This post is from the non-Reed Smith side of the blog.
We’ve had the United States’ amicus brief opposing Medtronic’s petition for certiorari in Stengel for a few weeks now. See 2014 WL 2111719 (May 20, 2014). We turned it over on our desk the way you flip a sofa cushion with a red wine stain. You simply can’t stand to look at it. But, like with that stain – you know it’s there and you know you are going to have to deal with eventually. So, why not today?
You don’t have to look too hard on this site to find our repeated criticism of Stengel v. Medtronic Inc., 704 F.3d 1224 (9th Cir. 2013), the wrongly decided Ninth Circuit opinion allowing the plaintiff to allege state law failure-to-warn claims based on the purported failure to report adverse events to the FDA. (see post here) Finding that the state law claim was neither different from nor additional to federal regulations requiring adverse event reporting, the court concluded plaintiff alleged a “parallel” claim that was neither expressly nor impliedly preempted. Putting aside our view on parallel claims generally, certainly a claim that the manufacturer did something wrong by failing to report adverse events to the FDA is nothing more than a private party trying to enforce federal drug and medical device regulations, which the FDCA prohibits and which is preempted under Buckman.
Our dislike of Stengel may be outdone by our new dislike for the government’s position on Stengel. If we had to sum it up, the U.S. brief says essentially two things: every court to have considered Riegel (including the Stengel court) has gotten it wrong and Riegel should be overturned. The government reaches this conclusion by eliminating and/or ignoring the differences between §510k clearance and pre-market approval and the differences between drug and device law – essentially the entire reason why Riegel was decided differently than Lohr v. Medtronic, 518 U.S. 470 (1996) and Wyeth v. Levine, 555 U.S. 555 (2009). And based on its interpretation of Riegel preemption – plaintiffs would be allowed to bring direct failure to warn claims — something that every court since Riegel has found preempted.
The U.S. brief does not mince words regarding the Riegel progeny:
The courts of appeals, in every case since Riegel involving a device subject to premarket approval, have tacitly dispensed with the first step of a proper Section 360k(a) preemption analysis – i.e., asking whether FDA has established device-specific requirements on the same subject as the relevant state requirement.
2014 WL 2111719 at *15.
First, isn’t that exactly what Riegel says? “Premarket approval, in contrast [to §510k clearance], imposes ‘requirements’ under the MDA.” Riegel, 552 U.S. 312 at 322. Can it really be that every court to have read and considered Riegel got it wrong? Or is it more likely that the current administration is pushing a new interpretation with a more pro-plaintiff slant? We’ll let you decide.
Second, isn’t that exactly the opposite of what the government told the Court in Riegel? We already covered, on November 29, 2007, what the United States told the Court about express preemption in Riegel so we’ll just pull it out for you here:
- FDA pre-market approval (“PMA”) of a Class III device imposes preemptive federal “requirements” as meant by §360k(a) of the Medical Device Amendments (“MDA”) to the FDCA. Riegel SG/FDA Merits br. at 6, 11 (“FDA’s premarket approval gives specific content to those general requirements as applied to a particular device”), 24 (requirements “include the specific design and labeling requirements imposed as part of the PMA process”).
- PMA is in no way comparable to the less rigorous form of marketing clearance at issue in Medtronic, Inc. v. Lohr, 518 U.S. 470 (1996). Riegel SG/FDA Merits br. at 3, 7-11, 13, 22-23.
- Lohr recognized the FDA’s “significant role” in defining the preemptive scope of the MDA and gave “significant weight” to the Agency’s construction of the statute (so the Court should listen to the government just as much here). Riegel SG/FDA Merits br. at 23-24.
- PMA is the Agency’s judgment about the safety and efficacy of the device, and once PMA is granted, the manufacturer cannot make any safety-related changes to its device without a supplemental application requiring further FDA approval. Riegel SG/FDA Merits br. at 3, 13 (“respondent could not have lawfully marketed a product that deviated from the approved version nor made any changes affecting the safety or efficacy of the device, including labeling changes, without first submitting a supplemental application to FDA”), 15.
- Common law product liability claims can impose “requirements” under §360k(a), Riegel SG/FDA Merits br. at 16-19 (rejecting various arguments that would exclude state tort claims, generally, from preemption), and claims alleging that the PMA device is “defective,” would impose requirements “inconsistent’ with the FDA’s requirements imposed via PMA. Thus the conflicting tort claims are expressly preempted. Id. at 4-5, 7, 20-22.
- FDA PMA involves rigorous review of the adequacy of product labeling/warnings. Riegel SG/FDA Merits br. at 12-14, 22.
Compare the two briefs (Riegel in 2007 and Stengel in 2014) and it’s immediately apparent that the government has a Levine-sized inconsistent positions problem on its hands. But unlike Levine, where the government at least tried to justify its changed views, this time around the government hides its head in the sand, ostrich style, and fails to acknowledge that it was for preemption before it was against it.
In our opinion, the government’s current brief goes to undue lengths to gloss over the 510k/PMA distinction. For instance, the government states: “Riegel reaffirmed that distinction between ‘manufacturing and labeling requirements applicable across the board to almost all medical devices’ and ‘requirements specific to the device in question.’” 2014 WL 2111719 at *10. But, it’s not that hard. If you read Riegel, the distinction was actually being drawn between the vast majority of devices which enter the market via §510k clearance and the relatively small number of devices which undergo PMA. See Riegel, 552 U.S. 312 at 322-23. Why this attempt to obscure the very real differences? Because the Supreme Court was clear in Riegel that most state claims involving PMA devices are preempted.
There are two linchpins of the MDA’s express preemption provision – a state cannot maintain a requirement that is (1) “different from or in addition to” any
requirement applicable to the device which (2) “relates to the safety and effectiveness of the device.” 2014 WL 2111719 at *2 (emphasis added). PMA is all about safety and effectiveness. PMA devices undergo a rigorous “federal safety review.” Riegel, at 323. Therefore, PMA devices are entitled to
[T]he attributes that Lohr found lacking in § 510(k) review are present [in premarket approval]. While § 510(k) is focused on equivalence, not safety, premarket approval is focused on safety, not equivalence. While devices that enter the market through § 510(k) have never been formally reviewed under the MDA for safety or efficacy, the FDA may grant premarket approval only after it determines that a device offers a reasonable assurance of safety and effectiveness, §360e(d). And while the FDA does not require that a device allowed to enter the market as a substantial equivalent take any particular form for any particular reason, the FDA requires a device that has received premarket approval to be made with almost no deviations from the specifications in its approval application, for the reason that the FDA has determined that the approved form provides a reasonable assurance of safety and effectiveness.
Id. What part of the PMA process being a “safety review,” id., doesn’t the government understand?
Blurring the line between Riegel and Lohr is just one part of an overall approach by the government to eliminate Riegel preemption without actually saying it.
Its next step is to so severely limit the holding in Riegel as to make it inconsequential. The government suggests that Riegel only preempts failure to warn claims premised on a challenge that “FDA should have conditioned its approval on adopting some other . . . labeling.” 2014 WL 2111719 at *12. When the allegation is that the manufacturer should have warned of new information learned post-marketing, that conduct is not governed “by the terms of the device’s premarket approval, but rather by FDA’s general regulations” regarding adverse-event reporting and labeling revision. Id. Really? We can’t think of a single case where failure to warn wasn’t based on an allegation that the manufacturer should have alerted doctors to newly learned safety information. This is the exception swallowing the rule. Moreover, if the allegation is that the manufacturer failed to provide information to the FDA during the premarket approval process and that such information would have altered the FDA’s approved labeling – isn’t that a fraud-on-the-FDA claim? We don’t need Riegel to preempt that, we have Buckman.
And, it’s Buckman that gives the government some trouble in its support of the Ninth Circuit’s decision. The claim upheld by the Ninth Circuit is failure to warn premised on a failure to advise the FDA of adverse events. Even the U.S. has to admit that because a failure to report refers “to the agency decision-making process,” it “may implicate Buckman.” Id. at *19. Wow, what a concession – that failure to report information to the FDA may be impliedly preempted because the FDCA is a “critical element,” of the claim and it therefore counts as a prohibited attempt at private enforcement of the FDCA. Buckman, 531 U.S. 341, 353 (2001). Indeed, as a practical matter basing tort liability on a mere failure to report would have worse consequences for the policy reasons that underlie Buckman – that risk-averse manufacturers could inundate the FDA in a flood of “information” that the Agency doesn’t want or need, id. at 351 – than Buckman itself, the facts of which involved harder to make allegations of affirmative misrepresentations.
So Buckman looms as a massive obstacle to the government’s anti-preemption arguments. That obstacle is all the more massive, as the government’s current tolerance of state-law fig leafs for what are essentially FDCA-based claims, 2014 WL 2111719 at *31, conflicts directly with what the Solicitor General told the Court about FDCA enforcement hiding inside of claims that “sound traditional” back in 2007, in its amicus brief in the Warner-Lambert v. Kent, case. See our discussion of the government’s Kent brief here. Yet nowhere does the government acknowledge, let alone explain, this massive reversal of position.
Faced with a Buckman dilemma, the government contorts express preemption so far as to completely nullify Riegel. It contends that the Ninth Circuit need not have premised plaintiff’s failure to warn claim on a failure to report, but rather could have allowed a direct failure to warn claim. Riegel, as held by every court in the country – including the Ninth Circuit in Stengel — preempts claims of direct failure to warn. End of story. Unless you are the current federal
administration, in which case you suggest the wildly unbelievable conclusion that in a case involving a PMA device:
Respondents may properly proceed on the theory that petitioner should have invoked the CBE process to update its device’s labeling to reflect new information bearing on the safety of the device. Such a claim would mirror the failure-to-warn claim against the prescription drug manufacturer that this Court held was not impliedly preempted in Wyeth v. Levine.
Id. at *19-20. We pause for a moment to allow the absurdity to sink in. The government is arguing that a plaintiff in a PMA device case be allowed to bring a claim that the manufacturer should have revised its label. And, in doing so says that PMA devices are no different than prescription drugs. But again, Riegel says exactly the opposite:
If . . . the pre-emption clause permits tort lawsuits for medical devices just as they are (by hypothesis) permitted for drugs and additives; and if . . .Congress wanted the two regimes to be alike; Congress could have applied the pre-emption clause to the entire FDCA. It did not do so, but instead wrote a pre-emption clause that applies only to medical devices.
Riegel, 552 U.S. at 327.
What’s left of Riegel when the government is done with it. Almost nothing. Their interpretation would re-open the door to virtually every claim against a PMA device that courts have found preempted for the last six years. We can only hope their ploy doesn’t work, that the Supreme Court grants certiorari, and that it sees the Stengel claim for what it is – a non-parallel violation claim both expressly and impliedly preempted. We see some reason for optimism. The government’s position is so extreme that it almost invites review, despite the request that it be denied. Doesn’t the assertion that all the lower courts have gotten preemption wrong for six years almost beg for review? In fact, we can’t help wondering if that’s what’s really at work. Perhaps the government secretly wants the Court to review its position, but for political reasons can’t say so. Taking this kind of provocative position may be one way getting review without asking for it, at least if your name is “United States of America.”
And we should know sooner rather than later. The Stengel docket confirms that the cert petition has been distributed for the June 19th conference, meaning we should hear one way or another next Monday at 9:30.
We don’t think we can stomach much more. We’re going to flip the sofa cushion back over and allow ourselves a temporary reprieve from the awful stain of this brief. Anyone have an industrial size can of Oxi-Clean?