Friday, February 01, 2013
Spherical Error Revisited
Every now and then we run into a decision that we think is wrong in so many different ways that we call it an example of “spherical error” – that is, error no matter how one looks at it. We only do that rarely. Back in May, 2010 we first applied that description to In re Gadolinium-Based Contrast Agents Products Liability Litigation, 2010 WL 1796334 (N.D. Ohio May 4, 2010). Then in September of that year we likewise identified spherical error in Sheet Metal Workers Local 441 Health & Welfare Plan v. GlaxoSmithKline, PLC, 737 F. Supp.2d 380 (E.D. Pa. 2010) – or at least the Pennsylvania part of it, which dealt with the law we knew best.
That’s it – two cases earned that sobriquet in the blog’s six-plus years of existence.
Make that three.
The recent decision in Arters v. Sandoz Inc., 2013 WL 308768 (S.D. Ohio Jan. 25 2013), likewise bollixed up the law in so many different ways that we think it rises (falls?) to the level of spherical error.
Arters involved a question that a lot of courts have considered lately with almost (but not quite) uniformly good results – preemption in the context of generic drugs following PLIVA v. Mensing, 131 S. Ct. 2567 (2011). As in many generic preemption cases, the defendant sought to have the case dismissed under Rule 12.
That – Rule 12 – is the first error in Arters. The decision states:
A motion for judgment on the pleadings pursuant to Rule 12(c) should not be granted “unless it appears beyond doubt that the plaintiff can prove no set of facts in support of his claim which would entitle him to relief.” Conley v. Gibson, 355 U.S. 41, 45–46, 78 S.Ct. 99, 2 L.Ed.2d 80 (1957).
Arters, 2013 WL 308768, at *1.
Someone must have been the second coming of Rip Van Winkle and slept through the last five years. In a rather well-publicized decision, back in 2007, the Supreme Court expressly overruled Conley on specifically this “no set of facts” point. See Bell Atlantic Corp. v. Twombly, 550 U.S. 544, 561-63 (2007). Twombly? Maybe you’ve heard of it. If so, you’re already one up on Arters. Anyway, the Supreme Court in Twombly explicitly “retired” the Conley standard relied upon in Arters:
On such a focused and literal reading of Conley’s “no set of facts,” a wholly conclusory statement of claim would survive a motion to dismiss whenever the pleadings left open the possibility that a plaintiff might later establish some set of undisclosed facts to support recovery. . . . It seems fair to say that this approach to pleading would dispense with any showing of a reasonably founded hope that a plaintiff would be able to make a case; Mr. Micawber’s optimism would be enough.
Seeing this, a good many judges and commentators have balked at taking the literal terms of the Conley passage as a pleading standard. [many citations omitted] We could go on, but there is no need to pile up further citations to show that Conley’s “no set of facts” language has been questioned, criticized, and explained away long enough. . . . [A]fter puzzling the profession for 50 years, this famous observation has earned its retirement. The phrase is best forgotten as an incomplete, negative gloss on an accepted pleading standard.
Twombly, 550 U.S. at 561-63 (emphasis added). Then, a couple of years later, the Court made clear that Twombly’s interrment of the Conley language cited in Arters applies across the board to all types of cases. See Ashcroft v. Iqbal, 556 U.S. 662, 684 (2009).
So that’s where the spherical error in Arters gets rolling. The opinion starts off with the wrong basic standard for review.
And it’s downhill from there. Spherical error keeps rolling.
As we mentioned, Arters is a preemption case. So where does the decision go next? Why, straight to that old saw, the purported “presumption against preemption”:
In considering whether these claims are preempted by the FDCA, some deference to state law is appropriate: “[I]n all preemption cases, and particularly in those in which Congress has legislated in a field which the States have traditionally occupied, we start with the assumption that the historic police powers of the States were not to be superseded by the Federal Act unless that was the clear and manifest purpose of Congress.” Levine, 555 U.S. at 565 (internal quotations omitted).
Arters, 2013 WL 308768, at *4.
Wrong again. Arters is a generic drug case – it’s governed by Mensing, not Levine. And it’s safe to say that the presumption against preemption was a major bone of contention in Mensing – and a 5-4 majority of the Court chose not to apply any such presumption. Four justices in Mensing went full bore the other direction: rejecting any such presumption and expressing the opposite view, that the Supremacy clause is something called a “non obstante” provision, and consequently, if anything, there should be a presumption in favor of preemption:
[T]he provision suggests that courts should not strain to find ways to reconcile federal law with seemingly conflicting state law. . . . A non obstante provision thus was a useful way for legislatures to specify that they did not want courts distorting the new law to accommodate the old.
Mensing, 131 S. Ct. at 2580 (emphasis added) (plurality opinion). Now, this piece of Mensing wasn’t a majority opinion (Justice Kennedy did not join this section), but only four justices wanted to apply a positive presumption against preemption of the sort invoked in Arters – and that’s a minority.
So, before we even get to the merits, Arters has directly contradicted three controlling Supreme Court opinions – taking an erroneous pro-plaintiff position as to each.
Why are we not surprised that the merits aren’t any better?
The plaintiff in Arters asserted a “design defect”/“remove from the market” claim à la Bartlett v. Mutual Pharmaceutical Co., 678 F.3d 30 (3d Cir.), cert. granted, 133 S.Ct. 694 (U.S. 2012). 2013 WL 308768, at *4. Of course, the vast majority of precedent is to the contrary. Another Ohio case, Fulgenzi v. PLIVA, Inc., 867 F. Supp.2d 966 (N.D. Ohio 2012), held, first, that design defect claims were improperly pleaded under the Ohio Product Liability Act (“OPLA”), id. at 972-73, and further that plaintiff couldn’t replead because there was no way to construct an unpreempted “design defect” claim:
Even if the pleading deficiencies could be remedied, dismissal of these claims would still be appropriate because, regardless of how Plaintiff attempts to cast these claims, they are, at the core, failure-to-warn claims that are clearly preempted. . . . [A] generic pharmaceutical product must be the same as the referenced listed drug in active ingredients, safety and efficacy and hence, as was the case with labeling, federal law pre-empts state laws imposing the duty to change a drug’s design upon generic manufacturers.
867 F. Supp.2d at 973-74 (citations and quotation marks omitted). Fulgenzi exemplifies decisions reached by many other courts. See Demahy v. Schwarz Pharma, Inc., 702 F.3d 177, 186 (5th Cir. 2012) (design defect claims are “failure-to-warn claims under different names” and preempted) (applying Louisiana law); Frazier v. Mylan Inc., ___ F. Supp.2d ___, 2012 WL 6641626, at *5-6 (N.D. Ga. Dec. 18, 2012) (“a generic drug is designed to be a copy of a reference listed drug and it must be identical in active ingredients and efficacy” rejecting Bartlett because generic “was not free to unilaterally pursue a safer alternative design . . . to comply with state law while also being in compliance with federal law”); Strayhorn v. Wyeth Pharmaceuticals, Inc., ___ F. Supp.2d ___, 2012 WL 3261377, at *10, 12 9-10 (W.D. Tenn. Aug. 8, 2012) (“design claim is “at [its] core” a warning claim; Bartlett “offered little explanation for accepting the failure to withdraw theory”); Lashley v. Pfizer, Inc., 877 F. Supp.2d 466, 478, 480 (S.D. Miss. 2012) (design defect claim without alternative design is really a preempted warning claim; Bartlett “unpersuasive and not a proper basis for relief”); Metz v. Wyeth, 872 F. Supp.2d 1335, 1341 (M.D. Fla. 2012) (generic drug “was required to be the bioequivalent to the reference listed drug”); Eckhardt v. Qualitest Pharmaceuticals, Inc., 858 F. Supp.2d 792, 801 (S.D. Tex. 2012) (“Generics were required to produce a drug that was equivalent to the brand-name drug and were not free to unilaterally pursue a safer alternative design in order to comply with state law”); In re Pamidronate Products Liability Litigation, 842 F. Supp. 2d 479, 484 (E.D.N.Y. 2012) (“the federal duty of ‘sameness’’’ also applies in the context of generic drug design, and federal law preempts state laws imposing a duty to change a drug’s design on generic drug manufacturers”); Jacobsen v. Wyeth, LLC, 2012 WL 3575293, at *9 (E.D. La. Aug. 20, 2012) (“federal law requires a generic drug to be of the same design as its brand name counterpart”); In re: Accutane Products Liability Litigation, 2012 WL 3194952, at *2 (M.D. Fla. Aug. 7, 2012) (“defective design claims are similarly preempted under Mensing because the generic drug was required to be the bioequivalent to the reference listed drug”); Aucoin v. Amneal Pharmaceuticals, LLC, 2012 WL 2990697, at *8-9 (E.D. La. July 20, 2012) (“federal law requires a generic drug to be of the same design as its brand name counterpart”); Fullington v. PLIVA, Inc., 2012 WL 1893749, at *5 (E.D. Ark. May 23, 2012) (design defect claim based on inherent risk really a preempted warning claim); Johnson v. Teva Pharmaceuticals USA, Inc., 2012 WL 1866839, at *4 (W.D. La. May 21, 2012) (design defect claims preempted because plaintiff “cannot show that an alternative drug design was available to the Generic Defendants”); Cooper v. Wyeth, Inc., 2012 WL 733846, at *6 (M.D. La. March 6, 2012) (“if state law could require a generic drug manufacturer to wholly withdraw from the market based on the unreasonable danger of the product . . . it necessarily must repudiate the label approved by the FDA”); In re Darvocet, Darvon & Propoxyphene Products Liability Litigation, 2012 WL 718618, at *3 (E.D. Ky. March 5, 2012) (failure to withdraw claim was “based on the allegedly defective design of the drug, which the Generic Defendants, bound by their ‘ongoing federal duty of sameness,’ were powerless to change”); Lyman v. Pfizer, Inc., 2012 WL 368675, at *4 (D. Vt. Feb. 3, 2012) (design defect claims “are preempted as well by Mensing’s logic,” since the “ ‘federal duty of sameness’ . . . applies to the design or composition of the drug as well as to its labeling”); In re Fosamax (Alendronate Sodium) Products Liability Litigation, 2011 WL 5903623, at *6 (D.N.J. Nov. 21, 2011) (“FDA requires generic Fosamax to have the same active ingredient as Fosamax”); Stevens v. PLIVA, Inc., 2011 WL 6224569, at *2 (Mag. W.D. La. Nov. 15, 2011) (“generic pharmaceutical product must be the same as the referenced listed drug . . . in active ingredients”), adopted, 2011 WL 6224556 (W.D. La. Dec. 2, 2011).
Arters’ response to this massive opposing precedent? First, they’re not binding, so they can be ignored. 2013 WL 308768, at *5 (“[n]one of the cases . . . bind this Court on the issue of whether design defect claims are preempted by the FDCA”). Second, if it’s not a failure to warn claim, it’s distinguishable. Id. (“other district courts within the Sixth Circuit have reached the opposite conclusion. . . . These Courts have considered different state laws and different factual situations. Here, plaintiffs' design-defect claim rests on the alleged violation of an entirely different duty than their failure to warn claim”).
Well, although the first is true, one would expect Arters at least to address the other courts’ reasoning – that bioequivalence means the “same” design as well as warnings – but no, not one word about how there isn’t “sameness” in design claims. As for the second, that’s just false. While some other Sixth Circuit district courts (Strayhorn (Tennessee) and Darvocet (Kentucky)), were under other states’ laws, Fulgenzi – which Arters notably fails to cite – simply isn’t. Fulgenzi analyzed the identical question under identical state law. Arters sweeps the most directly-on-point precedent silently under the rug.
But the scope of the error in Arters is even worse than ignoring uncomfortable precedent. Rather, Arters should never have been out on that limb, and thus ruling contrary to both Mensing and the great weight of precedent, in the first place. The decision puts the cart before the horse. There has to be a state law claim first – before it can be preempted. See Purvis v. Teva Pharmaceuticals, USA, Inc., ___ F. Supp.2d ___, 2012 WL 5364392, at *4 (M.D. La. Oct. 30, 2012) (design defect claim without alternative design not viable; Bartlett not persuasive); Grinage v. Mylan Pharmaceuticals, Inc., 840 F. Supp.2d 862, 870 (D. Md. 2011) (plaintiff failed to allege a proper design defect claim under state law; no need to reach preemption); Fisher v. Pelstring, 817 F. Supp.2d 791, 817-18 (D.S.C. 2011) (design defect failed when plaintiff failed to demonstrate an alternative feasible design). This is what “failure to state a claim” is all about.
Unlike Bartlett, where (non-drug) New Hampshire law could at least be plausibly read as allowing a “design defect” based on pure risk/utility balancing without an alternative design, Arters arose under Ohio law. 2013 WL 308768, at *5. Ohio simply does not recognize a design defect without a feasible alternative design:
A product is not defective in design or formulation if, at the time the product left the control of its manufacturer, a practical and technically feasible alternative design or formulation was not available that would have prevented the harm for which the claimant seeks to recover compensatory damages without substantially impairing the usefulness or intended purpose of the product.
Ohio Rev. Code §2307.75(F) (emphasis added). That goes double for prescription medical products:
An ethical drug or ethical medical device is not defective in design or formulation because some aspect of it is unavoidably unsafe, if the manufacturer of the ethical drug or ethical medical device provides adequate warning and instruction . . . concerning that unavoidably unsafe aspect.
Ohio Rev. Code §2307.75(D) (emphasis added). While Arters cites §2307.75 generally, 2013 WL 308768, at *5, it inexplicably ignores the two most directly on-point subsections that establish the unavailability of an OPLA design defect cause of action in the absence of any alternative design. Nor is there any conceivable other avenue open in Arters. OPLA is “intended to abrogate all common law product liability claims or causes of action.” Ohio Rev.Code §2307.71(B).
In short, the “design defect” claim purportedly at issue in Arters does not exist under relevant state law. A “design defect” claim absolutely, by statute, requires an alternative design in Ohio. Plaintiff cannot assert an Ohio claim that a product – particularly a prescription medical product – can simply be removed from the market. Because there is no Bartlett rationale under Ohio law, it was error for Arters even to consider preemption of that claim. “[W]e do not ordinarily decide constitutional issues if a case can be resolved on non-constitutional grounds.” Sistrunk v. City of Strongsville, 99 F.3d 194, 197 (6th Cir. 1996).
In reaching and resolving this constitutional issue, the District Court ignored the basic and longstanding rule of construction that when two questions are raised, and resolution of the non-constitutional question makes unnecessary a decision on the constitutional question, only the non-constitutional question should be decided.
Nagy v. Farmers Insurance Exchange, 758 F.2d 189, 192 (6th Cir. 1985).
And that’s not the end of it. Arters goes on to hold that something it calls “off-label promotion” also survives preemption. 2013 WL 308768, at *6. Arters states that there’s some sort of “duty not to promote a dangerous use of [the drug] that is inconsistent with the FDA-approved label.” Id. Where does this novel duty come from? Arter cites no law at all, stating only that the defendant allegedly “promoted a use of [the drug] that was in violation of Ohio law.” Id.
Off-label use isn’t illegal in Ohio. The only off-label use ever banned in Ohio has to do with (you guessed it) abortion. See Planned Parenthood Southwest Ohio Region v. DeWine, 696 F.3d 490, 496 (6th Cir. 2012) (discussing Ohio Rev. Code §2919.123(A)). Otherwise, off-label use is perfectly legal under Ohio common law.
[A]lthough the FDA regulates, among other things, drugs and medical devices, the FDA does not regulate the practice of medicine. . . . [C]ourts addressing issues of FDA-regulated drugs have determined that physicians are permitted to use federally approved drugs as they, in their medical judgment, see fit. . . . FDA approved indications were not intended to limit or interfere with the practice of medicine nor to preclude physicians from using their best judgment in the interest of the patient The Food, Drug and Cosmetic Act does not limit the manner in which a physician may use an approved drug. Once a product has been approved for marketing, a physician may prescribe it for uses or in treatment regimens or patient populations that are not included in approved labeling. Thus, the decision whether to use a drug for an off-label purpose is a matter of medical judgment, not of regulatory approval.
Klein v. Biscup, 673 N.E.2d 225, 231 (Ohio App.) (citations and quotation marks omitted), review denied, 667 N.E.2d 987 (Ohio 1996).
It is well established that the U.S. Food and Drug Administration neither regulates the practice of medication nor restricts uses of those of the “official” approval, but merely approves or disapproves the use and marketing of medical devices and drugs. The Ohio State Medical Association has taken the position that “off-label use” is neither “experimental” nor “investigational.” The off-label use of a medical device is merely a matter of medical judgment . . . is not barred by the U.S. Food and Drug Administration.
State ex rel. Bax Global, Inc. v. Industrial Com’n, 2007 WL 512529, at *5 (Ohio App. Feb. 20, 2007).
So if not off-label use itself, what could be a “violation” of the law? Off-label promotion? For one thing, no federal statute or regulation prohibits that, as the Second Circuit pointed out recently in United States v. Caronia, ___ F.3d ___, 2012 WL 5992141, at *3 (2d Cir. Dec. 3, 2012) (“[t]he FDCA and its accompanying regulations do not expressly prohibit the 'promotion' or 'marketing' of drugs for off-label use”). For another, assuming a plaintiff could nonetheless tease a “plausible” violation out of various FDA regulations: (1) off-label promotion is not inherently false or fraudulent, as we discussed at length here; and (2) there’s not a scrap of Ohio precedent holding that off-label promotion also somehow violates state tort law. It’s not the role of federal courts sitting in diversity to dream up new state-law tort claims.
[F]ederal courts must proceed with caution when making pronouncements about state law. Sitting in diversity, we are not commissioned to take a position regarding the advisability or fairness of the state rule to be applied, but must determine the issue as would the highest court of the state. This Court’s proper reluctance to speculate on any trends of state law applies with special force to a plaintiff in a diversity case. . . . When given a choice between an interpretation of state law which reasonably restricts liability, and one which greatly expands liability, we should choose the narrower and more reasonable path.
Combs v. International Insurance Co., 354 F.3d 568, 577 (6th Cir. 2004) (citations and quotation marks omitted); see Ventas, Inc. v. HCP, Inc., 647 F.3d 291, 328 n.15 (6th Cir. 2011 (quoting and following Combs); Kurczi v. Eli Lilly & Co., 113 F.3d 1426, 1429 (6th Cir. 1997) (“A federal court in a diversity case is not free to engraft onto ... state rules exceptions or modifications which may commend themselves to the federal court, but which have not commended themselves to the State in which the federal court sits”) (quoting Day & Zimmermann, Inc. v. Challoner, 423 U.S. 3, 4 (1975)).
Moreover, there’s another United States Supreme Court case holding, in the specific context of off-label use, that supposed state-law claims asserting FDA regulatory violations are in fact nothing more than improper attempts to sue over alleged violations of a statutory scheme that doesn’t permit private enforcement. See 21 U.S.C. §337(a) (reserving FDCA enforcement exclusively to the federal government). However far Buckman Co. v. Plaintiffs Legal Committee, 531 U.S. 341 (2001), might or might not otherwise extend, it forbids state-law claims such as those in Arters that allege improprieties involving off-label use:
[Plaintiff’s claims would ] delay health care professionals’ ability to prescribe appropriate off-label uses. . . . [I]t is clear that the [Lohr] claims arose from the manufacturer’s alleged failure to use reasonable care in the production of the product, not solely from the violation of FDCA requirements. In the present case, however, the [plaintiffs’] fraud claims exist solely by virtue of the FDCA disclosure requirements. . . . [W]ere plaintiffs to maintain their fraud-on-the-agency claims here, they would not be relying on traditional state tort law which had predated the federal enactments in questions. On the contrary, the existence of these federal enactments is a critical element in their case.
531 U.S. at 351-53 (numerous citations omitted). Without the FDA-approved label, there’s no such thing as “off-label” use – let alone “promotion” – which is another regulatory term. Thus, a tort claim for “off-label promotion” as discused in Arters necessarily runs afoul of Buckman’s “critical element” test.
If that weren’t enough, the Ohio Supreme Court has also banned this sort of disguised statutory enforcement claim where, as with the FDCA, the statute in question bars private enforcement. In Hernandez v. Martin Chevrolet, Inc., 649 N.E.2d 1215 (Ohio 1995), the court considered the effect of a claimed violation of OSHA, a federal statute that, like the FDCA, explicitly barred its use (the so-called “disclaimer”) in private civil litigation. The court, in a rare unanimous decision, said “no” to the claim:
This statutory disclaimer clearly indicates that Congress did not intend OSHA to affect the duties of employers owed to those injured during the course of their employment. . . . Negligence per se decreases the elements that a plaintiff must prove in a negligence action. Thus, a plaintiff's case is significantly aided if negligence per se is established. If we held that a violation of OSHA constitutes negligence per se, we would allow OSHA to affect the duties owed by individuals to those injured in the course of their employment. Such a holding would be contrary to the intent of the legislation. Accordingly, we hold that a violation of OSHA does not constitute negligence per se.
Id. at 1216-17 (citations omitted). As for "intent" in the FDCA context, as Buckman held, §337(a) is “clear evidence that Congress intended that the MDA be enforced exclusively by the Federal Government.” 531 U.S. at 352.
And finally, while it only makes the rubble of Arters bounce, as Caronia, supra, pointed out, there is nothing in the FDCA itself about off-label promotion (or, indeed, off-label use). Ohio law also prohibits violations of mere governmental regulations from being the basis of an negligence per se claim. See Chambers v. St. Mary’s School, 697 N.E.2d 198, 203 (Ohio 1998) (“we hold that the violation of an administrative rule does not constitute negligence per se”).
Spherical error. In one relatively short opinion (seven Westlaw pages), the Arters decision manages to:
(1) employ a “retired” standard for deciding a rule 12 motion, directly contrary to two of the most-widely cited Supreme Court decisions in recent years;
(2) apply a presumption against preemption directly contrary to the holding in a third controlling Supreme Court decision;
(3) hold that a design defect/failure to withdraw claim was not preempted when that claim cannot possibly exist under the controlling state (OPLA) statute;
(4) hold that a design defect/failure to withdraw claim was not preempted ignoring (and in some cases not even acknowledging) the reasoning of the vast weight of contrary precedent;
(5) hold that an off-label promotion claim is not preempted, directly contrary to the holding of a fourth controlling Supreme Court decision;
(6) entertain an off-label promotion claim that no Ohio court has permitted, contrary to the court’s limited role when sitting in diversity; and
(7) entertain an off-label promotion claim that’s barred by two Ohio Supreme Court decisions that limit the scope of negligence per se claim under controlling state law.