Indiana Law Review Volum e 46 2013 Num ber 1 NOTES A CRITICAL ANALYSIS OF PLIVA, INC. V. MENSING M ATTHEW J. C LARK * INTRODUCTION A wealthy business executive gives her pharmacist a prescription from her doctor for a drug to help treat her disease. The pharmacist informs her that the doctor has prescribed the brand-name of the drug, but she could save 80% of the cost by getting the generic of the same drug. The business executive informs the pharmacist that she always buys brand-name. The pharmacist replies that there is absolutely no difference between the drugs other than the com pany that manufactures them; both of the drugs are equally effective. The business executive refuses the offer, convinced that the brand-name has to be better. Later, a blue-collar worker gives the pharmacist a prescription for the same drug to treat the same disease. The pharmacist informs him of the same information she gave the business executive. The man’s face lights up at the sound of saving 80% of the price, and the pharmacist fills his order with the generic drug. A s a result of taking the drug for several years, both the business executive and the blue-collar worker develop a debilitating neurological disorder. Unfortunately, the warnings on the drugs’ labels did not adequately inform practitioners of the dangers of taking the drug for more than one year. Even more unfortunate is both the brand-name manufacturer and the generic manufacturer had an abundance of medical information that established there was an extremely high risk of consumers developing the neurological disorder if the drug was taken longer than one year. D espite this information, neither manufacturer changed its label or even sought to supplement its label with a warning regarding this risk of the drug. Because of this lack of warning, the business executive files a successful state failure-to-warn claim against the brand-name drug manufacturer. After hearing of the business executive’s success in receiving compensation for her injury, the * J.D. Candidate, 2013, Indiana University Robert H. McKinney School of Law; M.F.A., 2006, Chapman University, Orange, California; B.A., 2001, Ball State University, Muncie, Indiana. First and foremost, I would like to thank my wife, Jennifer, for all her support, encouragement, and understanding. I would like to thank my family for their support as well. I am grateful to everyone who assisted in the development of this Note, especially all the editors of the Indiana Law Review. I would also like to give a special thanks to David Meehan for his efforts in getting this Note published. I dedicate this work to my son and best little buddy, Flynn. 174 INDIANA LAW REVIEW [Vol. 46:173 blue-collar worker decides to sue the generic drug manufacturer. To his surprise, his lawyer informs him that he does not have a failure-to-warn claim against the generic drug manufacturer. T he worker is confused because his situation is identical to the business executive’s situation. The only distinction is that he took the generic form of the drug rather than the brand-name version. The lawyer informs the worker that taking the generic drug was his mistake. The lawyer explains that as a result of the Supreme Court’s decision in PLIVA, Inc. v. M ensing; state failure-to-warn claims against generic drug1 manufacturers are preem pted by federal drug regulations because the federal regulations require that a generic drug’s label be the same as its equivalent brand- name drug’s label at all times. In other words, if a brand-name drug2 manufacturer fails to warn of a danger, then the generic equivalent, in order to keep the label the same, must fail to warn of the danger too. It would be unfair if a person injured from taking a generic drug could sue the generic drug manufacturer for failing to warn because it was just fulfilling its federal duty. 3 A brand-name manufacturer, on the other hand, can “unilaterally” change its label to add or update warnings on its label; hence, a person injured by a brand-name drug can sue the manufacturer. 4 W ith a puzzled look on his face, the worker inquires whether he can sue the brand-name drug manufacturer because his generic drug had to have the same label. The lawyer sadly informs him that the worker’s state, like most states, does not allow a person injured by the use of a generic drug to sue the manufacturer of its brand-name equivalent. The lawyer regretfully tells the worker that, unlike5 the business executive, he has no remedy for his injury. 6 As the lyric from the Genesis song “Land of Confusion” says, “This is the world we live in.” An injured person’s ability to sue turns on whether the drug7 he or she had taken was either brand-name or generic. The Supreme Court’s8 1. 131 S. Ct. 2567 (2011). 2. Id. at 2577-78. 3. See id (“It was not lawful under federal law for the Manufacturers to do what state law required of them.”). 4. Id. at 2581. 5. Kellogg v. Wyeth, 762 F. Supp. 2d 694, 706-07 (D. Vt. 2010); see also Foster v. Am. Home Prods. Corp., 29 F.3d 165, 171 (4th Cir. 1994) (“As Wyeth has no duty to the users of other manufacturers’ products, a negligent misrepresentation action cannot be maintained against it on the facts of this case.”). But see Conte v. Wyeth, Inc., 85 Cal. Rptr. 3d 299, 320-321 (Cal. Ct. App. 2008) (holding that a brand-name drug manufacturer’s duty to use care extends to users injured by the generic equivalent if the doctor relied on the brand-name drug’s warning when prescribing the brand-name or generic drug). 6. See PLIVA, 131 S. Ct. at 2592 (Sotomayor, J., dissenting) (“The majority’s pre-emption analysis strips generic-drug consumers of compensation when they are injured by inadequate warnings.”). 7. GENESIS, Land of Confusion, on INVISIBLE TOUCH (Atlantic Recording Corp. 1986). 8. PLIVA, 131 S. Ct. at 2592 (Sotomayor, J., dissenting) (disagreeing with the majority’s finding that an individual’s right to a remedy “turns on the happenstance of whether her pharmacist 2013] PLIVA, INC. V. M ENSING 175 decisions in Wyeth v. Levine, which held that federal law does not preempt state9 failure-to-warn claims against brand-name drug manufacturers, and PLIVA, Inc.10 v. M ensing has set up this bizarre reality. As Justice Sotomayor, writing for the dissent in PLIVA , points out, the majority “decision leads to so many absurd consequences.” One absurd result of the PLIVA decision is that A mericans are11 supposed to believe that Congress intended to deprive people injured by generic drugs of a remedy while simultaneously promoting the use of generic drugs. 12 Another absurd result is that in a market with generic drugs constituting 75% of prescription drugs, and in which many brand-name manufacturers leave the13 market once their patent expires, “there will [now] be no manufacturer subject to failure-to-warn liability.”14 This decision is especially unfortunate in that it comes at a time when people in the United States are protesting on W all Street about corporate accountability and economic inequality. Although the hypothetical given in this introduction15 does not have to be cast in terms of a wealthy individual compared with a middle class individual, it logically could be a likely effect because wealthier individuals, for the most part, are probably more able and willing to pay the higher cost of a brand-name drug. Thus, this decision might result not only in a drop in demand for generics, but could also add a little fuel to the fire on class distinction16 amongst people in this country. This may be overstating the situation somewhat, but these are the types of ramifications that can come from absurdity. On the other hand, the PLIVA decision will most definitely add to the distrust and opposition of corporations. As Justice Sotomayor explains, “[M ]any generic manufacturers . . . are huge, multinational companies . . . [that have] sold an estimated $66 billion of drugs in [the United States] in 2009.”17 This Note analyzes the PLIVA, Inc. v. M ensing decision. A lthough Justice Thomas, writing for the majority, admitted that its decision could be seen as making “little sense” to the plaintiffs, he informed them that the majority’s hands were tied by the supremacy of federal law. Despite the majority’s apparent18 dislike for the result of its decision, this Note argues that the majority’s hands were not as tied down by federal regulation and the supremacy of federal law as it so determined. Part I discusses the background of the case, starting with how filled her prescription with a brand-name drug or a generic”). 9. 555 U.S. 555 (2009). 10. Id. at 580-81. 11. PLIVA, 131 S. Ct. at 2592 (Sotomayor, J., dissenting). 12. Id. at 2583-84, 2592. 13. Id. at 2583. 14. Id. at 2593. 15. Andrew Ross Sorkin, On Wall Street, A Protest Matures, N.Y. TIMES, Oct. 4, 2011, at B, available at http://query.nytimes.com/gst/fullpage.html?res=9C04E1D91E31F937A35753C1A9679 D8B63. 16. PLIVA, 131 S. Ct. at 2593 (Sotomayor, J., dissenting). 17. Id. at 2584. 18. Id. at 2581-82 (majority opinion). 176 INDIANA LAW REVIEW [Vol. 46:173 the FDA approves drugs and how changes are made to drug labels and then recapitulates the Court’s decisions in Levine and PLIVA. Part II of this Note continues with the analysis of the PLIVA decision. Finally, Part III summarizes the analysis by concluding that the Court’s deference to the FDA’s interpretation of its regulations, rather than the regulations themselves, was ultimately responsible for denying the plaintiffs in PLIVA compensation for their injuries. I. B ACKGROUND A. The Drug Approval Process Prior to 1906, food and drug safety was overseen only by the states through regulations and common law liability. After the C ivil W ar, the drug industry19 saw a boom in manufacturing and “secret formula” drugs, meaning drugs whose ingredients were not disclosed to the public. Because “neither medicine nor20 pharmacy had a firm scientific basis” during this time, public concern grew over the quality of drugs. In addition, because of the inexpensive manufacturing cost21 for secret formula drugs, “pharmacists adulterat[ed] legitimate drugs in an attempt to compete” with “grocers and other uneducated formulators.” For example,22 public opposition existed during this time over “addictive ‘soothing syrups,’”23 which were “[r]ecommended for ‘teething babies’” and contained morphine sulfate. Obviously, these products were effective but addictive, and at the same24 time, imitations existed “with no effective active ingredients.” Fearing these25 altered and misbranded drugs were traveling through interstate commerce, state regulators urged the federal governm ent to aid in the protection of consumers from fraudulent drugs. In 1906, Congress responded by passing the Pure Food26 and Drugs Act, which established the Food and D rug Administration (FDA)27 28 and made the manufacturing or shipping of altered or misbranded drugs illegal.29 In the 1930s, sulfa was “one of the first effective anti-infective drugs . . . 19. Wyeth v. Levine, 555 U.S. 555, 565-66 (2009); see also Mary J. Davis, The Battle Over Implied Preemption: Products Liability and the FDA, 48 B.C. L. REV. 1089, 1100 (2007) (“State regulators encouraged . . . the national government to create a federal agency to aid in regulation because of concerns over the States’ inability to reach the interstate sale of fraudulent products.”). 20. FOOD & DRUG LAW INST., FDA: A CENTURY OF CONSUMER PROTECTION 30-31 & 273 n.31 (Wayne L. Pines ed., 2006). 21. Id. at 30. 22. Id. 23. Id. 24. Id. 25. Id. 26. Davis, supra note 19, at 1100. 27. 21 U.S.C. § 14 (repealed 1938). 28. U.S. Food & Drug Admin., History, FDA.GOV, http://www.fda.gov/AboutFDA/ WhatWeDo/History/default.htm (last updated July 29, 2010). 29. Wyeth v. Levine, 555 U.S. 555, 566 (2009). 2013] PLIVA, INC. V. M ENSING 177 developed,” but it could not be given to children because sulfa’s composition as a “bulky powder” required oral administration via a large capsule. In 1937, a30 chemist discovered that sulfa could be made into a liquid formulation by dissolving it in diethylene glycol, which is known today as antifreeze. The only31 premarket test performed was a taste test. A fter it was sold on the market,32 numerous “infants suffered slow, painful death[s] as the diethylene glycol . . . produced irreversible liver toxicity.” In 1938, in part to respond to this tragic33 disaster, Congress went further for the protection of consumers by enacting the Federal Food, Drug, and Cosmetic Act (“FDCA”). The FDCA not only34 prohibited the sale of altered or misbranded drugs, but it also required the FDA’s approval of any new drug before sale on the market. As a result, drug35 manufacturers wanting to market a new drug had to submit to the FDA a new drug application (“NDA”) that included investigative reports about the drug and proposed labeling. Although the manufacturer was prohibited from distributing36 the drug until it received approval from the FDA, the burden, prior to 1962, was on the FDA to prove the drug was unsafe. If the FDA determined the drug was37 safe for its intended use as shown on the label, the manufacturer could sell the new drug on the market.38 Despite the FDCA requirement for premarket approval, drug testing was not very intensive. Clinical trials of new drugs were not used for evaluation,39 labeling was more “promotional” rather than informative, and the FDA only had40 a small staff to review the safety of drugs. In 1961, the manufacturers of41 thalidomide, a drug used in Europe and Japan to help pregnant wom en manage m orning sickness, was seeking approval in the United States. W hile the drug42 was still pending approval, thousands of children in Europe and Japan were born with birth defects to mothers who had taken thalidomide. Subsequently,43 30. FOOD & DRUG LAW INST., supra note 20, at 161. 31. Id. at 161-62. 32. Id. at 162. 33. Id. 34. Davis, supra note 19, at 1100; Federal Food, Drug, and Cosmetic Act, ch. 675, 52 Stat. 1040 (1938) (codified as amended at 21 U.S.C. §§ 301-399(d) (2006)). 35. Id. 36. Wyeth v. Levine, 555 U.S. 555, 566 (2009). 37. Id. at 566-67; see also Charlotte J. Skar, Products Liability—Conflict Preemption: The United States Supreme Court Denies Preemption Defense for Drug Manufacturers Using FDA- Approved Warning Labels Wyeth v. Levine, 129 S. Ct. 1187 (2009), 86 N.D. L. REV. 405, 409 (2010). 38. Kellogg v. Wyeth, 612 F. Supp. 2d 421, 424 (D. Vt. 2008). 39. FOOD & DRUG LAW INST., supra note 20, at 165. 40. Id. 41. Id. 42. Id. at 19, 166. 43. Id.; see also U.S. Food & Drug Admin., History, FDA.GOV, http://www.fda.gov/ AboutFDA/WhatWeDo/History/Overviews/ucm056044.htm (last updated Dec. 14, 2011). 178 INDIANA LAW REVIEW [Vol. 46:173 Congress passed amendments to the FDCA in 1962 that further strengthened control over new drugs. The 1962 amendments required not only that the new44 drug be safe but also that it be effective for its intended use. It also shifted the45 burden of proving the safety and effectiveness of the new drug to the manufacturer. As part of the NDA, a manufacturer m ust include “full reports46 of investigations” into the drug’s safety and effectiveness gathered from clinical47 trials and other adequate data; a list of the drug’s ingredients; a full statement of the drug’s composition; a description of how the drug was made, processed, and packaged; samples, if needed; and the manufacturer’s proposed labeling. The48 FDA must disapprove the application if the FDA finds that the investigations lacked adequate tests, the reports indicate that the drug is unsafe for its intended use as described in the label, the methods used to manufacture and pack the drug were inadequate to preserve the drug’s purity, the information gathered from the reports was inadequate to make a determination of the drug’s safety or effectiveness, or the labeling is false or misleading. 49 In addition to the 1962 amendments, Congress added a saving clause, which50 stated, “a provision of state law would only be invalidated upon a direct and positive conflict with the FDCA.” Later, in 1976, Congress enacted an express51 preemption clause for medical devices as part of the M edical Device Amendments but declined to do so for prescription drugs.52 53 In an effort to make low-cost generic drugs more available to consumers, in 1984 Congress enacted the Drug Price Competition and Patent Term Restoration A ct, often referred to as the Hatch-W axman Amendments, to the FDCA. In54 55 the amendments, Congress established an abbreviated new drug application (“ANDA”) to allow generic drug manufacturers to gain FDA approval by showing that its new drug was essentially the same as the “bioequivalent” of the listed 44. FOOD & DRUG LAW INST., supra note 20, at 166. 45. Kellogg v. Wyeth, 612 F. Supp. 2d 421, 424 (D. Vt. 2008). 46. Skar, supra note 37, at 409-10. 47. 21 U.S.C. § 355(b)(1) (2006). 48. Id.; Kellogg, 612 F. Supp. 2d at 424. 49. 21 U.S.C. § 355(d), amended by Pub. L. No. 112-144, § 3187, 126 Stat. 993 (2012); Kellogg, 612 F. Supp. 2d at 424-25. 50. “A saving clause is generally used in a repealing act to preserve rights and claims that would otherwise be lost.” BLACK’S LAW DICTIONARY 1146 (9th ed. 2010). Here, the saving clause is saying that a state law would be repealed by the FDCA only if the state law presented a direct and positive conflict with the FDCA. 51. Wyeth v. Levine, 555 U.S. 555, 567 (2009) (internal quotation marks omitted). 52. Medical Device Amendments of 1976, Pub. L. No. 94-295, 90 Stat. 539 (codified as amended at 21 U.S.C. § 360(k) (2006)). 53. Levine, 555 U.S. at 567. 54. Drug Price Competition and Patent Term Restoration Act of 1984, Pub. L. No. 98-417, 98 Stat. 1585 (codified as amended at 21 U.S.C. § 355 (2006)). 55. PLIVA, Inc. v. Mensing, 131 S. Ct. 2567, 2583 (2011) (Sotomayor, J., dissenting). 2013] PLIVA, INC. V. M ENSING 179 drug, i.e., the approved brand-name drug. Two drugs “are ‘bioequivalent’ if56 they are given at the same dose, contain the same active ingredient, and reach the same level at the site of action. Two bioequivalent drugs . . . are absorbed the same way by the body and result in the same clinical response in the patient.”57 Therefore, along with other requirements, a generic drug manufacturer must show that its drug has the same active ingredients as the brand-name drug, “the route of administration, the dosage form, and the strength of the new drug are the sam e as the brand-name drug,” its drug is the bioequivalent of the brand-nam e drug, and that proposed labeling of the new drug “is the same as the labeling58 approved for the [brand-name] drug,” with some exceptions. By establishing59 the ANDA, Congress was able to increase the availability of generic drugs because generic manufacturers did not have to conduct the costly clinical trials to get approval and thus could “bring [their] drugs to market . . . less expensively.”60 In implementing the Hatch-W axman Amendments, the FDA requires the generic drug’s proposed labeling be the same as the labeling of its brand-name equivalent except for some allowed differences. These exceptions include61 differences because of “expiration date, formulation, bioavailability, or62 pharmacokinetics, labeling revisions made to comply with current FDA labeling63 guidelines or other guidance.” If any drug manufacturer obtains newly acquired64 information about the drug after it has been on the market that shows “reasonable evidence of an association of a serious hazard with a drug,” the drug manufacturer must revise the label. All drug manufacturers are prohibited from65 distributing “a ‘misbranded’ drug, . . . including a drug whose ‘labeling is false or misleading in any particular.’” 66 56. Id.; 21 U.S.C. § 355(j) (2006). 57. Melanie McLean, ONADE Participates in Bioequivalence Workshop, U.S. FOOD & DRUG ADMIN., http://www.fda.gov/AnimalVeterinary/NewsEvents/FDAVeterinarianNewsletter/ucm 222135.htm (last updated Jan. 26, 2011). 58. PLIVA, 131 S. Ct. at 2583 (Sotomayor, J., dissenting); 21 U.S.C. § 355(j)(2)(A)(ii)-(iv). 59. PLIVA, 131 S. Ct. at 2574 (alteration in original); 21 U.S.C. § 355(j)(2)(A)(v). 60. PLIVA, 131 S. Ct. at 2583 (Sotomayor, J., dissenting); see also Mensing v. Wyeth, Inc., 588 F.3d 603, 606 (8th Cir. 2009) (noting that Congress passed the Drug Price Competition and Patent Term Restoration Act in order “to bring more affordable generic drugs to [the] market”). 61. Kellogg v. Wyeth, 612 F. Supp. 2d 421, 426 (D. Vt. 2008); 21 C.F.R. § 314.94(a)(8)(iii) (2011). 62. Bioavailability means a drug’s absorption into the blood. PHARMACOTHERAPY: A PATHOPHYSIOLOGIC APPROACH 53 (Joseph T. DiPiro et al. eds., 6th ed. 2005). 63. Pharmacokinetics is “the absorption, distribution, metabolism, and elimination of drugs in patients requiring drug therapy.” Id. 64. 21 C.F.R. § 314.94(a)(8)(iv) (2011). 65. Id. § 201.80(e); accord id. § 201.57(c)(6)(i). 66. Mensing v. Wyeth, Inc., 588 F.3d 603, 606 (8th Cir. 2009) (quoting 21 U.S.C. §§ 331(a)- (b), 352(a) (2006), amended by Pub. L. No. 112-144, § 3187, 126 Stat. 993 (2012)), vacated in part, reinstated in part, 658 F.3d 867 (8th Cir. 2011). 180 INDIANA LAW REVIEW [Vol. 46:173 FDA regulations establish a few ways in which a drug manufacturer may change its labeling. For major changes, defined as having “a substantial potential to have an adverse effect on the identity, strength, quality, purity, or potency of the drug,” a manufacturer must submit a supplemental application and receive67 FDA approval of the label change before it can distribute the drug. For68 moderate changes, defined as having a “moderate potential to have an adverse effect,” which includes changes in the labeling “[t]o add or strengthen a . . .69 warning,” a manufacturer can make the change and distribute the drug prior to70 obtaining approval from the FDA by submitting a supplemental application labeled “Changes Being Effected” (C B E ). In summary, to make any major71 changes to the drug, the manufacturer must get prior approval from the FD A , whereas for a moderate change that includes updating warnings, a manufacturer may change the label prior to FDA approval through the CBE process. B. The W yeth v. Levine Decision Diana Levine, a musician, developed gangrene in her arm as a result of an injection of the brand-name drug Phenergan, which required the amputation of her right forearm. Phenergan can cause gangrene if it makes contact with a72 patient’s arterial blood. The drug was adm inistered to Levine via an IV-push73 method, meaning it was injected into her arm with a needle, and thus the needle either directly made contact with an artery, or the drug escaped the vein and entered an artery. Levine sued the manufacturer, W yeth, for failing to have a74 strong enough warning of the danger of administering the drug via an IV-push method when it could have been more safely administered via an IV-drip method, where the drug is mixed “into a saline solution in a hanging intravenous bag and slowly descends through a catheter inserted in a patient’s vein.”75 W yeth argued that federal regulations preempted Levine’s suit because it76 67. 21 C.F.R. § 314.70(b)(1). 68. Id. § 314.70(a)-(b). 69. Id. § 314.70(c)(1). 70. Id. § 314.70(c)(6)(iii)(A). 71. Id. § 314.70(c)(3)-(6)(iii)(A). 72. Wyeth v. Levine, 555 U.S. 555, 559 (2009). 73. Id. 74. Id. 75. Id. at 559-60. 76. The Supreme Court has developed three ways in which to find preemption of state law. Caleb Nelson, Preemption, 86 VA. L. REV. 225, 226 (2000). First, express preemption occurs when Congress has enacted a provision that specifically states that it is preempting state law. Id. Second, field preemption occurs when nothing in a federal statue explicitly states it is preempting state law, but the statutory scheme so dominates a field that it must be implied because “‘Congress left no room for the States to supplement it.’” Id. at 227 (quoting English v. Gen. Elec. Co., 496 U.S. 72, 79 (1990)). Finally, conflict preemption may exist if state law and federal law either conflict in a way that it is physically impossible to comply with both, or if state law “‘stands as an obstacle to 2013] PLIVA, INC. V. M ENSING 181 was impossible for W yeth to comply with both state law— requiring a safer label— and federal law— prohibiting W yeth from “unilaterally” changing its label. W yeth also argued that allowing a state failure-to-warn claim created an77 “‘obstacle to the . . . full purposes and objectives of Congress’ because it substitute[d] a . . . jury’s decision about drug labeling for” that of the FDA’s judgment. 78 The Court began its analysis of the case with what it considered to be the “two cornerstones of . . . pre-emption [sic] jurisprudence” : Congress’s intent79 is the ultimate guiding light, and when a case involves the police powers of the states, the Court needs “clear and manifest” proof that it was Congress’s intent to preempt state law. Using this guidepost, the Court held that it was not80 “impossible for W yeth to comply with both federal and state [law].” Although81 W yeth needed FDA approval of a label change, it could have unilaterally changed its label before receiving approval using the CBE process. Newly acquired82 information, which the CBE process requires to make a change, did not have to come from new clinical trials, but instead also could come from new analysis of old data. In addition, before Levine’s injury, she had submitted evidence of83 twenty incidents in which patients had developed gangrene after receiving Phenergan from the IV-push method. The Court also found that unilaterally84 changing the label did not make it autom atically misbranded because W yeth would be fulfilling its federal duty to have adequate warnings. To show85 impossibility, W yeth needed “clear evidence that the FDA would not have approved a change,” and W yeth had not done so in this case. The Court stated: 86 [T]hrough many amendments to the FDCA and to FDA regulations, it has remained a central premise of federal drug regulation that the manufacturer bears responsibility for the content of its label at all times. It is charged both with crafting an adequate label and with ensuring that its warnings remain adequate as long as the drug is on the market.87 The Court also held that state failure-to-warn claims did not serve as an obstacle to Congress’s purposes in enacting federal drug labeling regulations. 88 the accomplishment and execution of the full purposes and objectives of Congress.’” Id. at 227-28 (quoting Boggs v. Boggs, 520 U.S. 833, 844 (1997)). 77. Levine, 555 U.S. at 563, 570. 78. Id. at 563-64 (quoting Hines v. Davidowitz, 312 U.S. 52, 67 (1941)). 79. Id. at 565. 80. Id. 81. Id. at 571. 82. Id. 83. Id. at 569. 84. Id. 85. Id. at 570. 86. Id. at 571-72. 87. Id. at 570-71. 88. Id. at 573. 182 INDIANA LAW REVIEW [Vol. 46:173 Instead, “Congress enacted the FD CA to bolster consumer protection.” 89 Congress must have been aware of state tort actions, but despite this information, it never enacted a federal remedy for injured consumers. The Court reasoned90 that this was evidence of Congress’s belief that state tort actions provide injured consumers with enough relief, as well as evidence of Congress’s possible recognition of the fact that state tort actions help to protect consumers by incentivizing “manufacturers to produce safe . . . drugs” with “adequate warnings.” Thus, “Congress did not intend FDA oversight to be the exclusive91 means of ensuring drug safety and effectiveness.”92 As further evidence that state tort actions are not an obstacle, the Court examined the FDA’s long history of supporting state tort law. The FDA had93 consistently presented federal drug labeling regulations “as a floor upon which states could build.” In fact, the FDA had previously stated that it did not believe94 state tort law would “‘be at odds with [its] regulations,’” and that it did not want95 to stop states from “‘imposing additional labeling requirements.’” The Court96 believed this to be evidence of the FDA’s belief that state tort law served a “complementary” position with federal regulation. In addition, the Court found97 that this belief supported the reality that manufacturers were in the better position to monitor their drugs because they had better access to data, and the FDA lacked the resources necessary to oversee the 11,000 drugs on the market. Finally, the98 Court stated that “[f]ailure-to-warn actions, in particular, lend force to the FDCA’s premise that manufacturers . . . bear primary responsibility for their drug labeling at all times.” As a result, the Court concluded that federal law did not99 preempt state failure-to-warn claims against brand-name manufacturers.100 C. The PLIVA, Inc. v. M ensing Decision Gladys M ensing suffers from diabetic gastroparesis, a disorder that slows101 the digestion of food and, as a result, can worsen diabetes by making it harder to 89. Id. at 574. 90. Id. at 574-75. 91. Id. at 574. 92. Id. at 575. 93. Id. at 577-79. 94. Id. at 577. 95. Id. at 578 (quoting Prescription Drug Product Labeling; Medication Guide Requirements, 63 Fed. Reg. 66,378, 66,384 (Dec. 1, 1998)). 96. Id. 97. Id. 98. Id. at 578-79. 99. Id. at 579. 100. Id. at 581. 101. Mensing v. Wyeth, Inc., 588 F.3d 603, 605 (8th Cir. 2009), vacated in part, reinstated in part, 658 F.3d 867 (8th Cir. 2011). 2013] PLIVA, INC. V. M ENSING 183 control blood glucose. In 2001, M ensing’s doctor prescribed the brand-name102 drug Reglan to treat her gastroparesis, and her pharmacist, following M innesota law, substituted the Reglan with its generic bioequivalent, metoclopramide. 103 M ensing took the generic metoclopramide for four years as prescribed.104 Julie Demahy suffers from gastroesophageal reflux, a disease that can105 cause frequent heartburn and regurgitation and, in severe cases, can cause narrowing of the esophagus. In 2002, Demahy’s doctor prescribed Reglan as106 well, and her pharmacist, following Louisiana law, substituted it with generic metoclopramide. Like M ensing, Demahy took the drug for four years as107 prescribed.108 M etoclopramide is a drug that speeds digestion of food by “enhancing . . . contractions of the esophagus, stomach, and intestines,” as well as blocking109 dopamine receptors in the brain, which helps to prevent nausea and vomiting. 110 111 B ecause metoclopramide acts on dopamine receptors, it can affect the body’s extrapyramidal system, which is responsible for controlling fine motor skills. 112 One type of severe extrapyramidal symptom is tardive dyskinesia, a severe neurological disorder that is “characterized by grotesque involuntary movements of the mouth, tongue, lips, and extremities, involuntary chewing movements, and a general sense of agitation.”113 The FDA approved Reglan in 1980 for short-term use only, with no indication over twelve weeks. In 1985, generic manufacturers were receiving114 approvals to make generic metoclopramide. Although the drugs were intended115 for short-term use, data revealed that doctors were prescribing R eglan and metoclopramide for longer than one year. Despite this information, from 1985116 to 2005 (the years during which M ensing and Demahy were taking metoclopramide), the labels for Reglan and generic metoclopramide presented the 102. Living with Diabetes, AM. DIABETES ASS’N, http://www.diabetes.org/ living-with- diabetes/complications/gastroparesis.html (last visited Jan. 16, 2012). 103. Mensing v. Wyeth, Inc., 588 F.3d at 605. 104. Id. 105. Demahy v. Actavis, Inc., 593 F.3d 428, 430 (5th Cir. 2010), overruled by PLIVA, Inc. v. Mensing, 131 S. Ct. 2567 (2011). 106. McNeil v. Wyeth, 462 F.3d 364, 366 (5th Cir. 2006). 107. Demahy, 593 F.3d at 430. 108. Id. 109. McNeil, 462 F.3d at 366. 110. Id. 111. See Mayo Clinic, Metoclopramide (Oral Route), MAYO CLINIC, http://www.mayoclinic. com/health/drug-information/DR600921 (last visited Jan. 16, 2012). 112. McNeil, 462 F.3d at 366. 113. Id. 114. See Mensing v. Wyeth, Inc., 588 F.3d 603, 606 (8th Cir. 2009), vacated in part, reinstated in part, 658 F.3d 867 (8th Cir. 2011). 115. PLIVA, Inc. v. Mensing, 131 S. Ct. 2567, 2572 (2011). 116. McNeil, 462 F.3d at 369. 184 INDIANA LAW REVIEW [Vol. 46:173 risk of developing tardive dyskinesia at about .2% . During this period,117 however, studies showed that 29% of patients taking metoclopramide for several years developed tardive dyskinesia. One study in 1994 even found that 27%118 of patients taking metoclopramide for longer than thirty days developed tardive dyskinesia. In 2004, the manufacturer of Reglan requested and received an119 approval from the FDA for a change in the label that stated “‘[t]herapy should not exceed [twelve] weeks in duration.’” The previous version stated less120 strenuously that therapy longer than twelve weeks was not recommended. 121 Finally, in 2009, acting on its own initiative, the FDA ordered brand-name and generic manufacturers of metoclopramide to add a black box warning— the FDA’s strongest warning— to their labels stating that “[t]reatment with m etoclopramide can cause tardive dyskensia,” and that “[t]reatment . . . longer than [twelve] weeks should be avoided in all but rare cases.” 122 After taking generic metoclopramide for four years, both M ensing and Demahy developed tardive dyskinesia. Both women sued the generic123 manufacturers of metoclopram ide for failing to provide an adequate warning as the risk of developing tardive dyskinesia was much higher than indicated on their labels. The generic manufacturers argued federal drug regulations preempted124 state failure-to-warn claims because the federal regulations required them to have the same label “as their brand-name counterparts.” As a result, the125 manufacturers argued it was impossible for them to modify their labels to comply with state law.126 The Court framed the issue of the case as whether generic manufacturers could make changes to their label after initial approval of an ANDA. The FDA127 interpreted its regulations to mean a generic manufacturer has a “duty of sameness,” meaning its label must be the same as its brand-name counterpart128 at all times. The Court began its analysis by indicating that it would follow129 Auer v. Robbins in deferring to the FDA’s interpretation. A uer held that a130 131 117. Id. at 370; Mensing v. Wyeth, Inc., 588 F.3d at 606. 118. PLIVA, 131 S. Ct. at 2572 (citing McNeil, 462 F.3d at 370 n.5). 119. McNeil, 462 F.3d at 370 n.5. 120. PLIVA, 131 S. Ct. at 2572-73 (alteration in original) (quoting Brief for the United States as Amicus Curiae Supporting Respondents at 8, PLIVA, 131 S. Ct. 2567 (Nos. 09-993, 09-1039, 09-1501)). 121. Id. 122. Id. at 2573. 123. Id. 124. Id. 125. Id. 126. Id. 127. Id. at 2574. 128. Id. at 2574-75 (internal quotation marks omitted). 129. Id. 130. 519 U.S. 452 (1997). 131. PLIVA, 131 S. Ct. at 2575. 2013] PLIVA, INC. V. M ENSING 185 federal agency’s interpretation is “‘controlling unless plainly erroneous or inconsistent with the regulation[s]’ or there is any other reason to doubt that they reflect the FDA’s fair and considered judgment.”132 The plaintiffs, M ensing and Demahy, first argued that the generic drug manufacturers could have used the CBE process to change their label. The133 FDA, however, determined that the CBE regulation only allowed the generic manufacturers to use that process to change their labels to match their brand- name counterpart that had recently changed its label. The Court simply stated134 that it deferred to the FDA’s interpretation because it was not “‘plainly erroneous or inconsistent with the regulation.’” The plaintiffs also argued that the135 manufacturers could have utilized “Dear Doctor” letters, letters sent to health care professionals, that could have advised the professionals of the additional warnings. The FDA argued that “Dear Doctor” letters qualified as “labeling”136 under 21 U.S.C. § 321(m). The Court, again, stated that it deferred to the FDA137 for the same reasons as its prior interpretation. 138 Although the FDA interpreted its regulations as prohibiting generic manufacturers from unilaterally changing their labels to strengthen a warning, it also interpreted its regulations as imposing an affirmative duty on generic manufacturers to propose stronger warnings to the FDA if needed. According139 to the FD A , this duty exists under 21 C.F.R. § 201.57(e), which states, “‘[L]abeling shall be revised to include a warning as soon as there is reasonable evidence of an association of a serious hazard with a drug.’” If the FDA agreed140 with the generic m anufacturer, it would then work with the brand-name manufacturer to update the label. The FDA believed this process allowed a141 generic manufacturer to simultaneously maintain its duty of sam eness and its statutory obligation under 21 U.S.C. § 352(f)(2) to not distribute a drug misbranded with inadequate labeling. 142 Assuming this duty existed, the Court still found preem ption. The Court143 132. Id. (quoting Auer, 519 U.S. at 461-62). 133. Id. 134. Id. 135. Id. (quoting Auer, 519 U.S. at 461). 136. Id. at 2576. 137. Id.; see 21 U.S.C. § 321(m) (2006) (stating that “‘labeling’ means all labels and other written, printed, or graphic matter (1) upon any article or any of its containers or wrappers, or (2) accompanying such article”). 138. PLIVA, 131 S. Ct. at 2576. 139. Id. 140. Id. (quoting 21 C.F.R. § 201.57(e) (2011)). 141. Id. 142. Id. (quoting 21 U.S.C. § 352(f)(2) (2006), amended by Pub. L. No. 112-144, § 3187, 126 Stat. 993 (2012)) (“[A] drug is ‘misbranded . . . [u]nless its labeling bears . . . adequate warnings against . . . unsafe dosage or methods or duration of administration or application, in such manner and form, as are necessary for the protection of users.’”) (omissions in original). 143. Id. at 2577. 186 INDIANA LAW REVIEW [Vol. 46:173 explained that state law required the manufacturers to use a different label, but federal law required generic manufacturers to keep the label the same as its brand-name drug counterpart and to propose a different label to the FD A . 144 Proposing a change in the label did not satisfy the state duty because “[s]tate law demanded a safer label; it did not instruct the [m ]anufacturers to communicate with the FDA about the possibility of a safer label.” Rather than beginning its145 analysis of these requirements with the two cornerstones of preemption cases, as it did in Levine, the Court just reiterated the basic idea of preemption— federal law is “the supreme Law of the Land.” Although the Court admitted it was146 ultimately possible for the generic manufacturers to comply with both state law and federal law if the manufacturers had proposed stronger warnings and the FDA then approved of it and ordered the change, the Court thought that these additional actions made conflict preemption meaningless because many situations could be made possible by the actions of third parties. As a result, the Court147 stated the test for impossibility was “whether the private party could independently do under federal law what state law requires of it.” Therefore,148 because a generic manufacturer could not use the CBE process to unilaterally change its label prior to FDA approval, it could not independently comply with what state law requires. Acknowledging that its decision from the perspective149 of the plaintiffs made “little sense” in light of the Court’s decision in Levine, this inability to use the CBE process, from the Court’s perspective, distinguished the plaintiffs’ case from Levine. Realizing, too, that compensation turned on which150 version of the same drug was taken by the injured plaintiffs, the Court concluded that federal drug regulation had dealt the plaintiffs an “unfortunate hand,” but151 reminded them it was up to Congress and the FDA to change the law. 152 II. A NALYSIS OF THE PLIVA D ECISION Because the Court in Levine found that federal law did not preempt state failure-to-warn claims as a brand-name drug m anufacturer could use the CBE process to unilaterally strengthen the warnings on its label, the dispositive issue153 in PLIVA was whether a generic drug manufacturer could utilize the CBE process as well. In deciding this issue, the Court used the deference standard154 144. Id. at 2578. 145. Id. 146. Id. at 2577 (internal quotation marks omitted). 147. See id. at 2578-79. 148. Id. at 2579. 149. Id. at 2575, 2578, 2581. 150. Id. at 2581. 151. Id. 152. Id. at 2581-82. 153. Wyeth v. Levine, 555 U.S. 555, 573 (2009). 154. PLIVA, 131 S. Ct. at 2574. 2013] PLIVA, INC. V. M ENSING 187 established in Auer. Therefore, it makes sense to review how the Court155 developed this deference and subsequent decisions interpreting this standard of deference. A. Auer Deference In Auer v. R obbins, the Supreme Court held that a federal agency’s interpretation of its own regulation is “controlling unless plainly erroneous or inconsistent with the regulation,” or if there is some “reason to suspect that the156 interpretation does not reflect the agency’s fair and considered judgment.” In157 Auer, St. Louis sergeants sued the city commissioners for overtime pay under the Fair Labor Standards Act of 1938 (FLSA). In response, the commissioners158 argued the sergeants were not entitled to overtime pay because they were exempt from such pay under the FLSA. The Secretary of Labor had established159 regulations that determined exem pt status, one of which was the “salary-basis test.” According to the salary-basis test, an employee qualified for exemption160 if he received “a predetermined amount” as compensation that is “not subject161 to reduction because of variations in the quality or quantity of the work performed.” The sergeants argued they failed this test because their salary162 could be subject to reductions for disciplinary infractions based on the “quality or quantity” of their work. Thus, the primary issue of the case was whether a163 hypothetical possibility of a reduction in pay qualified as being “subject to” such reductions. 164 At the Court’s request, the Secretary of Labor filed an amicus brief interpreting the salary-basis test. Specifically, the Secretary of Labor165 interpreted the “subject to” language to mean that there was “an actual practice of” deductions in compensation, or that there was a policy that made such deductions significantly likely. The Court deferred to the Secretary’s166 interpretation finding it was not “‘plainly erroneous or inconsistent with the regulation’” (“Auer deference”). The Court reasoned that the Secretary’s167 155. Id. at 2575. 156. 519 U.S. 452, 461 (1997) (internal quotation marks omitted). 157. Id. at 462. 158. Id. at 454-55; Fair Labor Standards Act of 1938, ch. 676, 52 Stat. 1060 (codified as amended at 29 U.S.C. §§ 201-219 (2006)). 159. Auer, 519 U.S. at 455. 160. Id. at 454-55. 161. Id. at 455. 162. Id. 163. Id. (internal quotation marks omitted). 164. Id. at 459 (internal quotation marks omitted). 165. Id. at 461. 166. Id. 167. Id. (quoting Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 359 (1989)). 188 INDIANA LAW REVIEW [Vol. 46:173 interpretation of “subject to” easily fell within the phrase’s ordinary meaning. 168 The Court found that the police manual containing the rule violations applied to all employees, some of whom were not paid salary. As a result, it was unclear169 whether the “pay deductions [were] an anticipated form of punishment for employees in [the sergeants’] category.” Furthermore, the Court found that the170 single deduction in pay for one sergeant did not establish a significant likelihood of deductions. Finally, the Court also held there was no reason to doubt the171 Secretary’s interpretation as reflecting the agency’s fair and considered judgment because it did not come as a “‘post hoc rationalizatio[n]’” in response to a “past agency action [under] attack.”172 The Supreme Court further clarified its Auer deference in C hristensen v. Harris County, in holding the Auer deference to be only necessary when the173 agency’s regulation is ambiguous. In Christensen, deputy sheriffs sued Harris174 County, Texas for violating the FLSA by making them use compensatory time they had accumulated by working overtime. Under the FLSA, counties are175 allowed to compensate em ployees for overtime work by giving them compensatory time— time off work with full pay— instead of having to pay the higher hourly wage rate, provided the employee agrees to it. Yet, when an176 employee reaches the maximum hours of compensatory time allowed, the employer must pay the em ployee the overtime rate for the overtime hours. 177 Because Harris County was afraid they would not be able to afford the overtime pay for the deputy sheriffs who had accrued the m aximum amount of compensatory time, the County developed a policy under which it could order the deputy sheriffs to use their compensatory time. The deputy sheriffs argued that178 the County’s policy violated 29 U .S.C. § 207(o)(5) of the FLSA, which provides that an em ployer could not deny a request of compensatory time off unless it would “unduly disrupt the operations of the public agency” because that179 provision provided the only way to use compensatory time absent an agreement.180 Although the Court agreed with the principle “that when a statute limits a 168. Id. 169. Id. at 461-62. 170. Id. at 462 (emphasis omitted). 171. Id. 172. Id. (alternation in original) (quoting Bowen v. Georgetown Univ. Hosp., 488 U.S. 204, 212 (1988)). 173. 529 U.S. 576 (2000). 174. Id. at 588. 175. Id. at 578, 581. 176. Id. at 578-79. 177. Id. at 579-80. 178. Id. at 580-81. 179. Id. at 582. 180. Id. at 581-82. 2013] PLIVA, INC. V. M ENSING 189 thing to be done in a particular mode, it includes a negative of any other mode,”181 the Court rejected the deputy sheriffs’ argument that the “thing to be done” was the use of compensatory time. Rather, the Court found the “thing to be done”182 was the approval of the request, meaning the county could not reject a request for any reason other than undue disruption of operations. As a result, § 207(o)(5)183 restricted the employer’s ability “to prohibit the use of compensatory time” but did not restrict the employer from compelling the use of compensatory time. 184 The Court found support for its interpretation “in two other features of the FLSA”— one being that the FLSA allows an employer to reduce the hours an employee works, and the other being that 29 U.S.C . § 207(o)(3)(B) allows an employer to “cash out accumulated compensatory time by paying the employee his regular hourly wage for each hour accrued.” 185 The deputy sheriffs also argued for deference to the Department of Labor’s interpretation of its regulations as prohibiting an employer from requiring the use of compensatory time without obtaining prior consent of the employee. The186 C ourt held Auer deference was not applicable because the agency’s regulation was not ambiguous. The Court reasoned the Secretary of Labor’s regulations187 clearly permitted compelled compensatory time. The regulation implementing188 § 207(o)(5) prohibits an employer from using compensatory time to avoid paying overtime compensation. The Court read this as confirming § 207(o)(5)’s189 purpose of safeguarding the employee from not being compensated at all for overtime work. Another regulation stated that the agreement between the190 employer and employee regarding the use of compensatory time instead of overtime pay “may include other provisions governing the preservation, use, or cashing out of compensatory time . . . consistent with [§ 207(o)].” The Court191 found this regulation unambiguously permissive, meaning nothing within it suggested that compelled com pensatory time had to have been included within an agreement. The Court further held that “[t]o defer to the agency’s position192 would be to permit the agency, under the guise of interpreting a regulation, to create de facto a new regulation.”193 The two prior cases detail the proper analysis to give to an interpretation of 181. Id. at 583 (internal quotation marks omitted). 182. Id. (referring to the definition of “thing to be done” set forth in 29 U.S.C. § 207(o)(5) (2006), preempted by Jones v. United States, 88 Fed. Cl. 789 (Fed. Cl. 2009)). 183. Id. 184. Id. at 585. 185. Id. 186. Id. at 586. 187. Id. at 588. 188. Id. 189. Id. at 584 (paraphrasing 29 C.F.R. § 553.25(b) (1999)). 190. Id. 191. Id. at 587-88 (alteration in original). 192. Id. at 588. 193. Id. 190 INDIANA LAW REVIEW [Vol. 46:173 an agency regulation, beginning with whether the regulation in question is ambiguous and next providing a flushed-out analysis of whether the interpretation is plainly erroneous or inconsistent or whether some other reason exists to doubt the interpretation. Unfortunately, since Auer, most courts rarely conduct this type of thorough analysis when reviewing an agency’s interpretation of its own regulation. PLIVA is a case on point because the majority opinion only states194 that it defers to the FDA’s interpretation regarding the CBE regulation without providing any reasoning behind its decision. Because the Court does not195 provide any analysis, it appears that the Court is assuming the regulation is am biguous. In addition, the lack of analysis suggests that the Court is relying heavily on the plainly erroneous standard of the Auer test because it does not take much for an interpretation to overcome it. Based on precedent and the importance of the issue, the majority in PLIVA should have provided a more thorough analysis, starting with whether the FDA regulations were ambiguous. 1. The FDA’s Regulations Are Unambiguous.— M uch like the analysis in Christensen that turned on the Secretary of Labor’s use of the word “may” in its regulation, PLIVA, too, should have turned on the FDA’s use of a particular word in its regulation. Before 2007, 21 C.F.R. § 201.57 specified the requirements for label content and form for both brand-name and generic drug manufacturers;196 however, in 2006, the regulation was amended to divide the label content and form requirements between the two, with § 201.80 applying to generic drug manufacturers and § 201.57 applying to brand-name manufacturers. Despite197 this amendment, the language regarding the revision of warnings on the label remains the same for generic drug manufacturers under both versions. The198 FDA’s regulation requires warnings on the drug’s label “be revised . . . as soon as there is reasonable evidence of an association of a serious hazard with a drug.” In PLIVA, the FDA interpreted this regulation to mean only that a199 generic m anufacturer had “to propose [a] stronger warning label[ ]” to the FDA if it deemed one was necessary. If the FDA agreed, it would work with both200 the brand-name and generic drug manufacturer to change the label, and then the generic drug manufacturer could use the CBE process to adhere to the FDA’s direction and match the brand-name drug’s change in label. 201 This interpretation does not make sense with the language in § 201.80(e). 202 194. See Claire R. Kelly, The Brand X Liberation: Doing Away with Chevron’s Second Step as Well as Other Doctrines of Deference, 44 U.C. DAVIS L. REV. 151, 204-06 (2010) (discussing a review of district courts showing they applied the doctrine in roughly half of the applicable cases from 2008-2009). 195. PLIVA, Inc. v. Mensing, 131 S. Ct. 2567, 2575-76 (2011). 196. See 21 C.F.R. §§ 201.56 to -.57(e) (1998). 197. Id. §§ 201.56 to -.57(c)(6), 201.80(e) (2007). 198. Compare id. § 201.57(e) (1998), with id. § 201.80(e) (2011). 199. Id. § 201.80(e) (2011); PLIVA, 131 S. Ct. at 2576. 200. PLIVA, 131 S. Ct. at 2576. 201. Id. at 2575-76. 202. See Kellogg v. Wyeth, 612 F. Supp. 2d 421, 435-36 (D. Vt. 2008). 2013] PLIVA, INC. V. M ENSING 191 The dictionary definition of “revise” means “to make a new, amended, improved, or up-to-date version of.” In contrast, “propose” means “to form or put forward203 a plan or intention,” or “to engage in talk or discussion.” The two verbs cannot204 be reconciled because putting forth a new plan or intention, in this case an intention to strengthen a warning, does not result in any change of the warning. Proposing a strengthening of a warning is strictly preparatory. Actually changing the warning results in a revision. In other words, to “propose” occurs before the action of changing the warning, and to “revise” occurs during and after the change in the warning. As the Court pointed out in Christensen, the word “may” is clearly permissive, and similarly here, the word “revise” clearly means an actual change. The requirement under former § 201.57(e) and current § 201.80(e) requires a revision, not a discussion about a revision. 205 Furthermore, the amendment to m ove generic drug manufacturers’ label content and form requirements to § 201.80 affirms the argument that “revise” really m eans “revise” and not “to propose.” Again, under § 201.80(e) the FDA kept the same language regarding revisions under the warning section of the labeling as it had when the regulation applied to both brand-name and generic drugs under § 201.57(e) before the amendm ent. If the FDA interpreted206 “revise” to mean “revise” for brand-name manufacturers but to mean “propose” for generic manufacturers, then the FDA would have changed the wording in its 2006 amendments. In other words, surely the FDA, when given the chance to rewrite its regulation in which it interprets a single word to have two different meanings for two different parties, would take advantage of that opportunity and apply the actual m eaning and words it had intended from the start for each manufacturer. A counterargument to this analysis is that § 201.80 does not necessarily apply to generic drug manufacturers, but rather it applies strictly to the generic drug’s label itself. In that respect, the regulation does not address who should implement the revision but rather only provides that the label needs to be changed when reasonable evidence exists suggesting it should be updated with a new warning; however, the FDA’s interpretation undermines this argument. The FDA determined a generic drug manufacturer had a duty to report the need for a stronger warning based on the language in § 201.57(e) (now codified at § 201.80(e)). In support for this proposition, the FDA stated in its amicus brief207 to the Court that the language in § 201.57(e) “reflect[ed] the ‘central premise of federal drug regulation that the manufacturer bears responsibility for the content 203. MERRIAM-WEBSTER, http://www.merriam-webster.com/dictionary/revise (last visited Oct. 29, 2011). 204. MERRIAM-WEBSTER, http://www.merriam-webster.com/dictionary/propose (last visited Oct. 29, 2011). 205. See 21 C.F.R. § 201.57(e) (1998); id. § 201.80(e) (2011); Kellogg, 612 F. Supp. 2d at 436. 206. See 21 C.F.R. § 201.57(e) (1998); id. § 201.80(e) (2011). 207. PLIVA, Inc. v. Mensing, 131 S. Ct. 2567, 2576 (2011) (referring to the FDA’s interpretation of 21 C.F.R. § 201.57(e) (1998) (codified at 21 C.F.R. § 201.80(e) (2011)). 192 INDIANA LAW REVIEW [Vol. 46:173 of its label at all times.’” As this statement shows, the FDA interprets §208 201.57(e), and thus § 201.80(e), as applying to the manufacturer. In addition, other language within § 201.80(e) suggests the FDA is addressing generic drug manufacturers in regards to the requirements within the section. For example, the regulation explains that certain serious problems might need to be placed in a boxed warning and that the FDA itself will determine when it is necessary and where it will be placed on the label. This language suggests the other language209 within the section, specifically the language regarding the revising of labels, is directed to generic drug manufacturers because if it was not, the FDA would have stated that it would revise the label to add new warnings. In other words, it would be odd for the FDA explicitly to state its responsibilities in some parts of the section and in other parts state its responsibilities implicitly. Another counterargument is that the FDA construes § 201.80(e) to mean both a proposal and a revision, meaning that the generic drug manufacturer is supposed to propose a change in its label, and it actually accomplishes the revision by updating its label to match the brand-name drug. In other words, the generic drug manufacturer fulfills the mandate of revising its label per § 201.80(e) when it updates its label to match the brand-name drug label. Yet, this argument is undermined with the remaining language in the requirement. Again, § 201.80(e) mandates that the drug’s label “be revised . . . as soon as there is reasonable evidence of an association of a serious hazard with a drug.” If the210 revision applies within the context of updating the generic drug’s label to match the brand-name drug label, then “reasonable evidence” necessarily means the direction from the FDA to the generic drug manufacturer to change its label and/or the brand-name drug’s actual change to its label. It makes no sense for the FD A to refer to either action— its own order to do something or a brand-nam e drug manufacturer’s change in label— as “reasonable evidence.” Only two actions could prompt the generic drug manufacturer to update its label; thus, it is reasonable to expect the FDA to explicitly state those occurrences when a generic drug manufacturer needed to revise its label. In addition, § 314.97 of the FDA’s regulation requires generic drug manufacturers to follow the requirements under § 314.70 when submitting supplemental applications or other changes. Section 314.70 details how a211 generic drug manufacturer can make changes to its approved application. The212 regulation requires a generic drug manufacturer to notify the FDA of any changes via the prior approval process, CBE process, or in an annual report, depending on the type of change. It also requires that when a generic drug m anufacturer is213 208. Brief for the United States as Amicus Curiae Supporting Respondents at 12, PLIVA, Inc. v. Mensing, 131 S. Ct. 2567 (2011) (Nos. 09-993, 09-1039, 09-1501) (quoting Wyeth v. Levine, 555 U.S. 555, 570-71 (2009)). 209. 21 C.F.R. § 201.80(e) (2011). 210. Id. 211. Id. § 314.97. 212. Id. § 314.70. 213. Id. 2013] PLIVA, INC. V. M ENSING 193 making a change in accordance with an FDA regulation or guideline, the manufacturer is required to make the change in a way that provides for the least burdensome way of notifying the FDA of the change. In other words, if a214 generic drug manufacturer is making a change that falls under the CBE procedure, then it must make the change using that procedure. Thus, if a215 generic manufacturer had reasonable evidence of a hazard that required it to revise its warning on its drug’s label, as required by § 201.80(e), then to comply with § 314.70(a)(3) the generic manufacturer would have to submit the change to the FDA via the CBE process because it is a change that either adds or strengthens the warning, one of the changes falling under the CBE process. 216 Again, if the FD A ’s interpretation is correct that the generic drug manufacturer cannot make a change to its label with the CBE process unless to match its brand-name drug label and thus could only propose a change to the FDA upon reasonable evidence of hazard, then § 314.70(a)(3) is meaningless to generic drug manufacturers. The FDA’s interpretation means the generic drug217 manufacturer is not required to notify the FDA of a change in the least burdensome way. In fact, the FDA’s interpretation requires the generic drug manufacturer to notify the FDA of a change in the most burdensom e way, as evidenced by the requirement to work with the brand-name manufacturer to change its label first. Like the Court in Christensen found that to defer to the218 Secretary of Labor’s “interpretation ‘would be to . . . create de facto a new regulation,’” here too, the FDA’s interpretation creates a new regulation because it requires a unique process of notification for a particular section of a regulation, § 201.80(e), that is not found in the words of the regulation.219 The FDA bases its interpretation that a generic drug manufacturer can only use the CBE process to match a brand-name’s drug label on the definition of an abbreviated application, which includes the application under § 314.94 and supplements to it. Under § 314.94(a)(8)(iv), a generic drug’s label must be the220 same as its brand-name drug that it is using to gain approval. B ecause supplements are a part of the abbreviated application, the FDA interprets this to mean the generic’s drug label must always be the same as the brand-name drug’s label. 221 This interpretation, however, is undermined by language in the same section that the FDA purports requires the constant “sameness.” Under § 314.94(a)(8)(iv), generic and brand-name drug labels are allowed to have differences, one of which being “labeling revisions made to comply with current 214. Id. § 314.70(a)(3). 215. Id. 216. See id. § 314.70(c)(6)(iii)(A). 217. See Kellogg v. Wyeth, 612 F. Supp. 2d 421, 436 (D. Vt. 2008). 218. See PLIVA, Inc. v. Mensing, 131 S. Ct. 2567, 2575-76 (2011). 219. See Kellogg, 612 F. Supp. 2d at 436 (quoting Christensen v. Harris Cnty., 529 U.S. 576, 588 (2000)). 220. Brief for the United States, supra note 208, at 16; 21 C.F.R. § 314.3(b). 221. Brief for the United States, supra note 208, at 15-16; 21 C.F.R. § 314.3(b). 194 INDIANA LAW REVIEW [Vol. 46:173 FDA labeling guidelines or other guidance.” Section 201.80 is an effective222 FDA guideline that details labeling requirements, and prior to that § 201.57 was an effective guideline that detailed labeling requirements. In other words, the guideline dictating when a warning on a label needs to be revised has always been effective since at least 2001 when Gladys M ensing received generic metoclopramide. As a result, complying with § 201.80(e) is an allowed223 difference in labeling according to § 314.94(a)(8)(iv). In summary, the regulations regarding the revising of warnings for generic drug manufacturers are clearly unambiguous. Section 314.3 defines an abbreviated application that generic drug manufacturers use for approval as the application described in § 314.94 and supplements to it. Section 314.94 requires that the labeling for the generic drug be the same as the brand-name drug it is seeking approval under except for certain differences, such as labeling revisions done in compliance with other FDA guidelines. Section 201.80, an aforementioned FDA guideline, details the requirements for the content and form of generic drug labels, one of which being that the label must “be revised to include a warning as soon as there is reasonable evidence of an association of a serious hazard with a drug.” In making this change to the label, § 314.97224 requires generic drug manufacturers to follow the requirements for supplements to an approved application under § 314.70, which requires applicants to make a change in the least burdensome way as allowed under the section. Because adding or strengthening a warning is allowed under the CBE process, the generic manufacturer is required to use that process to make the change. 2. The FDA’s Interpretation Is Inconsistent with Its Regulation.— Even if the FD A ’s regulations are ambiguous, the FDA’s interpretation that a generic drug manufacturer could only use the CBE process to m atch a brand-name drug’s label is inconsistent with its regulations because it conflicts with a guiding225 principle of FDA regulation: to “protect the public health by ensuring that . . . drugs are safe and effective.” Congress empowered the Secretary of Health and226 Human Services to promulgate regulations to effectuate that purpose. As the227 FDA confirmed in PLIVA, and as the Supreme Court articulated in Levine, “a228 central premise” of FDA regulation in advancing this purpose is “that the manufacturer bears responsibility for the content of its label at all times.” The229 Court found this premise of FDA regulation evident in the CBE process because it put “ultimate responsibility” on the manufacturer for its label and provided a procedure in which it could add safety information before receiving FDA 222. 21 C.F.R. § 314.94(a)(8)(iv) (emphasis added). 223. See PLIVA, 131 S. Ct. at 2573; 21 C.F.R. § 201.57(e); id. § 201.80(e). 224. 21 C.F.R. § 201.80(e). 225. PLIVA, 131 S. Ct. at 2575. 226. 21 U.S.C. § 393(b)(2)(B) (2006). 227. Id. § 371(a). 228. PLIVA, 131 S. Ct. at 2576. 229. Wyeth v. Levine, 555 U.S. 555, 570-71 (2009). 2013] PLIVA, INC. V. M ENSING 195 approval. The premise is further supported under the FDA’s regulation §230 314.80, which requires both brand-name and generic drug manufacturers to review adverse drug experiences and report them to the FDA.231 If the FDA’s interpretation is correct, then a generic manufacturer effectively has no responsibility for the content of its label at any time because having responsibility for content means being held accountable for its substance. 232 Here, any inadequacy with the substance of the label’s content can be directed away from the generic drug manufacturer if it can show it matches the brand- name drug label. Thus, the FDA’s interpretation actually means a generic drug manufacturer only has responsibility in confirming that its label is the same as its brand-name drug counterpart at all times. If a generic manufacturer could use the CBE process unilaterally to update its warning in compliance with § 201.80(e), then at some point in time it would be true that the generic manufacturer has actual responsibility for its label content because it could make a change to its label. It is im plied in Levine that the Court found this ability to make a change to the label before FDA approval as a touchstone for responsibility.233 The responsibility of keeping the label the same as the brand-name drug does not advance the purpose of ensuring a drug is safe; in fact, this duty decreases a drug’s safety because it safeguards a generic manufacturer from liability and disincentives the generic manufacturer from proposing a change. Furthermore,234 if the brand-name drug leaves the m arket, as is often the case, the responsibility for providing adequate warnings is entirely on the FDA. This proposition goes235 against the Court’s finding in Levine that the reason the FDA puts the onus on manufacturers is because the FDA lacks the capability to oversee the 11,000 drugs on the market, and thus, the manufacturers are in the best position to monitor the information regarding their drugs. Therefore, not only is the236 FD A ’s interpretation inconsistent with its regulations, but it is also inconsistent with the FDCA. Furtherm ore, the FDA’s interpretation of an “ongoing . . . sameness” in the warning labels is inconsistent with the regulation because it is contrary to the237 section of federal regulation describing the content of an abbreviated application. As stated before, § 314.94(a)(8)(iv) allows many differences between the generic drug label and the brand-name drug label, including differences in “expiration date, formulation, bioavailability, . . . pharmacokinetics, [and] labeling revisions 230. Id. at 571. 231. 21 C.F.R. § 314.80 (2011); id. § 314.98. 232. See MERRIAM-WEBSTER, http://www.merriam-webster.com/dictionary/responsible (last visited Nov. 5, 2011) (defining “Responsible” as “liable to be called to account as the primary cause, motive, or agent” and “able to answer for one’s conduct and obligations”). 233. Levine, 555 U.S. at 571. 234. PLIVA, Inc. v. Mensing, 131 S. Ct. 2567, 2592-93 (2011) (Sotomayor, J., dissenting). 235. See id. at 2593. 236. Levine, 555 U.S. at 578-79. 237. PLIVA, 131 S. Ct. at 2574-75. 196 INDIANA LAW REVIEW [Vol. 46:173 made to comply with current FDA labeling guidelines or other guidance.” 238 Besides the general tenor of non-sameness this language exudes, § 314.94(a)(8)(iv) makes reference to revisions in compliance with current FDA labeling guidelines, which implies changes throughout its time on the market, as what is “current” is constantly evolving. Section 314.94(a)(8)(iv) precedes this list of allowed differences by explicitly saying, “differences between the [generic drug manufacturer’s] proposed labeling and labeling approved for the [brand- name] drug may include.” Just as the Court in Christensen found the word239 “may” is unambiguously permissive, here, too, “may” indicates that the FDA240 can disapprove of a label revision but is in no sense required to disapprove of it. Thus, to say there is an ongoing sameness requirement is contrary to the language in § 314.94(a)(8)(iv). B. The Supremacy Clause Containing a Non Obstante Provision Unlike Levine, the Court did not begin its analysis with a presumption against preemption. Instead, a plurality of the Court relied on a theory that the241 Supremacy Clause contained a non obstante provision— a theory presented in a law review article written by University of Virginia law professor Caleb Nelson. The plurality used this theory to bolster their conclusion that the Court242 should not consider the actions of the FDA in determining whether it was possible for a generic drug manufacturer to comply with both state and federal law.243 According to Professor Nelson, two legal principles of legislative drafting were well established by the late eighteenth century: one, that newer laws abrogated older, conflicting laws, and two, a presumption against implied repeals, meaning a new law should not be read to conflict with an older law if it could be harmonized. The presumption against implied repeals reflected the courts’244 traditional reluctance to find that a new law repealed an older law. These two245 principles conflicted when state legislatures actually wanted the new law to abrogate conflicting laws, and they did not want to distort the new law so as to harmonize it with the older law. To solve this problem, state legislatures would246 add a non obstante clause, taken from the Latin word for “notwithstanding,” to the new law that said it would apply “notwithstanding any provisions to the contrary in prior laws.” A non obstante clause informed courts that the state247 238. 21 C.F.R. § 314.94 (2011). 239. Id. § 314.94(a)(8)(iv) (emphasis added). 240. Christensen v. Harris Cnty., 529 U.S. 576, 588 (2000). 241. See PLIVA, 131 S. Ct. at 2575-81; Levine, 555 U.S. at 565. 242. PLIVA, 131 S. Ct. at 2579-80 (plurality opinion); see generally Nelson, supra note 76. 243. PLIVA, 131 S. Ct. at 2579-80 (plurality opinion). 244. Nelson, supra note 76, at 235-41. 245. Id. at 241. 246. Id. 247. Id. at 238-41. 2013] PLIVA, INC. V. M ENSING 197 legislature did not want it to try and harmonize the new law with the older law. 248 In other words, a non obstante clause told the court not to apply the presumption against implied repeals and instead give the new law its ordinary meaning. 249 The Supremacy Clause of the United States Constitution states that the “Constitution, and the Laws of the United States . . . and all Treaties . . . shall be the supreme Law of the Land; and the Judges in every State shall be bound thereby, any Thing in the Constitution or Laws of any State to the Contrary notwithstanding.” This clause m eans state laws that interfere with, or are250 contrary to, federal law must yield to federal law. In other words, federal law251 nullifies contrary state law. Professor N elson argues that the end phrase of the252 Suprem acy Clause is a non obstante provision and is intended to fulfill its traditional purpose— telling the “court[] not to apply the . . . presumption against implied repeals.” 253 Professor Nelson first notes that the Court has interpreted the Supremacy Clause to mean that federal law is a part of state law and thus forms one jurisprudence within the state. Second, the section of the Supremacy Clause254 describing federal law as supreme “substitute[d] a federal rule of priority for the traditional temporal rule of priority.” This means that rather than having the255 temporal rule of priority in which new laws abrogate old, conflicting laws, the federal rule of priority provides that a federal law will reign supreme over a state law, even if a state law comes later in time. Because the Supremacy Clause256 means that federal law and state law form one jurisprudence, and federal law always reigns supreme, “courts are always bound to apply the federal portion of in-state law.” 257 Although the Suprem acy Clause mandates courts to apply federal law, the federal rule of priority occurs only when state law forces the court to choose between them. Like the test for implied repeals, the test for preemption is258 whether both state law and federal law can be followed or whether they create contradictory rules. As a result, the last section of the Supremacy Clause works259 like a traditional non obstante provision because without it courts might seek to avoid finding a contradiction. Rather than having each federal statute or treaty260 contain a non obstante provision, the founders established a “global non obstante 248. Id. at 241-42. 249. Id. 250. U.S. CONST. art. VI, cl. 2. 251. Gibbons v. Ogden, 22 U.S. 1, 210-11 (1824). 252. Id. at 210. 253. Nelson, supra note 76, at 232, 245-46. 254. Id. at 246-49, 246 n.61. 255. Id. at 250. 256. Id. at 250-51. 257. Id. at 252 (internal quotation marks omitted). 258. Id. at 251. 259. Id. at 252. 260. Id. at 255. 198 INDIANA LAW REVIEW [Vol. 46:173 provision” in the Supremacy Clause. Nelson argues the non obstante provision261 of the Supremacy C lause forces the Court to abandon its longstanding presumption against preemption because it actually instructs the Court not to try to harmonize state and federal law. 262 As stated before, the plurality in P LIV A used this theory of the Supremacy Clause containing a non obstante provision to support its finding of impossibility. The plurality reasoned that because the non obstante provision263 of the Supremacy Clause instructed the Court not to distort federal law so as to reconcile it with state law, the Court should “look no further than the ordinary meaning of [the] federal law.” The plurality thought that considering the264 possible actions of the FDA and brand-name drug manufacturers in response to a generic drug manufacturer’s proposal for stronger warnings went beyond the ordinary meaning of the federal law because the supremacy of federal law would always be dependent on these third-party actions. Thus, the non obstante265 provision supported the Court’s conclusion that determinations of impossibility should depend only on whether a generic drug manufacturer could independently do what state law required of it. Because a generic drug manufacturer could266 not independently strengthen the warnings on its label without prior approval from the FDA, the Court determined it could not satisfy its state law duty to have a safer label, and thus state law was preempted.267 As the dissent in PLIV A points out, the plurality’s new theory of the Supremacy Clause goes against more than a half century of precedent of applying a presumption against preemption. The dissent believed that if the Court had268 used its typical presumption against preemption, as it did in Levine, then it would have not found impossibility. Although Professor Nelson’s article is very269 persuasive, the plurality missed several key points within the article. Of course, the majority is free to interpret the article as it pleases because it is not an authoritative document the m ajority must follow to the letter, like the Constitution; however, when the majority is effectively overturning a half-century of precedent, and potentially denying plaintiffs suffering a debilitating neurological disorder any remedy, based on one law review article, it is hard to argue that the author’s other views contained in the article are not important or informative. W ith this in mind, contrary to the dissent’s belief, if the plurality had closely followed Professor Nelson’s article on which it so heavily relied in reaching its decision, the plurality should not have found impossibility, even viewing the Supremacy Clause as containing a non obstante provision. 261. Id. 262. Id. at 304. 263. PLIVA, Inc. v. Mensing, 131 S. Ct. 2567, 2579-80 (2011) (plurality opinion). 264. Id. at 2580 (internal quotation marks omitted). 265. Id. 266. Id. 267. Id. at 2581 (majority opinion). 268. Id. at 2590-91 (Sotomayor, J., dissenting). 269. Id. at 2591-92. 2013] PLIVA, INC. V. M ENSING 199 The overall purpose of Professor Nelson’s article is to simplify the Court’s preemption analysis. He sets out to convince readers that the Court’s current270 preemption analysis, which contains three types of preemption (express, field, and implied), is confusing and preempts too much state law. Nelson argues that,271 because the Supremacy Clause contains a non obstante provision, the doctrine of preemption should work like the “traditional doctrine of repeals,” and as such, the Court should not use a presumption against preemption. Thus, the only test for272 preemption is that the “[c]ourts are required to disregard state law if, but only if, it contradicts a rule validly established by federal law.” In other words, courts273 should disregard state law only in situations when they are forced to choose between both laws.274 Nelson refers to this test as the “logical-contradiction test,” and he warns that it is not the sam e as the physical impossibility test that the Court uses to find conflict preemption under the umbrella of im plied preemption. Nelson states275 that many situations will prove physically possible, while still forcing the Court to choose between laws, such as a federal law giving a person a right to join a union and a state law prohibiting that person from joining a union. Both laws276 may physically be possible to comply with if the person does not join a union, but a court is still forced to choose between them because enforcing the state law would require disregarding the federal law.277 In PLIVA, the Court does not use the notion of the Supremacy Clause containing a non obstante provision to overhaul its preemption analysis, but instead uses it only to buttress its new theory that proving physical impossibility within the context of an implied preemption analysis requires a determination of whether a party can take unilateral action. The Court does not use the non278 obstante provision as a springboard for finding a contradiction between the laws, but instead uses it to establish how to determine whether something is physically impossible. The Court does exactly what Nelson cautions not to do by279 confusing the concept of physical impossibility with the concept of contradiction. If the Court had used the logical-contradiction test that necessarily follows from the Supremacy Clause’s mandate to apply federal law, the Court would not have found a contradiction, and thus, no preemption of state law. For starters, federal law requires a drug m anufacturer to have adequate warnings on its labels. As the Court pointed out, M innesota and Louisiana state law also280 270. Nelson, supra note 76, at 231, 260. 271. Id. at 225-35. 272. Id. at 232, 245-46. 273. Id. at 260. 274. Id. at 251, 261. 275. Id. at 260-61. 276. Id. 277. Id. at 261. 278. PLIVA, Inc. v. Mensing, 131 S. Ct. 2567, 2589-90 (2011) (Sotomayor, J., dissenting). 279. Id. at 2579-80 (plurality opinion). 280. 21 U.S.C. § 352(a), (f) (2006); 21 C.F.R. § 201.57(e) (1998); id. § 201.80(e) (2011). 200 INDIANA LAW REVIEW [Vol. 46:173 require that a drug manufacturer have labels with adequate warnings. No281 contradiction exists here because clearly both federal and state laws express the same requirement. In other words, the Court is not forced to choose between state law and federal law. Furthermore, assuming that federal law does prevent a generic drug manufacturer from using the CBE process to unilaterally change its drug label, and instead requires it to propose a label change to the FDA, this still does not reveal a contradiction with state law. The Court found that the state law duty required generic drug manufacturers “to use a different, stronger label than the label they actually used.” Yet, this duty does not force the Court to choose282 between both federal and state law. State law only mandates that a drug manufacturer have an adequate label and does not prescribe how a manufacturer is to affect that change. Federal law, on the other hand, does instruct a283 manufacturer as to how it can m ake a change to its label. In contrast, a284 contradiction between state law and federal law would exist if the state law required a manufacturer to not only have adequate warnings, but it also required the manufacturer to unilaterally update its label to address any new or stronger warnings. This would force the Court to choose between the state law and federal law because, according to the Court, federal law would not allow a generic drug manufacturer to unilaterally update its label.285 Although the Court states there is a contradiction between the state law duty for adequate labeling and the federal law duty of sameness for a generic drug manufacturer regarding its label, this is not a contradiction because it is not a286 duty of sameness no matter what. The duty of sameness, according to the FDA, is not telling a m anufacturer that despite whatever evidence it finds that the current label is inadequate and not safe that it must keep the label the same and sit idle. A s the FDA argues, and the Court assumes to be true, the FDA287 requires the generic drug manufacturer to seek help from the FDA in effecting a change to make the label adequate. This is how the FDA argues that a generic288 drug manufacturer fulfills its other duty to have adequate warnings under 21 U.S.C. § 352(f)(2) regarding misbranding. If a generic drug manufacturer’s289 action of seeking help from the FDA to update a warning is good enough to fulfill 281. PLIVA, 131 S. Ct. at 2577; see also MICHAEL K. STEENSON, MINNESOTA PRACTICE SERIES: PRODUCTS LIABILITY LAW §§ 4.1, 16.6 (West 2011); LA. REV. STAT. ANN. §§ 9:2800.54, preempted by Green v. BDI Pharm., 803 So. 2d 68 (La. Ct. App. 2001), 9:2800.57 (West 2011), preempted in part by PLIVA, 131 S. Ct. 2567. 282. PLIVA, 131 S. Ct. at 2577. 283. See STEENSON, supra note 281, at §§ 4.1, 16.6; LA. REV. STAT. ANN. §§ 9:2800.54, 9:2800.57 (West 2011). 284. 21 C.F.R. § 314.70 (2011). 285. See PLIVA, 131 S. Ct. at 2575, 2578. 286. PLIVA, 131 S. Ct. at 2576-78. 287. Id. at 2576. 288. Id. 289. Id. 2013] PLIVA, INC. V. M ENSING 201 its federal duty of having adequate warnings, why would that not be good enough to fulfill the same state duty to have adequate warnings? Although Professor Nelson argues the Supremacy Clause’s non obstante provision instructs courts not to use a presumption against preemption, he makes a qualification regarding this premise. He warns about taking the non obstante provision’s mandate “too far.” T he goal in deciding preemption cases is “to290 give effect to congressional intent.” Although the non obstante provision of the291 Suprem acy Clause instructs courts not to apply a presumption against preemption, it does not prevent courts from applying other rules of statutory interpretation. Nelson points out that even without a presumption, “there may292 well be other reasons to believe that Congress did not intend a particular statute to have much preemptive effect,” as evidenced, for example, in either the context of the federal law or even within a “pattern of legislation.” 293 One reason Congress might not have intended the FD C A or the federal regulations implementing it to have a preemptive effect on state law is that it did not want to deny a person any compensation for harm caused by an inadequate warning. As the dissent in PLIVA points out, 75% of prescriptions in the U nited States are for generic drugs. W ith so many consumers using generic drugs, it294 is not plausible that Congress intended to deny such a large portion of the population a remedy if harmed by an unsafe or ineffective drug. M oreover, the purpose of the Hatch-W axman Amendments is to increase Americans’ accessibility to inexpensive drugs. W ith no right to compensation for an injury,295 people might be inclined to seek expensive brand-name drugs, thus decreasing the demand for generic drugs. This in turn could have an adverse effect on296 consumer health as many consumers would likely curb their medication plans to adjust to the increase in prices. A s N elson suggests, other evidence that Congress did not intend to have preem ptive effect on state law is observed through its pattern of legislation. A s the Court stated in Levine, Congress enacted the FDCA “to bolster consum er protection against harmful products.” The Court found further support for this297 proposition in the 1962 amendments to the FDCA, which shifted the burden of proof of a drug’s safety to the manufacturers; this strengthened consumer298 protection because manufacturers have better access to information regarding 290. Nelson, supra note 76, at 294. 291. See id. at 292. 292. Id. at 294. 293. Id. 294. PLIVA, Inc. v. Mensing, 131 S. Ct. 2567, 2583 (2011) (Sotomayor, J., dissenting). 295. Id.; see also Mensing v. Wyeth, 562 F. Supp. 2d 1056, 1059-60 (D. Minn. 2008) (stating that the “primary purpose” of the Hatch-Waxman Amendments “was to increase the availability of low cost generic drugs” via the abbreviated application process), rev’d, 588 F.3d 1056 (8th Cir. 2011), vacated in part, reinstated in part by Mensing v. Wyeth, Inc., 658 F.3d 867 (8th Cir. 2011). 296. PLIVA, 131 S. Ct. at 2592-93 (Sotomayor, J., dissenting). 297. Wyeth v. Levine, 555 U.S. 555, 574 (2009). 298. Id. at 567. 202 INDIANA LAW REVIEW [Vol. 46:173 their drugs than the FDA. This, in conjunction with state failure-to-warn299 claims, which provide the incentive for manufacturers to pay attention to data and studies regarding their drugs, increases the chances of discovering and disclosing drug risks. The Court stated that the importance of the state failure-to-warn300 claims in this process was evidenced within the same 1962 amendments to the FDCA by the enactment of the saving clause that stated, “a provision of state law would only be invalidated upon a direct and positive conflict with the FDCA.” 301 Ultimately, the Court found that because Congress had never enacted an express preemption provision, like it had done for medical devices, along with its apparent awareness of state failure-to-warn claims, Congress did not intend for federal law to preempt state law in this area. 302 Finally, the Court’s acceptance of the FDA’s interpretation of a constant duty of sameness in PLIVA is inconsistent with its established opinion in Levine that the Congressional purpose of the FDCA is to bolster consumer protection. In Levine, the Court rejected the FD A ’s interpretation that the point of the FDCA was to “entrust an expert agency to make drug labeling decisions that strike a balance between competing objectives.” H owever, an interpretation requiring303 generic manufacturers to have the same label at all times as their brand-nam e counterpart supports the FDA’s interpretation in Levine because it effectively makes the FDA the decision maker for generic drug labeling. The Court arguably is now interpreting the point of the FDCA, at least as it applies to generic drug manufacturers, as to establish an expert, decision-making agency. Again, all the non obstante provision is telling the Court is “not to assume automatically that Congress did not want to displace state law[ ].” If the304 majority in PLIVA had kept this in the forefront of its analysis, the other normal tools of statutory construction would have led the majority to the same conclusion reached in Levine. C ONCLUSION A lthough Justice Thomas believed that federal regulation had dealt the plaintiffs an “unfortunate hand,” the reality is the judicial deference given to305 the FDA’s interpretation of its regulations dealt the plaintiffs the unfortunate hand. This is evident in the inconsistency between the Court’s interpretation of state law and the FDA’s interpretation of its regulations regarding label content. The Court stated that M innesota law required that “‘where the manufacturer . . . of a product has actual or constructive knowledge of danger to users, the . . . 299. Id. at 578-79. 300. Id. 301. See id. at 567 (internal quotation marks omitted). 302. Id. at 574-75. 303. Id. at 573 (internal quotation marks omitted). 304. Nelson, supra note 76, at 295. 305. PLIVA, Inc. v. Mensing, 131 S. Ct. 2567, 2581 (2011). 2013] PLIVA, INC. V. M ENSING 203 manufacturer has a duty to give warning of such dangers.’” The Court stated306 that under the applicable Louisiana law, “‘a manufacturer's duty to warn includes a duty to provide adequate instructions for safe use of a product.’” From this307 language, the C ourt interpreted M innesota and Louisiana state law as “demand[ing] a safer label.” Both laws actually use somewhat broad language308 in regards to the manufacturer’s duty: M innesota’s duty “to give” a warning and Louisiana’s duty “to provide” a warning. One can imagine the many ways a drug manufacturer could fulfill this duty. For example, a drug manufacturer could post the added warning on its website. Yet, the Court infers from this language a commandment to literally have a different label with nothing short of that satisfying the duty. On the other hand, FDA regulations specifically applying309 to label content require a manufacturer to revise its label. Revise has a narrow310 meaning, yet the Court defers to the interpretation of the FDA and gives it a broad meaning that includes merely proposing a change, which, as stated earlier, does not even fall within the definition of revise. It seems apparent then that had the311 Court interpreted the FDA regulations without any deference, the Court would have interpreted “revise” to mean generic drug m anufacturers had to literally change the label. In other words, if “to give” and “to provide” are interpreted by the Court to be commands to have a different label, then surely “to revise” would mean the same thing. The Court apparently finds value in the legal teachings of the eighteenth and early nineteenth century, as shown in its reliance on Professor Nelson’s theory of the Supremacy Clause. In expounding the meaning of the Supremacy Clause312 to the plaintiffs, however, the majority could have also harkened back to Justice M arshall’s declaration in M arbury v. M adison to explain the Court’s role in313 preemption cases— that “[i]t is emphatically the province and duty of the judicial 306. Id. at 2573 (omissions in original) (quoting Frey v. Montgomery Ward & Co., 258 N.W.2d 782, 788 (Minn.1977)); see also STEENSON, supra note 281, at §§ 4.1, 16.6 (stating that the Minnesota Supreme Court has embraced the principle that “[o]ne who supplies directly or through a third person a chattel for another to use, is subject to liability . . . for bodily harm caused by the use of the chattel . . . if the supplier . . . fails to exercise reasonable care to inform them of its dangerous condition or of the facts, which make it likely to be so”). 307. PLIVA, 131 S. Ct. at 2573 (quoting Stahl v. Novartis Pharm. Corp., 283 F.3d 254, 269–70 (5th Cir. 2002); see also LA. REV. STAT. ANN. §§ 9:2800.54, preempted by Green v. BDI Pharm., 803 So. 2d 68 (La. Ct. App. 2001), 9:2800.57 (West 2011) (“A product is unreasonably dangerous because an adequate warning about the product has not been provided if . . . the product possessed a characteristic that may cause damage and the manufacturer failed to use reasonable care to provide an adequate warning of such characteristic.”), preempted in part by PLIVA, Inc. v. Mensing, 131 S. Ct. 2567 (2011). 308. PLIVA, 131 S. Ct. at 2578 (emphasis added). 309. See id. 310. 21 C.F.R. § 201.57(e) (1998); id. § 201.80(e) (2011). 311. See PLIVA, 131 S. Ct. at 2576-78. 312. See id. at 2579-80. 313. 5 U.S. (1 Cranch) 137 (1803). 204 INDIANA LAW REVIEW [Vol. 46:173 department to say what the law is.” Unfortunately, this role seems to be greatly314 diminished when the Court exercises such a high degree of deference. It seems especially harmful and contradictory to the Court’s stated role when it is very likely the Court would have reached a different interpretation. This begs the question whether the Court is saying what the law is, or merely just saying what the law is not. There is a difference, and unfortunately Gladys M ensing and Julie Demahy learned this the hard way. 314. Id. at 177.