Fisher v. Bristol-Myers Squibb Co.

MEMORANDUM OPINION AND ORDER ASPEN, Chief Judge. Plaintiff Natalie Fisher (“Fisher”), on behalf of the estate of her deceased son, has sued the Bristol-Myers Squibb Company, Mead Johnson Laboratories, and a number of their subsidiaries and affiliates (collectively, “the defendants”). She alleges that the defendants manufactured and distributed a harmful drug called Stadol, and she seeks recovery under a variety of tort, warranty, and fraud theories. Presently before us is Fisher’s motion to certify a class of persons similarly situated to her son. For the reasons set forth below, the motion is denied. [367] I. Background1 Stadol is the brand name the defendants have assigned to butorphanol tartrate, a narcotic agent prescribed by physicians to relieve moderate to severe pain. The defendants supply Stadol in two forms: (1) an injectable form administered intravenously or intramuscularly by a physician, and (2) a nasal spray, which can be self-administered by the patient. The product literature the defendants provided to physicians has evolved somewhat over the years, but every description contains warnings about the possibility that Stadol might be abused by patients (especially patients with a history of drug dependency) because of its narcotic properties. See Fisher Aff. Ex. 3 (containing the entries for Stadol in the Physicians Desk Reference for the years 1991-98). From 1993 onwards, each entry contained an identical section on “Drug Abuse and Dependence” that noted the existence of reports of patients becoming addicted to Stadol, but there was no explicit warning to physicians about the possibility of addiction, or about how severe such any such addiction might be. Id. Physicians were warned, however, that “[sjpecial care should be exercised in administering butorphanol to emotionally unstable patients and to those with a history of drug misuse. When long-term therapy is necessary, such patients should be closely supervised.” Id. Starting in December, 1991, and continuing until his death in August, 1995, Bruce Fisher’s doctors prescribed Stadol to him in order to alleviate pain from his migraine headaches. During this period he became physically and mentally addicted to Stadol. Bruce attempted to break his addiction, but despite obtaining professional help he was unable to do so. As a result of this inability (and the depression and withdrawal symptoms he was suffering because of his efforts) Bruce committed suicide on August 8, 1995. Fisher, as administratrix of Bruce’s estate, seeks damages for her son’s death on the grounds that the defendants’ were negligent in failing to provide adequate warnings about the risk of addiction, that the inadequate warnings rendered Stadol a defective product, that the defendants violated their implied warranties of merchantability and fitness, and that the defendants’ misrepresentations and omissions regarding the dangers of the product were fraudulent. More importantly for our present purposes, Fisher also seeks to bring all of these claims on behalf of a class of “all persons throughout the United States from 1991 to the present, who are or were addicted to Stadol and/or died as a direct and proximate result of said addiction.” Pl.’s Br. at 2. II. Discussion A. Rule 23 Standard Federal Rule of Civil Procedure 23(a) specifies four preliminary requirements that any proposed class must meet: “One or more members of a class may sue or be sued as representative parties on behalf of an only if (1) the class is so numerous that joinder of all members is impracticable, (2) there are questions of law or fact common, to the class, (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class, and (4) the representative parties will fairly and adequately protect the interests of the class.” Fed.R.Civ.P. 23(a). If the numerosity, commonality, typicality, and adequacy requirements are satisfied, then we must also decide whether the class qualifies under one of the three subsections of Rule 23(b). In the instant case, the plaintiff seeks certification under Rule 23(b)(3), which authorizes class actions where the “questions of law or fact common to the members of the class predominate over any questions affecting individual members, and ... a class action is superior to other available methods for the fair and efficient adjudication of the controversy.” Fed.R.Civ.P. 23(b)(3). The plaintiff bears the burden of showing that the proposed class meets the requirements for certification. Retired Chi [368] cago Police Ass’n v. City of Chicago, 7 F.3d 584, 596 (7th Cir.1993). In evaluating a motion for class certification we do not examine the merits of the case. Id. at 598. The defendants present challenges to the plaintiffs proposed class on every single one of these elements, but we find it unnecessary to analyze them all. Even assuming arguen-do that the class would satisfy the four elements of Rule 23(a), we cannot permit class certification because we conclude that the proposed class does not satisfy either of the requirements of Rule 23(b)(3) — predominance and superiority. B. Predominance of Common Questions “No precise test governs the determi- • nation of whether common questions of law or fact predominate.” 5 James Wm. Moore et al, Moore’s Federal Practice § 23.46[1] (3d ed.1997). In a recent opinion, the Supreme Court indicated that the purpose of the predominance requirement is to “test[ ] whether proposed classes are sufficiently cohesive to warrant adjudication by representation,” Amchem Prods. v. Windsor, — U.S. —, —, 117 S.Ct. 2231, 2249, 138 L.Ed.2d 689 (1997), but this does not help us to apply the rule in a given ease. We believe the proper approach is to try to determine if enough economies of time and effort would accrue to the judicial system from resolving the common questions on a class basis to justify the diminution of individual autonomy that inherently results from the certification of a class. Cf. 7A Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, Federal Practice & Procedure § 1777 (2d ed.1986) (urging courts to balance “the value of allowing individual actions to be instituted so that each person can protect his own interests” against “the economy that can be achieved by allowing a multiple party dispute to be resolved on a class action basis”). 1. Tort and Warranty Claims a. Variety of State Law Because this is a diversity case, the plaintiffs’ claims will be governed by the legal standards of the various states in which their injuries occurred. Recent circuit court opinions, including one from the Seventh Circuit, have expressed serious reservations about classes whose members will be governed by the tort laws of 50 different states. See Castano v. American Tobacco Co., 84 F.3d 734, 741-43 (5th Cir.1996); Georgine v. Amchem Prods., Inc., 83 F.3d 610, 627 (3d Cir. 1996), aff'd, — U.S. —, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997); In re American Med. Sys., Inc., 75 F.3d 1069, 1085 (6th Cir.1996); In re Rhone-Poulenc Rorer Inc., 51 F.3d 1293, 1300-01 (7th Cir.1995). Particularly with respect to negligence claims, it may be impossible to craft jury instructions that take into account the nuances in the laws of the various states. See In re Rhone-Poulenc, 51 F.3d at 1300. These cases do not go so far as to say that no nationwide mass tort class action could ever be certified, and indeed a few courts have decided to certify such classes notwithstanding choice of law problems. See, e.g., In re School Asbestos Litig., 789 F.2d 996 (3d Cir.1986); In re Copley Pharmaceutical, Inc., 161 F.R.D. 456 (D.Wyo.1995). Nevertheless, nationwide class certification is usually denied in cases involving allegedly defective pharmaceutical products. See In re American Med. Sys., 75 F.3d at 1089 & n. 24 (referring to a “national trend to deny class certification in drug or medical product liability/personal injury eases,” and collecting cases). In the face of this concern, the plaintiff offers only her opinion that the claims “can be handled in an acceptable manner by discerning general and manageable patterns,” PL’s Reply Br. at 18,2 and refers us to a handful of cases where courts certified nationwide classes notwithstanding variations [369] in applicable law. We do not find this cursory analysis persuasive. The comments of one leading treatise regarding the variety of products liability laws found in the 50 states are worth quoting at length: There is no monolithic products liability law in the United States. In fact, there is probably no significant aspect of products liability which is precisely the same in all American jurisdictions. Someone seeking an example of the diversity of our nation and the continuing supremacy of the system of federalism as embodied in the Tenth Amendment to the United States Constitution need look no farther than products liability.... [T]his Balkanization creates enormous confusion when comparing the detailed aspects of the law of one state with that of another. Strict liability in California is different from strict liability in New York, strict liability in Kansas, and strict liability in Alabama. In Michigan, while no separate strict liability cause of action exists, one commentator claims that Michigan’s breach of implied warranty is really strict liability. In Delaware, there is no strict liability at all .... In [New York], a products liability plaintiff may sue in negligence, breach of implied warranty of merchantability, breach of implied warranty of fitness for a particular purpose, breach of express warranty, strict liability in tort, as well as intentional or negligent misrepresentation. In [Connecticut, Kansas, Oregon, or Washington], that same plaintiff would have only a single cause of action under the state products liability act. Somewhere else, such as Indiana, that identical plaintiff could sue in negligence or strict liability but not warranty, while a few miles away in Michigan, he or she could only sue in negligence or warranty. 1 Louis R. Frumer & Melvin I. Friedman, Products Liability § 2.01 (1997) (citations omitted). We believe this kind of variation in state laws would make it difficult, if not impossible, to craft comprehensible jury instructions for the tort and warranty claims.3 Some courts may be undaunted by these difficulties and certify nationwide products liability classes anyway, but we think it will be rare to find such a court in the Seventh Circuit now that the court of appeals has indicated its distaste for “Esperanto” jury instructions. See In re Rhone-Poulenc, 51 F.3d at 1300. Certainly the class proponent in such a case faces a very difficult challenge of showing that variations in state law will not combine with factual variations in the plaintiffs’ cases to overwhelm any common issues presented. Cf. Georgine, 83 F.3d at 618 (“[Factual] differences, when expotentially magnified by choice of law considerations, eclipse any common issues in this ease.”). Fisher has not even made a credible attempt at such a showing. b. Causation Although we would normally consider each of the tort and warranty theories of liability separately in order to determine whether common issues predominate, we need not do so in this case because they all share a common Achilles heel: they all require the plaintiffs to make a showing that Stadol (and/or the defendants’ warnings about its use) caused the injuries of which they complain. See Restatement (Second) of Torts §§ 281,402A (1965); W. Page Keeton et al., Prosser and Keeton on the Law of ToRts § 41, at 263 (5th ed.1984); id. § 102, at 710; 2 Frumer & Friedman, supra, § 12.05[1]. As we see it, this determination involves three steps, each of which will require a detailed examination of the facts of each plaintiffs case. First, as a threshold matter, [370] even to quality for membership in the proposed class every plaintiff will need to establish that he or she was in fact addicted to Stadol. Addiction is not a simple concept. To find that a person is addicted to a particular substance, the factfinder must examine both biological and psychological evidence.4 This will require each plaintiff to provide a substantial amount of information about his or her medical history, emotional condition, and lifestyle, the significance of which will then be debated by experts. If a given plaintiff can convince the jury that he was addicted to Stadol, step two of the causation battle will require him to show that his physician would not have prescribed Stadol if the defendants had provided adequate warnings.5 See Thomas v. Hoffman-LaRoche, Inc., 949 F.2d 806, 812 (5th Cir. 1992) (“[I]n a prescription drug failure to warn case, the plaintiff must establish that an adequate warning would have convinced the treating physician not to prescribe the product for the plaintiff.”). This will depend on such factors as: (1) how carefully the prescribing physician read the Stadol product literature; (2) whether the doctor’s reading made him aware of the risk of addiction; (3) whether the doctor was otherwise aware of the risk that Stadol might be addictive, such as by exposure to other trade literature or even just on the basis of his own medical knowledge; and (4) whether the doctor was aware of alternative analgesic products that could have been used instead of Stadol without the same risks. The defendants will presumably present an array of challenges to the credibility of each physician’s answers to these questions. The third step in the causation inquiry will require the plaintiffs to prove that their Sta-dol addictions caused the damages they suffered. This step will present a huge number of individual questions because of the variety of potential harms plaintiffs might claim they suffered as a result of using Stadol: assorted physical side effects from prolonged use, the costs of professional help needed to break the addiction, pain and suffering related to withdrawal symptoms, emotional damages, and so forth.6 While differences in the amount of damages inflicted on each potential class member would not justify us in refusing to certify the class, see De La Fuente v. Stokely-Van Camp, Inc., 713 F.2d 225, 232 (7th Cir.1983), differences in the kind of damages each person suffered are a different matter because the causation inquiry will be different for each of them. A plaintiff claiming, say, that his Stadol addiction caused him to suffer cardiovascular problems (a possibility suggested by the Physicians Desk Reference) will need to make a factual presentation entirely different from a plaintiff claiming that his addiction forced him to incur the costs of psychiatric treatment for clinical depression. And even after the necessary facts related to each claim are established, they will be subject to conflicting expert testimony as to how they should be interpreted. [371] The array of individual questions involved in this three part causation inquiry— especially when combined with the underlying variation in applicable law — overwhelms the common issues suggested by Fisher, such as whether the defendants’ warnings adequately disclosed the product’s risks and whether Stadol is in fact a potentially addictive drug. While these questions are indeed common to the class, obtaining favorable answers to them advances each plaintiff only a short distance toward his or her goal of obtaining money damages. The alleged inadequacy of the warnings is the cornerstone of the plaintiffs case, but resolution of this issue seems likely to require just a brief presentation of expert testimony from each side. We therefore deny the motion for class certification with respect to the plaintiffs tort and warranty claims. 2. Consumer Fraud In addition to plaintiffs tort and warranty theories of liability, she also claims that the defendants’ inadequate warnings constituted consumer fraud. Fisher’s own consumer fraud claim arises under the Illinois Consumer Fraud Act (ICFA), 815 ILCS 505/1 et seq., which requires a plaintiff to show: (1) that the defendants engaged in a deceptive act or practice; (2) intent on the defendants’ part that the plaintiffs rely on the deception; and (3) that the deception occurred in the course of conduct involving trade or commerce. See Adler v. William Blair & Co., 271 Ill.App.3d 117, 207 Ill.Dec. 770, 648 N.E.2d 226, 233 (1995). The plaintiff need not show that she personally relied upon the deception, see id., but she must establish that the defendants’ acts were a proximate cause of her damages, see Connick v. Suzuki Motor Co., 174 Ill.2d 482, 221 Ill.Dec. 389, 675 N.E.2d 584, 593 (1997); Martin v. Heinold Commodities, Inc., 163 Ill.2d 33, 205 Ill.Dec. 443, 643 N.E.2d 734, 746-47 (1994). a. Variety of State Law Every state has some type of consumer fraud statute, so we are faced with the same kind of question we encountered with respect to the tort and warranty claims, namely, are there meaningful variations in the consumer fraud laws of the 50 states? The plaintiff claims that there are not, and bolsters this assertion by referring us to an article surveying consumer fraud statutes. See Pl.’s Reply Br. at 16-17.7 This article concludes that all of the state statutes have “similar substantive provisions” and that they can easily be broken down into three or four groups to take into account the most important nuances. Sandra Benson Brantley & Beverly C. Moore, Jr., Commonality of Applicable State Law in Nationwide or Multistate Class Actions — Deceptive Trade Practices, 18 Class Action Reports 188, 194 (1995). Though this article persuades us that there is substantial similarity in the states’ respective articulations of the basic elements of consumer fraud, troublesome variations remain. Two of the most significant are that 10 states require the plaintiff to show that the defendant acted with scienter when it perpetrated the deceptive act or practice, and four states require the plaintiff to show that she relied on the deception. See id. at 198-99. While it might be possible to craft a special verdict form in such a way that the jury could indicate whether the defendants acted with scienter without allowing this issue to cloud their judgment about the liability requirements for the other 40 states, the matter of individual reliance is obviously not susceptible to class treatment. Citizens of the four states that require reliance (Arizona, Georgia, Indiana, and Wyoming) would have to be either (1) excluded from the class, or (2) left to make a showing of reliance on their own in later proceedings. Another commentator cautions that “substantial problems can arise in truly nationwide class actions if the court utilizes each state’s [Uniform Deceptive Acts and Prac [372] tices] statute because of variations among state statutes as to statute of limitations, notice requirements, special preconditions to class actions ... and the like.” Jonathan Sheldon, Unfair and Deceptive Acts and Practices § 8.3.5 (3d ed.1991). The statutes of limitations imposed on consumer fraud claims vary widely in length and begin to run at different times (some states from the date of the act or practice, some from the date of discovery by the plaintiff). See Brantley & Moore, supra, at 212-14. Six states require the plaintiff to send a detailed demand letter to the defendant prior to filing suit, and six bar class actions for consumer fraud entirely. See id. at 214-15. Furthermore, there is little consensus regarding the types and quantities of damages that are permitted for consumer fraud claims. See id. at 200-04. While these variations in state law are not as overwhelming as those related to the tort and warranty claims, we have serious reservations about certifying a nationwide class, especially since the plaintiff has made no effort to eiqilain how these variations could be handled. b. Causation Even assuming variations in state law could be overcome, we must also consider whether common factual questions would predominate if we certified the class specified by the plaintiff. The basic factual questions of liability — whether the defendants’ conduct was deceptive and whether they intended it to be so — are obviously common to the class. See Rohlfing, 172 F.R.D. at 340 n. 11. But each plaintiff must also show that the deceptive acts caused his or her injuries. See Connick, 221 Ill.Dec. 389, 675 N.E.2d at 593; Haesche v. Kissner, 229 Conn. 213, 640 A.2d 89, 94 (1994); Ellis v. Northern Star Co., 326 N.C. 219, 388 S.E.2d 127, 131 (1990); Massachusetts Farm Bur. Fed’n v. Blue Cross, 403 Mass. 722, 532 N.E.2d 660, 665 (1989); see also Sheldon, supra, § 8.2.2.8 & n. 81 (3d ed.1991) (collecting state cases supporting the proposition that “the consumer has the clear responsibility to prove the amount of [his or her] damages, and how they are related to the deceptive conduct”). For the same reasons discussed above with respect to the tort and warranty claims, this causation inquiry will involve innumerable individual questions. And the amount of damages suffered by each plaintiff is, by its very nature, a question that will require individualized (and probably lengthy) proceedings. In light of these substantial individual questions of causation and damages, we cannot conclude that the plaintiff has carried her burden of showing that common questions predominate with respect to the consumer fraud claim as a whole. But common questions would predominate if we were to narrow our focus to the basic liability questions alone: we have already observed that, unlike tort and warranty law, the law governing liability for consumer fraud is reasonably similar from state to state. The Federal Rules give us the power to certify a class for the limited purpose of resolving basic liability issues alone, see Fed.R.Civ.P. 23(c)(4)(A), but it remains to be considered whether doing so would be a superior method of resolving the parties’ controversy. C. Superiority of Class Action The final element a class action proponent must show is that a class action “is superior to other available methods for the fair and efficient adjudication of the controversy.” Fed.R.Civ.P. 23(b)(3). In making this assessment, Rule 23 expressly invites courts to consider “the interest of members of the class in individually controlling the prosecution or defense of separate actions.” Fed.R.Civ.P. 23(b)(3)(A). The defendants argue that a class action would not be a superi- or method of resolving the controversy in this ease because most of the members of the proposed plaintiff class would have claims large enough to permit them to bring suits on their own, and that they therefore have strong interests in individually controlling the prosecution of their cases. The Third Circuit recently observed that “claims for personal injury and death ... have a significant impact on the lives of the plaintiffs and [they] frequently receive huge awards in the tort system. Plaintiffs have a substantial stake in making individual decisions on whether and when to settle.” See Georgine, 83 F.3d at 633 (citation omitted). This exact language was quoted with approval by the [373] Supreme Court in Amchem, — U.S. at —, 117 S.Ct. at 2246, which further observed: “While the text of Rule 23(b)(3) does not exclude from certification cases in which individual damages run high, the Advisory Committee had dominantly in mind vindication of ‘the rights of groups of people who individually would be without effective strength to bring their opponents into court at all.’ ” See also Mace v. Van Ru Credit Corp., 109 F.3d 338, 344 (7th Cir.1997) (“The policy at the very core of the class action mechanism is to overcome the problem that small recoveries do not provide the incentive for any individual to bring a solo action ----”); Rhone-Poulenc, 51 F.3d at 1299 (“In most class actions — and those in which the rationale for the procedure is most compelling — individual suits are infeasible because the claim of each class member is tiny relative to the expense of litigation.”). The plaintiff responds to this argument with the unexplained assertion that “[a]ll of the proposed class member’s [sic] damages were not so great that individual suits would be feasible.” Pl.’s Reply Br. at 19. While it is possible that some class members will not have huge claims for damages, the nature of the complaint and the definition of the class make it unlikely that many class members will have de minimis claims. The plaintiffs theory is that the defendants’ wrongful conduct caused each of the prospective class members to become addicted to a dangerous narcotic. Common sense tells us that if this is proven virtually every member of the class will be entitled to significant damages. The remaining question is whether the efficiency to be gained by resolving the basic liability questions involved in the consumer fraud claims on a class basis outweighs the interest of the plaintiffs in controlling their own cases. Neither party provides any analysis of this issue. It is certain that resolving these questions on a class basis would have some efficiency benefits, but we think these benefits would be small for two reasons. First, as is made clear in our earlier discussion of predominance, each prospective plaintiff is going to be involved in extensive individualized proceedings regardless of whether we certify a class on the consumer fraud liability questions or not. Each plaintiff will have to prove causation and damages for their fraud claim in a separate proceeding, and they will have to bring entirely separate lawsuits if they desire to pursue other tort and warranty theories. In this setting, little efficiency is gained by singling out one or two discrete issues for class treatment. Second, we believe that most of the efficiency gains that would result from certifying a consumer fraud liability class could also be obtained through the use of the doctrine of collateral estoppel. This doctrine provides that a “right, question or fact distinctly put in issue and directly determined by a court of competent jurisdiction ... cannot be disputed in a subsequent suit between the same parties or their privies.” Montana v. United States, 440 U.S. 147, 153, 99 S.Ct. 970, 59 L.Ed.2d 210 (1979). The Supreme Court has made clear that the doctrine can be used “offensively” — that is, by a plaintiff against a defendant — even if the party asserting the estoppel was not a party to the prior lawsuit in which the issue was resolved. See Parklane Hosiery Co. v. Shore, 439 U.S. 322; 331, 99 S.Ct. 645, 58 L.Ed.2d 552 (1979); see also 18 Moore et al., supra, § 132.04[2][c] (generally discussing offensive collateral estoppel). The main limitation on the doctrine, that the party against whom the estoppel is asserted have a full and fair opportunity to litigate the issue in question, see Montana, 440 U.S. at 153-54, 99 S.Ct. 970, would not preclude its operation here because the defendants will obviously be given a full and fair opportunity to litigate the consumer fraud claim in Fisher’s case. Although application of this doctrine is a matter of discretion for each court in which it is asserted, see Parklane, 439 U.S. at 331, 99 S.Ct. 645, and thus it is not certain that every prospective . plaintiff could benefit from it,8 we think that [374] most courts would permit its application, thereby achieving some of the same efficiency gains that a class action would achieve. In sum, the plaintiff has not convinced us that the efficiency to be gained from certifying a class limited to the consumer fraud liability questions would outweigh the fairness interests of the prospective plaintiffs in controlling their own cases. Consequently, we do not think a class action is a superior method of resolving this controversy. III. Conclusion For the foregoing reasons, plaintiffs motion for class certification is denied. It is so ordered. Footnotes [1] In evaluating a motion for class certification we generally take the plaintiff's allegations as true so as to avoid prejudging the merits of the case, see Hardin v. Harshbarger, 814 F.Supp. 703, 706 (N.D.Ill.1993), but we are permitted to "probe behind the pleadings” if doing so will facilitate our resolution of the motion, General Tel. Co. v. Falcon, 457 U.S. 147, 160, 102 S.Ct. 2364, 72 L.Ed.2d 740 (1982). [2] In this respect the plaintiff's arguments are distinguishable from those in the School Asbestos case, which is frequently cited by courts that have elected to certify nationwide mass tort classes despite the need to apply different state tort laws, because there the plaintiffs supported . their arguments for certification with an "extensive analysis of the variances in products liability among the jurisdictions.” School Asbestos, 789 F.2d at 1010. In any event, School Asbestos hardly constitutes a ringing endorsement of the practice. Although the Third Circuit found that the district court's decision to certify a nationwide class was not an abuse of discretion, it expressed "some doubt” as to whether the variations in state laws would be manageable. Id. [3] One might think that this concern would not apply with as much force to the plaintiffs' warranty claims since those claims arise under the Uniform Commercial Code, but courts and commentators alike have noted that the treatment of warranty claims from state to state is far from uniform. See, e.g., Walsh v. Ford Motor Co., 807 F.2d 1000, 1016-17 (D.C.Cir.1986) (Ruth Bader Ginsburg, J.) (expressing doubt that state warranty laws are uniform and finding that the plaintiffs had not carried their burden of establishing that they are); 1 Frumer & Friedman, supra, § 2.01 n. 1 ("[A]lthough 49 states have adopted article 2 of the Uniform Commercial Code, from the standpoint of products liability, the warranty sections are often treated differently from state to state.”). [4] One source defines "addiction" as "a state of heavy dependence on a drug, defined by some authorities as a state of physical dependence characterized by tolerance or withdrawal and by others in a wider sense that includes also emotional dependence, i.e. compulsive or pathological drug use, also referred to as abuse or habituation.” Dorland’s Illustrated Medical Dictionary at 25 (28th ed.1994). We offer this definition for illustrative purposes only, and take no position on whether or not the definition of addiction should include the "wider sense” of emotional dependence. In fact, we suspect the definition of addiction is a question about which the parties will disagree, and on which the states' laws may differ. [5] Some states employ a rebuttable presumption that when a defendant has given an inadequate warning about a product, the plaintiff's injury would have been avoided if an adequate warning had been given. See 2 Frumer & Friedman, supra, § 12.05[3]. Since the presumption is rebuttable it does not preempt the causation inquiry, and it serves as a good example of the kind "nuance” in state law that might be very important to the outcome of a particular plaintiff’s claims. [6] We suspect that Natalie Fisher's claim will be unusual, if not unique, in that she is claiming that Stadol addiction caused her son’s suicide. This will require her to confront an extraordinarily difficult and highly personal causation question: would Bruce have killed himself even if he had never become addicted to Stadol? While few (if any) other members of the proposed class will have to confront questions quite this painful, they will all have vexing proof problems related to whether their alleged injuries were caused by Stadol or some other factor. [7] Plaintiff's argument on this point requires us to handle the question of whether to certify a consumer fraud class differently than we did in Rohlfing v. Manor Care, 172 F.R.D. 330 (N.D.Ill.1997). In Rohlfing, the plaintiff only sought certification of a nationwide class seeking relief under the ICFA, see id. at 339; he never argued that other states’ consumer fraud laws were substantially similar to Illinois’, so we never considered whether certification would be appropriate on that basis. [8] The Supreme Court has said that the "general rule should be that in cases where a plaintiff could easily have joined in the earlier action ... a trial judge should not allow the use of offensive collateral estoppel.” Parklane, 439 U.S. at 331, 99 S.Ct. 645. We think it is clear from our earlier discussion of predominance that the non-Illinois plaintiffs could not "easily” join in Fisher's lawsuit since different law will apply to their claims. As for the Illinois plaintiffs, how easily [374] they could have joined Fisher's case will have to be determined case-by-case, based on such factors as how similar their claims are to Fisher's and whether they knew about her suit.