Perhaps this sample might not be reflective of the entire universe of women who have abortions?

I would also not say that the methodology of this study is ironclad. The authors’ methodology consists of scanning anonymous blog posts for words such as “but”:

Women used negating when saying, “can’t,” “not,” “couldn’t,” and “never.” Examples of countering were most apparent when women used the word “but.” Entertaining often occurred when women used the words “possibility” and “could have.” Finally, we identified where and how competing discourses interpenetrated.

The sample size consists of 98 blog posts. You might think this is a small sample size, but actually it turns out that the authors applied their methodology on only 54 blog posts, decided they’d had enough, and then decided to skim the rest for good quotes:

Data were analyzed until the point of theoretical saturation (i.e., no new thematic categories were present in the blog posts; Strauss & Corbin, Citation1990), which occurred after the 54^th blog post. However, we continued to analyze the remaining blog posts in an effort to verify that our analysis of the discourses evident in the 54 posts accurately reflected all of the posts within the entire data set. Further, we wanted to extract the best exemplars from the entire case study and desired that quotations within all posts be considered for representation.

And again, I am totally lost on how this shows that pro-life doctors who never even prescribe mifepristone have standing to seek an order banning all women from taking mifepristone.

More on this study later.

The court then claims that “Plaintiff medical associations have associational standing via their members’ third-party standing to sue on behalf of their patients.” (p. 9). This section reveals that the court does not understand what “third-party standing” means.

“Third-party standing” is something that plaintiffs sometimes have to show in addition to—not instead of—Article III standing. Plaintiffs always, 100% of the time, have to show what the Supreme Court has called the “irreducible constitutional minimum” of standing under Article III—(1) a concrete and particularized injury, that is (2) caused by the defendant’s action, and (3) redressable by the requested relief. In some cases, a plaintiff that has standing asserts that some action harming them is illegal because it violates someone else’s rights—in which case they bear the additional burden of showing they are entitled to assert the rights of that third party, i.e., third party standing.

Lawsuits by abortion providers challenging abortion restrictions are a classic illustration of the third-party standing doctrine in action. In most cases, especially pre-Dobbs, abortion restrictions target abortion providers rather than women seeking abortions: if a provider performs an abortion in violation of state law, the provider is punished. Abortion providers who sue states seeking to enjoin abortion restrictions have obvious Article III standing. They face a concrete and particularized injury (the state is going to shut them down and incarcerate their doctors if they perform illegal abortions); it’s caused by the state, which enforces the law; and an order enjoining the state from enforcing the law would redress the injury.

In challenging those laws pre-Dobbs, the providers would claim that the laws are unconstitutional because they violate their patients’ constitutional right to an abortion. Thus, although the providers were the direct targets of the law, the providers’ legal theory was that the laws were unconstitutional because they violated the constitutional rights of third parties—i.e., their patients. In this context, courts would consider whether the providers were permitted to assert the legal interests of their patients under a doctrine known as “third-party standing,” or whether instead the plaintiffs had to be women seeking abortions. So, for plaintiffs who already had Article III standing, courts would consider whether the plaintiffs also had third-party standing.

The district judge doesn’t understand this. He says: “The injuries suffered by patients of the Plaintiff medical associations’ members are sufficient to confer associational standing” (p. 10). In other words, he thinks that if hypothetical patients have Article III standing, this means that the doctor-members of the plaintiff organizations can also assert “third-party standing” without a showing that the doctor-members themselves were injured. Standing does not work this way, this is completely wrong.

Even if standing did work this way, the court’s application of the doctrine would still be wrong. Let’s start with the black-letter law. To establish third-party standing, a plaintiff must typically show two things: (1) the party asserting the right has a “close” relationship with the person who possesses the right, and (2) there is a “hindrance” to the possessor’s ability to protect his own interests. In assessing third-party standing, courts also consider whether there is a conflict of interest between the party asserting the right and the party possessing the right.

The “right” at issue is the right to be safe from dangerous drugs. So, I guess, the women who face “certainly impending” harm from mifepristone are women who are considering taking mifepristone and must be protected from their own bad decisions.

Do the plaintiffs have a “close relationship” with those women? Obviously not! They are other doctors’ patients. The plaintiff-doctors don’t even know who they are!

The court holds otherwise, finding that the plaintiff-doctors have the requisite “close relationship” to women who, in the future, will take mifepristone prescribed by other doctors, have complications, and then switch to the plaintiff-doctors (p. 10). I find it bizarre to suggest that the plaintiffs have a “close relationship” with this totally conjectural class of patients. But even assuming these hypothetical patients ever actually materialize, by the time the plaintiff-doctors see the patients, they have already taken the mifepristone and wouldn’t benefit from an order banning it. It’s literally never the case that any of the plaintiffs’ actual patients could benefit from an order banning mifepristone.

I am also wowed by this statement by the court (p. 10):

Here, the physician-patient dynamic favors third-party standing. Unlike abortionists suing on behalf of women seeking abortions, here there are no potential conflicts of interest between the Plaintiff physicians and their patients.

Really?! There are “no potential conflicts of interest” between women who want to use mifepristone to have an abortion and doctors who are seeking to ban them from using mifepristone?!

The court also finds that that the plaintiffs’ patients have a “hindrance” to asserting their own rights. It claims that “women who have already obtained an abortion may be more hindered than women who challenge restrictions on abortion” because “adverse abortion experiences that are often deeply traumatizing pose a hindrance to a woman’s ability to bring suit.” (p. 11). The court doesn’t seem to realize that “women who have already obtained an abortion” don’t have standing based on the prior abortion. The plaintiffs aren’t seeking damages to redress prior harms; they are seeking an injunction to redress future harms. These women don’t have standing unless they wish to obtain an additional abortion, in which case there is a rather obvious conflict of interest between those women and the plaintiffs who seek to ban them from getting the medical care they want.

Satisfied that there is associational standing, the court then turns to organizational standing. The court finds that “Because of FDA’s failure to require reporting of all adverse events, FDA’s actions have frustrated their ability to educate and inform their member physicians, their patients, and the public on the dangers of chemical abortion drugs” (p. 12). And, “Plaintiffs have re-calibrated their outreach efforts to spend extra time and money educating their members about the dangers of chemical abortion drugs” (p. 13).

My prior blog post explained why this theory of standing directly contradicts Supreme Court precedent and can’t possibly be right. First, there’s a Supreme Court case, Clapper v. Amnesty International USA, explicitly rejecting this diversion-of-resources theory unless the diversion of resources is curing the members’ injury caused by the challenged law (in which case organizational standing adds nothing to associational standing).

The district court fails to cite Clapper in its discussion of organizational standing. Instead it offers this citation to a post-Clapper Fifth Circuit case (pp. 11-12):

See also Zimmerman v. City of Austin, Tex., 881 F.3d 378, 390 (5th Cir. 2018) (changing one’s “plans or strategies in response to an allegedly injurious law can itself be a sufficient injury to confer standing”).

Looks promising! But here’s what Zimmerman actually says:

Finally, while changing one’s campaign plans or strategies in response to an allegedly injurious law can itself be a sufficient injury to confer standing, the change in plans must still be in response to a reasonably certain injury imposed by the challenged law.

Oh.

In addition to directly contradicting Supreme Court precedent, the district court’s theory implies that any organization can immediately manufacture standing to challenge any rule in the entire world it thinks would harm “safety.” It’s simple:

1. Federal agency does something that the organization thinks is “unsafe.”

2. Organization now has to start educating people about safety, which it wouldn’t have to do without the rule.

3. Organization has less money to do other things.

And actually, I don’t even see why the rule would have to be a safety rule, presumably it could be any rule:

1. Federal agency enacts rule.

2. Organization doesn’t like rule, spends money educating people about why rule is bad.

3. Organization has less money to do other things.

Presto, standing! The district court doesn’t identify any kind of limiting principle. It just blows past this.

The court then spends some time explaining why it thinks the injuries are “concrete and redressable” The Supreme Court has held that an injury cannot give rise to standing unless it is “certainly impending,” and the court finds that this standard is satisfied (pp. 13-17). The court’s reasoning boils down to this: “Plaintiffs’ declarations list specific events where Plaintiff physicians provided emergency care to women suffering from chemical abortion.” (p. 14). “Plaintiffs have good reasons to believe their alleged injuries will continue in the future, and possibly with greater frequency than in the past.” (p. 15). That doesn’t sound “certainly impending” to me, but the court assures us that “the ‘certainly impending’ standard for an ‘imminent’ injury is not as demanding as it sounds.” (p. 14).

The problem here is that the organizations have no idea which doctor-members will be harmed, or who the hypothetical future patients will be; they’re simply saying that statistically, it’s likely that some unspecified doctors will be harmed in the future by encountering future patients who show up in the emergency room. The Supreme Court has rejected this exact theory of standing in Summers v. Earth Island Institute, a 2009 decision by Justice Scalia:

The dissent proposes a hitherto unheard-of test for organizational standing: whether, accepting the organization's self-description of the activities of its members, there is a statistical probability that some of those members are threatened with concrete injury … This novel approach to the law of organizational standing would make a mockery of our prior cases, which have required plaintiff-organizations to make specific allegations establishing that at least one identified member had suffered or would suffer harm.

The court fails to cite this binding Supreme Court precedent in its decision.

Timeliness

It gets worse.

The FDA approved mifepristone in 2000. According to the FDA’s regulations, a plaintiff seeking to challenge an agency decision is required to file a citizen petition raising its arguments to the agency. See 21 C.F.R. § 10.45(b). If the citizen petition is denied, the plaintiff can sue, subject to a six-year statute of limitations.

In this case, two of the plaintiff organizations submitted citizen petitions challenging the FDA’s approval of mifepristone in 2000. In March 2016, the FDA denied the petitions. This gave the organizations until March 2022 to sue. They sued in November 2022. So the lawsuit is time-barred.

Not so fast, says the district court.

The court identifies several reasons why the suit is timely, starting with the “reopening” doctrine. Under that doctrine, if an agency reopens a prior rule or adjudication in a new proceeding, the clock to challenge the prior rule or adjudication restarts.

It’s hard to show that the “reopening” doctrine is satisfied, though. Agencies amend their rules all the time. Ordinarily, they start from the assumption that the underlying rule is legally sound, rather than reinventing the wheel. The D.C. Circuit has therefore made it very difficult to show that the “reopening” doctrine is satisfied:

Under the reopening doctrine, the time for seeking review starts anew where the agency reopens an issue by holding out the unchanged section as a proposed regulation, offering an explanation for its language, soliciting comments on its substance, and responding to the comments in promulgating the regulation in its final form.

“Reopening” is even less common where, as here, an agency is conducting adjudications. While it is possible for an agency to reconsider a prior adjudication and then stick to it, the adjudicative process typically involves building on, rather than reconsidering, prior decisions. When the Supreme Court cites a case, this ordinarily means that it accepts that case as binding precedent rather than that it is silently reconsidering whether the case is correct and silently concluding that it is, in fact, correct. It is the same for agencies.

In this case, the district court concludes that the FDA reopened its 2000 approval of mifepristone not once but twice, in both 2016 and 2021.

In 2016, the FDA approved a supplemental application to alter certain of the conditions of mifepristone’s approval. Essentially, the FDA concluded that mifepristone had a good safety track record, and some safety restrictions on mifepristone could therefore be relaxed. The FDA increased the gestational age limit from 49 to 70 days, reduced the number of required in-person clinic visits to one, and allowed healthcare providers other than doctors to prescribe it.

In 2019, the plaintiffs filed a citizen suit challenging the 2016 decision, which was denied in 2021. So the plaintiffs’ suit to challenge the 2016 decision is timely.

But did the 2016 decision “reopen” the approval from 2000? Of course not:

• The 2016 decision does not suggest anywhere that it was reopening the 2000 decision.

• No one asked the FDA to reopen the 2000 decision.

• The whole premise of the 2016 decision was that mifepristone was safe—i.e., the conclusion of the 2000 decision—which is why the FDA relaxed safety restrictions.

In 2021, FDA decided to exercise enforcement discretion during COVID with respect to the in-person dispensing requirement. The plaintiffs never filed a citizen petition in response to this decision and it’s going to expire anyway, so I don’t really understand why we are even talking about this, but may as well pose the question … did the 2021 decision “reopen” the approval from 2000? Of course not:

• The FDA simply said it would decline to enforce the in-person dispensing requirement because of COVID. There is not the slightest indication in this enforcement decision that the FDA was “reopening” its decision finding mifepristone safe 21 years earlier.

• Also, the FDA’s reasoning was that there was little evidence that in-person dispensing increased safety, a further indication that it was not “reopening” the underlying approval of mifepristone.

So why does the district court think that the 2016 and 2021 decisions “reopened” the 2000 decision?

Well, the district court certainly doesn’t cite any language from either of these decisions indicating that the FDA was reopening its 2000 decision, because no such language exists and it is obvious from these decisions that the FDA did not think it was reopening anything. Instead here’s what the district court says (p. 22):

FDA’s 2016 and 2021 Changes thus significantly departed from the agency’s original approval of the abortion regimen. FDA repeatedly altered its original decision by removing safeguards and changing the regulatory scheme for chemical abortion drugs. Sierra Club, 551 F.3d at 1025; Nat’l Biodiesel, 843 F.3d at 1017.

This is wrong.

No case has ever held that if an agency merely “departs from” or “alters” a prior scheme, that means the agency has reopened the underlying scheme. Also, this makes absolutely no sense. To repeat, if the agency decides mifepristone is safe, and then decides to relax safety restrictions, the whole premise of the subsequent decision is that mifepristone is safe. This is the opposite of “reopening” the prior decision.

But the district court does cite two cases, Sierra Club and National Biodiesel. What do those cases hold?

Those cases recite a different version of the “reopening” doctrine. They hold that if an agency significantly alters the stakes of judicial review—such that a challenger wouldn’t have had an incentive to challenge the prior rule, but now does— the reopening doctrine applies. Quoth Sierra Club (cleaned up):

A constructive reopening occurs if the revision of accompanying regulations significantly alters the stakes of judicial review, as the result of a change that could have not been reasonably anticipated … Just as the court in Kennecott agreed with industry that the agency had constructively reopened a regulation when it incorporated amended regulations that expanded available remedies and thus altered its financial incentives for challenging the regulation, so too here from the perspective of environmental petitioners' interests and allocation of resources the general duty may not have been worth challenging in 1994, but the revised regulations gave that duty a new significance

And National Biodiesel (cleaned up):

The CARBIO proposal is, the argument goes, a constructive reopening of the Rule, which occurs if the revision of accompanying regulations significantly alters the stakes of judicial review as the result of a change that could have not been reasonably anticipated. We have described the magnitude of alteration required to invoke this doctrine as a sea change, and have declined to apply it when the basic regulatory scheme remains unchanged.

That version of the “reopening” doctrine clearly doesn’t apply here, because of course the plaintiffs had an incentive to challenge the 2000 rule—their whole premise is that the 2000 rule put an unsafe drug on the market! And the district court doesn’t even suggest that the “alters the stakes” doctrine applies here. Instead, the court just blithely cites these cases, which recite a different and irrelevant doctrine, and then declares that because the safety standards are different now than before, the “reopening” doctrine applies, despite the total absence of any indication in the agency’s decisions that it was reopening anything.

The court then offers another “reopening” theory. In 2019, the plaintiffs filed a citizen petition challenging the FDA’s 2016 decision, which the FDA denied in 2021. The court cites language in the FDA’s 2021 denial as evidence that the “reopening” doctrine applies (p. 22):

Additionally, FDA’s response to the 2019 Petition explicitly states FDA “undertook a full review of the Mifepristone REMS Program” in 2021. ECF No. 1-44 at 7 (emphasis added); 17 see also Peavey, 128 F. Supp. 3d at 100–02 (agency reopened decision by conducting “thorough review” of the merits, even where the order did not state it was a “reconsideration” and did not reference prior decision). And FDA even granted the 2019 Petition in part. ECF No. 1-44 at 3.

This paragraph is gibberish. Even the plaintiffs’ lawyers don’t make this argument (I don’t know if it can really be called an “argument”). To give some background, “REMS” stands for “Risk Evaluation and Mitigation Strategy.” Under a new regulatory framework established in 2007, the FDA is permitted to impose safety restrictions to ensure the safe use of drugs. The manufacturers of certain drug are required to submit a “REMS” to the FDA to mitigate the risk of their drugs. Mifepristone is one of those drugs. Hence Danco, which manufactures Mifepristone, submitted a REMS application, which the FDA approved in 2011 (with modifications). In the 2019 Petition, the petitioners urged the FDA to retain the REMS, which the FDA partially did (hence the decision to “grant[] the 2019 Petition in part”).

The FDA’s 2021 petition denial decision reports that in 2021, FDA “undertook a full review of the Mifepristone REMS Program.” OK, but that is completely separate from the two decisions that the district court claimed “reopened” the approval: i.e., the 2016 approval of the supplemental application and the 2021 decision to exercise enforcement discretion during COVID. Hence, the plaintiffs did not make this argument in their brief. My guess is that the district judge did CTRL-F for “review” and similar keywords in the 2021 petition denial, hit on the phrase “full review,” and thus decided, hey, let’s go with that.

The district court then offers two alternative justifications for saying the statute of limitations hasn’t expired, which are, I think, even weaker than the “reopening” doctrine, which is saying something.

First, the district court invokes the “equitable tolling” doctrine (p. 23). This is another example of litigation from the bench, as the plaintiffs rightfully do not make this argument in their brief.

The Supreme Court has stated that a “litigant is entitled to equitable tolling of a statute of limitations only if the litigant establishes two elements: (1) that he has been pursuing his rights diligently, and (2) that some extraordinary circumstance stood in his way and prevented timely filing.”

The doctrine does not apply here. The plaintiffs received the denial of their citizen petition challenging the approval of mifepristone in March 2016. They had six years—a pretty generous period—to sue. They have never offered any explanation for waiting to sue until November 2022 rather than suing by March 2022. Hence, (1) the plaintiffs did not pursue their rights diligently: they didn’t sue for over six years. And (2) no extraordinary circumstance stood in their way: there was literally nothing standing in the way of suing. See, sometimes law is easy!

And yet the district court finds that equitable tolling applies. It relies on the fact (p. 24) that the FDA took 14 years to respond to the plaintiffs’ citizen petition (from 2002 to 2016). OK, this is a long delay, but I am totally at a loss for why this establishes that equitable tolling applies to give the plaintiffs more than six years from 2016 to file their lawsuit.

The district judge also points out that the FDA took 2 years to respond to the plaintiffs’ separate citizen petition challenging the 2016 decision (from 2019 to 2021). The plaintiffs undoubtedly filed suit within 6 years of that denial so this has nothing to do with anything.

Essentially the district court’s reasoning is this: the FDA seems slow and bad, so as a kind of revenge against the FDA, let’s allow plaintiffs to bring suits that are plainly time-barred! This isn’t really “law” in any conventional sense of the word.

The district court has one more arrow in its quiver, and it is worth quoting in full (pp. 24-25):

Additionally, statutes of limitations “are primarily designed to assure fairness to defendants,” and “to promote justice by preventing surprises through the revival of claims that have been allowed to slumber until evidence is lost, memories have faded, and witnesses have disappeared.” Clymore v. United States, 217 F.3d 370, 376 (5th Cir. 2000), as corrected on reh’g (Aug. 24, 2000) (internal marks omitted). But it “has not been argued, and cannot seriously be, that the government was unfairly surprised” when Plaintiffs filed this suit. Id. Plaintiffs have been reasonably diligent in pursuing their claims. See, e.g., ECF No. 1-4 at 6 (after years of waiting for FDA to respond to the Petition, Plaintiff “called upon” FDA to issue a response in 2005 and again in 2015). And the public interest in this case militates toward resolving Plaintiffs’ claims on the merits. Accordingly, Plaintiffs’ challenges to FDA’s Pre-2021 Actions concerning chemical abortion drugs are not time-barred.

Let’s not apply a six-year statute of limitations because, well who really needs statutes of limitations? And wouldn’t the world be better without them? I’ve never seen such reasoning in a judicial opinion.

Exhaustion

An FDA regulation, 21 C.F.R. § 10.45(b), says: “A request that the Commissioner take or refrain from taking any form of administrative action must first be the subject of a final administrative decision based on a [citizen petition] … before any legal action is filed in a court complaining of the action or failure to act.”

In their 2019 citizen petition, the plaintiffs didn’t ask the FDA to overturn the 2000 approval of mifepristone. They didn’t suggest that the “reopening” doctrine might apply. To the contrary, the petition said: “The undersigned submit this petition to request the Commissioner of Food and Drugs to: … retain the Mifeprex Risk Evaluation and Mitigation Strategy (REMS), and continue limiting the dispensing of Mifeprex to patients in clinics, medical offices, and hospitals, by or under the supervision of a certified prescriber.” Hence, their claim is not exhausted.

When I first saw that the district court granted the plaintiffs’ preliminary injunction motion, this was actually the first thing I looked at. I was genuinely curious how the district court would get around this failure to exhaust. The answer is that … it didn’t. Apparently unable to think of any response to this argument, even a creative one, the judge just elects to ignore it and rule for the plaintiffs anyway. It’s nice to be a judge.

The judge does, however, address a different exhaustion problem, related to the plaintiffs’ argument under the Comstock Act. The Comstock Act is a federal criminal law from the 1800s that declares abortion drugs to be “nonmailable matter.” The plaintiffs claim that, because of the Comstock Act, the FDA shouldn’t have approved mifepristone. However, the plaintiffs have never made this argument to the FDA at any time.

So how does the district court get around exhaustion? The court cites (p. 27) a 41-year-old Fifth Circuit case that the plaintiffs do not cite in their briefs: Myron v. Martin, 670 F.2d 49, 52 (5th Cir. 1982). Here’s what that case says:

The general rule of nonreviewability is not absolute. In exceptional circumstances a court will review for the first time on appeal a particular challenge to an agency's decision which was not raised during the agency proceedings. Board of Public Instruction v. Finch, 414 F.2d at 1072-73. In Finch this court found an exception to the general rule after determining that the action of the agency was (1) in excess of statutory authority, (2) likely to result in individual injustice, (3) disruptive of the legislative scheme and (4) contrary to an important public policy extending beyond the rights of the individual litigants. Id. at 1073. After considering each of the Finch factors, however, we feel that they are inapplicable in the instant case.

This is an impressive find from the district court! In the 41 years since Myron, this test has been applied in only one decision: an unpublished 2019 district court decision in a social security case, where the court declined to find an exception to exhaustion. And the Finch case, which was decided in 1969, was literally the only time any court within the Fifth Circuit has found an exception to exhaustion under this rule.

But hey, binding precedent is binding precedent, right?

Not exactly. Finch is a case in which there was no agency rule requiring exhaustion. The court relied only on the “general rule of appellate review from decisions of an administrative agency,” and concluded that it could invent a discretionary exception to that discretionary exhaustion doctrine. This principle does not apply when there is an agency rule requiring exhaustion.

Indeed, in Darby v. Cisneros, 509 U.S. 137 (1993), the Supreme Court held that exhaustion applies in Administrative Procedure Act cases only “to the extent that it is required by statute or by agency rule as a prerequisite to judicial review.” And here, there is such a rule. So these random cases from 1982 and 1969 are totally irrelevant.

Anyway, latching onto Finch, the district court opines that mifepristone abortion is “contrary to public policy” and would work an “injustice” to women who choose to take mifepristone and perhaps even to their unborn children (pp. 28-30). It is here where the judge throws in a reference to the argument that the unborn children are entitled to constitutional protection under the Equal Protection Clause (p. 30). This an impressive amount of juice to squeeze from this now-overruled doctrine applied a single time in an irrelevant context 54 years ago.

The court also throws in an argument that exhaustion wasn’t required because the FDA acted on the petitions too slowly (p. 30). I have no idea why these delays somehow retroactively excuse the failure to raise the Comstock Act argument to the agency in the first place. Finally, the court holds that exhaustion would be “futile” because of public statements by Joe Biden since Dobbs was decided (pp. 30-31). Again, I have no idea how this excuses the plaintiffs’ failure to raise their arguments to the FDA during the Trump Administration.

I cannot help but quote the district court’s final flourish on exhaustion (p. 31):

e. The Comstock Act was raised with Sufficient Clarity

Finally, the Comstock Act issue was “raised with sufficient clarity.” Ross, 976 F.3d at 942. This is because: (1) the 2019 Petition requested FDA to retain the in-person requirement for dispensing of chemical abortion drugs; and (2) the Comstock Act issue was also raised by the United States Postal Service and the Department of Health & Human Services on July 1, 2022, “[i]n the wake of” Dobbs.27 The Office of Legal Counsel specifically mentioned FDA’s regimen for chemical abortion drugs when concluding “the mere mailing of such drugs to a particular jurisdiction is an insufficient basis for concluding that the sender intends them to be used unlawfully.” OLC Memo at *1. This shows not only that the issue was raised with sufficient clarity, but also the futility of raising the issue before the agency. Therefore, Plaintiffs’ failure to exhaust their claims does not preclude judicial review.

Yup, the district court thinks that the Comstock Act was raised with “sufficient clarity” to the FDA based on the conjunction of (1) a citizen petition that said literally nothing about the Comstock Act, and (2) an inquiry by other agencies after the citizen petition was decided.

The Merits

Finally, the district court concludes that the plaintiffs win on the merits.

The court begins by holding that mailing mifepristone would violate the Comstock Act (pp. 32-36). I’m not persuaded by this for the reasons stated by the Office of Legal Counsel. But even assuming this is correct, I doubt the FDA’s statutory obligations include assessing whether mailing a drug would violate a federal criminal statute, and there are serious separation-of-powers problems with effectively permitting the plaintiffs to privately enforce a criminal statute over the objections of the United States. Finally, at most, the Comstock Act would prevent the mailing of mifepristone and wouldn’t cast doubt on the FDA’s conclusions regarding health and safety. So the Comstock Act can’t justify the district court’s decision to overturn the FDA’s approval of mifepristone. The district court blows past all of this; it just says that the Comstock Act applies, hence the FDA loses.

Next, the district judge finds that the FDA’s actions violated the Administrative Procedure Act.

The district court opens by holding that the agency’s 2021 actions related to mifepristone (apparently both the petition denial and the enforcement discretion during COVID) violate the APA. Why? From 2000 to 2016, the FDA required providers to report all serious adverse events resulting from mifepristone, whether fatal or not. Based on 16 years of data, the FDA concluded that adverse safety events were very rare, and therefore concluded that only fatal events need be reported. This seems reasonable and doesn’t violate any statute. Also reasonably, in post-2016 decisions, the FDA relied on all available data, including the data regarding fatal events from 2016 to 2021. The district court, however, holds that the agency’s reliance on that post-2016 data is arbitrary and capricious because it reflects “a predetermined conclusion in search of non-data — a database designed to produce a null set” (p. 39). I don’t understand what that means. The FDA is allowed to rely on data that is reported and it’s allowed to conclude that some reporting isn’t necessary.

Next, the court concludes that the FDA’s 2000 approval violated its Subpart H regulations and this violation wasn’t cured by subsequent statutory amendments (pp. 39-48). This analysis is wrong too, but this is a technical issue and this post is too long already, so you will have to take my word for it. I’d like to focus on only one portion of this discussion, in which the court concludes that the FDA acted arbitrarily and capriciously in concluding that mifepristone abortions provide a “meaningful therapeutic benefit” relative to surgical abortions (pp. 44-47).

The FDA concluded that “medical abortion through the use of Mifeprex provides a meaningful therapeutic benefit to some patients over surgical abortion,” because it “avoided an invasive surgical procedure and anesthesia in 92 percent” of patients in the trial. This seems pretty obvious, right? Aren’t women better off if they have the option of avoiding surgery, and the sedation and anesthesia that comes with it? Isn’t there a reason so many doctors recommend this option and so many patients take advantage of it? And isn’t it at least non-arbitrary that the agency would reach this conclusion?

Nah. The court cites some studies cited by the plaintiffs, some of which were never submitted to the agency, purporting to show that bleeding was more likely to occur from a mifepristone abortion than a surgical abortion (pp. 45-46). The court also relies on websites not cited by the plaintiffs that he apparently just Googled (p. 46 n.42). The court doesn’t suggest that the agency overlooked, or improperly analyzed, a study that was submitted to it. It doesn’t even address the agency’s actual reasoning at all. It just points to a few studies cited by the plaintiffs purporting to show that mifepristone is dangerous, and holds that this is enough to overturn the FDA’s decision. This isn’t how administrative law works.

The court also says this (p. 46):

Other studies show eighty-three percent of women report that chemical abortion “changed” them — and seventy-seven percent of those women reported a negative change.

What support does the court cite for this proposition? Why, it’s the study of the 98 54 anonymous blog posts on the pro-life website.

Wait a second. The website is called “Abortion Changes You.” The URL is abortionchangesyou.com. So yes, Judge Kacsmaryk really is quoting the statistic that “eighty-three percent of women report that chemical abortion ‘changed’ them,” without mentioning that the entire sample consists of anonymous bloggers on a website called “Abortion Changes You”!!!! This is roughly like reporting a statistic that “83% of people are fans of Judge Kacsmaryk” without mentioning that the entire sample consisted of posters on JudgeKacsmarykFanClub.com.

Even the authors of the study, to their credit, feel compelled to note this issue when they recite the statistic: “Most women (N = 81; 83%) reported that their medication abortion changed them, which is not surprising given the name of the website: Abortion Changes You.” Touché!

Oh and also, this study is from 2021—21 years after the purportedly arbitrary and capricious agency decision that the judge is overturning. So apparently, the FDA violated its regulations because, back in 2000, it failed to fire up its DeLorean, accelerate to 88 mph, and travel to 2021 to educate itself on this apparently seminal study showing that anonymous posters to AbortionChangesYou.com felt changed by abortion.

Skipping forward a bit, the judge also second-guesses every aspect of the agency’s safety analysis from 2000. For example, in 2000, the FDA declined to require doctors to perform an ultrasound to assess gestational age before prescribing mifepristone. Instead it concluded that “a provider can accurately make such a determination by performing a pelvic examination and obtaining a careful history,” and declined “to mandate how providers clinically assess women for duration of pregnancy and for ectopic pregnancy.” This does not sound “arbitrary and capricious” to me. And, the FDA then mandated reporting of fatal and non-fatal adverse safety events for the next 16 years, and found an extremely low rate of adverse events, fully vindicating its reasoning.

No, says the district court: this just isn’t safe enough! All women must have ultrasounds (pp. 51-53)! There’s no statute that requires this; the district judge simply declares he is better at determining what is safe for women than both the FDA and the women’s doctors. The judge is buoyed by “myriad stories and studies,” such as an anonymous anecdote from a woman who says she didn’t get an ultrasound when she went to Planned Parenthood (p. 52).

Essentially the judge’s reasoning is as follows:

• More restrictions would add safety.

• The plaintiffs have submitted anecdotes regarding accidents the restrictions would have prevented.

• Therefore, it is arbitrary and capricious not to have the safety restrictions.

• And, therefore, the correct remedy is to ban all women from taking mifepristone.

This is not how administrative law works.

The judge then proceeds to an extraordinary argument that even the plaintiffs don’t make. He explains that in 2000, “FDA privately delivered to the [applicant] a set of proposed restrictions to rectify the safety issues” (p. 55). But then, after the FDA’s letter was leaked and public controversy ensued, the FDA chose not to implement those restrictions (p. 56). The judge then invokes the doctrine that an agency must offer a reasoned justification for changing positions (p. 56). It is true that when an agency makes a decision that conflicts with an actual prior decision of the agency—a rule, an adjudication—it has to explain that change. That doctrine doesn’t extend to the situation where an agency’s final decision purportedly conflicts with a private letter sent during negotiations that is leaked to the press.

The court also says it is entitled to overturn an agency decision any time “the agency’s decision was in any material way influenced by political concerns” (p. 56). It then complains about meetings the FDA and HHS might or might not have arranged back in 2000 (p. 57) and concludes that Clinton-era “political concerns” influenced the FDA, hence warranting wiping mifepristone off the market. To state the obvious, press reporting suggesting an agency may have been influenced by “political concerns” does not permit a district court to overturn an agency decision. I look forward to the court applying this rule during the next Republican administration!

Not surprisingly, the court also concludes that the FDA’s 2016 decision to relax certain restrictions associated with mifepristone was also arbitrary and capricious (pp. 58-60). Once again the court relies on its view that the agency didn’t have enough safety information, notwithstanding the exhaustive agency record and the agency’s thoroughly explained conclusion that it had enough safety information.

I will skip over the court’s discussion of irreparable harm and the public interest, which includes assertions such as “Though eugenics were once fashionable in the Commanding Heights and High Court, they hold less purchase after the conflict, carnage, and casualties of the last century revealed the bloody consequences of Social Darwinism practiced by would-be Übermenschen” (p. 64). I’ll let the administrative law professors address the mind-bending assertion that the judge is merely “staying” the 23-year-old approval of mifepristone (p. 66).

***

I will conclude with a note of optimism. Decisions like this one are rare. We should be proud of the exceptionally high quality of the federal judiciary and its commitment to the rule of law. Individual decisions like these should not change that. I do not agree with suggestions that the Administration or anyone else should violate judicial decisions. Trust the process.

1. In the interest of brevity (this post is very long), I’m going to skip a few things, like the zone-of-interests analysis, the “enforcement discretion” analysis, and a few others.