Friday, December 29, 2006

Territoriality and Punitive Damages

It’s often hard to know how seriously to take the issue of sovereign state power any more, with various attempts at nationwide class actions and arguments that one state can pass a law – like a consumer protection statute – and expect it to be applied to transactions that take place on the other side of the country.

But every now and then an opinion comes down that makes us stop and think that maybe there’s still something to that rickety 217-year-old structure after all. See Johansen v. Combustion Engineering, Inc., 170 F.3d 1320, 1333 (11th Cir. 1999) (“punitive damages must be based upon conduct in a single state – the state where the tortuous conduct occurred”); In Re Brand Name Prescription Drugs Antitrust Litigation, 123 F.3d 599, 613 (7th Cir. 1997) (“[a] state cannot regulate [prescription drug] sales that take place wholly outside it”); Yu v. Signet Bank/Virginia, 82 Cal. Rptr. 2d 304, 314 (Cal. App. 1999) (federalism “prohibit[s] an award of punitive damages herein to punish or deter respondents’ conduct with respect to consumers in states other than California”); American Libraries Ass’n v. Pataki, 969 F. Supp. 160, 176 (S.D.N.Y. 1997) (“comity. . .mandates that one state not expand its regulatory powers in a manner that encroaches upon the sovereignty of its fellow states”); Geressy v. Digital Equipment Corp., 950 F. Supp. 519, 521 (E.D.N.Y. 1997) (“punitive damage can not be awarded to punish or deter acts in other states which do not affect the forum state”).

We recently became aware of another such case, Republic Services, Inc. v. Liberty Mutual Insurance Co., 2006 WL 3500875 (E.D. Ky. Dec. 1, 2006). This case has nothing to do with drugs, medical devices, or even personal injury. It involved a rather run-of-the-mill contract dispute, dressed up like a tort case, about whether certain workers compensation claims had been mishandled. Because of the tort theories, however, punitive damages were asserted, and that’s where things got interesting.

For some reason the case was filed in Kentucky, even though very few of the affected claims involved Kentucky claimants. For the big claims, it turned out that there weren’t any at all. Only five Kentucky claims met the dollar criteria that the plaintiffs put on their own theories. Discovery took place and it turned out that “none [of those five] resulted in any disallowance, or damage” to the plaintiff. 2006 WL 3500875, at *2.

Relying upon a less-well-known portion of the celebrated decision in State Farm Mutual Automobile Insurance Co. v. Campbell, 538 U.S. 408 (2003), the court in Republic Services dismissed the punitive damages claim as a matter of law because none of the challenged conduct involved the state in which the suit was pending. The court followed the federalist reasoning in Campbell to its logical conclusion – “the jury in this case cannot award punitive damages against [defendants] because Kentucky does not have a legitimate interest in punishing the [defendant’s] extraterritorial conduct.” 2006 WL 3500875, at *2.

Campbell had reaffirmed what the Court termed a “basic principle of federalism” that each state only has the power to punish conduct occurring within its borders, and cannot act as an officious intermeddler against conduct in other, co-equal states. “[E]ach State may make its own reasoned judgment about what conduct is permitted or proscribed within its borders, and each State alone can determine what measure of punishment, if any, to impose on a defendant who acts within its jurisdiction.” Campbell, 538 U.S. at 422. Thus, to permit one state “to operate beyond the jurisdiction of that State” and to award punitive damages against conduct that occurred entirely outside its boundaries would “throw[] down the constitutional barriers by which all the States are restricted within the orbits of their lawful authority.” Id. at 421 (citation and quotation marks omitted).

That was enough for the court in Republic Services. Because there were no disallowances or damages in the five Kentucky claims, “there [wa]s simply no specific proof that any of the alleged wrongdoing occurred in Kentucky.” Without in-state conduct, there was no basis for a court sitting in Kentucky to punish conduct elsewhere. Thus the court granted summary judgment and dismissed the punitive damages claim altogether. 2006 WL 3500875, at *3-4.

That’s of interest to us because of the increased aggregation of pharmaceutical and medical device product liability litigation. Multidistrict litigations and similar agglomerations are created – and plaintiffs nationwide are drawn to these litigation “honeypots.” The hope is that “safety in numbers” will allow plaintiffs to keep weak cases alive, avoid having to much work throughout protracted litigation, and generally use the fact of aggregation to increase settlement value.

Especially after Lexecon Inc. v. Milberg Weiss Bershad Hynes & Lerach, 523 U.S. 26 (1998), prevented plaintiffs in transferred MDL cases from threatening to take cases to trial in the transferee jurisdiction, there’ve been quite a few filings that make no geographic sense. Example: Arizona plaintiff brings suit against New Jersey company in a federal court in Louisiana. Nothing in the facts concerning either the plaintiff or the defendant has anything to do with Louisiana, but because that’s where the multidistrict litigation is situated, that’s where suit is brought.

There’s been very little anyone on the defense side can do as a practical matter to deter this sort of piling on – and to be sure there are good logistical reasons why some defendants positively encourage it in some cases. The significance of the Republic Services case, and of the territorial analysis in Campbell generally, is that there's at least a potential price to be paid for the luxury of filing in a far-flung jurisdiction simply because other plaintiffs have. That price can be preclusion of claims for punitive damages.

Wednesday, December 27, 2006

Amending California's "Mini MDL" Statute

Readers of this blog are familiar with the MDL Panel. And readers probably know that some, but not all, states have “mini MDL” procedures for coordinating cases in state court systems. This post criticizes, and proposes an amendment to, one aspect of California’s “mini MDL” statute.

California's process for creating statewide coordinated proceedings is among the most highly developed in the country. Unlike the process in many states, California's "mini MDL" structure was created by statute. The Code of Civil Procedure permits parties to file a petition with the chair of the Judicial Council of California asking that cases pending in different courts and sharing a common question of law or fact be coordinated. See Cal. Code Civ. Proc. §§ 404-404.9. Code of Civil Procedure § 404.1 lists seven factors to be considered in making the coordination decision.

The California statute incorrectly requires the coordination motion judge to consider one irrelevant, and potentially harmful, factor in deciding whether to coordinate cases — “whether the common question of fact or law is predominating and significant to the litigation.” See Cal. Code Civ. Proc. § 404.1. In the federal system, the MDL Panel is empowered to coordinate "civil actions involving one or more common questions of fact." See 28 U.S.C. § 1407. The statutory language thus permits coordination if the cases contain as few as one common question of fact. Although the MDL Panel has wavered on the issue, the Panel has never held that common questions must predominate to allow coordination. See, e.g., In re Diet Drugs (Phentermine, Fenfluramine, Dexfenfluramine) Prods. Liab. Litig., 990 F. Supp. 834, 835-36 (J.P.M.L. 1998) (rejecting the argument that common questions failed to predominate, but not holding that common questions must predominate). In practice, the MDL Panel seems to balance the existence of common questions with other considerations. As more convenience and fairness is achieved through transfer, less commonality is required. See, e.g., In re Westinghouse Electric Corp. Uranium Contracts Litig., 405 F. Supp. 316, 319 (J.P.M.L. 1975).

The standard for certifying a lawsuit as a class action is, of course, quite different. Under Federal Rule of Civil Procedure 23(b)(3), a court will certify a class action only if "the court finds that the questions of law or fact common to the members of the class predominate over any questions affecting only individual members." Fed. R. Civ. P. 23(b)(3).

The distinction between cases that contain a single common question of fact, on the one hand, and predominating common questions, on the other hand, makes a difference. The MDL Panel frequently transfers cases for coordinated proceedings because the cases contain one or more common questions of fact, but, because common questions do not predominate, the cases cannot be certified as class actions. See, e.g., In re Bridgestone/Firestone, Inc. Tires Prods. Liab. Litig., 288 F.3d 1012 (7th Cir. 2002) (reversing order certifying class in MDL proceeding). This result is entirely proper: There are many types of cases, such as product liability cases seeking to recover for personal injuries, in which federal courts almost never certify classes. See, e.g., In re American Med. Sys., Inc., 75 F.3d 1069 (6th Cir. 1996). Even though those cases cannot be certified as class actions, however, transferring the cases to a single court for coordinated discovery proceedings might nonetheless promote efficiency. Documents would be produced only once; corporate witnesses would be deposed only once; interrogatories would be asked and answered only once; and so on. Even though the cases are not suitable for class certification, the parties and the judicial system might well benefit from coordinated pretrial proceedings.
The standards governing most state "mini MDL" processes (whether imposed by statutes, court rules, or ad hoc procedures) mirror the federal system: Courts do not assess whether common questions predominate when deciding whether to coordinate cases. See generally Mark Herrmann, Geoffrey J. Ritts & Katherine Larson, Statewide Coordinated Proceedings: State Court Analogues to the Federal MDL Process (2d rev. ed. 2004) (surveying the law in all 50 states).

The California mini MDL statute, however, is different. California Code of Civil Procedure § 404.1 requires the coordination motion judge to consider “whether the common question of fact or law is predominating. . . ." (Emphasis added.) The California mini MDL statute thus interposes at the coordination stage a legal issue that is irrelevant to whether the cases should be coordinated and relevant only to the later, independent question whether putative class actions should be certified.

California's statutory flaw creates two related problems. First, defendants who favor coordination but oppose class certification may be reluctant to invoke the California coordination statute. Defendants may fear that the arguments they are required to make to seek coordination may later be used against them at a class certification hearing. If this concern discourages defendants from invoking the coordination process, then cases in which coordination would promote judicial efficiency will not be coordinated, wasting both the parties' and judicial resources.

On the other hand, if defendants do seek coordination in these cases, then plaintiffs may (as they have in past litigation) use the defense arguments pressed at the coordination stage at a later class certification hearing. If a judge were to grant class certification based on a defendant's supposed admission at the coordination stage that common questions "predominate," then a class might be certified in an inappropriate case. The resulting proceeding could harm either defendants who find themselves defending against improperly aggregated claims or absent class members who find themselves bound to a judgment tried by an inappropriate class representative.

There is no reason for California to interject the question whether common questions "predominate" into the process for seeking coordinated proceedings. The legislature should amend California Code of Civil Procedure § 404.1 to delete that one factor, thus assuring that parties will not be discouraged from seeking coordination in appropriate cases and that the California state judicial system will achieve maximum efficiency when handling related cases.

Wednesday, December 20, 2006

Preemption of "Black Box" Warning Claims

One of the high-profile ways in which the FDA has been strengthening product warnings lately has been to require more “black box” labeling – the FDA’s strongest form of warnings. Among other consequences, products bearing black box warnings cannot be advertised directly to consumers. Thus, the Agency’s decision to require a black box warning on a drug or medical device is a big deal. That’s the way it's always been. Nobody but the FDA has ever - in history - ordered a black box warning. See Beach, “Black Box Warnings In Prescription Drug Labeling: Results of a Survey 0f 206 Drugs,” 53 Food & Drug L.J. 403 (1998) (reviewing 375 black box warnings concerning 206 drugs; determining that every one of them was ordered by FDA).

Predictably, the plaintiffs’ litigation response to this type of FDA activity has been to second guess it. “It’s about time,” the refrain goes. “The FDA should have done this earlier – before I got hurt.” It turns out that the FDA anticipated just such second guessing from the outset, when it first set up the current black box warning regime. Twenty-seven years ago, long before plaintiffs are willing to admit that the FDA ever favored preemption, the Agency came down hard on the side of administrative exclusivity. Only the FDA, and not Monday-morning quarterbacking plaintiffs, can authorize a black box warning.

Given how long the Agency has taken this position, it’s rather surprising the dearth of legal precedent on preemption of black box warning claims. There’s really only one case that’s ever addressed it, Ehlis v. Shire Richwood, Inc., 233 F. Supp.2d 1189 (D.N.D. 2002), aff’d, 367 F.3d 1013 (8th Cir. 2004), and even there the court really didn’t analyze the issue very thoroughly. Instead of relying upon what the FDA had specifically decided about black box warnings, that court found the claim preempted as a de facto attempt to assert a private right of action under the FDCA. Id. at 1197.

Given the lack of precedent, it’s useful to go over the regulatory basis for preemption of black box warning claims. Basically, what happened is that the FDA was afraid, unless it closely regulated the use of these warnings, that they would be overused and their impact reduced.

In 1979, the Agency created “boxed warnings” specifically for “special problems” of the most serious nature. Thus, it promulgated a regulation stating: “Special problems, particularly those that may lead to death or serious injury, may be required by the Food and Drug Administration to be placed in a prominently displayed box. The boxed warning ordinarily shall be based upon clinical data, but serious animal toxicity may also be the basis of a boxed warning in the absence of clinical data. If a boxed warning is required, its location shall be specified by the Food and Drug Administration.” 21 CFR 201.80(e), see 21 C.F.R. §201.57(c)(1) (for products approved after June 2001). Thus, it’s always been up to the FDA both to require a boxed warning and to determine where on the label it would go.

The meaning of these provisions is hardly in doubt. During their initial notice and comment period , the FDA was “asked whether a manufacturer may include a boxed warning without prior FDA approval.” 44 Fed. Reg. 37434, 37448 (FDA Jun. 26, 1979). Without equivocation or qualification, the FDA said “no.” “[T]o ensure the significance of boxed warnings in drug labeling, they are permitted in labeling only when specifically required by the FDA.” Id. “Instead, “the decision as to whether a warning is legally required for the labeling of a drug must rest with the agency.” Id. at 37447. Thus, the Agency “has emphasized that, to ensure the significance of boxed warnings in drug labeling, they are permitted in labeling only when specifically required by FDA.” Lars Noah, “The Imperative to Warn; Disentangling the ‘Right to Know’ from the ‘Need to Know’ about Consumer Product Hazards,” 11 Yale J. Reg. 293, 331 (1994).

In prohibiting any boxed warnings other than those it authorizes, the FDA has been concerned that any new boxed warning potentially dilutes the strength of others. It adopted a policy of “restraint in requiring warnings to be boxed because overuse of the box will ultimately lead to reducing its effect.” 51 Fed. Reg. 43900, 43902 (FDA Dec. 5, 1986). Unapproved emphasis “might detract from other language of equal or greater importance, and thus mislead or confuse physicians.” 44 Fed. Reg. 37434, 37440 (FDA June 28, 1979). The FDA continues to follow this policy to this day. Right now, the FDA’s website states that the Agency “reserves this format, with the warning placed in a prominently displayed box, for the most serious warnings necessary to ensure the continued safe use of the product.” < http://www.fda.gov/medwatch/report/DESK/actions.htm >.

If the FDA’s reasoning sounds familiar, that’s because it is. The same concern about dilution of black box warnings through overuse that the Agency has expressed for a quarter century has recently come to fore in global fashion. The adverse effects of overwarning that helped drive the FDA to a more pro-preemption position generally, see 71 Fed. Reg. 3922, 3935 (FDA Jan. 24, 2006), are no different than those that led the Agency to restrict the use of black box warnings in 1979.

Zyprexa protective order update

We posted on Monday about the two cover stories in the New York Times about Zyprexa that were based on internal Eli Lilly documents that might have been disclosed in violation of a protective order. We withheld judgment on whether the documents had been disclosed lawfully or not, because we just don't know; we are outsiders to the Zyprexa litigation.

The court now appears to have sided with Lilly. On Monday, Judge Cogan (in Judge Weinstein's absence) entered an "Order for Mandatory Injunction" enjoining the lawyer who gave the documents to the Times from further disseminating the documents and requiring him to return the documents and to preserve certain information relating to how he obtained and distributed the documents.

Here's a link to the order.

If we learn more, you'll learn more.

Tuesday, December 19, 2006

Reprocessing single-use medical devices

This article from the Minneapolis Star-Tribune reminded us of the simmering controversy about reprocessing medical devices that are labeled for single-use. Folks in the device and health care industries know the story: manufacturers produce certain medical devices labeled for single use only. Hospitals use the devices once; reprocessors sterilize the devices; and hospitals buy the sterilized devices and re-use them in new patients. Commentators raise issues about both whether the devices are safe when re-used and whether patients should be told as part of the informed consent process that they are being treated with reprocessed devices that were labeled for single use.

This is a tricky issue for us to discuss -- not because we don't have opinions, but because of who we are and what we do for a living. We typically represent drug and device manufacturers in product liability cases. But we both have colleagues who represent hospitals or other health care providers. And there's always a chance that a reprocessor will call us one day, and who knows where zealous advocacy will lead then? So we won't say publicly anything about the merits of the on-going debate.

We'll say only this: As litigators, we're called into action after smoldering issues like this one have ignited into legal infernos. And, if that ever happens here, all of the players in the single-use device controversy will look back with regret on the things they've said so stridently before the litigation began. A manufacturer's protestation that a device labeled for single-use is unsafe if reprocessed, a hospital's statement about the need to control costs, a reprocessor's claim that manufacturers are just protecting profits -- all of these comments can come back to haunt all of the players collectively.

Take it from a couple of guys who have been there and done that: Turn the volume down a notch, folks. It's surely possible to discuss these issues without making accusations that may ultimately redound to everyone's collective detriment. Be careful what you say in public; some day, you may be held to your words.

Monday, December 18, 2006

Take our second thesis -- please! (Protective orders)

Cover stories in both yesterday's and today's New York Times criticize Eli Lilly's marketing of Zyprexa based on internal documents provided by a plaintiff's lawyer. The lawyer appears to have a clear conscience -- his name and photograph appear in Sunday's article, so no court would have any trouble tracking him down. Lilly, on the other hand, appears to be outraged. The documents were originally provided to plaintiffs' lawyers under seal, and Lilly insists that the release of the documents is "illegal." As to this particular release of documents, we (or at least the one of us typing these words) have no idea whether the documents were released lawfully or not. But it seems as though, in general, we've seen this film before.

Doesn't this happen in virtually all of the mass torts? Courts enter protective orders; defendants produce documents subject to the protective orders; documents are leaked to the press; night follows day.

What surprises us is how infrequently courts seem to care about this. Although documents are often leaked in violation of mass tort protective orders, courts rarely conduct full-scale investigations to find and punish the person who violated the court's order. Why not?

We'd like to see an academic (who might have the time to pursue this) research the mass tort cases (since about 1985 or so, when modern mass torts sprung into being) to determine the percentage of cases in which protective orders have been violated and the number of times courts have tracked down and punished the violator. If, as we suspect, protective orders are routinely ignored, but violators are rarely punished, that fact should influence how courts conduct future cases. If courts can't honestly tell defendants that protective orders will be enforced, what should follow from that? Should particularly sensitive documents not be produced at all? Should document depositories be protected more carefully? Are there other implications for future cases?

We understand that some lawyers who leak confidential documents may believe that they're serving the public good. But other leakers are unfairly trying their cases in the court of public opinion, with the aim of either pressuring a defendant to settle or poisoning the minds of potential jurors. If a lawyer wants to give documents to a newspaper to serve the public good, then the lawyer should do so lawfully. If documents should be unsealed, file a motion to unseal them. Don't simply leak them illegally to the press.

Again, we don't know anything at all about the propriety of the leak in the Zyprexa case. But we've seen and read about leaks in an awful lot of mass tort cases in the past. It's time for the judicial system to decide how it will respond in cases where leaks in fact violate court orders.

Friday, December 15, 2006

Drug Preemption Cases

This is another "better here than in a heap on my floor" posts. This post provides links to all court decisions (of which we're aware) on the issue of drug preemption that have been handed down since the FDA published the so-called "Preemption Preamble" in January 2006.

As readers of this blog surely know, in January the FDA included certain statements about preemption in the preamble to its rule on the content and format for prescription drug labeling. See Department of Health and Human Services, Food and Drug Administration, Requirements on Content and Format of Labeling for Human Prescription Drug and Biological Products, 71 Fed. Reg. 3922 (2006). Here's a link to the FDA's press release about that rule, which in turn links to the preamble and rule itself.

We have collected below the opinions that courts have handed down in drug preemption cases since the publication of the preemption preamble. Please note that a parenthetical description of a case as "finding no preemption" in the context of one drug does not mean that the decision would necessarily be an adverse preemption precedent in the context of a different drug. Several of these decisions find no preemption for the particular drug involved in the case, but note that the court might find preemption for a drug that had a different regulatory history. Here are the cases, from January 2006 to date:

1. McNellis v. Pfizer, Inc., No. 05-1286, 2005 U.S. Dist. LEXIS 37505 (D.N.J. Dec. 29, 2005); motion to vacate denied, cert. for app. granted, 2006 U.S. Dist. LEXIS 70844 (D.N.J. Sept. 29, 2006) (finding no preemption in a Zoloft/suicide case; certifying interlocutory appeal).

2. Abramowitz v. Cephalon, Inc., No. BER-l-617-04, 2006 WL 560639 (N.J. Super. Ct. Bergen Cty. Mar. 3, 2006) (finding preemption in an Actiq/tooth decay case). We're sorry that we can't provide a link for this decision, but we have only a Westlaw copy of the opinion, and we're not sure whether copyright law permits us to post a Westlaw copy of a decision.

3. Laisure-Radke v. Par Pharm., Inc., No. C03-365RSM, 2006 WL 901657 (W.D. Wash. Mar. 29, 2006), reconsid. denied (May 3, 2006) (finding no preemption in a generic Prozac/suicide case).

4. Colacicco v. Apotex, Inc., 432 F. Supp. 2d 514 (E.D. Pa. 2006) (finding preemption in a generic Paxil/suicide case).

5. Jackson v. Pfizer, Inc., 432 F. Supp. 2d 964 (D. Neb. 2006) (finding no preemption in a Zoloft and Effexor/suicide case).

6. In re Bextra and Celebrex Mktg. Sales Practices & Prod. Liab. Litig., No. 05-1699 CRB, 2006 WL 2374742 (N.D. Cal. Aug. 16, 2006) (finding preemption of failure-to-warn, but not false advertising, claims in the Celebrex/cardiovascular risk MDL).

7. Ackermann v. Wyeth Pharmaceuticals, No. 05-84, slip op. (E.D. Tex. Sept. 6, 2006) (finding preemption of failure to warn claims in Effexor/suicide case), opinion withdrawn as moot in light of order granting complete summary judgment on other grounds (E.D. Tex. Dec. 20, 2006).

8. Conte v. Wyeth, Inc., No. CGC-04-437382, slip op. (Super. Ct. S.F. Cty., Cal. Sept. 14, 2006) (finding preemption in a Metoclopramide/gastroesophageal reflux disease case).

9. Perry v. Novartis Pharms. Corp., No. 05-5350, 2006 U.S. Dist. LEXIS 75319 (E.D. Pa. Oct. 16, 2006) (finding no preemption in an Elidel/lymphoma case).

10. Levine v. Wyeth, No. 2004-384, 2006 Vt. LEXIS 306 (Vt. Oct. 27, 2006) (finding no preemption in a Phenergan/IV push case).

11. Weiss v. Fujisawa Pharmaceutical Co., No. 5:05-527-JMH, slip op. (E.D. Ky. Nov. 28, 2006) (finding no preemption in an Elidel/lymphoma case).

We hope that your office floor can now be a little cleaner, too.

Thursday, December 14, 2006

The FDA's Amicus Briefs on Preemption

As everyone who litigates drug and medical device cases knows, the FDA has filed a number of briefs as amicus curiae over the past few years. It strikes us that it would be a good idea to collect them all in one place. Not only is this collection a resource in its own right, but a review of these briefs will show that, (1) by and large, the FDA has been much more consistent than the plaintiffs’ bar gives the Agency credit for, and (2) the advent of the FDA’s pro-preemption advocacy, far from being a figment of the change in political administrations that took place in January, 2001, actually began a year earlier, while the Democrats still ran the Agency.
What happened was not so much that the FDA changed its position, but that the type of lawsuits being filed plaintiffs have been infringing more and more on the FDA’s prerogative to approve prescription medical products and to control their labeling.

The first FDA amicus brief in this series argued in favor of preemption of what had become known as “fraud-on-the-FDA” claims. Under a fraud on the FDA theory, courts applying state law supposedly had the power to ignore in-force FDA decisions if those decisions were deemed “fraudulently” obtained by the defendants. Obviously, such claims posed a direct threat to, and inherent conflict with, the Agency actions challenged as fraudulent. When the question of preemption of fraud on the FDA claims reached the Supreme court in early 2000, the Court asked the Agency for its views. The Department of Justice (the FDA’s lawyers) filed an amicus brief urging that the Court take the case and find fraud on the FDA claims preempted on the basis of that conflict (the Agency argued against express preemption). See Brief For United States As Amicus Curiae, Buckman Co. v. Plaintiffs’ Legal Committee, No. 98-1768 (U.S. filed June 7, 2000).

The Supreme Court agreed with the FDA and accepted the appeal. The FDA/DOJ filed a second amicus brief in the Buckman case on the merits, and once again it argued that fraud on the FDA claims were preempted as inherently conflicting with federal agency decisions. See Brief For United States As Amicus Curiae Supporting Petitioner, Buckman Co. v. Plaintiffs’ Legal Committee, No. 98-1768 (U.S. filed Sept. 13, 2000). The Supreme Court unanimously agreed. Buckman Co. v. Plaintiffs’ Legal Committee, 531 U.S. 341 (2001).

Also in 2000 the FDA elected to file a brief supporting preemption of common-law tort claims in a case where the plaintiffs had moved for injunctive relief under state law to force a manufacturer to make changes in FDA-approved labeling and to issue “Dear Doctor” letters. Finding these claims were a direct challenge to the FDA’s control over labeling content and over issuance of such letters to physicians, the FDA argued that these claims could not stand in the face of the Agency’s approval of the existing warnings and the FDA’s power to order notices to the public, and that the FDA had primary jurisdiction over such matters. See Statement Of Interest Of The United States, Bernhardt v. Pfizer, Inc., 00 Civ. 4042 (LLM) (S.D.N.Y. filed Nov. 13, 2000). The court agreed that the FDA had primary jurisdiction and did not reach preemption. Bernhardt v. Pfizer, Inc., 2000 WL 1738645 (S.D.N.Y. Nov. 22, 2000).

After the change in administrations, a similar situation arose in 2002, where plaintiffs obtained an injunction under state law against a manufacturer’s continuation of FDA-approved advertising. In re Paxil Litigation, 2002 WL 1940708 (C.D. Cal. Aug. 16, 2002). Once again the Agency found the claim a direct threat to its superintendence of what regulated manufacturers told the public about regulated products and argued in favor of preemption. See Brief Of The United States, In re Paxil Litigation, No. CV 01-07937 MRP (C.D. Cal. filed Sept. 5, 2002). The district court, while not finding preemption, reconsidered its order and withdrew the objection. In re Paxil Litigation, 2002 WL 31375497 (C.D. Cal. Oct. 18, 2002).

At approximately the same time the FDA also intervened in litigation in California seeking to enforce a state proposition that mandated birth defect warnings under certain circumstances, even with respect to FDA-approved product labeling. The litigation sought to force manufacturers of smoking cessation devices to include such warnings. The Agency, however, had nixed these warnings because it considered the risk of birth defects from use of these products to be much less of a risk to the developing fetus than the risks posed by cigarette smoke if pregnant women, deterred by the state-law warnings from using the devices, continued to smoke. When this litigation reached an appellate stage in 2002, the FDA filed an amicus brief taking the position that, once the Agency had ruled that no warning was appropriate, state law could not hold the manufacturer liable for following the FDA’s direction in preference to the state proposition. See Amicus Curiae Brief of the United States of America in Support of Defendants/Respondents, Dowhal v. SmithKline Beecham Consumer Health Care LP, No. A094460 (Cal. App. filed March 25, 2002).

The intermediate appellate court disagreed with the FDA and held that the state proposition had been exempted by preemption by a federal statute. Dowhal v. SmithKline Beecham Consumer Healthcare, 100 Cal. App. 4th 8, 122 Cal. Rptr. 2d 246 (2002). On further appeal to the California Supreme Court, the FDA submitted a second amicus brief in the same case making the same arguments. See Amicus Curiae Brief Of The United States Of America, Dowhal v. SmithKline Beecham Consumer Healthcare, LP, No. A094460 (Cal. filed July 18, 2003). Siding with the FDA, the California Supreme Court reversed and held that the litigation was preempted – not even by statute could Congress approve state law prevailing in a direct conflict with supreme federal law. Dowhal v. SmithKline Beecham Consumer Healthcare, 32 Cal. 4th 910, 12 Cal. Rptr. 3d 262, 88 P.3d 1 (2004).

The next FDA intervention in state-law tort litigation occurred in state court in Tennessee. In medical device litigation, the plaintiffs contended that the product in question was approved by the Agency through one method of review, and the defendant contended that a different method of review had been used. The FDA filed a brief, siding with the defendant, stating how the product was approved, and arguing that plaintiff’s contrary contentions misapplied the FDCA and were therefore preempted. See Statement Of Interest Of The United States Of America, Murphee v. Pacesetter, Civ. No. ct-005429-00-3 (Tenn. Cir. 30th Dist. Dec. 12, 2003). The litigation did not produce any opinion resolving the issue.

Shortly after the turn of the millennium, plaintiffs started filing suits alleging that selective serotonin uptake inhibitors (“SSRIs”), medication approved to treat clinical depression, caused suicide. Since depression – the condition being treated – also caused suicide, these claims were problematic from a scientific standpoint. The FDA reviewed this issue several times and until recently refused to permit suicidality warnings on SSRIs as to any category of user (and as of this moment still requires only pediatric warnings). Ignoring the FDA’s resolution of this issue, plaintiffs alleged in state-law SSRI litigation that the warnings the FDA had prohibited were mandated by state tort law. Beginning in 2002, the FDA filed several amicus briefs arguing that its decisions not to require suicide/suicidality warnings preempted claims asserting a state-law duty to provide those same warnings. See Brief For Amicus Curiae The United States Of America, Motus v. Pfizer, Inc., Nos. 02-55372, 02-55498 (9th Cir. filed Sept. 3, 2002); Amicus Brief For The United States, Kallas v. Pfizer, Inc., No. 2:04CV0998 PGC (D. Utah. filed Sept. 15, 2005); Brief for Amicus Curiae United States of America, Colacicco v. Apotex Corp., C.A. No. 05-5500-MMB (E.D. Pa. filed May 10, 2006); Brief of United States as Amicus Curiae in Support of Defendants-Appellees, Colacicco v. Apotex Corp., No. 06-3107 (3d Cir. filed Dec. 4, 2006). The court in Motus avoided the preemption issue. Motus v. Pfizer Inc., 358 F.3d 659 (9th Cir. 2004). The Kallas litigation settled before there was any decision. The district court in Colacicco agreed with the FDA and found the SSRI claim preempted, Colacicco v. Apotex, Inc., 432 F. Supp.2d 514 (E.D. Pa. 2006) – a ruling that is currently on appeal.

In a recent case that did not involve a purported drug risk previously passed upon by the Agency, the FDA has taken the position that, unless it would have found the scientific basis insufficient, there is no preemption prior to the FDA actually taking a regulatory position. See Letter Brief of the United States as Amicus Curiae, Perry v. Novartis Pharmaceuticals, C.A. No. 05-5350 (E.D. Pa. filed Sept. 21, 2006). The FDA distinguished the SSRI cases as involving state-law challenges to actual FDA decisions, and the court ultimately ruled the same way. Perry v. Novartis Pharmaceuticals, ___ F. Supp.2d ___, 2006 WL 2979388 (E.D. Pa. Oct. 16, 2006).

In the only instance where the FDA has reversed a specific prior legal position in an amicus filing has to do with express preemption under the Medical Device Amendments (“MDA”) and pre-market approved devices – the most rigorously reviewed type of medical device. Acknowledging that it was reversing a 1998 position against preemption, the FDA determined in 2004 that the its pre-market approval process produced “device specific” requirements that preempted most product liability claims against this category of product. See Letter Brief of United States as Amicus Curiae, Horn v. Thoratec, No. 02-4597 (3d Cir. filed May 14, 2004). The court, which had invited the FDA’s participation, agreed with the Agency and affirmed preemption, notwithstanding the regulatory reversal of position. Horn v. Thoratec Corp., 376 F.3d 163 (3d Cir. 2004).

We’ve provided links to all of these amicus briefs. We believe that they demonstrate how the Agency’s trend in favor of preemption began before the change of administration, and thus cannot be waved away as political. Rather, the FDA’s advocacy of preemption is almost entirely a reflection on the types of claims being brought against regulated manufacturers. We also believe that, with the exception of Horn, the FDA has been quite consistent in its positions. The Agency’s position is, and has been that tort claims which either ignore FDA decisions, or threaten to undermine the Agency’s authority over the approval or labeling of regulated products, are preempted. That’s the position that the Agency has taken in these amicus briefs, and it’s also the position that it has taken in its 2006 Final Rule. See 71 Fed. Reg. 3922, 3933-36, 3967-69 (FDA Jan. 24, 2006).

Nagareda on preemption

Professor Richard Nagareda, who's developed quite a name for himself as a complex litigation scholar, just put out an article for the inaugural issue of a new publication, the Journal of Tort Law. The article, FDA Preemption: When Tort Law Meets the Administrative State, analyzes FDA preemption generally, but it focuses in particular on litigation asserting that there is a causal link between ingesting certain antidepressants and developing suicidal thinking or behavior. We'd love to describe the article in a little more detail; in fact, we'd love to say whether we agree or disagree with some or all of Nagareda's ideas. But we're keenly aware that the words we post in this blog could (inappropriately) be cited against us or our colleagues some day, and so we are necessarily terribly guarded when we speak on topics that come close to home.

Here's a link to the Berkeley Electronic Press abstract of the article. If you register there, you can download the entire article. If you have a reaction to the article, please post your thought as a "comment" below, and our on-line community can discuss Nagareda's thesis. Your on-line co-hosts may be constrained from participating too actively in the discussion, but we can at least follow along with interest, can't we?

Just to show off (a Curmudgeonly podcast)

We've never listened to podcasts.

Heck, back in November, we weren't sure what a podcast was.

But two weeks ago, the ABA asked one of us to do a podcast. Herrmann recorded Chapter 3 ("What They Didn't Tell You in Law School") of his book, The Curmudgeon's Guide to Practicing Law. As of today, we're not simply able to do a podcast. We've come within a hair's breath of actually linking from our blog to a podcast; we're feeling less and less like luddites every day. (The hair's breath is that we couldn't link directly to the podcast; you'll have to click once more after this link to hear the podcast itself.) This was originally taped for a segment on ABA Net's Section of Litigation and billed as something "every law student and new lawyer ought to read or hear." Listen in, if the spirit moves you.

Tuesday, December 12, 2006

Consumer reaction to off-label use

We saw this morning the attached description of consumer awareness, and approval, of off-label use of prescription drugs. (This squib came from a service available at owner-managedcare@wire-source.com, but we weren't sure that a link would take you directly to this item, so we're reproducing the item here, with attribution.) For lawyers trying cases that involve off-label use of a drug or device, it may be helpful to know jurors' uneducated reaction to off-label use. Here's the squib:

Study: Patients Unsure About Viability Of Off-Label Prescriptions

ROCHESTER, NY (Managed Care Wire): U.S. adults are either divided or ambivalent when it comes to the issue of off-label prescription drug use, according to the latest Wall Street Journal Online/Harris Interactive Health-Care Poll. While it is legal for doctors to prescribe prescription drugs for diseases other than those for which they have been approved by the U.S. Food and Drug Administration (FDA), there are strict rules governing the marketing of a drug for treatment of a disease for which it hasn’t been approved. About one-third (27 percent) of adults say that prescribing drugs off-label should be allowed, while one-third say in most cases, the risks of doing so outweigh the benefits.
Interestingly, half (50 percent) of all adults incorrectly believe that once a drug is approved, a doctor can prescribe it only for the diseases for which it has been approved by the FDA, and another quarter (25 percent) are unsure. Nearly half (48 percent) think that doctors should not be allowed to prescribe drugs for off-label uses and even more (69 percent) think drug companies should not be allowed to encourage off-label use. However, a plurality (55 percent) agrees that if doctors aren’t allowed to prescribe freely, it will be much more difficult to find new and innovative ways to treat diseases.
Approximately one in three (31 percent) adults believes that in most cases, the benefits of off-label prescription drug use outweigh the risks, but an equal proportion (34 percent) believes the opposite is true. The greatest level of agreement amongst adults in this poll is on the potential prohibition of off-label prescription drug use, except as part of a clinical trial, with nearly two-thirds (62 percent) agreeing with such restrictions.
Payers looking for ways to contain the rising costs of healthcare can take away one clear message from the study’s results. "Precedent has been set for prohibiting payment for certain types of surgical procedures except to patients who are enrolled in a clinical trial," explains Katherine Binns, Division President of Healthcare Research for Harris Interactive. "These findings suggest that U.S. adults do not want to hamper innovation, but would also be supportive of increased oversight and limitations being applied to the off-label use of prescription drugs."

Monday, December 11, 2006

Judge Weinstein proposes amending CAFA

In an order entered on Thursday, December 7, in the Zyprexa litigation, Judge Weinstein requested further briefing on a motion to remand. Along the way, he noted his frustration with the fact that the Class Action Fairness Act of 2005 allows removal only of certain class actions and mass actions, but not of all state court cases that might usefully be included in a federal MDL proceeding. In Judge Weinstein's words:

"It may be useful for Congress to consider expanding the Class Action Fairness Act from class actions to at least some national MDL, non-Rule 23, aggregate actions. As use of the class action device to aggregate claims has become more difficult, MDL consolidation has increased in importance as a means of achieving final, global resolution of mass national disputes. See Part II, supra. Much the same concerns which animated CAFA's preference for a single, federal forum apply to national MDL aggregate actions."

Here's a copy of Judge Weinstein's order in its entirety.

Saturday, December 09, 2006

Take our thesis -- please! (Class action notice)

We're busy guys. We can just barely manage the full-time practice of law, blogging, being decent husbands and fathers, and the rest, without writing sophisticated legal articles in our spare time. So here's an idea for a law review article that would make a real contribution to the legal profession. We'll never get around to writing this. So take our thesis (and write this article) -- please!

The article would answer the question that clients regularly ask when class action settlement notices are about to be mailed: What will the response rate be? Defense lawyers have different ways of evading this question. No one provides details, but all lawyers give the same basic answer: The response rate will be "low," "extraordinarily low," "nil," or "next to nothing." With luck, clients don't press further.

Surely there's an empirical answer to this question.

Litigants have now implemented thousands of class action settlements, and the response rate to each of those notice campaigns is a historical fact. The response rate surely varies based on identifiable factors that include, for example: (1) nature of the product (or security, or whatever) involved in the underlying litigation, (2) nature of the notice process, including the direct mail element, internet-based notice, size and scope of print media notice campaign, etc., (3) the value of the settlement to the class members (because people are presumably more likely to fill out the forms needed to claim a $1000 check than a $1.75 check), (4) the burden imposed by the claims process (such as the number and complexity of forms that must be completed, the need to have saved receipts or other proof of purchase, etc.), (5) size (and notoriety) of the litigation underlying the settlement, and (6) many other factors that the person who chooses to write this article will surely identify.

So here's the idea: Obtain all available data for known class action settlements that required class members to respond to notice campigns. Sort the data based on the types of criteria that we've identified above. And tell the world the response rates based on those criteria.

That data would be invaluable. It would allow litigants to calculate more precisely the value of the settlement to the class and the cost of the settlement to the defendant. It would also let courts decide whether a proposed settlement is fair and whether a proposed notice campaign is sufficient for its purpose.

Come on, academia! You guys have time to do this type of analysis, and you could produce an article that would cross the divide between pure academic abstraction and useful information that lawyers and judges could use regularly.

Take our thesis -- please!

Friday, December 08, 2006

Colacicco appellate briefs

We told you that we would use this blog as a file drawer as well as a way to communicate. For future reference, here are links to all of the briefs filed on behalf of the pharmaceutical manufacturers in the Third Circuit in Colacicco: the Joint Brief for Appellees, the FDA's amicus brief, PhRMA's amicus brief, and PLAC's amicus brief. Now you'll know where to find them.

MDL Panel statistics on mass torts

There's a noteworthy article in the December 2006 issue of The American Lawyer about the supposed death of mass torts. The entire article is interesting, but here's the statistical analysis of the MDL Panel that caught our eye:

"In the panel's first 15 years of existence, the judges heard arguments to transfer 32 mass torts matters. They denied the transfer of 19 of them--including several proposed asbestos MDLs and at least a half-dozen pharmaceutical cases--leaving the cases to be litigated in multiple courtrooms. Then trends changed. As mass tort litigation burgeoned, the MDL panel judges not only heard more requests, they also transferred a higher percentage of them. From 1986 through 2000, the panel denied only nine of the 68 MDL motions it heard. And in the last six years, only six of the 62 cases the panel considered were denied MDL status."

Clients often ask about the statistical tendencies of the MDL Panel; it's nice to have the numbers gathered in one place in print.

Heedless Use of Heeding Presumptions

We are often mystified that courts rotely apply purportedly “general” product liability concepts to litigation involving prescription medical products without stopping to think whether these concepts make sense where: (1) the product cannot be marketed unless and until a federal agency approves them as safe and effective, and (2) because of its inherent risks, the product cannot be used by anyone except upon a doctor’s prescription.

One of these “cross-over” concepts is the so-called “heeding presumption” – adopted in a number of states – that “presumes” (in the absence of contrary evidence) if the defendant had given an adequate warning, the plaintiff would have heeded and followed it. This proposition has serious problems generally, so serious that the language giving rise to it was intentionally removed from the Restatement (Third) of Torts: Products Liability. That, however, is something for a different post.

In the case of prescription medical products, such a “heeding presumption” rarely, if ever, makes any sense at all. Yet, all too often we come across cases like Tingey v. Radionics, 2006 WL 2258872, at *10 (10th Cir. Aug. 8, 2006), and In re Diet Drug Litigation, 895 A.2d 480, 490-91 (N.J. Super. Law 2005), that apply the “heeding presumption” in prescription medical product liability litigation primarily because some earlier court applied it to an entirely different product that anybody could go out and buy at a hardware store.

Conversely it’s gratifying to read cases like Lineberger v. Wyeth, 894 A.2d 141, 145, 149-50 (Pa. Super. 2006), and Koenig v. Purdue Pharma Co., 435 F. Supp.2d 551, 556-57 (N.D. Tex. 2006), where the courts get it. These courts take the time to understand that not all products are the same – that drugs and devices are qualitatively different from step ladders and automobiles.

So what’s this big qualitative difference? Mainly, it’s that there are two fundamentally different kinds of warnings. Most warnings concern a product’s use – that if you use (or don’t use) the product in a certain way, you are likely to get hurt; and if you follow the warning, you won’t. Examples are not driving a riding lawnmower parallel to a slope (because you’ll tip over) or only handling asbestos while using a respirator (because breathing asbestos can do nasty things to you).

While so-called “use” warnings occasionally arise as to prescription medical products – overdose instructions come to mind – that type of warning is not what most litigation involving these products is about. Rather, with prescription-only products, most of the relevant risks arise whenever the product is used. A warning about an inherent risk – a so-called “risk warning” – serves an entirely different purpose.

With inherent risks, people are warned so they can decide whether that risk outweighs the benefits that might be gained from using the product. The only way to avoid the risk is not to use the product at all. All prescription medical products have inherent risks – which is why the FDA requires a physician’s prescription in the first place. For a case discussing the distinctions between these two types of warnings in detail, read Thomas v. Hoffman-LaRoche, Inc., 949 F.2d 806, 814 (5th Cir.1992).

What happens when a heeding presumption is imposed concerning a risk warning? In that situation, “heeding” the warning can only mean not using the product at all, because that’s the only way to avoid the risk. So applying a “heeding presumption” to an inherent risk lets plaintiffs argue, in effect, that the product should never have been sold – that every “reasonable” person would have “heeded” an adequate warning and not used it. But we know that’s not simply not true. For example, millions still smoke cigarettes, although those risks have been public knowledge for decades. Millions more use drugs and medical devices even though they carry the FDA’s strongest “black box” warnings. A “heeding presumption” in this situation is contrary to fact and to common sense.

But where a prescription medical product is involved there’s even more. Making the decision whether a product’s risks outweigh its benefits for society as a whole is precisely what we created the FDA to do. The Agency evaluates safety and efficacy for the products it approves and decides whether those products should be marketed at all and, if so, for what types of patients and for what types of medical conditions. Making that same decision on individual basis is the job of the prescribing doctor.

Doctors do for each of their patients what the FDA does for everybody – on an individual basis, they balance the risks of a product against its expected health benefit. Thus, the prescribing physician only “heeds” warnings in the sense of considering them while deciding how to prescribe the product and whether or not to tell the patient about the risk. It is unrealistic to presume physicians would “heed” any warning either in the sense of never using the product or that they would automatically telling every patient about every risk. Eck v. Parke, Davis & Co., 256 F.3d 1013, 1021 (10th Cir. 2001) (wrong to “construe [a treater’s] ‘heeding’ an adequate warning to mean [s/he] would have given the warning”).

So if an inherent risk of an FDA-approved prescription-only product is omitted (or not adequately discussed), does that mean that the law should “presume” that nobody would have used it – that everyone would have “heeded” a better warning and used something else (or have been content to use nothing)? Of course not. The FDA’s approval means that the product can be used for its approved uses. Unless the FDA determines the “new” risk is so serious as to affect that approval, it will deal with the risk by approving a new warning. Unless a “new” risk is great enough to tip the prescribing decision for a particular patient, at most a prescribing doctor will tell the patient about it. The prescriber might not even do that if the new risk is not viewed as very likely in that particular patient.

For these reasons, we believe that – logically and legally – the “heeding presumption” has no legitimate role in virtually all prescription medical product liability litigation.

Thursday, December 07, 2006

Interlocutory appeals in MDLs

We love choice of law issues for federal questions in MDLs; it gives us something to think about.

Readers of this blog know the basic issue. A lawsuit is filed in federal court in, say, New York. The New York court enters an order (on a federal question) that is appropriate under Second Circuit law. The MDL Panel then transfers the case to, say, Chicago, and Seventh Circuit law on the relevant point differs from Second Circuit law. Should the Chicago court reconsider the order entered in New York?

Whether or not it does, the Chicago court then rules on other substantive issues that arise under federal law (such as, say, preemption). The Chicago court applies Seventh Circuit precedent. (On that issue, at least, the law has been relatively uniform for the last two decades. Since In re Korean Air Lines Disaster, 829 F.2d 1171 (D.C. Cir. 1987), MDL transferee courts have generally applied local federal precedent on matters of federal law.)

After pretrial proceedings are concluded, the MDL Panel transfers the case away from Chicago and back to New York for trial. Second Circuit law on the substantive issue differs from Seventh Circuit law. Should the New York court reconsider the order entered in Chicago?

These issues are generally resolved under the "law of the case" doctrine, which is a tad amorphous to begin with. And we hate even to type these next words, but consider this: Federal precedent on the law of the case doctrine varies by circuit. Should courts apply Second Circuit or Seventh Circuit precedent on "law of the case" when deciding the issues discussed above? One scholar grappled with this question years ago: Joan Steinman, "Law of the Case: A Judicial Puzzle in Consolidated and Transferred Cases and in Multidistrict Litigation," 135 U. Pa. L. Rev. 595 (1987).

But here's what grabbed our eye. In the Enron MDL, Judge Harmon recently refused to certify an interlocutory appeal in part on the ground that certifying an appeal in an MDL wouldn't do much good, since the constituent cases would ultimately be remanded for trial in different circuits:

"because this is a multidistrict litigation, many of the consolidated member suits arose in
other Circuit Courts of Appeals. . . . Thus the Fifth Circuit's determination of the questions
may not be controlling."

In re Enron Corporation Securities, Derivative & ERISA Litigation, MDL No. 1446, slip op. at 5 (S.D. Tex. Nov. 21, 2006).

Isn' there a decent argument grounded in law of the case that cuts in favor of certifying an appeal in this situation? The law of the case doctrine instructs courts to defer to some extent to decisions made by coordinate courts. Thus, trial courts defer to decisions made by other trial courts. But appellate courts also defer to decisions made in a single case by other appellate courts. See Wright, Miller & Cooper, Federal Practice and Procedure: Jurisdiction 2d Sec. 4478.4 at 780 (2002). Thus, if Judge Harmon had certified the appeal to the Fifth Circuit, other circuit courts might well have deferred to that decision under the law of the case doctrine, making it more likely that all of the Enron cases would ultimately be controlled by a single set of rules.

In future MDLs, defendants should consider asking MDL transferee courts to certify appeals for the express purpose of obtaining appellate precedents that may apply unusually broadly because of the deference they might deserve under the law of the case doctrine.

Saturday, December 02, 2006

Choice of law in class actions (St. Jude Medical)

It won't surprise anyone to learn that we're not big fans of Judge Tunheim's recent class certification decision in In re St. Jude Medical, Inc. Silzone Heart Valves Products Liability Litigation. Here's the link, for ease of reference. Judge Tunheim adopted the distinct minority view, holding that the District of Minnesota could apply Minnesota consumer protection statutes to a nationwide class of recipients of certain prosthetic heart valves who had neither undergone explant surgery nor manifest an injury that would permit the filing of a personal injury lawsuit in the patient's state of residence. (Another "no injury" class! But we'll save that rant for another day.)

Most courts have of course held that choice-of-law issues preclude certifying nationwide classes; the Seventh Circuit squarely so held in Firestone Tires. Judge Tunheim begs to differ, finding that St. Jude Medical has its headquarters in Minnesota, and its relevant corporate conduct occurred in that state, so Minnesota law applies to, say, an operation in which a New York physician implanted a heart valve into a New York patient in a New York hospital.

We won't trot through all the issues here, but we'll note one possible way of arguing against certification of a nationwide class in these situations. (The argument borrows in part from a concept raised by Judge Easterbrook in his decision in Firestone Tires.)

Suppose a New York resident wins the lottery. He spends his winnings foolishly, buying three cars at his local car dealership in a single day. At car dealerships on a single street in New York, he buys a Ford, a Toyota, and a Hyundai.

As always happens in these hypotheticals, the New Yorker realizes that he was induced to buy all three cars by consumer fraud. He sues the three car manufacturers in New York state court.

Is there really anyone -- anyone -- who thinks that this hypothetical lawsuit would be governed by the laws of three different jurisdictions -- the jurisdictions in which the corporate defendants have their headquarters and in which their relevant corporate conduct occurred? If so, then the presiding judge must apply Michigan consumer protection law to the purchase of the Ford, Japanese consumer protection law to the purchase of the Toyota, and South Korean consumer protection law to the purchase of the Hyundai. Surely the consumer would be startled by this result, and the presiding judge probably wouldn't be too pleased with the task confronting him, either. But that's where most of Judge Tunheim's analysis leads.

Judge Tunheim did suggest one other factor permitting him to apply Minnesota law to the St. Jude case -- the named plaintiff chose to sue in a Minnesota court. That is surely true, but it poses its own problems. First, does "choice-of-law" really mean "the law the plaintiff chooses"? Doesn't it mean "the law applicable to the facts of the case, whatever the plaintiff's or defendant's preference"? (Those questions may not be precisely as rhetorical as the way we just phrased them, but you get our drift.)

Second, what will Judge Tunheim's class notice tell absent class members in this situation? "The named plaintiff chose to sue in Minnesota, so Minnesota law will apply to your claims. This will help those absent class members who live in states with consumer protection laws weaker than Minnesota's, but will hurt class members who live in states with consumer protection laws stronger than Minnesota's. If you live in any of the following states, you'd be better off opting out of the class and pursuing a statewide class action in your home state." We doubt we'll see that class notice, but isn't it what the rules (and due process) would require?

Finally, are the laws of the United States really meant to permit nationwide class actions against domestic companies, such as Ford (or St. Jude) but to forbid nationwide class actions against foreign companies, such as Toyota and Hyundai? That's an unusual form of discrimination.

Until someone answers our questions, we're counting on the Eighth Circuit, once again, in the St. Jude MDL.