We confess, we can’t think of any good reason for admitting evidence concerning product risks that the plaintiff in a particular case never actually encountered – yet plaintiffs try it with a straight face all the time. It’s another example of plaintiffs throwing mud against the wall to see if it will stick; anything to divert attention from a weak merits case concerning the injuries actually being claimed.
Sometimes plaintiffs argue that, even though they never encountered a particular risk, it was so severe and downright scary that if they had only been warned about that risk, they wouldn’t have used the product. Trouble is, while that argument could be causal in a “but for” sense (assuming plaintiffs aren’t lying), it’s not causal in the legal, or proximate, sense. One of the major tort treatises explains the difference:
More centrally, the injury suffered must be within the class of injury that the warning requirement was meant to avoid. For example, the plaintiff, if properly warned that asbestos might cause cancer, might have ceased to work around asbestos. A failure to give such a warning could result in liability if the plaintiff did develop cancer as a result of asbestos exposure. But the failure to provide such a warning would not result in liability if the plaintiff, not being warned, kept her job and lost a hand in a job-related machine accident. In that example, failure to warn would be a cause in fact – the plaintiff would have been elsewhere, not working at the machine, if a proper warning had been given – but it is not a proximate legal cause. It is not, in other words, within the risk that a warning was designed to avoid.
Dan B. Dobbs, The Law of Torts, at 1018 (2001).
Federal appellate decisions similarly reject this sort of “causation” over and over, initially in cases involving non-prescription products. Most commonly, asbestos plaintiffs have been precluded from introducing evidence of cancer when they had no such injury, or even a significant risk. See O’Banion v. Owens Corning Fiberglass Corp., 968 F.2d 1011, 1013 (10th Cir. 1992) (“evidence of cancer is so prejudicial that in the absence of expert medical testimony that ‘a reasonable degree of medical certainty’ exists that the plaintiff will develop cancer, such evidence should be excluded”) (applying Oklahoma law); Smith v. A.C. & S, Inc., 843 F.2d 854, 859 (5th Cir. 1988) (cancer evidence in non-cancer case “is highly inflammatory and understandably incites the passions and fears of most reasonable individuals”) (applying Louisiana law); Jackson v. Johns-Manville Sales Corp., 750 F.2d 1314, 1321 (5th Cir. 1985) (same) (applying Mississippi law); In re Related Asbestos Cases, 543 F. Supp. 1152, 1160 (N.D. Cal. 1982) (excluding reference to cancer “in any cases in which the plaintiff has not contracted cancer”).
The same result has occurred in automotive product liability. For example, in Kane v. Ford Motor Co., 450 F.2d 315 (3d Cir. 1971), the plaintiff started out claiming one type of brake defect in a car, but then shifted to a different defect claim at trial. The Third Circuit affirmed exclusion of such evidence, because it was “irrelevant to appellant’s theory of the case.” Id. at 316. See Fields v. Volkswagen of America, Inc., 555 P.2d 48, 58 (Okla. 1976) (excluding recall where “the defect mentioned in the recall letter is not the defect claimed to have cause the [plaintiff’s injury]”); Johnson v. Ford Motor Co., 988 F.2d 573, 578-80 (5th Cir. 1993) (affirming exclusion of five other incidents that different car models and different claims of mechanical defect); Olson v. Ford Motor Co., 410 F. Supp. 2d 869, 873-75 (D.N.D. 2006) (excluding automotive recalls where the reason for recall differed from plaintiff’s alleged product defect); Jordan v. General Motors Corp., 624 F. Supp. 72, 77 (E.D. La. 1985) (same).
But we’re more interested in prescription medical product liability litigation.
In our sandbox, the New York Court of Appeals rejected failure to warn of risks the plaintiff had not suffered in Martin v. Hacker, 628 N.E.2d 1308 (N.Y. 1993), where the plaintiff claimed that a drug should have been contraindicated in patients with a history of depression – but personally had no history of depression. “Since it is undisputed that [plaintiff] had no history of mental depression, we are not concerned with the adequacy of the Contraindications section of the insert. Id. at 1313. See McFadden v. Haritatos, 448 N.Y.S.2d 79, 81 (N.Y.A.D. 1982) (“Since there was no showing” that plaintiff “had a history of that condition,” the allegedly inadequate warnings “are not applicable”).
Other state appellate courts agree. A Texas appellate court in Ethicon Endo-Surgery, Inc. v. Meyer, 249 S.W.3d 513 (Tex. App. 2007), entered judgment n.o.v. on a warning claim, where the prescriber knew of the risk that the plaintiff encountered − rejecting allegations of inadequate warnings about diabetes (“[n]othing in the record suggests that [plaintiff] was diabetic”), and several general surgical risks (“[a]ssuming that the general . . . risk factors . . . could form the basis of an appropriate warning, there is no evidence that the lack of such a warning caused [plaintiff’s] injury”). Id. at 519.
In King v. Danek Medical, Inc., 37 S.W.3d 429 (Tenn. App. 2000), one of Bexis’ Bone Screw cases, the plaintiffs’ expert attempted to opine about a condition that the plaintiffs did not have. Same result:
[Plaintiffs’ expert] did not attempt to testify that either [plaintiff] developed [that condition] as the result of their implants. . . . The theoretical possibility that some [products] of other manufacturers can cause various bone problems, without proof that it occurred to these plaintiffs because of the products of these defendants, cannot defeat summary judgment.
Id. at 446-47.
Likewise, in Peterson v. Parke Davis & Co., 705 P.2d 1001 (Colo. App. 1985), the court affirmed rejection of a proposed jury instruction that would have required the defendant to warn about every possible drug risk:
[T]he request for an instruction that [defendant] had a duty to warn of all known dangers was properly denied. In a failure to warn case, the plaintiff has the burden of proving that the manufacturer gave inadequate warning of the danger which caused the injury.
Id. at 1004.
Here in Pennsylvania, plaintiffs tried warning claims about the risk of injuries they never suffered in both the breast implant and fen-phen mass torts – failing both times. In fen-phen litigation, plaintiffs alleging primary pulmonary hypertension (“PPH”) were precluded from arguing that better warnings about valvular heart disease (“VHD”) (and vice versa) would have made a difference, even when a prescriber would not have been deterred by better warnings about the injury (PPH) the plaintiff actually had:
In these circumstances, the relationship between the legal wrong (the failure to disclose the risk of VHD) and the injury (PPH) is not directly correlative and is too remote for proximate causation. Therefore, as a matter of law, there is no proximate, causal connection between [defendant’s] failure to disclose the risk of VHD and [plaintiff’s] specific injury.
Cochran v. Wyeth, Inc., 3 A.3d 673, 681 (Pa. Super. 2010); accord Owens v. Wyeth, 2009 WL 3244890, at ??? (C.P. Phila. Co. Aug. 17, 2009) (“the only warnings properly at issue in a failure to warn case are those relating to the condition to which the plaintiff alleges to have suffered”), aff’d mem., 6 A.3d 572 (table), 2010 WL 2965014, at *5-6 (Pa. Super. 2010) (adopting trial court’s reasoning).
Likewise, in breast implant litigation, the Pennsylvania mass tort court held that liability for inadequate warnings about risks not encountered was too broad to be permitted:
If I accept plaintiffs’ argument, the law will be permitting recovery for a risk that the plaintiff assumed because the plaintiff might have made a different decision as a result of knowing of other risks for which the plaintiff did not experience any harm.
In re Silicone Breast Implant Litigation, 64 D. & C. 4th 21, 25-26 (C.P. Allegheny Co. 2003). Recovery for warnings about risks that never happened “would have a reach that extends far beyond the purposes for the [duty to warn] doctrine.” Id. at 26.
The same has proven true with mass torts in other state courts. In re NuvaRing Litigation, 2013 WL 1874321, at *23 (N.J. Super. Law Div. April 18, 2013) (“any failure to notify the FDA of increased [other] risks would not affect the adequacy of the warning with respect to the . . . cause of [plaintiff’s] death”); Hedrick v. Genentech Inc., 2011 WL 5902794, at ??? (Cal. Super. Oct. 20, 2011) (“liability should be limited when a plaintiff does not suffer the unwarned-of injury”) (applying Massachusetts law).
In federal court, the plaintiffs in Rivera v. Wyeth-Ayerst Laboratories, 283 F.3d 315, 321 (5th Cir. 2002) (applying Texas law), “never assert[ed] that they were part of a risk group that should have been warned.” Id. at 321. Their warning claims were therefore “absurd”:
To find causation, we would have to infer the absurd − for example, that an extra warning, though inapplicable to [the lead plaintiff], might have scared her and her doctor from [the drug]. Such reasoning is too speculative to establish Article III standing.
In Eck v. Parke, Davis & Co., 256 F.3d 1013 (10th Cir. 2001) (applying Oklahoma law), the plaintiff alleged that his prescriber “would have read and heeded a warning” about not to prescribe two drugs together. Turns out that didn’t happen. The other drug was prescribed later by somebody else. The warning to the first prescriber had nothing to do with plaintiff’s eventual injury. Therefore, no causation as a matter of law:
[I]t is undisputed that [plaintiff] did not ingest the [two drugs] concomitantly . . . until . . . his prescribing physician was [somebody else]. . . . That [the first prescriber] might have heeded a warning. . .about possible adverse effects were he to prescribe [the two drugs together] is of no significance given the facts before us.
Id. at 1020.
In Mills v. United States, 764 F.2d 373 (5th Cir. 1985) (applying Louisiana law), the adverse reaction that the plaintiff suffered was the subject of thorough warnings – so plaintiff claimed that warnings about other conditions were inadequate. The court held that, since, plaintiff didn’t have those injuries the alleged lack of warnings about them were irrelevant.
The question of the adequacy of the warnings must be confined to consideration of whether the warnings were sufficient to inform the plaintiff of the risk of the particular condition or disease which allegedly caused his injury or death. . . . [A] determination that warnings were inadequate with respect to some other condition does not bear on our conclusion that [plaintiff] was adequately informed of the risk of severe allergic reaction to the swine flu vaccine.
Id. at 379.
Likewise, in Novak v. United States, 865 F.2d 718 (6th Cir. 1989) (applying Ohio law), the plaintiff lacked any scientific evidence that the product could cause her condition, and therefore claimed inadequate warnings about a different risk she never suffered. Amazingly, the district court allowed the switch. The court of appeals did not, and reversed:
There was, therefore, no reason for the defendant to make the references deemed important or vital by the district court, and there was no duty on the part of the defendant to warn about any of these conditions. Particularly, the district court erred in finding the warning inadequate, negligent, and insufficient because it did not specifically caution those who may have experienced [the other condition]. It was not proven that [plaintiff] had actually suffered from [that condition]. . . .
Id. at 726. See In re Avandia Marketing, Sales Practices & Products Liability Litigation, 639 Fed. Appx. 874, 879 n.8 (3d Cir. 2016) (expert opinion had “no impact” on causation when based on a study that “dealt with a population of which [plaintiff] was not a part”) (applying Pennsylvania law); Coursen v. A.H. Robins Co., 764 F.2d 1329, 1336 (9th Cir. 1985) (“collateral misconduct unrelated to the specific injury suffered by the plaintiffs” properly excluded) (applying Oregon law).
Numerous other cases have reached the same conclusion. In Stahl v. Novartis Pharmaceuticals Corp., 2000 WL 33915848 (E.D. La. Nov. 29, 2000), aff’d, 283 F.3d. 254 (5th Cir. 2002), the plaintiff’s prescriber knew all about the drug’s alleged risks that he suffered. Id. at *4. Thus, plaintiff alleged inadequate warnings about other more serious conditions that he didn’t have. Summary judgment granted:
[T]his Court is unpersuaded by [plaintiff’s] complaints that the warnings did not warn of death, liver failure, and the need for transplant. [Plaintiff] has not suffered any of those injuries and is, therefore, precluded from imposing a duty to warn regarding those injuries not suffered. . . . [A] claimant cannot seek to impose a duty to warn on a manufacturer of damage not sustained.
Id. (citation and quotation marks omitted). See also Vakil v. Merck & Co., 2016 WL 7175638, at *6 (D.N.J. Dec. 7, 2016) (“Defendants cannot be held accountable for failing to warn Plaintiff of a symptom he never experienced”); Guidry v. Janssen Pharmaceuticals, Inc., 2016 WL 633673, at *4 (E.D. La. Feb. 17, 2016) (“The only specific . . . that supports the plaintiff’s failure-to-warn claim is that the FDA issued a safety announcement warning [about] ketoacidosis. But the plaintiff does not allege (at least coherently) that she ever suffered from ketoacidosis”; warning claim dismissed); Stephens v. Teva Pharmaceuticals, U.S.A., Inc., 2013 WL 12149265, at *3 (N.D. Ala. Oct. 31, 2013) (“warnings for a wide range of conditions from which decedent did not allegedly suffer . . . would not have had any effect on him”; motion to dismiss granted); Austin v. Bayer Pharmaceuticals Corp., 2013 WL 5406589, at *7 (S.D. Miss. Sept. 25, 2013) (“Plaintiff failed to state a claim for the failure to warn of side effects which Plaintiff did not suffer”); Harris v. Eli Lilly & Co., 2012 WL 6732725, at *3 (N.D. Ohio Dec. 28, 2012) (“For a plaintiff to succeed on an inadequate warning claim, the risk about which the manufacturer allegedly failed to warn must be the same risk which harmed the plaintiff.”) (following Ohio Rev. Code §2307.76(A)(1)); Tolliver v. Bristol-Myers Squibb Co., 2012 WL 3074538, at *4 (N.D. Ohio July 30, 2012) (same); Mason v. Smithkline Beecham Corp., 2010 WL 2697173, at *5 n.3 (C.D. Ill. July 7, 2010) (“a warning is only inadequate if it fails to list risks or side effects that do occur”); Harrington v. Biomet, Inc., 2008 WL 2329132, at *6 (W.D. Okla. June 3, 2008) (“no evidence” that two allegedly inadequate warnings caused plaintiff’s alleged injuries; summary judgment granted); Hernandez v. Ciba-Geigy Corp. USA, 200 F.R.D. 285, 295 (S.D. Tex. 2001) (dismissing plaintiffs who “do not plead that the[ir] children suffered any injuries or side effects of which [defendant] allegedly failed to warn”) (applying Texas law).
We’ve observed many times that multi-district litigation all too often becomes a breeding ground for bogus claims, and warning claims asserting injuries that plaintiffs don’t actually have is no exception. Plaintiffs in Pelvic Mesh litigation have repeatedly advanced similar bogus warning claims based on risks they had not encountered, and lost each time. The MDL judge granted an in limine motion against such evidence in Tyree v. Boston Scientific Corp., 2014 WL 5445769 (S.D.W. Va. Oct. 22, 2014):
I agree that evidence of complications that no plaintiff experienced is irrelevant and lacking in probative value. For the claims that require evidence of injury (strict liability for failure to warn, strict liability for design defect, and negligence), only the injuries experienced by the complainant are relevant. Strict liability . . . requires the plaintiff to show . . . that the defect was the probable cause of her injuries. . . . [E]vidence that the [the product] causes injuries not experienced by the plaintiffs has little probative value. Moreover, elaborating on injuries that the plaintiffs did not incur risks needless presentation of cumulative evidence.
Id. at *6 (citations and quotation marks omitted) (emphasis original). Accord Hall v. Boston Scientific Corp., 2015 WL 856786, at *5 (S.D.W. Va. Feb. 27, 2015) (“Evidence of complications that the plaintiff has not experienced is irrelevant and lacking in probative value. For the claims that require evidence of injury, only the injuries experienced by the complainant are relevant.”) (applying Wisconsin law); Eghnayem v. Boston Scientific Corp., 2014 WL 5465741, at *7 (W.D. Va. Oct. 28, 2014) (same) (applying Florida law); In re Ethicon Pelvic Repair Systems Products Liability Litigation, 2014 WL 505234, at *10 (S.D.W. Va. Feb. 5, 2014) (excluding purported cancer risk where plaintiff “does not have cancer or allege any injuries related to an increased risk of cancer”) (applying Texas law); In re Ethicon Pelvic Repair Systems Products Liability Litigation, 2014 WL 457544, at *2 (S.D.W. Va. Feb. 3, 2014) (“secondary infections were not ‘a fact in issue’ because [plaintiff] did not experience a secondary infection”) (applying Texas law).
Other MDL cases reaching the same result include:
In re Xarelto (Rivaroxaban) Products Liability Litigation, 2017 WL 1352860, at *4, 7 (E.D. La. April 13, 2017) (excluding expert testimony about cancer where “cancer was not an issue for either Plaintiff”) (defense testimony); Solomon v. Bristol-Myers Squibb Co., 916 F. Supp.2d 556, 564 (D.N.J. 2013) (“Plaintiff, however, did not suffer from transient ischemic stroke. Thus, this study is irrelevant to Plaintiff’s claim”) (applying Texas law); Carr-Davis v. Bristol Myers-Squibb Co., 2013 WL 322616, at *5 (D.N.J. Jan. 28, 2013) (“because the studies are not relevant to Decedent’s condition, then the failure to inform the physicians of such findings cannot establish causation”) (applying Missouri law); Cooper v. Bristol-Myers Squibb Co., 2013 WL 85291, at *6 n.11 (D.N.J. Jan. 7, 2013) (plaintiff “does not link those potential warnings to his personal circumstances”) (applying Alabama law); Thrope v. Davol, Inc., 2011 WL 470613, at *32 (D.R.I. Feb. 4, 2011) (“[n]othing in [the surgeon’s] account indicates that he handled the [product] in a manner that [defendant] failed to warn against”; judgment as a matter of law granted) (applying North Carolina law); In re Fosamax Products Liability Litigation, 2010 WL 1257299, at *5 (S.D.N.Y. March 26, 2010) (“Plaintiff cannot establish proximate cause without evidence [of a] failure to warn of the specific risk that allegedly materialized”) (applying Florida law); In re Rezulin Products Liability Litigation, 331 F. Supp.2d 196, 200-01 (S.D.N.Y. 2004) (alleged inadequate warning about risk plaintiff did not suffer held “quite beside the point”; “95 percent of the patients who took [the drug] suffered no liver injury” so “there is no evidence from which a trier of fact reasonably could conclude that these plaintiffs’ treating physicians would not have prescribed Rezulin even if plaintiffs are right as to the warning concerning liver toxicity”); In re Rezulin Products Liability Litigation, 2004 WL 1802960, at *3 (S.D.N.Y. Aug. 13, 2004) (“these plaintiffs do not claim to have suffered any liver injury” so “that there may have been a breach of the duty to warn of liver toxicity for which patients suffering from liver dysfunction may recover does not avail these plaintiffs”) (footnotes omitted), vacated in part on other grounds, 2004 WL 2009445 (S.D.N.Y. Sept. 8, 2004) (applying Mississippi and Texas law); Greiner v. Medical Engineering Corp., 99 F. Supp.2d 759 (W.D. La. 2000) (breast implants; “[a] claimant cannot seek to impose a duty to warn on a manufacturer of damage not sustained”), aff’d, 243 F.3d 200 (5th Cir. 2001); In re Norplant Contraceptive Products Liability Litigation, 1997 WL 81094, at *1 (E.D. Tex. Feb. 21, 1997) (“whether the [product] warnings were not adequate with respect to an injury not alleged is not relevant to whether physicians were adequately warned of Plaintiffs’ alleged injuries”).
Finally, one other argument that plaintiffs sometime advance about alleged failures to warn of risks that they didn’t actually encounter is that such non-causal failures to warn are relevant to “punitive damages.”
We don’t think so, and this time the argument is constitutionally based. Evidence of inadequate warnings about risks not encountered by plaintiffs is solely relevant to injuries suffered (if at all) by persons other than the plaintiff who is before the court. That’s a punitive damages no-no.
In considering punitive damages, the United States Supreme Court first held:
A defendant’s dissimilar acts, independent from the acts upon which liability was premised, may not serve as the basis for punitive damages. . . . Due process does not permit courts, in the calculation of punitive damages, to adjudicate the merits of other parties’ hypothetical claims against a defendant under the guise of the reprehensibility analysis.”
State Farm Mutual Automobile Insurance Co. v. Campbell, 538 U.S. 408, 422-23 (2003).
A few years later, the Court decided that, “[t]o permit punishment for injuring a nonparty victim would add a near standardless dimension to the punitive damages equation.” Philip Morris USA v. Williams, 549 U.S. 346, 354 (2007). Williams therefore flatly declared it unconstitutional to base punitive damages on allegations of harm to absent third parties (such as other people suffering different injuries):
We did not previously hold explicitly that a jury may not punish for the harm caused others. But we do so hold now. . . . [W]e believe the Due Process Clause prohibits a State’s inflicting punishment for harm caused strangers to the litigation.
Id. at 357. Thus, “it is particularly important that States avoid procedure that unnecessarily deprives juries of proper legal guidance. Id. at 355. It is “constitutionally important” to insure that “the jury will ask the right question, not the wrong one.” Id.
[W]here the risk of. . .misunderstanding is a significant one – because, for instance, of the sort of evidence that was introduced at trial or the kinds of argument the plaintiff made to the jury – a court, upon request, must protect against that risk. Although the States have some flexibility. . ., federal constitutional law obligates them to provide some form of protection in appropriate cases.
Id. at 357 (emphasis added).
Admission of warning/risk evidence beyond that involving the plaintiff in question constitutes a state-law “procedure” doing the exact opposite of what the Supreme Court required in Williams. With respect to punitive damages, such evidence is inherently suspect because by definition it places before the jury “injury . . . inflict[ed] upon those who are, essentially, strangers to the litigation.” Id. at 353.
In Skibniewski v. American Home Products Corp., 2004 WL 5628157 (W.D. Mo. April 1, 2004), after State Farm, but before Williams, the court excluded evidence of “any alleged side effect or risk of the products at tissue other than the side effect or risks that allegedly harmed plaintiff” both generally, and in response to a punitive damages argument. Id. at *12-13.
[F]ailure to warn of a medical condition plaintiff does not have cannot serve as the basis of a failure to warn claim and . . . evidence regarding injuries plaintiff does not have is irrelevant. Because plaintiff does not have [certain conditions], evidence of these should be excluded.
Id. at *7. See Ray v. Allergan, Inc., 863 F. Supp.2d 552, 565 (E.D. Va. 2012) (punitive damages argument based on all persons taking a drug for a condition, regardless of injury, required new trial under Williams); Schilf v. Eli Lilly & Co., 2010 WL 3909909, at *6 (D.S.D. Sept. 30, 2010) (evidence about a different drug “are irrelevant to Plaintiffs’ claim for punitive damages” under Williams); In re Fosamax Products Liability Litigation, 647 F. Supp.2d 265, 284-85 (S.D.N.Y. 2009) (conduct occurring after a plaintiff’s injury cannot be used to support punitive damages under Williams); In re Prempro Products Liability Litigation, 2007 WL 4189510, at *3 (Mag. E.D. Ark. Nov. 15, 2007), adopted, 2007 WL 4189497 (E.D. Ark. Nov. 21, 2007) (under Williams evidence “will not be permitted to testify regarding [defendant’s] general badness or badness in the specific areas which are not connected to [plaintiff’s] injury”). See also Branham v. Ford Motor Co., 701 S.E.2d 5, 24 (S.C. 2010) (punitive damages theory that “invited the jury to punish [defendant] for all [product-related] deaths and injuries, rather than the “harm to [plaintiff],” was improper under Williams); Pedroza v. Lomas Auto Mall, Inc., 2009 WL 1300944, at *4-5 (D.N.M. April 2, 2009) (conduct occurring after a plaintiff’s injury cannot be used to support punitive damages under Williams) (non-product liability case); Berardi v. Village of Sauget, 2008 WL 2782925, at *5-6 (S.D. Ill. July 17, 2008) (other allegedly similar incidents could not be admitted to prove punitive damages) (non-product liability case).
So there you have it. Evidence of alleged product defects, that did not give rise to the harm that the plaintiff actually encountered, should be irrelevant and inadmissible in product liability litigation, no matter what the rationale offered for their admissions. Such evidence is a smoke screen and a side show offered when a plaintiff is trying to distract attention from a lousy liability case.