One of the ways we feed this blog is doing what comes naturally to us lawyers – reading recent cases – and hoping that something inspires us. Sometimes that works. Sometimes that doesn’t. This week it worked too well. We’ve seen several decisions that bring back memories of stuff we’ve had to deal with over the years, and sometimes decades.
Taking the most recent decision first, the Second Circuit finally put an end to the long-running circus that’s been firearms litigation in New York state. This brings back memories because one of us was involved in the appeal the squelched what turned out to be the first ring of the circus. Back in 2001, after the Second Circuit punted the issue to the New York Court of Appeals, that court agreed with our analysis of both market share liability and negligence duty and held that you can’t ignore product identification and go suing an entire industry for what amounts to selling a non-defective, legal product that might be misused by those who buy it. See Hamilton v. Beretta U.S.A. Corp., 750 N.E.2d 1055 (N.Y. 2001).
This wasn’t really about guns for us – both of us live in large cities, after all, and we’ve seen the mayhem that those things cause. Rather, it was about theories of liability that would let plaintiffs sue entire industries indiscriminately when the product was neither illegal nor defective. It was also about judicial triumphalism, lawyers trying to get courts to step in and ban products that the other branches of government had determined should be legal.
Let’s see, Hamilton was a unanimous decision by the state’s highest court. That should pretty well kill the issue, right?
Good old Client #9 stepped up to the plate. He took the same negligent marketing allegations that the court had thrown out in Hamilton and gave them a new name – “public nuisance.” On behalf of the state, this time, he sued the same defendants over the same purported “facts” claiming that they should pay for the government’s costs of “abating” the nuisance and seeking judicial imposition of all kinds of sales restrictions on guns that the other branches of government had considered and rejected. Oh yes – and it still didn’t matter that the defendants were making and selling legal and non-defective products.
He lost too, in Spitzer v. Sturm, Ruger & Co., 761 N.Y.S.2d 192 (N.Y.A.D. 2003), the court held, among a lot of other things, that using public nuisance against makers of legal products would invite “a host of societal problems. . . into the courthouse”:
[A] common-law public nuisance cause of action. . .will, in our judgment, likely open the courthouse doors to a flood of limitless, similar theories of public nuisance, not only against these defendants, but also against a wide and varied array of other commercial and manufacturing enterprises and activities.
Id. at 196. The New York Court of Appeals, evidently thinking there was no reason to decide the same issue a second time under a new name, denied #9’s appeal. We could say a whole lot more about the problems with unleashing an amorphous tort theory like public nuisance upon the product liability world – but then again, we already have. See our prior posts, here and here critiquing various aspects of public nuisance litigation.
After the Spitzer case, gun litigation in New York now had two stakes through its heart. Surely that would be the end.
Back in the same federal district court that had been willing to let the original Hamilton case go forward, the City of New York picked up the fallen public nuisance standard, even though the New York state courts had told the state government (which usually brings nuisance actions) to, in effect, pound sand. The third ring of the circus was thus opened.
At this point Congress stepped in. It passed a statute called the “Protection of Lawful Commerce In Arms Act,” 15 U.S.C. §§7901–03, providing essentially that nobody could bring tort suits against gun manufacturers where the product wasn’t defective and caused injury only because of illegal use (this is a substantial oversimplification). Still the court refused to dismiss the litigation. That refusal to dismiss prompted an interlocutory appeal – and that produced the Second Circuit’s latest decision, City of New York v. Beretta USA, Inc., 524 F.3d 384 (2d Cir. April 30, 2008).
We like City of New York because the court rejects the same sort of arguments against the firearms statute that that we’ve seen raised against tort reform statutes generally – and also because of the court’s treatment of the public nuisance issue.
The first anti-tort reform argument is that “civil litigation. . .is not commercial in nature,” and thus not reachable by Congress’ Commerce Clause authority. Slip op. at 15. Anybody who’s seen lawyer advertising (who hasn’t, these days?) knows that this argument shouldn’t pass the red face test. The court rejected it more or less out of hand:
We find that Congress has not exceeded its authority in this case, where there can be no question of the interstate character of the industry in question and where Congress rationally perceived a substantial effect on the industry of the litigation that the Act seeks to curtail.
Slip op. at 18.
The second anti-tort reform argument is that passing a statute barring already filed lawsuits violated “separation of powers” because only courts have the power to order dismissal of actions before them. Id. at 19. The court rejected this argument, too. Congress can change the law generally; it just can’t tell courts what to do about specific cases. This statute prohibited all actions attacking the legal manufacture and sale of non-defective guns, with no limitation. That, too, was a legitimate congressional determination:
The [act] bars qualified civil liability actions, as defined in the statute. The definition. . .permissibly sets forth a new legal standard to be applied to all actions. . . . Because the [act] does not merely direct the outcome of cases, but changes the applicable law, it does not violate the doctrine of separation of powers.
Slip. op. at 19.
The third anti-tort reform argument is that the Tenth Amendment prohibits Congress from allowing state legislatures, as opposed to state courts, from recognizing certain specified exceptions to the litigation ban. Id. at 20. The act thereby “commandeer[ed]” the states, the plaintiff claimed. Id. at 21. This argument failed because Congress “impose[d] no affirmative duty of any kind on any” state. Id. We don’t claim to be Tenth Amendment scholars, but it does seem that giving a state the option of doing something is a lot different than demanding that a state do something.
The final anti-tort reform argument that the Second Circuit rejected is that a federal statute prohibiting the prosecution of certain types of tort causes of action violates potential plaintiffs’ “First Amendment right of access to the courts.” Id. at 22. Wow. If that one had succeeded, no tort reform effort would have been safe – even judicial abolition of causes of action would have been implicated. Fortunately, this unique view of the First Amendment didn’t go anywhere:
By its terms, the Act bars plaintiffs from courts for the adjudication of qualified civil liability actions, allowing access for only those actions that fall within the Act’s exceptions. We conclude that these restrictions do not violate plaintiffs’ right of access to the courts. . . . The right to petition exists in the presence of an underlying cause of action and is not violated by a statute that provides a complete defense to a cause of action or curtails a category of causes of action. . . . The [act] immunizes a specific type of defendant from a specific type of suit. It does not impede, let alone entirely foreclose, general use of the courts by would-be plaintiffs. . . . Congressional alteration of prior rights and remedies does not provoke right-of-access concerns because there is no fundamental right to particular state-law tort claims.
Slip op. at 23.
That’s music to our ears, folks. There is no constitutional right to sue.
There’s more, though, because now that the statute was constitutional, the plaintiffs still claimed that their claims escaped its reach. That argument was possible because Congress included an exception allowing suits where somebody intentionally sold a gun in violation of gun control statutes. The plaintiffs attempted to stuff their generalized public nuisance allegations through that loophole. After all, there was a public nuisance statute, wasn’t there?
New York’s general public nuisance statute, however, was neither expressly nor impliedly applicable to guns. The plaintiff’s contentions required the court to consider whether a general public nuisance statute was impliedly “applicable” to firearms – and by extension to the lawful marketing of other legal products. The court said no. “It is not disputed that [the public nuisance statute] is a statute of general applicability that has never been applied to firearms suppliers for conduct like that complained of.” Slip op. at 24-25. The court determined that such a general statute was neither something that “clearly can be said to regulate the firearms industry,” id. at 31, nor did it “actually regulate the firearm industry.” Id. at 33. Cf. Slip op. at 37 (dissent objecting to these conclusions).
A different construction of the general public nuisance statute would be “absurd” because it would allow litigation that interferes with precisely the conduct that Congress had, elsewhere in the same act, stated that it intended to protect. A general public nuisance statute was not “capable of being applied” to the legal marketing of guns:
In the broader context of the statute as a whole resort to the dictionary definition of “applicable” – i.e. capable of being applied – leads to a far too-broad reading of the predicate exception. Such a result would allow the predicate exception to swallow the statute, which was intended to shield the firearms industry from vicarious liability for harm caused by firearms that were lawfully distributed into primary markets.
Slip op. at 30. The court finished with an observation that very litigation in question had been specifically named during the congressional debate as an example of that the act would prohibit. Id. at 31-32 (quoting Sen. Wide-Stance (R-ID)).
That’s the first trip down memory lane. The second is the Kentucky Supreme Court’s recent decision in Hyman & Armstrong, P.S.C. v. Gunderson, 2008 WL 1849798 (Ky. April 24, 2008) (“Gunderson”). This decision reminds us of the old Bendectin litigation, which served as one of our introductions to the world of pharmaceutical mass torts. Among other things, the Supreme Court’s exposure to the questionable “theories” of plaintiff-side causation experts in Bendectin litigation caused the Court to declare war on “junk science.” Yes, Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U .S. 579 (1993), was a Bendectin case. It’s also an example of a phenomenon that we’ve discussed before of how plaintiffs’ insistence upon pursuing bad-fact cases can produce precedent of great benefit to future defendants.
It’s too bad that Daubert isn’t a constitutional principle, though, because Gunderson allowed liability based precisely on the sort of “a little of this and a little of that” expert causation testimony that Daubert and other Bendectin cases virtually laughed out of court. Gunderson involved the drug Parlodel. That drug’s been alleged to cause a variety of things – strokes, seizures, high (and low) blood pressure – but the evidence of causation has been so thin as to be routinely held inadmissible in federal court, where Daubert controls. See Rider v. Sandoz Pharmaceuticals Corp., 295 F.3d 1194, 1202 (11th Cir. 2002); Hollander v. Sandoz Pharmaceuticals Corp., 289 F.3d 1193, 1213 (10th Cir. 2002); Glastetter v. Novartis Pharmaceuticals Corp., 252 F.3d 986, 991-92 (8th Cir. 2001); Dunn v. Sandoz Pharmaceuticals Corp., 275 F. Supp.2d 672, 677-81 (M.D.N.C. 2003); Soldo v. Sandoz Pharmaceuticals Corp., 244 F. Supp.2d 434, 545-46 (W.D. Pa. 2003); Caraker v. Sandoz Pharmaceuticals Corp., 188 F. Supp.2d 1026, 1030-31 (S.D. Ill. 2001); Bendet v. Sandoz Pharmaceuticals Corp., 2001 WL 34134898, at *2 (E.D. Mo. July 27, 2001), rev’d on other grounds, 308 F.3d 907 (8th Cir. 2002) (summary judgment procedure); Brumbaugh v. Sandoz Pharmaceuticals Corp., 77 F. Supp.2d 1153, 1157 (D. Mont. 1999).
Why has this happened? Because just as in Bendectin, Parlodel has been studied epidemiologically in humans, and the results have been statistically insignificant:
The [clinical] study was commissioned by [defendant]. . .to allay concerns of the FDA that [the drug] was responsible for causing strokes and seizures in women. . . . The final report from the. . .study. . .concluded that women taking [the drug] had only a 1.61% greater risk of having a seizure than women not getting the drug, which is considered statistically insignificant.
Gunderson, 2008 WL 1849798, at *4 (emphasis added). Cf. Daubert, 509 U.S. at 582 (“[n]o study had found Bendectin to be a human teratogen”).
Gunderson, however, held that the same kind of potpourri of causation “evidence” (we put the term in quotes deliberately) that Daubert rejected was enough to permit causation under Kentucky law.
Most notably, the plaintiffs’ case in Gunderson included expert reliance upon adverse event experience reports collected by the FDA as “some evidence” of causation. 2008 WL 1849798, at *5. Well, this sort of anecdotal evidence is voluntarily reported to the FDA, and the FDA itself maintains that it’s useless as proof of causation, given how much publicity and other extraneous factors influence reporting. As a result, practically no other court in the country allows ADEs into evidence, unless notice is at issue. We’ve discussed the deficiencies of ADEs as causation evidence before, here. Unfortunately, we have to say that it looks like Parlodel plaintiffs have found their Old Kentucky Home.
The plaintiffs in Gunderson also presented other tried and true standbys of “a little of this and a little of that” causation.
#1 Their experts finagled the numbers of the published data until they claimed they had teased out a statistically significant (“2.86% (which is considered statistically significant) due to changing the data analyzed,” 2008 WL 1849798, at *5) increased risk. Cf. Daubert, 509 U.S. at 584 (plaintiffs’ “epidemiological analyses, based as they were on recalculations of data in previously published studies….”).
#2 Their experts relied on what even the court had to concede were “paradoxical” animal studies in the face of contrary human epidemiology.
Although Parlodel is known to lower blood pressure by dilating blood vessels, the [plaintiff’s] causation experts also relied on animal studies wherein Parlodel was shown to cause the paradoxical effect of vasoconstriction.
2008 WL 1849798, at *5. Bendectin experts did the same thing. See Daubert, 509 U.S. at 583-84 (holding that “animal-cell studies, live-animal studies, and chemical-structure analyses on which petitioners had relied could not raise by themselves a reasonably disputable jury issue regarding causation” in the face of contrary human epidemiology).
#3 Their experts also relied upon “evidence of the properties of the general class of drugs within which Parlodel is contained.” 2008 WL 1849798, at *5. Bendectin déjà vu again. See Daubert, 509 U.S. at 583 (noting the Bendectin plaintiff’s reliance upon “studies of the chemical structure of Bendectin that purported to show similarities between the structure of the drug and that of other substances known to cause birth defects”).
In short, the plaintiffs succeeded in Gunderson with the same causation dog and pony show that Daubert and numerous other Bendectin cases had held couldn’t come close to resembling an admissible causation case:
[T]he primary evidence at trial supporting the conclusion that Bendictin caused appellant’s birth defect . . . was so flawed [FN5] as to render his conclusions unreliable and therefore inadmissible under either Frye or Daubert.FN5. . . In effect, [plaintiff’s expert] worked backwards through the science, from the statistical results back to the original mere associations that led to the studies in the first place. This procedure cannot be fairly described as generally accepted methodology for purposes of the Frye standard. With respect to application of the Daubert formulation, numerous federal courts have determined that [plaintiff’s expert’s] testimony is not admissible.
Blum v. Merrell Dow Pharmaceuticals, Inc., 764 A.2d 1, 4 & n.5 (Pa. 2000) (citing eight other Bendectin cases reaching the same result).
Gunderson defends its result by stating that the numerous federal Parlodel decisions (including three courts of appeals) “incorrectly requir[ed] scientific certainty, which was not intended by Daubert.” 2008 WL 1849798, at *6. The court admitted that none of the bits and pieces of “evidence” that the plaintiffs brought forward was, by itself, enough to prove causation. Id. at *8 (“We agree with the trial court’s assessment that. . .the individual pieces of evidence the individual pieces of evidence may not conclusively prove general causation”). In other words, by itself, any one of these pieces was junk science. That didn’t matter, however, because the jury could consider all of it “together.”
[T]ogether they tend to show that Parlodel can cause postpartum seizures in women taking the drug. . . . “Although it is true that none of these bits of evidence establish conclusively that Parlodel can cause vasoconstriction and vasospasm, taken together they present a compelling picture, one which can support a reasonable scientific inference.”
We don’t think that this rationale holds water. It goes beyond “junk science.” It’s the judicial equivalent to throwing mud against the wall and seeing what sticks.
Look at it this way. Say you want to get from point A to point B. You go outside, and there in the front yard is an old car up on blocks. Now that may look like something that might get you where you want to go. It might even be something that would get you where you want to go if there was a little more to it. But a car on blocks isn’t going to get you anywhere. Now, to get somewhere, you might try to fix that old car on blocks, but bringing in a second old car on blocks isn’t going to get you any closer to where you want to go – nor is a third, or a fourth. All you’ll get is a junkyard full of old cars in your front yard, with none of them able to get you from point A to point B.
That’s what the bunch of pseudoscience in Gunderson – none of it, admittedly, sufficient to prove causation by itself – resembles: a bunch of old cars on blocks, none of which can get from point A to point B, that is, none of which can prove what must be proven to make out a case of scientific causation. It’s not just junk science, it’s junkyard science.
It’s also precisely what the Supreme Court rejected in Daubert back in the days of Bendectin litigation. The entire Daubert idea had its genesis in the Supreme Court’s decision that causation can’t be with “a bit of this and a bit of that” – ADEs, recalculated data, animal studies, chemically similar substances – in the face of contrary human epidemiological evidence. Gunderson may claim that it’s applying Daubert, 2008 WL 1849798, at *6, but we old Bendectin lawyers know better. That no more resembles Daubert than Dan Quayle resembled John F. Kennedy.
That’s the second stop on memory lane.
Our third stop is Swicegood v. Pliva Inc., 543 F. Supp.2d 1351, 2008 WL 916282 (N.D. Ga. April 2, 2008). This case deals with one of our (many) pet peeves – plaintiffs suing pioneer drug manufacturers on warning-based theories when all they took was generic drugs. It’s not as far down memory lane, since we only first encountered that issue in Colacicco, but, hey, it’s our blog.
Anyway, Swicegood recognizes that this type of allegation goes against the entire grain of product liability. That’s because product liability is supposed to involve the liability of the manufacturer of a product on the rationale that the manufacturer received economic benefit from selling the product and thus can be called upon to bear the burden of harm caused by product defects.
Pioneer drug manufacturers don’t receive squat in economic benefit from the manufacture of generic drugs. Quite the contrary, once a generic hits the scene, the pioneer drug is often finished as a significant source of profit for the company that created it. We’ve always thought that holding one company liable for the alleged defects in a competing product made by someone else was unfair and counterproductive (the actual manufacturer becomes a free rider with little incentive to make safer products) – but that unfairness is only magnified in the generic drug setting.
The Swicegood court keeps its eye on the product liability ball:
Georgia courts have held that unless the manufacturer’s defective product can be shown to be the proximate cause of the injuries there can be no recovery. Thus, the Plaintiff needed to establish that the product or products that allegedly caused the injury were, in fact, manufactured or supplied by the defendants in this case. . . . [We can] say with confidence that a manufacturer in Georgia can only be liable for strict products liability if it manufactured the allegedly harmful goods.
2008 WL 916282, at *2 (citation and quotation marks omitted). The plaintiff didn’t fare any better by changing the name of the claim to “negligence.” Negligent failure to warn of a product risk is still product liability and “a manufacturer may be liable under products liability when its product causes injury and was sold by th[at] manufacturer.” Id.
Creatively, the plaintiff also tried to gin up a misrepresentation claim based upon an FDA regulation, 21 C.F.R. §314.50(d)(5)(vi)(b) (impressive division into subparts, that), under which (plaintiff claimed) the defendant “voluntarily” became the “Referenced Listed Drug Holder” with respect to the generic product the plaintiff had actually took. 2008 WL 916282, at *3.
Voluntary? The alternative was to stop making the product at all.
The court found that this regulation created no duty to anybody taking an generic drug:
The Court agrees with the Defendants that the obligation is only voluntary in the sense that [they] chose to market their drug; once a manufacturer makes that decision, it is obligated to submit an application for approval to market the drug. The regulation does not require a name brand manufacturer to ensure that the generic brand’s label is accurate. Further, solely by providing the initial safety labeling, [defendants] did not assume the duty of labeling generic [products].
2008 WL 916282, at *3. Thus the “misrepresentation claims against [the] manufacturer properly collapse into the failure to warn claims.” Id. at *4.
Nor did case law arising from situations where the manufacturer’s identity was uncertain apply in the generic drug context, where “there is no dispute about who manufactured the harmful product.” Id. We think that’s exactly right. Why should a plaintiff taking a generic product get a litigation advantage – a shot at a second deep pocket – that a plaintiff taking the identical drug made by the pioneer manufacturer can’t get? It makes neither logical nor legal sense.
Name brand manufacturers undertake the expense of developing pioneer drugs. Generic manufacturers avoid these expenses by duplicating successful pioneer drugs and their labels. Further, the generic manufacturer benefits from exposure of the name brand drug, and consequently the generic manufacturer can ride on the coattails of its advertising. Additionally, perhaps most importantly, the name manufacturer has no control over the manufacturing process of its generic competitor. Taken together, these factors seem especially unfair to hold a name manufacturer liable for its generic competitors’ mistakes.
Id. at *6 (quoting in part Foster v. American Home Products Corp., 29 F.3d 165, 170 (4th Cir. 1994)).
The plaintiff even “contend[ed] that federal law, not just state law, creates a duty towards consumers of even generic prescription drugs.” Id. at *5. The court quite correctly responded, “the Plaintiff has failed to point to any specific tort liability created by the statutory scheme.” It’s quite well established by now that “there is no explicit private cause of action against manufacturers contained in the [FDCA], and no suggestion that the Act created an implied private right of action.” Medtronic, Inc. v. Lohr, 518 U.S. 470, 487 (1996).
We confess to being a little nervous after Clark v. Pfizer came down whether that case might be something more than an outlier. Reading Swicegood’s compelling logic convinces us that the majority rule that that pioneer manufacturers owe no informational duties to users of generic drugs is likely to remain that way. Just to make sure that everybody has all the relevant authority (it’s sometimes obscure, since the issue is rarely appealable – as, by definition, the generic manufacturer remains in the case), here’s everything we’re aware of: Foster, supra; Colacicco v. Apotex, Inc., 432 F. Supp.2d 514, 539-543 (E.D. Pa. 2006), aff’d on other grounds, 521 F.3d 253 (3d Cir. 2008); Leblanc v. Wyeth, Inc., 2006 WL 2883030, at *9 (W.D. La. Oct. 5, 2006); Goldych v. Eli Lilly & Co., 2006 WL 2038436, at *6 (N.D.N.Y. July 19, 2006); Tarver v. Wyeth, Inc., 2006 WL 1517546, at *2-3 (W.D. La. Jan. 26, 2006); Tarver v. Wyeth, Inc., 2005 WL 4052382, at *2-3 (W.D. La. Jun. 7, 2006); Block v. Wyeth, Inc., 2003 WL 203067, at *2-3 (N.D. Tex. Jan. 28, 2003); DaCosta v. Novartis AG, 2002 WL 31957424, at *9 (D. Or. March 1, 2002); Sharp v. Leichus, 2006 WL 515532, at *4-6 (Fla. Cir. Feb. 17, 2006), aff’d per curiam, 952 So.2d 555 (Fla. App. 2007); Kelly v. Wyeth, 2005 WL 4056740, at *2-5 (Mass. Super. May 6, 2005); Reynolds v. Anton, 2004 WL 5000272 (Ga. Super. Oct. 28, 2004); Sheeks v. American Home Products Corp., 2004 WL 4056060, at *1-2 (Colo. Dist. Oct. 15, 2004); Sloan v. Wyeth, Inc., 2004 Extra Lexis 202, at *11-12 (N.J. Super. Oct. 13, 2004); Beutella v. A.H. Robins, 2001 WL 35669202, at *2-3 (Utah Dist. Dec. 10, 2001).