McNair v. Johnson & Johnson

William M. Janssen[1]


Suppose there was a new smart-phone “app” that allowed for super high-precision satellite imaging (ideal for real estate, driving-direction, and phone-locator uses). Suppose that this new app was invented and sold by a major vendor (let’s call it “Grapes Inc.”) and that, during development, that inventor discovered – but did not disclose – that its app connected with satellites in a peculiar way that oddly multiplied the electromagnetic waves passing through the smart-phone. A year after this new app was launched, suppose a few small-time techies succeeded in reverse-engineering the app’s technology and began to sell “knock-off” versions of the app that were functionally identical duplicates of the original (an unsurprising development in our nefariously elusive cyberspace age). Suppose Grapes Inc. was unable (because of untraceability) or unwilling (because of the negligible lost-sales financial impact) to stop the “knock-off” versions of its app. Suppose that, a while later, consumers using “knock-off” versions of the app began to experience deadly cancers with unexpected frequency (a phenomenon also encountered by purchasers of the official Grapes Inc. app). Everyone brought lawsuits to seek compensation for their injuries. Customers who purchased the cheaper, “knock-off” version, however, were unable to find and sue the small-time techies who sold them their apps. So, they sued Grapes Inc. instead. They reasoned that, lying at the root of everyone’s injury, was a harmfully poor invention—coupled with that inventor’s failure to disclose what it knew about its design’s enhancement of electromagnetic waves. Because all user injuries were the same, whether inflicted by the official Grapes Inc. app or by some pirated “knock-off” version of that app, and because, in a way, the small-time techie app-pirates were sort of deceived too, the “knock-off” customers insist that Grapes Inc. ought to be held accountable for everyone’s harm (theirs included). Grapes Inc. now faces thousands of such claims from “knock-off” customers. Is it liable to them?

Such is the landscape of “innovator liability.” With the U.S. Supreme Court having now ruled that federal preemption forecloses failure-to-warn claims against generic drug makers (because those producers are not permitted to alter the warnings they supply),[2] customers claiming injury from generic drugs have turned to trying to sue the brand manufacturers for those injuries. Their argument echoes our allegorical Grapes Inc. “knock-off” customers’ reasoning – at the fountainhead of everyone’s injury is the same bad invention (or, at least, the same dangerous invention sold without a proper cautionary warning), and if the brand manufacturer is excused from liability for injuries caused by the “knock-off” versions of its invention, those injured by those “knock-off” versions would be left with no remedy at all. At base, it is liability-by-necessity.

In May 2018, the Supreme Court of Appeals of West Virginia rejected the theory of innovator liability in McNair v. Johnson & Johnson, denying a generic drug customer’s access to compensation from the brand version’s manufacturer.[3] The McNair opinion ranks as one of the top food and drug cases of 2018 for several reasons. First, it arrived late to the innovator liability party, which has been careening through the country since the mid-1990s. Thoughtful, sometimes clashing court decisions now abound. Consequently, the McNair court enjoyed the benefit of robust judicial percolation (by one well-regarded account, more than a hundred innovator liability rulings have been handed down since the debate began).[4] The court thus had the ability to consider and weigh numerous well-reasoned views on the topic before ruling. Second, the court’s decision was a sharply divided one – a 3-2 holding – with a vigorous, scolding dissent. Third, the coverage of the opinion was impressive: the majority comprehensively surveyed the field on the issue, methodically dismissing many of the leading arguments in favor of innovator liability, while the dissent lucidly invoked policy and simple justice in admonishing the lead opinion for its unfairness. All in all, it’s a noteworthy piece of analytical prose. Fourth, the divided opinion was handed down at a moment in the debate when innovator liability, once thought defeated, now seems a bit resurgent. Fifth, mindful of that possible ascendancy, the debate the McNair court explored takes on a special urgency: the implications of adopting innovator liability may be as profound and reverberating as the implications of rejecting it. The McNair court therefore arrives at a pivotal juncture in this debate, as the future trajectory of innovator liability hangs in uncertainty.


In 2012, Mrs. McNair was prescribed a course of the antibiotic Levaquin® to treat her pneumonia. Shortly after beginning her treatment on the drug, Mrs. McNair fell critically ill, ultimately was diagnosed with ARDS (acute respiratory distress syndrome), and later contended that she has been left to suffer a severe and permanent pulmonary impairment.

In 2014, she filed a West Virginia State court lawsuit against Levaquin®’s manufacturer (which the manufacturer subsequently removed to federal court), claiming that the company knew about, but failed to adequately disclose, ARDS risks from taking the drug. A motion for summary judgment followed. In the motion, the manufacturer offered evidence to establish that Mrs. McNair had actually not ingested its Levaquin® brand antibiotic, but rather generic levofloxacin tablets produced by Dr. Reddy’s Laboratories Ltd. Accordingly, because it was not the source of the product Mrs. McNair claimed to have injured her, the manufacturer sought judgment in its favor. Mrs. McNair opposed the motion, but not with evidence showing that she had taken Levaquin®. Instead, Mrs. McNair argued that even if she had, in fact, ingested the generic version of the drug, the Levaquin® manufacturer should remain liable because, abiding by the federal pharmaceutical laws, generic companies are dutybound to copy the brand drug’s design and warning label. Ergo, reasoned Mrs. McNair, the brand manufacturer that wrote the allegedly inadequate warning label was responsible for the generic company’s failure to impart to her and her doctor a proper ARDS warning.[5] The district court rejected this argument and granted summary judgment.

On appeal, the United States Court of Appeals for the Fourth Circuit confirmed that although West Virginia had not yet taken a position in the innovator liability debate, its case law had left “open the possibility that brand-name manufacturers may be liable for failure to warn when a plaintiff ingests the generic drug.”[6] Therefore, the Fourth Circuit certified the question of innovator liability to West Virginia’s Supreme Court of Appeals. The State court accepted the federal court’s certified question, and then answered it in the negative in the divided 3-2 McNair opinion.

The McNair Majority: No Innovator Liability in West Virginia

The majority began by recounting the familiar generic drug story – how the 1984 Hatch-Waxman Act opened the market for low-cost, post-patent replicas of pioneering companies’ branded drugs via the abbreviated new drug application process, whereby generic manufacturers could obtain approval to sell such replicas merely upon, first, a showing that those copies were chemically equivalent and bioequivalent to their branded predecessors and, second, a commitment to supply those generic drugs to consumers using the same labeling that accompanies the branded drugs.[7] This “ongoing federal duty of ‘sameness’” led the U.S. Supreme Court to two recent, decisive opinions impacting generic manufacturer products liability, both of which influenced the West Virginia court’s reasoning: because generic manufacturers are not at liberty to unilaterally change a generic drug’s labeling, failure-to-warn claims against those manufacturers are preempted by federal law[8] and, because generic manufacturers are not at liberty to change a generic drug’s design and composition, design defect claims anchored to the adequacy of a generic drug’s warnings are similarly preempted.[9]

Because the injured generic drug customer’s litigation options have, thus, become so narrowed, the majority turned at last to the certified question – could such a customer hold the brand manufacturer (the company whose products that customer did not use or ingest) liable for injuries caused by a “knock-off” company’s generic copy of the brand drug?

The majority first dismissed that possibility in strict liability claims. The opinion traced the jurisprudential underpinnings of strict product liability theory. Strict products liability, classically understood, rests on the notion that, by offering a product for sale, the seller of that product “impliedly represents” to consumers that its product is “reasonably suitable, safe and fit” for its intended use.[10] Liability in such circumstances is justified philosophically by choosing that the financial consequences of product injuries be borne by those who “put such products on the market rather than by the injured persons who are powerless to protect themselves,” and by the assumption that sellers can spread the compensation cost burden among the cohort of product purchasers by adjusting the product’s cost accordingly.[11] From these core, baseline strict liability principles, the majority drew this definitive conclusion: “Our products liability law is abundantly clear: liability is premised upon the defendant being the manufacturer or seller of the product in question.”[12]

The majority next dismissed innovator liability garbed in a negligent misrepresentation theory. “No action in negligence will lie” in West Virginia, quoted the majority, “without a duty broken.”[13] Here, Mrs. McNair made a vigorous stand. Every brand drug manufacturer, she reasoned, knows that when its patent expires, generics can enter that market and, when they do so, they must exactly copy the brand drug’s warning label. Because generic entry into the market and the generic copying of the brand label are both foreseeable, Mrs. McNair reasoned that it is then equally foreseeable that defects in the labeling will cause injury to brand and generic customers alike. Since “duty” follows from “foreseeability,” Mrs. McNair insisted that brand manufacturers thus owe a derivative obligation in tort to all post-patent generic consumers. The majority agreed that foreseeability of risk is an “important consideration” in setting the negligence duty obligation, but held that other, “broader policy considerations also enter the equation.”[14] Embracing reasoning from an earlier federal court of appeals, the majority concluded that a generic drug customer’s injuries “are not the foreseeable result of the brand manufacturers’ conduct, but of the laws over which the brand manufacturers have no control.”[15] Consequently, the majority held (this time citing reasoning from the Iowa Supreme Court) that: “We are unwilling to make brand manufacturers the de facto insurers for competing generic manufacturers. (Deep product jurisprudence is law without principle.)”[16]

The opinion then recounted the heavy anti-innovator-liability majority West Virginia would be joining, noting that all the federal courts of appeals that had considered the issue had ruled against innovator liability, as had two of four State high courts and one State legislature.[17] The opinion also demonstrated how rejecting innovator liability fit snugly with West Virginia’s “clear public policy,” citing recent legislative enactments from 2016 and 2017 which, the majority explained, confirmed the State’s commitment to allowing product liability only against the manufacturer or seller of the injury-causing product.[18] In the end, rejecting innovator liability aligned best with West Virginia’s products liability philosophy. “Finding the brand manufacturer liable for the ingestion of a generic drug ‘would sever the connection between risk and reward’ . . . that forms the basis of products liability law.”[19]

Finally, the majority weighed and rejected two creative final contentions offered by Mrs. McNair in support of innovator liability. First, Mrs. McNair proposed that her claimed defect, properly understood, did not lie with the drug she ingested (a drug made by someone other than the defendant brand manufacturer), but rather with the label that accompanied the drug she ingested (text which the defendant brand manufacturer wrote and owed a federal duty to write properly). The majority was unpersuaded: the drug’s label is not the “product,” reasoned the court, but merely “a way in which a plaintiff can claim that the product is defective.”[20] Second, Mrs. McNair argued that, in West Virginia, “[f]or every wrong there is supposed to be a remedy somewhere.”[21] Such a remedy, the majority decided, “must rest with Congress or the FDA.”[22] Manufacturers of branded prescription drugs do not choose the content of their competitors’ warning labels; rather, that is the consequence of complicated federal lawmaking that balanced the need to incentivize “pioneering research and development of new drugs” while at the same time “enabling competitors to bring low-cost, generic copies of those drugs to market.”[23] Here, the majority concluded, treading gingerly was the wisest course: “We refuse to interfere in the delicate calculus of Congress in crafting the Hatch-Waxman Act, which is consistent with our traditional reluctance to impose liability on a party in a heavily regulated industry.”[24]

The McNair Dissent: Rejecting Innovator Liability is “Wrong,” “Imprudent,” “Appalling,” “Injurious,” “Short-Sighted,” and “Ill-Advised”

Harsh words indeed. To the two justices in dissent, the issue of innovator liability posed a “deceptively simple question” capable of being answered easily using settled principles of West Virginia tort law.[25] Foreseeability guides the question of tort liability in the State, bounded by considering the likelihood of injury, the magnitude of the burden to guard against it, and the consequences of burdening the defendant with that duty. In the dissent’s view, the possibility of injury was “undeniably foreseeable,” since misrepresentations on a generic medicine’s labeling are “obviously extremely likely to injure consumers.” The burden of guarding against deficient warning labels ought to be minimal, given that the brand manufacturer already has the obligation under federal law to seasonably update its label with new serious hazard information. The consequences of imposing that burden on the manufacturer counsel in favor of doing so: such a burden ought to further incentivize the scrupulous updating of those labels to the benefit of all medicine consumers.[26]

The dissent’s conclusion rested ultimately on one “crucial question,” namely: “what entity is responsible for the omission of a warning on the drug’s label?”[27] That question, reasoned the dissent, is readily answered: “It was not the generic manufacturer, who was simply a conduit for the transmission of warning label information. The content of the label was exclusively within the control of the brand-name manufacturer. The right and obligation to include an appropriate warning label must be placed squarely at the feet of the entity responsible for the label’s content.”[28] The complex federal regulatory scheme for pharmaceuticals, offered the dissent, certainly poses a “quandary” and the majority’s answer to it was, condemned the dissent, “appalling and injurious to the public good.”[29] The brand manufacturer’s role in Mrs. McNair’s injury was not “merely peripheral;” to the contrary, “[i]t drafted and maintained control over the contents of the label providing guidance to the plaintiff.”[30] The primacy and exclusivity of the brand manufacturer’s labeling authority justified liability, and failing to impose it, feared the dissent, “will necessarily invite inattention and neglectfulness by brand-name manufacturers, a transgression for which our citizens will have no legal recourse.”[31]


Few would quibble with the U.S. Supreme Court’s judgment that preempting the product-injury claims of generic drug users while, at the same time, allowing similar product-injury claims of branded drug users “makes little sense” from the consumer’s perspective and deals generic users an “unfortunate hand.”[32]

The Food and Drug Administration reports that 90% of all prescription scripts written in the United States today are filled by generic drugs.[33] And that consequence is not always (or even often) the product of consumer choice. Indeed, many jurisdictions, like West Virginia, impose an affirmative obligation on pharmacists to fill brand-name drug prescriptions with a generic unless such a substitution is adjudged not suitable or ruled-out expressly by the prescriber.[34] Mindful of the ongoing innovator liability debate, how a patient’s prescription script is filled by his or her local pharmacist may well determine whether product liability can lie or not. It really is the very worst sort of Russian roulette. Until you consider the brand manufacturers’ predicament: they account for just 10% of all prescription medical sales (the remaining 90% are revenues earned mostly by competitors), but are now threatened with funding 100% of the compensation for all prescription drug injuries.

So, what is the solution?

Most courts that have considered the question of innovator liability have rejected it.[35] But, just the same, new adopters still appear. The Alabama Supreme Court became the first high court to embrace the theory in 2014, although its adoption was short-lived; the Alabama legislature abrogated its court’s ruling just a year later.[36] California’s Supreme Court adopted the theory in late 2017, and even broadened its reach to impose liability on a brand manufacturer long after that company had exited the business.[37] Massachusetts’ high court accepted the theory in early 2018, albeit in a constrained form (permitting liability only when the brand manufacturer’s labeling failure was found to be “reckless”).[38]

Lower courts have occasionally (and infrequently) allied with the theory, too. A Vermont federal district judge predicted in 2010 that his State would impose innovator liability when the brand manufacturer’s label inadequately informed physicians who later made imprudent prescriptions that were filled, ultimately, with the generic version of the medicine.[39] A Mississippi federal district judge predicted in 2013 that his State would impose innovator liability, but only if the brand manufacturer’s labeling was false (rather than merely incomplete).[40] And, finally, an Illinois federal district judge predicted in 2017 that his State would follow innovator liability, though that ruling was overturned by the Seventh Circuit Court of Appeals a summer later.[41]

Whether a trend line can be plotted through the myriad of cases rejecting innovator liability and those handful of cases embracing it is, as yet, difficult to see. In those cases where the theory found a welcoming reception, the justification either expressly or indirectly seemed saddled to necessity. The recent Massachusetts Supreme Judicial Court’s opinion is an able illustration: “Were we also to bar [generic consumers’] claims against brand-name manufacturers, we would only exacerbate the unfairness of this regulatory scheme.”[42]

The judiciary often delivers only rough-hewn justice in situations like these. A more nuanced, measured solution within the confines of this byzantine regulatory structure requires the thoughtful hand of legislators or regulators. Alas, the opportunity for creative solutions even there seems disappointingly thin:


  • Congress could amend the federal pharmaceutical laws to authorize generic manufacturers to alter their medicines’ labels when experience and science require it, with State tort regimes supplying the pathways to compensation when needed alterations fail to be made (but this change would impose on the generic manufacturers the costs of performing expensive validating science that, no doubt, would spike the costs of generic medicines, thus undermining one of the key objectives of the Hatch-Waxman Act (lower drug costs));
  • The federal pharmaceutical laws could remain as they are, but courts could exempt from preemption State law failure-to-warn claims based on a generic manufacturer’s failure to petition FDA for a labeling change when experience and science require it (but here, too, the specter of that liability would likely cause the generic manufacturers to launch into the expensive science that, like the first solution, would invariably cause precipitously rising generic drug prices);
  • Congress could create a generics claims system, with capped awards funded by imposing a sur-charge on each generic prescription filled (with recovery caps cabining the increase in generic drug prices, but also winnowing the fulsomeness of each claimant’s recovery);
  • Congress could enact a federal prohibition that forbids States from maintaining involuntary generic-substitution policies and, also, that forbids insurers from mandating generic-substitutions, thus allowing consumers to decide individually whether they prefer lower-priced generic drugs (with a concomitant loss of an ability to pursue bad-drug product liability) or higher-priced branded drugs (and enjoy access to product liability), but this solution would likely raise the costs of health insurance and would also most likely redound harshly and unfairly on those with limited financial means who may be forced into the lower-priced option by their economic circumstances;
  • Congress could formally legislate innovator liability, statutorily foisting on brand manufacturers the obligation to fund compensation for those injured by generic drugs (but such obligatory nationwide exposure would likely encourage brand manufacturers to dramatically increase their brand pricing to fund such sprawling liability, cause them to be more reluctant in pioneering new drugs, and even prompt their withdrawing of the branded NDAs as patent expiry dates approach – all of which would disserve the interests of national healthcare);
  • Congress could formally legislate against innovator liability, statutorily foreclosing brand manufacturer liability for generic drug user injuries, reasoning that such is the cost and consequence of the balance the Hatch-Waxman Act achieves (post-patent drug costs fall remarkably but bad-drug product liability is extinguished).


Perhaps there are still other solutions beyond those few cataloged here. Perhaps the disquieting uncertainty over how the various States are choosing to resolve the question of innovator liability (and its spectra of nearly bottomless product liability) will inspire a compromising of positions among all constituencies in this debate that could lead to a breakthrough idea that achieves a better balance all around. As the Supreme Court, and the West Virginia high court, both remarked, Congress and FDA are always free to design a new solution.

Until then, the judiciaries across the nation will be left to fill the lawmaking gap with rough justice. How responsibly they are in doing so will continue to depend on whether you are the one having a prescription script filled or the one supplying the pharmacy shelves with medicines.



[1]    William M. Janssen is a professor of law at the Charleston School of Law in Charleston, South Carolina, where he teaches products liability, mass torts, civil procedure, and constitutional law.

[2]    See PLIVA, Inc. v. Mensing, 564 U.S. 604, 612-21 (2011).

[3]    241 W. Va. 26, 818 S.E.2d 852 (2018).

[4]    See Drug and Device Blog, Innovator Liability Scorecard, (last visited Mar. 13, 2019); Drug and Device Blog, 50-State Survey of Innovator Liability, (last visited Mar. 13, 2019).

[5]    McNair v. Johnson & Johnson, 2015 WL 3935787, at *1-*3 (S.D. W. Va. June 26, 2015), appeal and certified question issued, 694 F. Appx. 115 (4th Cir. 2017), certified question answered, 818 S.E.2d 852 (W. Va. 2018).

[6]    McNair v. Johnson & Johnson, 694 F. Appx. 115 (4th Cir. 2017), certified question answered, 818 S.E.2d 852 (W. Va. 2018).

[7]    McNair, 818 S.E.2d at 856-58 (describing the Drug Price Competition and Patent Term Restoration Act (Hatch-Waxman Act), 98 Stat. 1585, codified at 21 U.S.C. § 355(j)(2)(A)).

[8]    See PLIVA, Inc. v. Mensing, 564 U.S. 604 (2011).

[9]    See Mutual Pharm. Co. v. Bartlett, 570 U.S. 472 (2013). In Bartlett, the Court acknowledged that chemical redesigns of generic drugs are foreclosed by Congress’s generic drug approval regime. Id. at 483-84. This, in turn, obviously seems to foreclose all design defect theories that rest on a showing by claimants of a proposed alternative drug design posited to be superior to the incumbent. See, e.g., Branham v. Ford Motor Co., 701 S.E.2d 5, 16 (S.C. 2010) (“in a product liability design defect action, the plaintiff must present evidence of a reasonable alternative design. The plaintiff will be required to point to a design flaw in the product and show how his alternative design would have prevented the product from being unreasonably dangerous.”) (tracking the recommendations set out in Restatement (Third) of Torts – Products Liability § 2(b) (1998)). Nonetheless, the Bartlett Court left open the question of whether other State law design defect theories, those not anchored on a proffer of a generic drug’s redesign, would escape federal preemption. See Bartlett, 570 U.S. at 487 n.4 (electing not to resolve the preemption effect on State law design defect theories that parallel the federal misbranding statute’s obligation to pull from the market a federally approved drug when information about that drug demonstrates that it is “dangerous to health”).

[10]  McNair, 818 S.E.2d at 859 (citations omitted).

[11]  Id. (citation omitted).

[12]  Id. at 860.

[13]  Id. at 861 (citation omitted).

[14]  Id. at 862 (citations omitted).

[15]  Id. (quoting In re Darvocet, Darvon, and Propoxyphene Prods. Liab. Litig., 756 F.3d 917, 944 (6th Cir. 2014)).

[16]  Id. at 862-63 (quoting Huck v. Wyeth, Inc., 850 N.W.2d 353, 380-81 (Iowa 2014)).

[17]  Id. at 863-64.

[18] 865-66 (citing W. Va. Code §§ 55-7-30 & 55-7-31).

[19]  Id. at 866.

[20]  Id. at 861 (citations omitted).

[21]  Id. at 866 (quoting Sanders v. Meredith, 89 S.E. 733, 736 (1916) (Lynch, J., dissenting)).

[22]  Id.

[23]  Id.

[24]  Id. at 866-67.

[25]  Id. at 867-88 (Workman, C.J., dissenting).

[26]  Id. at 868-69.

[27]  Id. at 870.

[28]  Id.

[29]  Id. at 872.

[30]  Id. (emphasis in original).

[31]  Id.

[32]  PLIVA, Inc. v. Mensing, 564 U.S. 604, 625 (2011).

[33]  See U.S. Food & Drug Admin., Generic Drugs, (last visited Mar. 13, 2019).

[34]  See W. Va. Code § 30-5-12b(b).

[35]  See supra n. 4 and accompanying text.

[36]  See Wyeth, Inc. v. Weeks, 159 So. 3d 649 (Ala. 2014), abrogated, Ala. Code § 6-5-530(a).

[37]  See T.H. v. Novartis Pharms. Corp., 407 P.3d 18 (Cal. 2017). See our summary in last year’s volume, Top Food and Drug Cases, 2017, & Cases to Watch, 2018 (FDLI 2018).

[38]  See Rafferty v. Merck & Co., 92 N.E.3d 1205 (Mass. 2018).

[39]  See Kellogg v. Wyeth, 762 F. Supp. 2d 694 (D. Vt. 2010).

[40]  See Chatman v. Pfizer, Inc., 960 F. Supp. 2d 641 (S.D. Miss. 2013).

[41]  See Dolin v. GlaxoSmithKline LLC, 269 F. Supp. 3d 851 (N.D. Ill. 2017), rev’d, 901 F.3d 803 (7th Cir. 2018), petition for cert. filed, No. 18-803 (U.S. Dec. 21, 2018).

[42]  Rafferty, 92 N.E.3d at 1218.