Photo of Bexis

Yours truly (trulies?) chaired the American Conference Institute’s preemption conference in Philly earlier in the week. We won’t try to be too modest here. It was a great seminar. But frankly that had little to do with us and everything to do with the all-star group of speakers that ACI assembled.
We continue to be flabbergasted by the compliments we get about the blog. Really. And by the end of the program we could hardly believe some of the things we heard (some of which we’re not at liberty to repeat). Thanks to everyone for the more-than-kind words.
The scary part is now we have to live up to all these heightened expectations. What have you done for us lately, bloggers?
Not much today, I’m afraid. All we’ve done is put together some of the thoughts we had while we were listening to the various ACI presentations. There’s no particular theme, except that we wanted to get these down before, in the press of other business, we forget about them.
So here goes.
Watch out for disguised “manufacturing” claims.
We’ve already predicted that, after Riegel, we expect to see plaintiffs retooling their allegations to squeeze them into any perceived loopholes that they can find. Indeed, savvy plaintiffs were doing that already. See Adkins v. Cytyc Corp., 2008 WL 2680474 (W.D. Va. July 3, 2008) (including non-preempted claim about face-to-face interaction between physician and sales representative of company selling PMA device).
We’d (and plaintiffs, we suspect) mostly thought about that in terms of supposedly “parallel” violation claims, but another way for the toothpaste to squeeze out of the tube could be the proliferation of bogus “manufacturing defect” claims that really aren’t. In opposing that, it’ll be important to keep courts focused on what “manufacturing” defects really are.
In both the second and third Restatements, and in almost every state we know of (but don’t expect another state-by-state chart on this), a “manufacturing defect” is narrowly defined as a product that is not what the defendant intended it to be. A pacemaker that, because of some error, has a plastic coating that was not cured for as long as the manufacturer intended has a manufacturing defect. But, on the other hand, a pacemaker about which it is alleged that the manufacturer’s intended curing process somehow violates FDA Good Manufacturing Practices does not have a “manufacturing defect” as defined by state law. In that situation, because the resulting device was exactly what the manufacturer intended it to be, there is no “manufacturing defect.”
So any claim that mentions “manufacturing” isn’t necessarily a proper state-law “manufacturing defect” claim. We’ll need to watch carefully for plaintiffs trying to pass off what are really Riegel-preempted design defect claims or Buckman-preempted “solely from the violation of FDCA requirements” claims as so-called “manufacturing” claims. In doing so, the well-defined state-law definition of “manufacturing defect” in most jurisdictions will be our friend.
Post-Riegel trial court preemption decision
Another thing that happened at the conference is that we swapped information. We came away with a preemption decision we hadn’t heard of before, called Mullin v. Guidant. We think it’s the first state trial court decision to decide a medical device preemption motion after Riegel. It’s not long or profound, but it was (as far as we know) first. We found out about it, so now you know about it, too.
Buckman parallel requirements analysis could bar post-approval claims
We’ve discussed at length our view that Buckman’s analysis of what is and isn’t a “parallel” claim means that preemption bars a pure regulatory violation claim without any state-law counterpart. We just weren’t sure before the ACI conference whether that limitation would, in practice, eliminate that many claims that plaintiffs would really bring.
Then we heard one of the panels discuss how one type of claim that plaintiffs would probably pursue, because it hadn’t been directly addressed in Riegel, was that the defendant somehow violated post-marketing regulatory obligations. That is, plaintiffs would claim that the label as approved was OK, but that the defendant had not lived up to its obligations thereafter, and that for that reason the label should have been changed.
Well, that kind of claim sounded a lot like a post-sale duty to warn or failure to recall claim to us. Thus the question: “If the allegation is that the defendant violated an obligation to change the label after plaintiff had been prescribed the drug, and state law doesn’t recognize any post-sale duty to warn, then does there exist any common-law claim for the violation allegations to run ‘parallel’ to?” The answer, of course, was no.
So it looks like our Buckman analysis might have practical benefit in defending against post-Riegel claims after all.
Getting preemption protection for products prospectively
We confess (boy, do we confess) that we’re not FDA regulatory lawyers. But thankfully there were some of the regulatory types at the conference. They had this tip for minimizing exposure to plaintiffs’ arguments based upon the CBE regulation (those of you not up on the jargon, that’s the “you could have added or strengthened a warning without FDA pre-approval” argument). It’s a tip that would be especially useful in the mass tort context.
We haven’t studied the regulations ourselves to verify it, but we trust the source – so here goes.
Companies can voluntarily choose to comply with the post-2006 “updated” form of labeling that includes the new “highlights” section. The post-2006 regulations provide that the CBE route for modifying labeling is entirely inapplicable to the “highlights” section. We’re told (this is what we haven’t verified) that the regs and commentary make clear that CBE is unavailable to change anything in the rest of the labeling that would, in turn, require a change to the “highlights” section. All contraindications, for example, must be stated in the “highlights.” Thus for drugs that use the post-2006 labeling, there’s no argument (like the Vermont courts adopted in Levine) that the manufacturer could have used CBE to add a contraindication.
Think about that.
Many, if not most, pharmaceutical mass torts involve risks that already appear on the labeling – if not at the very beginning, then at least sometime in the drug’s labeling history. Of course, the plaintiffs are going to claim that the warnings about such risks should have been strengthened. To try to escape preemption, in such cases, plaintiffs are going to argue that defendants could have used CBE to strengthen the existing warning.
Well, if the drug’s under the post-2006 labeling regime (which manufacturers can accede to voluntarily), then a label change of the type that mass tort plaintiffs would be advocating (for example, something that the FDA required later on) would almost always result in requiring the risk in question to go into the “highlights” as well. That’s because any risk serious (or widespread) enough to generate a mass tort would, almost by definition, be serious enough to warrant inclusion in the “highlights” section.
And if it’s serious enough to go into the “highlights” section, then under the FDA’s regulations, the CBE procedure is unavailable. If the CBE process isn’t available, then the requirement of FDA pre-approval probably means that the claim will be preempted. That means (at least the argument goes) that, in practice, a lot less mass tort litigation against drugs under the post-2006 label format should survive preemption.
We’d recommend that pharma manufacturers consult with their own regulatory counsel and verify this. If it’s true, then a conversion to the post-2006 regime could well have an extremely beneficial effect upon future (not existing) mass tort litigation.
And we keep repeating the “mass tort” adjective deliberately. There could be drawbacks in defending “one off” claims because drug manufacturers using the post-2006 labeling regime are, conversely, expected to reduce the clutter on their labeling by removing warnings that lack scientific basis. But in litigation, those laundry lists of everything that showed up during clinical trials are beneficial, since they provide a basis for arguing that the warning adequate as a matter of law as to any risk “specifically mentioned” in the labeling. So there’s a downside to being under the post-2006 labeling regime in peculiar individual cases.
Since we do mass torts, we’d love to see that tradeoff made, but it does need to be seriously considered.
And getting the FDA to issue a written order that the manufacturer remove this or that warning as unsupported would go a long way towards getting preemption to protect those claims as well.
In tough jurisdictions consider primary jurisdiction arguments in addition to preemption
We’ve already got Riegel. If we get a good result in Levine as well, then we will have succeeded in moving preemption beyond the “foundational” stage. That means we’ll then be asserting preemption in a lot more jurisdictions – ones unsympathetic to our side – where we wouldn’t have before.
One alternative that was bandied about at the ACI conference for possible use in such jurisdictions was FDA primary jurisdiction. In other words, it’s an argument that, rather than dismissing the plaintiff’s suit outright, plaintiff should take the allegations to the FDA and let the agency decide them. If a defendant thinks that, realistically, the odds of the FDA agreeing with a plaintiff’s labeling critique are “slim and none, and Slim just walked out the door,” then giving a hostile court the out of telling the plaintiff to go to the FDA rather than outright dismissal may increase the likelihood of success.
But again, be careful. The “let the FDA decide” argument is in tension with the Supreme Court’s “agency inundation” argument in Buckman – so if Buckman’s also a big part of your argument, it’s best to leave primary jurisdiction alone.
Preemption is a reaction to an increasingly intrusive litigation environment
In essence, the FDA didn’t turn to preemption because it wanted to, but because it had to. We don’t know exactly where they were from, but one of the ACI conference speakers had stats on federal court prescription drug and medical device product liability filings. Such filings rose from 2,700 in 2001 to a peak of 22,400 in 2004, before dropping back slightly to 17,000 in 2005. In other words, there was a five-fold increase in federal-court prescription medical product liability litigation over the five years leading up to the FDA’s early 2006 preemption preamble.
And we thought we had it bad during the Bone Screw days.
One thing is obvious from those stats – the plaintiffs’ bar has been trying to turn the products the FDA regulates into the next asbestos. But you can’t manufacture a 500%+ increase in litigation over five years without changing the nature of the litigation itself … expanding its scope into areas previously unknown, such as attacking warnings even though they were specifically mandated by the FDA.
If preemption does no more than reduce the number of filings in our neck of the woods back to the 2001 level, it will have accomplished a lot.
A Levine win, and the CBE change, would let us breathe a little easier after 2008
Everybody in the preemption fight – both sides – uses January 20, 2009 as a reference point. There will be a new administration, and thus the FDA will be under new management. We don’t do partisan politics on this blog, but it’s certainly possible that the FDA’s new management will be cozier with the other side than current management is. And Congress could be more responsive to the other side as well.
Not good – at least for the folks on our side of the “v”.
But does it spell disaster for preemption?
Not necessarily.
If we get a ruling in favor of preemption in Levine – the consensus at the ACI conference was optimistic on that score – and the FDA finalizes its revisions to the CBE regulation to bring it into line with what the agency originally intended, then there’s reason for hope.
Nor is it likely that a new administration would try to reverse the government’s amicus position in Levine itself, because a move that blatantly political would destroy the credibility of the new Solicitor General with the Supreme Court.
If we get implied preemption on a reasonably broad rationale in Levine, then even if Congress were to take away the preemption clause that underlies Riegel, device manufacturers could rely on the same kind of implied preemption. Indeed, to the extent there’s an actual conflict, even §510k device manufacturers (the big losers in Lohr) would be able to rely on implied preemption after a favorable decision in Levine.
Because implied preemption is dependent upon conflict rather than congressional intent, it’s difficult for the litigation lobby to touch it. Indeed, there’s persuasive precedent (Dowhal) holding that they can’t touch it – that even the most explicit savings clause imaginable cannot authorize an actual conflict with supreme federal law. The litigation lobby would have to amend the Supremacy Clause itself, something that’s never been done in the nation’s history.
If the FDA acts to restrict availability of CBE changes to new information the FDA has not yet reviewed, it’s going to be hard to justify changing it back (administratively or legislatively) to the absurd meaning plaintiffs try to give the current language.
It just doesn’t make any sense for the CBE process to permit (and thus obligate, under plaintiffs’ view of the world), a manufacturer to ignore an FDA-mandated label by (as soon as the day after the FDA’s decision) unilaterally changing the warning to something else. Plaintiffs’ construction of the current CBE regulation would let (and thus obligate) manufacturers to repeat such changes again, and again, and again, over and over, each time the FDA rejects the proposed change – all without any new scientific evidence to warrant any change at all. To bring about such a blatantly nonsensical change would require the litigation lobby to make a huge investment of political capital.
So if we have implied preemption and a sane CBE regime, the litigation lobby would be forced to strike at the FDA itself, perhaps by trying to eliminate the preemptive conflict by reducing FDA-approved warnings to something that state law can ignore at will. But taking away the FDA’s control over labeling is not far away from abolishing the FDA itself. Theoretically Congress could do that, but there’s reason to question whether even the litigation lobby would be able to flex the kind of raw, political muscle necessary to do something like that.
So if we get preemption now, we may just be able to keep it.