Notice & Comment

Hatch-Waxman and Defective Generic Drug Labels: The Disjunction Between Federal Drug Safety Law and State Warning Defect Law (Part II)

In an earlier post I outlined a regulatory anomaly regarding liability for defective drug labels.  Hatch-Waxman expedited entry of generics medications into the market, by relieving generic-brand manufacturers of the need to conduct further testing.  Concomitantly, the statute placed control of the contents of the drug label in the hands of the original recipient of new drug marketing approval, generally the brand-name drug manufacturer.  The U.S. Supreme Court held in PLIVA, Inc. v. Mensing, 564 U.S. 604 (2011), that given generic-drug manufacturers’ lack of any power to change the label accompanying their drugs, federal law preempted state law products liability suits against them for any defect in the drug’s warnings. Brand-name users could sue the brand-name manufacturer on warning defect claims, because brand name manufacturers could change the label provisionally before securing FDA approval, see Wyeth v. Levine, 555 U.S. 555 (2009).  Patients who took generics have sought to sue the brand-name manufacturers for the inadequate warnings, due to their control over the content of warnings accompanying generic drugs. These efforts have generally been rebuffed.

In this post I will explore several options for addressing this anomaly.

The Options

There are at least five options for addressing the cost of injuries suffered by generic drug users due to defective labels: (1) leave injured consumers to bear their own catastrophic loses, (2) impose liability on brand-name manufacturers, (3) impose liability on generic-brand manufacturers, (4) create a federally-funded compensation program, (5) establish a compensation scheme jointly funded by brand-name and generic-brand manufacturers (and perhaps other market participants as well).

I omit the option of imposing liability on pharmacists.  Most jurisdictions have refused to apply strict products liability to pharmacists.  Restatement (Third) Products Liability §6. Comm. h; see, e.g., Murphy v. E.R. Squibb & Sons, 40 Cal. 3d 672, 676-77, 710 P.2d 247 (1985).  Courts have expressed a concern that such potential liability would limit the availability of pharmaceuticals.  Such potential liability might cause pharmacists to dispense only a limited range of medications, either medications whose adverse consequences are well known or ones made by only the most solvent drug makers. Murphy, 40 Cal. 3d at 680-81; Restatement (Third) Products Liability §6. Comm. h.  But in assessing the fifth option listed above, I will consider the possibility of including pharmacists as one funding source for a mandatory compensation scheme.

(1)  Leaving the Loss on Injured Consumers: Barring Liability Suits for Defective Warnings.  As of now, most states do not recognize a cause of action against the brand-name manufacturer for defects in generic drug labels, thus requiring consumers harmed by generic drugs to bear their own losses.  As noted in my first post, such a result contravenes the basic underpinnings of strict products liability.  See, Escola v. Coca-Cola, 24 Cal. 2d. 453, 461-62, 150 P.2d 436 (1944) (Traynor, J. concurring); David G. Owen, Products Liability Law §5.4 at 280-82 (2005).  In general, strict products liability seeks to place the burden of product injuries on the actor that is the “least cost avoider,” i.e., the party best able to reduce the level of harm, and the optimal risk spreader, the party best able to spread the cost of injuries among a pool of consumers.  Generally, consumers who are not at fault, by altering or misusing a product, are considered the least appropriate parties to bear liability, due to their limited ability to reduce the level of injuries or spread catastrophic losses more broadly.  See Restatement (Third) Products Liability §2. Comm. a (“individual users and consumers [should] bear appropriate responsibility for proper product use”)(emphasis added).

The federal government may wish to subsidize generic-brand drug makers to make prescription drugs more affordable.  But if so, the cost of the subsidy should be spread across the general public, not borne by an unfortunate few who suffer catastrophic losses.  And while such catastrophic losses are not entirely random, most consumers who take a particular prescription drug, or prescription drugs in general, are at risk from some unidentified adverse effect of a medications they are taking.  This makes spreading the cost among the general public more appropriate.

Ironically, in states that leave the loss on the consumer, some of the cost of the adverse drug reactions may be spread among the public through health insurance, either Medicare, Medicaid, employer-based private insurance or non-employer-based private insurance.  In 2016, nationally, 9% of the public was uninsured.  Of the total population, 49% were insured through their employers, 19% through Medicaid, 14% through Medicare, and 7% through non-group plans. Henry J. Kaiser Family Foundation, Health Insurance Coverage of the Entire Population, accessible at, https://www.kff.org/other/state-indicator/total-population/?currentTimeframe=0&sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D. But spreading the cost of medications’ adverse effects by means of medical insurance results in the cost being spread haphazardly, varying depending on the nature of the injured consumer’s insurance and the extent of its coverage, and states’ decision regarding whether to “opt out” of the Affordable Care Act’s Medicaid expansion.  And, of course, in the United States there are significant concerns about the affordability of and the extent of coverage offered by medical insurance plans. Given the strains on the system of medical insurance and the problems of affordability, the burdens on the health insurance system should be lowered to the extent possible.  That system should not be used to relieve the pharmaceutical industry of its obligation to compensate consumers injured as a result of inadequate drug labelling.

Moreover, consumers may not know that they lack the standard products liability law remedies when they take generic drugs.  Thus, when they “decide” to use generics, their choice is often not fully informed.  And indeed, the consumer may even lack the power to demand a brand-name drug.  As Justice Sotomayor observed in her Mensing dissent, state laws empower pharmacists to substitute generics for their brand-name counterparts.  Mensing, 564 U.S. at 628, 643-44.  In some states such substitution may be accomplished without consulting the patient.  Id.

In short, leaving the burden of serious injuries from defective labels is unappealing and undermines the goals of strict products liability.

(2)  Imposing Liability on Brand-Name Manufacturers.  Under the generally-accepted “state of the art” defense, a drug warning will not be considered defective if its adverse effect could not reasonably have been discovered before sale.  See, e.g., Vassallo v. Baxter Healthcare Corp., 428 Mass. 1, 22-23, 696 N.E.2d 909 (1998); accord, Restatement (Third) Products Liability §6. comm. g.  Courts have recognized that the “state of the art” may be dependent on a manufacturer’s commitment to research and development focusing upon the health-related side-effects of their pharmaceuticals.  Vassallo, 428 Mass. at 21-22; Beshada v. Johns-Mansville Products Corp., 90 N.J. 191, 206-208, 447 A.2d 539 (1982), overruled in, Feldman v. Lederle Laboratories, 97 N.J. 429, 479 A.2d 374 (1984); see, Restatement (Third) Products Liability, §6, comm. g; id. §2, comm. m.

The brand-name manufacturer of a drug has the greatest control over the comprehensiveness of pre-market research regarding a pharmaceutical.  Given Hatch-Waxman’s legal regime, original applicants for new drug approval knowingly take on primary responsibility for assessing the risks and formulating appropriate warnings for the brand-name medication and any bio-equivalent generic ultimately marketed.  Generic drug manufacturers are entitled to assume that the original applicant has reasonably identified potential adverse effects and included them on the label.  By law, in submitting their abbreviated new drug application (ANDA) the generic manufacturers need focus only on ensuring that their drugs are therapeutically-equivalent to the brand-name drug.  To the extent that maximizing incentives for optimal levels of research is a primary goal of any system to address injuries resulting from inadequate generic drug warnings, brand-name manufacturers are the logical choice for liability regarding harms from adverse effects that should have been, but were not, discovered.  Moreover, such liability is not entirely foreign to modern tort law, see, Robert L. Rabin, Enabling Torts, 49 DePaul L. Rev. 435, 435-443, 446-48 (1999); T.H. v. Novartis Pharmaceuticals Corp., 4 Cal.5th 145, 166-74, 407 P.3d 18 (2017)(employing standard “duty of care” analysis to conclude that brand-name manufacturers’ duty to reasonably understand the risks of a medication and formulate appropriate warnings extended to the customers of the generic brands that would ultimately gain approval based on bio-equivalence).

Granted, brand-name manufacturers do not entirely lack safety incentives.  They are liable for the defective warnings on their own drugs, both before and after the period of patent exclusivity.  And during the period of patent exclusivity all (or virtually all) of the harms caused by the drug will be attributable to the brand-name product.  The risks of unanticipated liability may be particularly great during this period of patent exclusivity, because warnings are most likely to be inadequate during the first years a pharmaceutical is on the market.  Presumably over time manufacturers become more aware of the adverse effects of their medications.  (Of course, though the panoply of adverse effects may be less knowable in the early years of a drug’s use, during later years courts may consider ignorance of a drug’s side-effects less excusable, making imposition of liability for failure to warn more likely.)

Brand-name manufacturers’ monopoly for a significant period of time gives them a relatively unique opportunity to price their products to account for potential liability for those injured by generic substitutes in the future.  However, liability should be proportionate to the culpability of an actor’s conduct.  See, e.g., People’s Express Airlines v. Consolidated Rail Corp., 100 N.J. 246, 252-54, 495 A.2d 107 (1985)(discussing concern underlying traditional limitations on tort liability to ensure that the legal system does not impose “liability out of all proportion to the defendant’s fault”); Berman v. Allen, 80 N.J. 421, 432, 404 A.2d 8 (1979)(limiting extent of damages in wrongful birth cases); Strauss v. Bell Realty Corp., 65 N.Y.2d 399, 402, 482 N.E.2d 34 (1985)(limiting liability for physical harms resulting from citywide blackout to property owners in privity of contract with public utility); Nycal Corp. v. KPMG Peat Marwick LLP, 426 Mass. 491, 493-94, 688 N.E.2d 1368 (1998)(refraining from basing accountants liability on pure foreseeability). In the products liability context, such proportionality is furthered by making each manufacturer liable for its own products and only its own products.  This limitation is all the more urgent with respect to generic drugs, because brand-name manufacturers frequently leave the market once their patent monopoly ends and generics enter the market.  See Mensing, 564 U.S. at 629 (Sotomayor, J., dissenting)(citing Brief for Marc T. Law et al. as Amici Curiae 18 (one-third to one-half of generic drugs no longer have a marketed brand-name equivalent)).   Indeed, the brand-name manufacturer can withdraw its new drug approval, officially leaving the market.  See, Victor E. Schwartz, et al., Warning: Shifting Liability To Manufacturers Of Brand-Name Medicines When The Harm Was Allegedly Caused By Generic Drugs Has Severe Side Effects, 81 Fordham L. Rev. 1835, 1845-46 (2013).

Imposing liability upon brand-name manufacturers would require them to subsidize generic-brand manufacturers.  Already, of course, the generic-brand manufacturers been accorded a considerable advantage to the detriment of brand-name drug manufacturers, namely avoiding the expense of extensively testing of their medications.  In addition, when the brand-name manufacturer remains in the market after the patent exclusivity period ends, the brand-name and generic version of the product will be in direct competition.  Rarely is one market participant required to subsidize another by assuming one of it competitor’s major obligations under strict products liability law, namely liability for defective warnings.

Such a subsidy would increase potential costs to developers of new drugs, which might well further increase the cost of such pharmaceuticals during the period of patent exclusivity.  Given the lack of competition, the increased prices charged may be greater than they would be later, after the period of patent exclusivity ends, when competitive conditions prevail.  Moreover, given the prevalence of prescription-drug insurance, in many cases there is a disjunction between making the decision to use the drug and paying for it.   The former decision is made primarily by the doctor who prescribes the medication and the patient who “purchases” it.  The burden of paying for the medication rests largely on the insurer, who will pay the bulk of the drug’s cost.  This arrangement, which is atypical for consumer products, further reduces the demand-side constraints on pharmaceutical pricing.

Even if imposing liability on brand-name manufacturers is the best option prospectively, imposing it retroactively by making unanticipated changes in state products liability law (or by federal statute) would burden brand-name manufacturers after the period in which they could have priced their products accordingly has ended.  Given generic brand makers’ immunity from warning defects liability, brand-name manufacturers could charge the premium necessary to provide compensation to those harmed by defective generic drug warnings only during the period of patent exclusivity.

Ultimately, this option does not require congressional action and has at least been left open by the Supreme Court’s decision in Mensing.  And it relieves consumers (and the health insurance system) of the burden of bearing losses attributable to adverse reactions to medications.  It does so, however, at the cost of being unduly charitable to generic drug manufacturers.

(3) Imposing Liability on Generic Manufacturers.  Imposing liability on generic-brand manufacturers would entail judicial or congressional action to overrule Mensing, and thus permit state (or federal) imposition of liability on generic manufacturers.  Given that generic manufacturers cannot change drug labels immediately upon discovering their inadequacy, and at best must seek FDA action to do so, there is a fairness argument (in addition to a Supremacy Clause argument) for absolving them of liability.  See, Restatement (Third) of Torts: Liability for Physical and Emotional Harm §16(b) (“[i]f an actor’s adoption of a precaution would require the actor to violate a statute, the actor cannot be found negligent for failing to adopt that precaution.”)  However, strict products liability is not based on fault, Restatement (Third) Products Liability §4, comm. a, but rather on both the greater ability of product sellers to internalize the cost of injuries so that the product’s price reflects the injuries caused by a product and on their greater ability of sellers to spread the loss of injuries.  Thus, liability could reasonably be imposed upon generic drug manufacturers for warning defects that they are powerless to prevent.

Consider the analogy with Chavez v. Southern Pacific Transportation Co., 413 F. Supp. 1203, 1213-14 (E.D. Cal. 1976), admittedly a case regarding liability attendant engaging in an abnormally dangerous activity.  In Chavez several individuals present near a Southern Pacific railyard were injured in an explosion of military munitions being transported on a Southern Pacific train. They sued for damages.  In defense, Southern Pacific asserted that it was legally obligated to transport the munitions.  Nevertheless, the Court found Southern Pacific liable, even though the military itself would later be found immune from liability due to sovereign immunity, see, In re Bomb Disaster at Roseville, California, 438 F. Supp. 769 (E.D. Cal. 1977)).  (The Federal Tort Claims Act (FTCA), which waives sovereign immunity for tort claims, makes the Government liable only on the basis of fault.)  The Court observed that Southern Pacific’s argument would have merit were fault the basis of absolute liability.  However, it observed, absolute liability for damage caused by explosives is based on the importance of internalizing costs and optimally spreading of risks — not fault.  Chavez, 413 F. Supp. at 1213-14.  The case suggests the flaw in the generic-brand manufacturers’ successful argument in Mensing.  In a sense, the federal law limiting changes to drug labelling and the law of strict products liability do not irreconcilably conflict — the generic manufacturer can decide not to enter the market knowing that some of its liability will be largely uncontrollable.  It can also enter the market and price its products to account for the liability it must bear for adverse consequences about which it cannot warn.

Imposing liability on generic manufacturers would raise their costs, increasing the price of generics.  Such price increases, to account for injuries caused by the generic, could be particularly burdensome to consumers because access to pharmaceuticals is indispensable, and thus the demand for many pharmaceuticals is likely particularly inelastic.  See generally, Restatement (Third) of Torts: Products Liability §6, comm. b (discussing judicial concern about “possible negative effects of judicially imposed liability on the cost and availability” of pharmaceuticals and valuable medical technology).  Granted the prescription drug market is not one in which a disaggregated class of consumers must bargain with product sellers.  Private prescription drug plans, by aggregating consumers, probably possess significant bargaining power.  The federal government, which provides prescription drug coverage through Medicare Part D, could use its bargaining power to lower drug prices, were it not statutorily barred from doing so.  Juliette Cubanski and Tricia Neuman, Searching for Savings in Medicare Drug Price Negotiations (April 26, 2018), accessible at, https://www.kff.org/medicare/issue-brief/searching-for-savings-in-medicare-drug-price-negotiations/.

In addition, given that pharmacists may fill prescriptions with different generics at different times, a patient’s harm may result from taking several different generic brands of the same drug over time.  This situation resembles that created by DES in the 1970’s and 1980’s, see, e.g., Hymonowitz v. Eli Lilly & Co., 73  N.Y.2d 487, 508, 539 N.E.2d 1069 (1989); Sindell v. Abbott Laboratories, 26 Cal.3d 588, 607 P.2d 924 (1980); Enabling Torts, 49 DePaul L. Rev. at 451. Granted, most pharmaceuticals do not have DES’s latency period and intergenerational effects.  Nevertheless, removing generic drug-makers’ immunity from liability will result in the state courts having to address issues similar to those presented by DES.

It may be appropriate to impose some or all of the liability for defective generic drug labels on generic brand manufacturers, but such an approach may fail to fully address the unique role manufacturers of brand-name drugs play in the drug-labelling process.

 (4)  Create a federal compensation scheme funded by federal tax dollars.  Congress has created federal compensation schemes with regard to some pharmaceuticals in the past, see, e.g., National Childhood Vaccine Injury Act of 1986, Pub. L. 99–660, Title III, 100 Stat. 3755 (1986) (codified at 42 U.S.C. §§ 300aa-1 to 300aa-34).  As noted above, if we as a society wish to subsidize generic drugs, perhaps the fairest way to do so is to pay for harms suffered by individual consumer through general tax revenues.  The taxation will not be proportionate with the use of pharmaceuticals, as older Americans more heavily use pharmaceuticals than younger more healthy Americans.  And such an approach might also involve federalizing products liability law, at least with regard to warning defects on generic drug labels.  If the federal government decides to fund compensation for harms resulting from inadequate warnings on generic drugs, presumably it will seek to control the standards for liability so as to limit the strain the program’s resources.  In addition, the federal government will surely impose a degree of uniformity to ensure that individuals in different states receive similar treatment.  Of course, given the currently budget deficits and political climate, Congress is unlikely to enact legislation creating a compensation fund that will necessitate imposition of additional taxation.

(5) Mandate a compensation scheme jointly funded by brand-name manufacturers and generic manufacturers.  This approach would only modestly modify strict products liability law.  Instead of funding compensation through general tax revenues, those responsible for the label and the prescription drug would share financial responsibility.  This would thus internalize the cost of consumer injuries to the pharmaceutical industry while at the same time placing liability on those who can spread costs than better than injured consumers.  Brand-name manufacturers would be called upon to contribute because their decisions regarding research and development and the content of warning labels largely determines the content of generic drugs labels.  Imposition of some liability may encourage greater diligence on the part of the best “accident-avoiders.”  The generic manufacturers would be included because manufacturers should take responsibility for their own products, even if the defect is in a label they cannot easily change.  And the prices of generic drugs would more accurately reflect the harms they cause to consumers, the basic goal of products liability law.  Concerns regarding the strain on consumers caused by raising prices on a necessity prescription medication would remain.  See generally, Steve Morgan & Jae Kennedy, Prescription Drug Accessibility and Affordability in the United States and Abroad, Commonwealth Fund Publication 1408 (June 2010), accessible at, http://www.commonwealthfund.org/~/media/Files/Publications/Issue%20Brief/2010/Jun/1408_Morgan_Prescription_drug_accessibility_US_intl_ib.pdf .

Interestingly in the 1972 and 1978 amendments to the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA), namely the Federal Environmental Pesticide Control Act of 1972, Pub. L. 92-516, 86 Stat. 973 (1972)(1972 Act), and the Federal Pesticide Act of 1978, Pub. L. 95-396, 92 Stat. 819 (1978)(1978 Act), Congress sought to ensure that later applicants for approval to market a substance regulated under FIFRA who used an original applicant’s data would be obligated to provide compensation to the original applicant.  That statute perhaps did not recommend itself as a model because the 1972 structure resulted in approvals of insecticides, fungicides, and rodenticides coming to a halt, due to EPA’s inability to determine the appropriate among of compensation due the original applicant.  Congress’ 1978 fix, substituting binding arbitration for EPA’s determination of the amount of compensation due, was subject to a constitutional challenge as an unlawful delegation of adjudicatory authority.  The Supreme Court resolved the challenge in favor of the statute in 1985, see Thomas v. Union Carbide Agricultural Products, 473 U.S. 568 (1985), one year after Hatch-Waxman had been enacted.

Finally, as mentioned above, in the context of a federal compensation scheme, pharmacist might be asked to participate in funding the program.  Most product retailers are liable in strict products liability, so the immunity pharmacists enjoy is somewhat of an exception to the general rule.  Moreover, limitations on the liability of both generic and brand-name manufacturers for the inadequacy of generic brand drug labeling, could easily lead plaintiffs’ attorneys to develop creative negligence theories for holding pharmacists liable, even when the injury is attributable to an inadequate label over which the pharmacist has no control.  In addition, the structure of the pharmacy retail sector suggests some capacity to shoulder a portion of the burden of injuries to consumers.  The pharmaceutical retail market is quite concentrated in many major cities.  Corey Stern, CVS and Walgreens are completely dominating the US drugstore industry, Business Insider (July 29, 2015), accessible at http://www.businessinsider.com/cvs-and-walgreens-us-drugstore-market-share-2015-7. However, because pharmacies also generally also sell a variety of products, including health and beautify aids, foodstuffs and beverages, any share pharmacies must contribute could theoretically be subsidized by raising prices of products other than pharmaceuticals.  But it is not entirely clear in which direction the cross-subsidization would run; after all demand for medications is probably more inelastic than demand for health and beauty aids.  And pharmacies face more competition with respect to their sales of foodstuffs and beverages than with respect to their sales of pharmaceuticals.

Conclusion.  Perhaps this is a tale of unintended consequences.  The separate and overlapping system of product safety regulation and state strict products liability causes of action often mesh well together, but sometimes they do not.  With regard to defective drug labels, the combination of federal regulation and state products liability law has produced the worst solution — consumers bearing the cost of potentially catastrophic injuries from defective generic drug labels, even while those who take brand-name drugs can recover.  Congress has left this anomaly unresolved for seven years, as state supreme courts like the West Virginia Supreme Court in McNair v. Johnson & Johnson, 2018 WL 2186550 (W. Va. S. Ct. May 11, 2018), continue to grapple with the issue.

Print Friendly, PDF & Email