Cross-posted on ACSBlog.
A panel of the Second Circuit Court of Appeals in New York has just taken under consideration the Food and Drug Administration’s motion for a stay of a district court order directing the agency to make levonorgestrel-based emergency contraceptives available to women and girls of any age without a prescription and without other point-of-sale restrictions. In deliberating on this motion, the panel of judges should not, I am sorry to say, take anything the FDA has said in its briefs at face value. The government’s opening and reply briefs on the motion to stay are so full of misstatements and omissions that the court could badly err if it did not take everything the government says with a shaker full of salt.
One of the factors in deciding whether to grant a stay pending appeal is the likelihood that the moving party will succeed on the merits. The government devotes most of its briefs to this factor. It makes two arguments as to why the court of appeals should find that the government is likely to win on appeal and should thus stay the district court’s order on emergency contraception. Both arguments depend crucially on incomplete and inaccurate renderings of the law and facts of the case.
Before turning to these arguments, a bit of context is necessary. The levonorgestrel-based emergency contraception at the center of this legal dispute takes two forms. One, Plan B and its generic versions, requires two pills. The other, Plan B One-Step and its generic versions, requires one pill. Both involve the same total dose of levonorgestrel. Despite these obvious similarities, the FDA has worked very hard to treat these drugs very differently; it has made Plan B One-Step available without a prescription to all women and girls over the age of 15, it has apparently blocked nonprescription market access to generic versions of Plan B One-Step for girls under 17, and it has resisted requests to make Plan B and its generic versions available without a prescription to girls under age 17. The district court’s order would make all of these drugs (except Plan B, which is no longer marketed) available without a prescription; the FDA would like to keep treating them differently.
The government’s first argument as to why it will likely succeed on appeal is that the district court had no power to order Plan B One-Step to be made available without a prescription. The government explains that Plan B One-Step was the subject of a supplemental new drug application before the FDA, the application was denied in 2011, and the district court had no power to overturn that denial. But the district court did not overturn the denial of this application. Rather, the district court, in responding to a citizen petition filed in 2001, ordered FDA to allow all levonorgestrel-based emergency contraceptives – which would include Plan B One-Step – to be made available without a prescription. (The court allowed FDA one possible out: if the FDA found that there was a “significant difference” between the two-pill regimen of Plan B and its generic equivalents, and the one-pill regimen of Plan B One-Step and its generic equivalents, it could limit its approval of nonprescription status to the one-pill products.) The FDA’s governing regulations clearly allow the FDA to order switches from prescription to nonprescription status based on citizen petitions, not only based on applications from drug manufacturers. So it is simply obtuse, indeed almost willfully blind, for the government not to be able to discern the difference between overturning the FDA’s decision on a new drug application and ordering a switch to nonprescription status pursuant to a citizen petition.
The government also says that the citizen petition did not ask for any remedy with respect to Plan B One-Step. True enough, the petition, filed in 2001 – eight years before Plan B One-Step was even approved – did not identify Plan B One-Step by name. Rather, and more broadly, the citizen petition asked the FDA to allow nonprescription status for Plan B and “any new drug eligible for filing an abbreviated new drug application because of its equivalence” to Plan B. Plan B One-Step is clearly encompassed within this request for relief. Drugs are eligible for an abbreviated new drug application both when they are “bioequivalent” to an approved drug and involve the same route of administration, dosage form, strength, and labeling and when a person submits a petition to file an abbreviated application and the Secretary has not found that further studies are necessary to show safety and effectiveness. The government mentions only the first, not the second, route to an abbreviated new drug application. It asserts that Plan B One-Step was not encompassed within the citizen petition because its dosage form was different from that of Plan B – even though the same statutory provision the government cites and quotes for this purpose explicitly refers to the petition process I just described, which does not require identical dosing. The government simply, and conveniently for its argument, omits this portion of the statute from its quotation and indeed from its brief. Yet one must understand the full breadth of the routes to eligibility for filing abbreviated new drug applications in order to grasp the breadth of the citizen petition.
Not content with under-quoting the statute at issue, the government also says that the citizen petitioners “at no point” have suggested that their petition encompassed Plan B One-Step. This is not true. In a motion for contempt in 2010, filed because the FDA had not complied with a previous district court order to reconsider its stance on levonorgestrel-based emergency contraceptives, the citizen petitioners argued that the FDA itself had “acknowledged the equivalence of the two brands [Plan B and Plan B One-Step] for the purposes of the action requested by the Citizen Petition.” The petitioners were undeniably acknowledging that in light of technological developments that occurred while FDA was sitting on their petition, their petition encompassed Plan B One-Step as well as Plan B. The petitioners repeated this assertion in an amended complaint filed in 2012, in which they asked for relief pertaining to Plan B One-Step and its generic equivalents as well as Plan B and its generic equivalents. This is the relief they eventually obtained from the district court. It is not correct, in other words, to say that the petitioners “at no point” have said that their petition encompassed Plan B One-Step.
Moreover, as the petitioners pointed out in their 2010 motion for contempt, FDA had announced that it would comply with the district court’s ruling on the citizen petition by reviewing “a supplemental new drug application expected to be submitted by the sponsor of Plan B One-Step.” Putting it all together, then, the government three years ago said that the way to respond to the citizen petition was to wait for, and if one was submitted, to review an application for Plan B One-Step, but this week the government has told the Second Circuit that the citizen petition had nothing to do with Plan B One-Step. These are the kinds of shenanigans that made the district court in this case finally lose patience with – and, I dare say, trust in – the FDA.
The government’s second argument about why the district court was wrong on the merits also suffers from mistakes of commission and omission. The government chides the district court for overstepping its bounds in ordering the FDA to make levonorgestrel-based emergency contraceptives available to all women and girls without a prescription, arguing that the proper course was to remand the case to the agency so that the agency could engage in a full-fledged rulemaking process on that question. To be sure, this is ordinarily the course a court takes when it has found errors in an agency’s decision making. But this is not the ordinary case. The district court found – in conclusions nowhere directly challenged in the government’s briefs on appeal – that the FDA had acted not only arbitrarily and capriciously, but unreasonably and in bad faith as well. None of these findings finds their way into the government’s briefs on the motion for a stay; it is as if they don’t exist.
Yet, without acknowledging it, the government’s latest brief bristles with defiance toward the district court’s factual judgments regarding the necessity of a remand. The government says that “no part of the administrative record in this case” involves the Plan B One Step new drug application (false, as found by the district court). The government says that the FDA would benefit from evidence about the extent to which studies on older individuals can be extrapolated to younger adolescents (irrelevant, based on existing FDA policy on such extrapolation, as found by the district court). The government believes the FDA would benefit from studies establishing that “comprehension difficulties” would not hinder safety and effectiveness of Plan B (the FDA itself long ago waived any requirement for such studies, the district court found). It would also be useful, the government tells the court, to have evidence showing that “anticipated errors of taking the second dose later than instructed will not render the drug ineffective” (already found in scientific studies, said the district court). In laying out its plans for the remand it prefers, the government tries to sneak its factual disagreements with the district court through the back door and at the same time amply demonstrates why the district court doubted the government’s good faith in this case. The government’s plans for a remand look like nothing so much as a reprise of its previous failed encounters with emergency contraception.
Indeed, the government even suggests – after all this, after the findings of unreasonable delay and political meddling and bad faith – that it might not answer the citizen petition with respect to Plan B and its generic equivalents at all. Maybe, it hints, it just isn’t worth it to “commit limited resources” to this matter, since Plan B and its generic equivalents have a “rapidly declining market share” and now account for “only an estimated 15% of the levonorgestrel-based emergency contraceptive market.” On remand, in other words, the government might just decide to do nothing at all. These kinds of arguments do not warrant a remand; they warrant a kick in the pants.
If the government does decide to act on remand, it says, it must do so through notice-and-comment rulemaking; that is the only possible interpretation, apparently, of a statutory provision saying that the FDA “may by regulation” switch a drug to nonprescription status. Never mind that the FDA’s own rules allow the FDA, acting on its own or in response to a citizen petition, to switch a drug to nonprescription status, and do not specify a particular procedure for this decision. The government, in its opening brief on the motion for a stay, elided this fact through another selective quotation to governing text, observing that the FDA’s regulations provide that “[w]hen a citizen petition is filed, ‘[t]he Commissioner may initiate a proceeding to issue … a regulation.’” The government left off the remainder of the relevant sentence, which states in full: “The Commissioner may initiate a proceeding to issue, amend, or revoke a regulation or order or take or refrain from taking any other form of administrative action.”
Consistent with this regulation, when the district court ordered FDA to make Plan B available to girls 17 and over without a prescription in 2009, the government felt no need to undertake a rulemaking in order to comply with the court’s ruling. Instead, FDA acknowledged the court’s order, announced it would not appeal, and rested with the scientific findings that its own Center for Drug Evaluation and Research had made in 2005. The agency also solicited “an appropriate application” from the Plan B drug sponsor – not in order to reconsider its 2005 findings, but to ensure that the drug sponsor would alter its labeling and packaging consistently with the court’s order. The agency could do the same now, in response to the district court’s most recent order, as the FDA has done everything it needs to find that levonorgestrel-based contraceptives are safe and effective for all users.
Indeed, that is what the FDA did – just over a month ago – in response to a citizen petition asking it to change some of the drug product labeling for nicotine replacement products. The agency conducted its own review of the available literature and data on the safety of these products and concluded that some of the statements previously required in the labeling for these products were not necessary in order to ensure the safety and effectiveness of these products. The agency then invited drug sponsors to submit supplemental new drug applications consistent with its findings. The agency conducted no rulemaking and did not require supplemental applications before making its decision. In its briefs in the emergency contraception litigation, the government does not explain why its streamlined process on nicotine replacement products was acceptable; it doesn’t even mention it.
The remainder of the government’s arguments – that the stay won’t cause injury because none of the “named plaintiffs” has averred she plans to imminently use emergency contraception (does the government know what “emergency” means?); that the government itself will be harmed if the ordinary process of switching a drug to nonprescription status isn’t followed (it wasn’t followed here, the district court found, because the FDA refused to follow it); that the government sorely needs to protect the market interests of a single company that markets a single emergency contraceptive – are not worthy of an agency whose mission is to protect the public health.
The FDA should stop finding reasons not to do the right thing, and instead just do the right thing. If it won’t, the Second Circuit shouldn’t come to its rescue.