Photo of Bexis

This is more of a think piece, we won’t be citing many cases (by our standards) today.

There are two preemption principles at work in the Supreme Court’s recent Bartlett decision.  The first of these, which we’ll call “independence,” was articulated first in PLIVA v. Mensing, 131 S. Ct. 2567 (2011).  “[W]hen a party cannot satisfy its state duties without the Federal Government’s special permission and assistance, which is dependent on the exercise of judgment by a federal agency, that party cannot independently satisfy those state duties for pre-emption purposes.”  Id. at 2581.  That was reiterated in Bartlett in the design defect context, holding that “redesign was not possible” in part because that would require intervening FDA action.  Any redesign “would be a new drug that would require its own NDA to be marketed.”  Mutual Pharmaceutical Co. v. Bartlett, 2013 WL 3155230, at *8 (U.S. June 24, 2013).  And, as in Mensing, to the extent that the “design” claim in Bartlett devolved into a warning claim (because redesign was impossible), FDA regulations “prevented [defendant] from independently changing” its label.  Id. at *10 (quoting Mensing).

The Supreme Court has now enunciated and applied this “independence” principle twice.  Nor has it limited that principle to the FDCA.  See Id. at *10 n.3 (invoking Florida Lime & Avocado Growers, Inc. v. Paul, 373 U.S. 132 (1963) (a non-FDA case involving agricultural pricing)).  Moreover, since Bartlett (and Mensing) involved solely implied preemption, their principles apply regardless of the statutory scheme.  E.g.Bartlett, 2013 WL 3155230, at *6 (impossibility preemption applies “in the absence of an express preemption provision”).

The second preemption principle at work in Bartlett is new.  We’ll call it the “inaction” principle because it rejects a long-time plaintiff’s rejoinder to all forms of implied conflict preemption – that there’s no conflict because the defendant could always choose inaction (usually but not necessarily withdrawing from the market altogether) rather than violate state or federal law.  That’s been around since at least the mid-1980s, when we remember it vividly compared to “the free choice of coming up for air after being underwater.”  Palmer v. Liggett Group, Inc., 825 F.2d 620, 627 (1st Cir. 1987) – ironically by the same First Circuit that went the other way in the Bartlett litigation.

In Bartlett the Supreme Court definitively rejected inaction as a means of avoiding a preemptive conflict between state and federal law:

We reject this “stop-selling” rationale as incompatible with our pre-emption jurisprudence.  Our pre-emption cases presume that an actor seeking to satisfy both his federal- and state-law obligations is not required to cease acting altogether in order to avoid liability.  Indeed, if the option of ceasing to act defeated a claim of impossibility, impossibility pre-emption would be “all but meaningless.”

2013 WL 3155230, at *10 (quoting Mensing, 131 S. Ct. at 2579).  What’s more, the Bartlett court made absolutely clear that this rejection applied across the board – at least in “direct conflict” cases – repeatedly pointing out how, if allowed, the plaintiff’s “stop selling” rationale would have defeated preemption in practically all such implied preemption situations.  Id. (in “every instance”); see id. at *3 (argument would make preemption a “dead letter” and “work a revolution” in preemption precedent).

What do these two principles mean?

Potentially, quite a lot.  In the FDCA context, since implied and express preemption operate separately, with implied preemption not bound by express statutory preemption (or savings) language, we think that Bartlett could apply to essentially all design defect claims asserted against FDA-regulated products.  Specifically, we’re thinking of both innovator prescription drugs and §510k medical devices.  The independence principle applies to both.  Can any drug manufacturer change the composition (“design”) of its product in a way that would materially affect that product’s safety or effectiveness – and thus change it in a way that would be causal in product liability litigation?  We don’t think so, and here’s where we’d recommend consulting with FDA administrative law experts for more details.  Our non-expert understanding of FDA regulations is that, aside from so-called “minor modifications” that don’t affect safety/effectiveness, all changes to product composition for all prescription drugs, medical devices, and probably other FDA-regulated products require prior FDA approval, and thus would be preempted by the independence principle.  One of those regulations, 21 C.F.R. §314.70(b)(2)(i), the one for all prescription drugs, was specifically quoted with approval for this point in Bartlett.  “Once a drug − whether generic or brand-name − is approved, the manufacturer is prohibited from making any major changes to the ‘qualitative or quantitative formulation of the drug product, including active ingredients, or in the specifications provided in the approved application’.”  2013 WL 3155230, at *4 (quoting regulation) (emphasis added).

Thus, as in both Mensing and Bartlett (with respect to generic drugs), there is no CBE equivalent that would allow plaintiffs to avoid the implied preemption independence principle in any cases involving design defects.  Bartlett specifically mentions “brand name” drugs as covered by this principle, but so are §510(k) medical devices.  Remember that Medtronic, Inc. v. Lohr, 518 U.S. 470 (1996), only dealt with express, and not implied, preemption, and the two categories of preemption operate independently.  See, e.g., Buckman Co. v. Plaintiffs Legal Committee, 531 U.S. 341 (2001) (recognizing implied preemption in a §510(k) case).

But why stop there?  A lot of other products – airplanes, motor vehicles, ships, various other consumer products (including, most recently, tobacco products) – have to be approved by other federal agencies.  Here’s where we get fuzzy, because we’re less familiar with those regulatory schemes, but to the extent that (like FDA with respect to designs) these other products similarly require federal approval before design (or other) changes may be implemented, then they should also subject to preemption under the independence principle.  If compliance with state tort law would require a change that, under federal law, can only be made after obtaining prior federal agency approval, then the state claim would be preempted.

But why stop there?  The independence principle could well extend beyond product liability, and beyond tort law.  What else is there that requires prior federal  approval?  Transportation, communications, and energy facilities immediately come to mind, given their interstate nature.  We’re totally out of our comfort zone here, but anytime that federal approval exists, such that accommodating some state-law plaintiff asserting a conflicting state-law duty would require first going back to the relevant federal agency, then implied preemption under the independence principle is at least potentially in play.

Then consider the inaction principle.  Under Bartlett, preemption cannot be avoided by offering the defendant the Hobson’s choice of either paying damages under state law or ceasing the activity in question altogether.  Is the product, or activity, federally approved?  If so, then state law can’t just say, in effect, “pay or don’t play.”  This again would seem to have implications well beyond design defect claims where the defendant’s design is fixed by federal law, or even “categorical liability” claims (see Restatement (Third) of Torts, Products Liability §2, Reporters’ Notes at IV(D) (1998)) like the non-alternative design drug “design defect” claim at issue in Bartlett or the cigarette claims in Wood.  After Bartlett, to the extent a state-law plaintiff would put a federally-regulated defendant to the Hobson’s choice of either deviating from its federally-approved conduct or ceasing it altogether to avoid state liability, the inaction principle should impliedly preempt that claim.  As the Third Restatement points out, very few (if any) states recognize categorical liability theories, but to the extent they do, Bartlett establishes that preemption would bar such theories should they be asserted against a federally-approved product in such a way that would conflict with the terms of the federal approval.

We know of one example of this type of preemption in the innovator drug field − the discussion in Longs v. Wyeth, 621 F. Supp. 2d 504, 508-09 (N.D. Ohio 2009), which distinguished Levine and held a claim that a drug was “‘unreasonably dangerous . . . for which no warning would have been adequate” preempted.  That’s Bartlett in the innovator context.  Longs was, of course, reversed on this point in the execrable Wimbush v. Wyeth, 619 F.3d 632 (6th Cir. 2010) (same case; change of executor) decision.  But Wimbush is a dead letter after Bartlett, since it relied on the same “stop selling” argument that Bartlett has now repudiated.  619 F.3d at 643 (rejecting impossibility preemption against “a claim of negligence for bringing a drug to market in the first place”).  The inaction principle mandates preemption of this sort of claim.

Again, because the inaction principle is a matter of implied, not express, preemption, it can apply to activities not protected by express preemption.  Nor is the inaction principle limited to the FDCA or to any particular statute.  As an implied preemption principle, it is potentially applicable to protect the supremacy of federal regulation in any case where the federally regulated party would be put to the same sort of Hobson’s choice (pay or stop acting) by a state law claim.  One such situation that comes to mind, although very little such litigation is left, is the assertion of state-law public nuisance actions (in vogue briefly about a decade ago) against various actors notwithstanding their compliance with a governing federal regulatory scheme.

In sum, implied preemption – preemption by direct conflict − is a very powerful defense to litigation because its principles are not limited by any particular statutory language.  In Bartlett, the Court reiterated one principle, independence, and adopted another, inaction, that should have implications far beyond the specific fact pattern (generic drugs and design defects) of the Bartlett decision itself.