Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
8 Products Liability and Contraceptive Development This chapter examines recent trends in contraceptive products liability litigation and insurance and evaluates their effect on the pace of development of new contraceptives in the United States. The chapter begins with an overview of the legal environment in which contraceptive products are developed. This discussion is followed by a description of the trends in litigation involving contraceptives, a discussion of current products liability rules, and a summary of selected contraceptive cases. The chapter continues with a description of the products liability insurance environment and, finally, presents the committee's conclusions and recommendations. THE LEGAL TERRAIN When a pharmaceutical company or nonprofit organization contemplates development of a new contraceptive, it does not look at products liability rules and cases as isolated phenomena, but at the legal landscape they create. From the perspective of those developing or thinking about developing or marketing a new contraceptive, that landscape can appear intimidating. First, since all the pharmaceutical firms with the resources to make substantial investments in contraceptive development are national firms that market products on a national basis, they face the prospect of different rules with regard to products liability in each of the 50 states. In practice, there is a good deal of uniformity among the jurisdictions because of the adoption by courts of the Restatement (Secorut) of Torts (American Law Institute and National Conference of Commissioners on Uniform State Laws, 1965), a text containing a general 118
PRODUCTS VIABLY AND CO=~CEPT~E DEVE~PME ~1 19 statement and analysis of products liability law, and the Uniform Commercial Code (1976), a model set of statutes. Nonetheless, companies operating in a national market continue to face the costly uncertainty that arises because 50 different jurisdictions have the power to make and change products liability rules. Second, a manufacturer involved in a products liability action will almost always have its case decided by a judge and jury who, although scientifically untrained, must evaluate highly technical and complex scientific issues, often on the basis of only the conflicting testimony of experts retained by the parties. Moreover, judges and juries are much more inclined to be sympathetic to injured plaintiffs, especially mothers and children, than they are to corporate defendants. Third, if a manufacturer becomes involved in a products liability action, it is not shielded from liability by Food and Drug Administration (FDA) approval of the product. This is so despite the rigorous testing required by the FDA before a new product is marketed, the practice of the FDA to give great weight to safety considerations (see Chapter 7), and the warnings and instructions that the FDA requires be given. Fourth, both the number of suits and the size of the awards are widely perceived to have increased dramatically in recent years, and future trends are unpredictable. Developers and manufacturers of contraceptives feel that they are substantially more likely to be sued today than in the past and, if they lose, to have to pay (in real terms) more than they would have had to pay in the past. Litigation expenses can be high and are incurred even when claims are meritless. Commercial liability insurance has sometimes proven impossible for some developers and manufacturers to obtain, and very expensive for those who have been able to secure it. The adverse media attention often associated with products liability claims and contraceptive-related injuries may independently discourage manufacturers from research, development, and the marketing of contraceptive products. Apart from the effect of such publicity in stimulating legal claims, the manufacturer may suffer a loss of public confidence in its other unrelated products, even if the manufacturer is not found liable. SOURCES OF DATA In an attempt to assess the magnitude and frequency of contraceptive products liability claims in the United States over the past two decades, the committee Recent legislation in New Jersey, Ohio, Oregon, and Texas allows a manufacturer of an FDA- approved product to assert an "FDA defense" in response to a claim for punitive damages~hat is, generally the manufacturer cannot be held liable for punitive damages if the product has been manufactured and labeled in accordance with FDA standards unless the manufacturer withheld from or misrepresented to the FDA material and relevant information. Ohio Rev. Code Ann. §2307.80(C) (Page Supp. 1987); 1987 Or. Laws ch. 774 §5, Prod. Liab. Rep. (CCH) pare. 93,835; Tex. Civ. Pract. and Rem. Code Ann. §81.001 et seq. (Supp. 1989); 1987 NJ. Laws, ch. 197.
120 DEVELOPING NEW CO=RACE~IVES examined a number of data sources. Legal counsel and products liability attorneys were contacted at companies currently marketing contraceptive products in the United States: Ortho Pharmaceutical Corporation; G.D. Searle & Co.; Wyeth Laboratories; Syntex Laboratories, Inc.; Parke-Davis & Co.; Mead Johnson Laboratories; ALZA Corporation, Berlex Laboratories, Inc.; Schmid Laboratories, Inc.; Whitehall Laboratories, Inc.; and GynoPharma, Inc. Representatives of A.H. Robins Company were contacted regarding liability for the Dalkon Shield. Company annual reports and quarterly reports were also reviewed for information on liability claims and recent settlements for contraceptive products. Some drug companies provided fairly detailed information on the number of cases and the magnitude of the awards, while others were reluctant to provide any information. In order to identify court cases involving products liability of contraceptives, several computer searches were performed using Mead Data Central, Inc.'s LEXIS Service, a computer-assisted legal research service. The LEXIS computer data base consists of decisions of federal courts and most state courts. From the searches it was possible to identify contraceptive products liability cases in which judicial opinions were issued. The number of reported cases, however, is only a subset of the total number of such cases because LEXIS does not identify pending cases or cases decided or settled without a judicial opinion. LEXIS also does not generally produce references to state trial-level cases. Other traditional legal research sources, such as state and federal topical digests, law review articles, and articles containing case annotations, were also consulted to obtain and confirm data on products liability cases involving contraceptive products. Pharmaceutical Litigation Reporter, OBlGYN Litigation Reporter, and Jury Verdict Information Reports were also reviewed. A number of independent consumer organizations, lawyers' associations, and victims' network organizations were contacted for additional information on contraceptive liability cases. Clearly, the committee's search of available data sources did not yield all the products liability actions initiated against contraceptive manufacturers. It also did not identify many of the claims that were settled without initiation of litigation. Because the vast majority of cases are settled out of court for dollar amounts that are unreported and for reasons that are not publicly known, the committee was unable to obtain access to much of the potentially relevant data. In addition, cases decided at the state trial level, whether resulting in judgments for manufacturers or for injured plaintiffs, are usually unpublished; such cases may be published only if and when they are appealed. Nonetheless, the committee believes that the information assembled on the number of reported cases is broadly indicative of the much larger number of such cases settled without even a court filing, filed but settled, or tried but not appealed. The information gathered is important because it is available to contraceptive manufacturers and may affect their behavior (e.g., changes in labeling) and is also used in settlement decision making to assess the likelihood of success in a particular case.
PRODUCTS LIABlL~Y AND CONTRACEPTIVE DEVELOPMENT ~ 2 TRENDS IN LITIGATION INVOLVING CONTRACEPTIVES Intrauterine devices, om1 contraceptives, spermicides, diaphragms, and condoms have all been the subjects of products liability litigation. The IUD litigation has been characterized by extremes: massive liability in the case of one product, the Dalkon Shield, but, with one significant exception of a case still under appeal, limited recovery by plaintiffs in most cases involving other IUDs. The litigation concerning oral contraceptives is characterized by its long-running nature, the adjustments manufacturers have made in response to the lawsuits, and the difficult issues regarding adequate warnings. The one reported case involving a spermicide surprised the legal and scientific communities with a decision that was contrary to currently accepted theories of causation. Intrauterine Devices Figure 8.1 shows the number of products liability cases by year that were filed against manufacturers of IUDs and oral contraceptives and reported by the sources surveyed. Prior to 1970 there were no reported products liability cases involving manufacturers of IUDs. The number of reported IUD cases reached its peak during the 1984-1986 period, the same period in which several IUDs were withdrawn from the market and A.H. Robins Company filed for bankruptcy. Prior to 1986, Dalkon Shield cases predominated, representing well over half the 30 28 26 24 LL 22 CD 20 18 A: o 16 AL ~ 14 a: 11 12 o 10 m 8 z 6 4 2 O Pill Non-Dalkon Shield IUD _ · Dalkon Shield IUD ,~ Am\ Hi it's %~ ,, 4~~ 77 78 79 80 81 82 83 84 85 86 87 88 YEAR 71 72 73 74 75 76 FIGURE 8.1 Yearly reported oral contraceptive and IUD cases.
122 DEVELOPING NEW CO=RACE~IVES IUD cases reported by the data sources surveyed. Although the vast majority of cases reported in the sources we surveyed have been settled out of court, a number of cases have gone to trial and punitive damages have been awarded to plaintiffs. At least six firms have distributed IUDs since the introduction of the first IUD in the United States in 1964. In 1975, A.H. Robins Company stopped selling its controversial Dalkon Shield, and in 1985 Robins filed for bankruptcy after more than 4,000 lawsuits concerning the IUD had been filed against the company. Schmid Laboratories, Inc., withdrew the Saf-T-Coil in 1983. In 1985, Ortho Pharmaceutical Corporation discontinued marketing the Lippes Loop, and in 1986 G.D. Searle & Co. discontinued marketing the Copper-7 and the Tatum-T. With the exception of the Dalkon Shield, all those devices were withdrawn even though the FDA did not raise questions about their safety and very few successful lawsuits had been brought against the manufacturers. Only two firms are currently selling IUDs in the United States. ALMA Corporation has been marketing the Progestasert IUD since 1976, and GynoPharma, Inc., has been marketing the ParaGard IUD (Copper T380A) since 1988. Both {UDs are accompanied by detailed informed consent guidelines. Oral Contraceptives Figure 8.1 also shows the number of products liability cases by year that were filed against manufacturers of oral contraceptives and were reported in the sources surveyed. The patterns in reported oral contraceptive cases appear to correspond closely to the patterns in reported IUD cases, with the number of cases peaking during the 1984-1986 period. Assuming the number of reported oral contraceptive cases fairly represents the larger number of such cases filed that are settled or tried but not appealed, they present a picture of manufacturers continuing to market products despite numerous lawsuits. The litigation against manufacturers of oral contraceptives has persisted over the past 17 years. In sharp contrast to the situation with respect to II3Ds, however, oral contraceptives continue to be marketed by seven companies, although over the past two decades manufacturers have complied with an FDA request to remove some standard and high-dose oral contraceptive formulations from the market. Despite the litigation costs, these products continue to be profitable because the market is relatively large and the monthly cost of the pills is high. Other Contraceptives Our review of litigation sources identified one case brought against a spermicide manufacturer (1984), one case brought against a diaphragm manufacturer (1988), one case brought against a condom manufacturer (1981), and no cases reported concerning other contraceptive drugs or devices. The plaintiff in the spermicide case was awarded over $5 million in damages. The plaintiff in the diaphragm
PRODUCTS [lABIL~Y AND CONTRACEPTIVE DEVELOPMENT 123 case received damages of $1.5 million. The condom case, in which the plaintiff sought damages for the "wrongful birth" of twins allegedly caused by the product's failure, was eventually dismissed and the plaintiffs received no damages. To assist in understanding the impact of such litigation on contraceptive development, we provide a brief introduction to products liability law and substantive summaries of selected cases involving IUDs, oral contraceptives, and other contraceptives. PRODUCTS LIABILITY RULES The legal rules regarding products liability that have been applied to the manufacturers of contraceptives are the same ones applied to all other manufacturers, whether of lawnmowers, bicycles, or electric drills. These rules generally have their source in common law, not statutory law; that is, they are rules made for the most part by judges rather than laws enacted by Congress or state legislatures. The objectives of products liability rules are to compensate people injured by unsafe products, to deter the marketing of dangerous or defective products, and to resolve fairly disputes between persons injured by a product and the manufacturer of that product (Smith, 1987~. If, in creating or applying the common law liability rules, courts fail to properly balance the interests of the injured person and the manufacturer, or fail to consider adequately the impact of their rulings on the interests of persons beyond the immediate parties to the litigation, then it is the responsibility of Congress or the state legislatures to correct the balance or account for the wider interests by enacting appropriate laws. Any nonfederal corrective legislation has to be enacted state-by-state, since products liability rules-as with tort law generally-are the subject of the laws of each of the 50 states and the District of Columbia. A federal products liability law, which would preempt state laws, has been proposed (U.S. Congress, 1987), but political agreement on the content of such a measure-and indeed on the wisdom of such a law whatever its form and substance has been lacking. In the absence of federally mandated uniformity, state legislatures may choose to adopt the Uniform Product Liability Act2 (U.S. Department of Commerce, 1979; see also Schwartz, 1980), which was drafted to serve as a set of model standards for state products liability law. A manufacturer of a product may be held liable to an injured user of the product under any one of five legal theories: express warranty, implied warranty, fraudulent misrepresentation, negligence, and strict liability. In most lawsuits, 2Section 106 of the Uniform Product Liability Act provides that a product seller cannot be found liable on the basis of defective design or failure to warn if the product conformed to an applicable administrative or legislative regulatory standard. This provision does not apply, however, if the claimant can prove by a preponderance of the evidence that a reasonably prudent product seller could and would have taken additional precautions.
124 DEVELOPING NEW CONTRACEPTIVES plaintiffs endeavor to base their claims on two or more of these theories. The elements of the five theories of liability are summarized below. Warranty A warranty claim may be based on an express or an implied warranty theory, or both. An express warranty is a written or oral affirmation of fact or promise made by the seller of the product to the buyer about the condition, efficacy, or safety of the product. An express warranty claim arises when a plaintiff-buyer asserts that the purchased product does not conform to the defendant-seller's representations. Thus, the arguments in many express warranty cases concern (1) whether the seller actually made a statement of fact to the buyer about the product if the seller merely expressed an opinion, an express warranty claim is not supportable; (2) the meaning or interpretation of any such statement; (3) whether such statement was true or false; and (4) whether the product caused the plaintiffts harm. An implied warranty is a representation by the seller that is implied in a contract for the sale of the product that the product is "merchantable," that is, `'fit for the ordinary purposes for which such goods are used." To recover under the theory of implied warranty, the seller must be a merchant within the statutory definition, and in most states manufacturers have been held to be so (see Gillespie v. Thomasville Coca-Cola Bottling Co., 17 N.C. App. 545, 195 S.E.2d 45, cert. denied, 283 N.C. 393, 196 S.E.2d 275 [19731~. Because all sales are contracts, this theory presents a broad avenue for recovery against sellers of defective goods. Although recovery under a warranty theory against the manufacturer of a defective product in many states has been available in common law for roughly 100 years (Birnbaum, 1980), recovery in most states today is governed by the relevant provisions of the Uniform Commercial Code, as adopted by each state. Fraudulent Misrepresentation Fraudulent misrepresentation is akin to the warranty theory of liability but has an additional requirement that the plaintiff prove fraud and deceit. In a products liability suit based on fraudulent misrepresentation, the plaintiff must prove the following elements: (1) the defendant made a false representation about the product; (2) the defendant knew the representation was false; (3) the defendant intended to induce the plaintiff to act or refrain from acting on the basis of the representation; (4) the plaintiff justifiably relied on the representation; and (5) the plaintiff was injured thereby [American Law of Products Liability 3d (1987), Vol. 2, §25: 1 (Lawyers Co-op. Pub. Co., Rochester, N.Y.~. Some courts have held that the plaintiff must prove fraud and deceit by "clear and convincing" evidence, rather than by the usual standard of "more likely than not." For this reason, and because of the difficulty of proving that the defendant knowingly misrepresented a fact about the product, products liability lawsuits are not usually based on fraudulent misrepresentation (Id. §25.2~.
PRODUCTS LIABl~Y^DCO=RACEPT~EDEVELOPME ~125 Negligence In products liability suits based on negligence theory, plaintiffs must show that the defendant-manufacturer owed the plaintiff a `'duty of care"; the defendant breached this duty of care (acted "unreasonably," was "negligently; the plaintiff was injured; and the defendants lack of care was the proximate cause of the injury (Harper et al., 1986~. As these rules have been developed in products liability cases, it can be said that today a manufacturer owes a duty of care to avoid an unreasonable risk of harm to the user of the product (Harper et al., 1986~. A manufacturer can breach this duty of care in three broad respects: (1) by adopting a design for the product that causes it to be unreasonably dangerous; (2) by making mistakes or omissions in the manufacturing process that result in a properly designed product becoming unreasonably dangerous; (3) or by failing to provide adequate warnings about the product's hazards and instructions concerning its use (Ha~per et al., 1986~. In most jurisdictions the manufacturer must warn only the medical profession, not the patient~he "learned intermediary" rule- and the adequacy of the warning is a matter frequently in contention (Ha~per et al., 1986~. Strict Liability in Tort The use of strict liability theory in products liability cases is a relatively recent development, dating from a 1963 decision of the California Supreme Court (Greenman v. Yuba Power Products, Inc., 59 Cal. 2d 57, 377 P.2d 897 [19631) and the publication in 1965 of section 402A of the Restatement (Second) of Torts (American Law Institute and National Conference of Commissioners on Uniform State Laws, 1965~. Under a strict liability theory an injured user of a product may recover from the manufacturer without showing that the manufacturer was negligent (breached a duty of care). A strict liability claim may be based on three independent theories: design defect, which includes defective testing; manufacturing defect; and failure to warn. To recover under any of these theories, the plaintiff must show something more than that he or she used the product and was injured by it- but both courts and commentators show a remarkably high degree of uncertainty over the meaning and scope of the doctrine (see Harper et al., 1986; Keeton, 1984; Owen, 1980; Schwartz, 1979~. Section 402A of the Restatement (Second) of Torts has had a profound impact on the development of strict liability in the courts. Although intended as a summary statement of common law rules and in no way binding on courts, the Restatement has been widely adopted by many jurisdictions. Section 402A imposes liability on a manufacturer if it sells a product "in a defective condition unreasonably dangerous to the user or consumer . . . for physical harm thereby caused to the ultimate user or consumer...." That the manufacturer exercised "all possible care" in designing and manufacturing the product does not absolve it from liability.
126 DEVELOPING NEW CO=RACE~IVES Two explanatory notes or comments to section 402A, comment j and comment k, have played important roles in cases involving drugs and medical devices. Comment j states that the manufacturer may prevent the product from being considered "unreasonably dangerous" within the meaning of section 402A by giving appropriate directions or warnings, and (as in negligence cases) a much- contested issue is the adequacy of the warning. Comment k sets forth a special risk-benefit policy applicable to certain drugs that are considered unavoidably unsafe. In particular, provided that the product is properly prepared and marketed and a proper warning is given, a drug that puts the user at a significant risk of harm is not to be considered unreasonably dangerous if the consequences of not using the drug also entail substantial risks; thus, the manufacturer should not be held responsible. The rationale of comment k, as explained therein, is that the seller-manufacturer should not be held strictly liable for unfortunate consequences attending a product's use, "merely because he has undertaken to supply the public with an apparently useful and desirable product, attended with a known but apparently reasonable risk." Courts in products liability cases involving drugs frequently employ comment k, and in such cases the only significant issue is usually the adequacy of the warning. CASE STUDIES Depending on the facts of the case, one or any combination of these five legal theories can be used by an injured plaintiff as a basis for liability claims against a contraceptive manufacturer. A review of selected contraceptive cases provides insight into the dynamic nature of the legal process. Many of the principles of these cases are reviewed by plaintiffs and defendants to assess the likelihood of success of potential cases. Often, manufacturers appear to adjust product warnings in response to case opinions. Frequently these cases show an alertness by plaintiffs' attorneys to locating and using new scientific studies that may not have achieved general scientific acceptability to contend for more extensive warnings. Cases involving IUDs, oral contraceptives, and other contraceptives are discussed separately, since each product's litigation has exhibited its own distinguishing characteristics. The Dalkon Shield Intrauterine Device The development and subsequent withdrawal from the market of the Dalkon Shield merits separate discussion for several reasons: it shows the risks associated with nonregulation of contraceptives because, at the time of its introduction, the Dalkon Shield was not subject to premarketing approval by FDA; it may have influenced the calculations of liability insurers in establishing premiums for
PRODUCTS [lABl~TY AND CONTRACEPTIVE DEVELOPMENT 127 contraceptive manufacturers generally and manufacturers of IUDs in particular; and it illustrates the operation of the legal system in addressing the problems raised by a defective product, particularly the compensation of injured persons and the imposition of penalties on the manufacturer of a defective product. The IUD that was to become known as the DaLkon Shield was invented in 1968 by Irwin S. Lerner, who modified an earlier design by Dr. Hugh J. Davis (Palmer v. A.H. Robins Co., Inc., 684 P.2d 187 [solo. 19841~. Later that year, Davis began a one-year test of the shield at the family planning clinic he directed. According to Davis, 640 insertions resulted in 5 pregnancies, the equivalent of a 1.1 percent rate during the testing period (Davis, 1970~. In June 1970, A.H. Robins Company purchased all rights to the shield, although two members of its medical department had questioned the validity of Davis's study (Palmer at 195~. The first reported that the pregnancy rate had climbed to 5.5 percent after 14 months, and the second stated that the period of the study was "not long enough . . . to project with confidence to the population as a whole" (Palmer at 195~. A.H. Robins Company began national marketing of the shield in January 1971.3 In promotional advertising to both the medical profession and the public, Robins made the following claims for the shield: "the modern superior I.U.D."; "lowest pregnancy rate [of] 1.1 %"; and "provides safe, sure, sensible contraception" (Palmer at 195-196~. In June 1971, a quality-control supervisor at the subsidiary that assembled the shield reported to management that he had performed an experiment that demonstrated that the multifilament tailsaing on the shield would wick fluid its entire length, thereby drawing bacteria from the vagina into the uterus of a woman wearing it. No changes were made in the product (Palmer at 195-196~. Prom June 1972 to November 1973, Robins received 22 reports of spontaneous septic abortions in shield users, one of which resulted in death, but the firm continued to advise physicians, as late as April 1973, to leave the shield in place in the event of pregnancy (Palmer at 196~. In May 1974 the Centers for Disease Control reported to the FDA the results of a survey of physicians regarding IUD-related disease and injury. The survey found that, among patients who conceived with the IUD in place, the incidence of complications was 61.6 percent for the Dalkon Shield, 29.6 percent for the Lippes Loop, and 6.9 percent for the Saf-T-Coil; the majority of these complications were related to pelvic infection (Kahn and Tyler, 1976~. One month later, Robins 3As discussed in Chapter 7, in 1971 the Federal Food, Drug and Cosmetic Act did not require approval by FDA before a medical device could be marketed. FDA could take enforcement action against a device if it could establish that the device was adulterated or misbranded (21 U.S.C. §331 (a)- (c), 351, 352 [1970]). ~ 1976, Congress enacted the Medical Device Amendments, which require premarketing approval for such devices as the Dalkon Shield (Pub. L. No. 94-295, codified at 21 U.S.C. §360-360K).
|28 DEVELOPING NEW CO=~CE~~ES proposed to the FDA that it voluntarily suspend sales of the Dalkon Shield, and the FDA accepted the proposal. In January 1975, Robins recalled all unsold Dalkon Shields in the United States (A.H. Robins, 1983~. The first verdict in a Dalkon Shield case was returned in February 1975 against Robins for $10,000 in compensatory damages and $75,000 in punitive damages (Couric, 1986~. During 1980, Robins incurred over $4 million in litigation expenses and settlements related to the Dalkon Shield (A.H. Robins, 1980~. In September 1980, Robins issued a letter to physicians recommending removal of all Dalkon Shields that were still being worn. After paying $250 million to settle approximately 4,400 suits and after juries in 11 cases had awarded $24.8 million in punitive damages against it, Robins petitioned for protection from creditors under chapter 11 of the Bankruptcy Code on August 21, 1985 (Bureau of National Affairs, 1987~. Approximately 320,000 claims were filed by April 30, 1986, the deadline set by the bankruptcy court for filing Dalkon Shield-related claims (Bureau of National Affairs, 1987~. In December 1987, Judge Robert Merhige, Jr., of the U.S. District Court for the Eastern District of Virginia, ordered the company to set aside $2.475 billion in its bankruptcy plan of reorganization to compensate claimants injured by the shield (A.H. Robins, 1988~. It is very unlikely that the Dalkon Shield would be approved for distribution and sale today, and it is certain that it would not be approved on the basis of the single premarketing test conducted in 1968. The history of the shield illustrates the operation of the legal rules of products liability, functioning in the absence of premarketing review by FDA, to achieve the objectives of compensation, deterrence, and dispute resolution by causing a defective device to be taken off the market and by providing a mechanism for compensating thousands of women injured by it. Other Intrauterine Devices With the exception of one recent case, there has been very limited recovery by plaintiffs from manufacturers of IUDs other than the Dalkon Shield. In the cases summarized below, the manufacturer-defendants prevailed. The fast two cases, involving the Copper-7 IUD, show judges and juries struggling with the difficult issue of what caused the plaintiff's injury. The third and fourth cases, involving the Lippes Loop, show plaintiffs seeking recovery from the manufacturer under different theories of liability. In Marder v. GD. Searle & Co. (630 F. Supp. 1087 [D. Md. 19861) aff'd sub nom. Wheelahan v. GD. Searle and Co., 814 F.2d 655 [4th Cir. 19871), lawsuits by 17 plaintiffs were consolidated into a single case. The plaintiffs alleged that they had suffered three kinds of injuries from wearing the manufacturer-defendant's Copper-7: pelvic inflammatory disease (PID), ectopic pregnancy, and perforation
PRODUCTS LIABILITY AND CO=RACEPT~E DEVELOPMENT 129 of the uterus. The court held a trial on the general issue of causation; that is, the jury was presented with the question of whether the Copper-7 could have caused injuries of the types alleged, rather than whether it caused particular injuries in identified plaintiffs. When the jury could not reach a verdict, the court declared a mistrial. The court then resolved the issue in favor of the defendant on the ground that the plaintiffs had shown no more than a mere possibility that their injuries were caused by the IUD. In Romero v. G.D. Searle & Co. (15 Prod. Safety & Liab. Rep. [BNA] 669 ED. N.M. 19871), after the plaintiff had become pregnant she was fitted with a Copper-7 manufactured by the defendant. Her baby was born prematurely while the IUD was in place, and the baby suffered brain damage. The plaintiff alleged that the Copper-7 was the cause of both the premature birth and the brain damage; the jury returned a verdict for the defendant, apparently because the plaintiff failed to carry her burden of proof on the issue of proximate causation. Beyette v. Ortho Pharmaceutical Corp. (823 F.2d 990 [6th Cir. 19871) concerned the insertion of a third consecutive Lippes Loop in the plaintiff in 1975. At that time, the package insert stated that the risk of PID with a Lippes Loop was essentially the same as without it. In 1977, in response to an FDA directive, Ortho altered the package insert to warn of an increased risk of PID, and it brought this change to the attention of the plaintiff's physician. The physician, however, did not advise the plaintiff of the change. In 1978, the plaintiff's physician diagnosed her as having severe cervicitis but did not remove the loop. He saw her four more times between July 1978 and January 1979 but did not advise her of the increased risk of PID or remove the loop. In June 1979, the plaintiff was admitted to the hospital with acute PID, and a complete hysterectomy was performed. She sued Ortho on negligence (failure to warn) and express warranty grounds and won a verdict against Ortho at the trial level. The court of appeals reversed. It held that, since Michigan followed the learned intermediary rule, Ortho's duty to warn extended only to the physician, and the company had discharged that duty. Its failure to give an adequate warning prior to the 1975 insertion was not the proximate cause of the plaintiff's injury because of her physician's knowledge and failure to act. Ortho's statement in the package insert about the risks of PID did not constitute a warranty because it was cast in terms of an estimate, and a statement of opinion or estimate is not a basis for a warranty. In Collins v. Ortho Pharmaceutical Corp. (231 Cal. Rptr. 396 [19861), the plaintiff had been fitted with a Lippes Loop manufactured by the defendant, had suffered pelvic disease, and had undergone a hysterectomy in 1980. The basis for her suit was that the defendant was strictly liable for injuries because the IUD was defectively designed. The California trial court granted summary judgment for the defendant, and the court of appeals affirmed. The court held that, whereas the FDA has approved a prescription drug or device but allows it to be distributed with appropriate warnings of foreseeable risk, such a drug or device falls within
~ 30 DEVELOPING NEW CO=RACE~IVES comment k of section 402A and thus the manufacturer cannot be held liable under a theory of strict liability for defective design.4 It should be noted that this decision is not representative of the law in other jurisdictions; in most states, FDA approval does not bar a claim based on design defect, but does serve as evidence in support of a manufacturer's defense. On January 31, 1986, G.D. Searle & Co., the largest maker of IUDs, announced that it was ceasing to manufacture the Copper-7 and the Tatum-T. The company's stated reasons for its decision were the "high cost of defending unwarranted litigation" and its inability to obtain liability insurance at a reasonable cost (Goldberg, 1986:52,54~. At the date of the company's decision, 800 suits were filed against Searle, of which about 500 were settled (Reporter on Human Reproduction and the Law, 1986~. As of September 1988, Searle had won 14 of 18 cases that had gone to trial. However, in September 1988 a federal jury in Minnesota awarded a St. Paul woman $8.75 million in damages from Searle on the ground that the company was negligent in testing and marketing its Copper-7 RJD (Kocieml~a v. GD. Searle & Co., No. 3-85-1599 ED. Minn. Sept. 13, 1988], 1988 U.S. Dist. LEXIS 10580~. This was the first case of punitive damages ($7 million) awarded against a manufacturer of an FDA-approved IUD. It is too early to predict the impact of this case, which is being appealed, on future liability claims against Searle. Oral Contraceptives To date, all the oral contraceptive cases that have been reported involve the warning given by the manufacturer concerning side effects of the oral contraceptive. The issues that are litigated fall into three categories: to whom the warning must be given, the adequacy of the warning, and whether an inadequate warning caused the injury. Nearly all the cases adopt the reamed intermediary rule: since oral contraceptives are prescription drugs, the manufacturer has discharged its duty to warn the consumer if it informs the medical profession, including both prescribing and treating physicians, of their potential risks and contraindications (see MacDonald v. Ortho Pharmaceutical Corp., 394 Mass. 131, 475 N.E.2d 65 [O'Connor, J. dissenting], cert. denied 106 S.Ct. 250 [19853~. This information is normally communicated through package inserts and in the Physicians' Desk Reference, a physician's guide to prescription and nonprescription pharmaceuticals. Courts in 4Compare Ortho Pharmaceutical Corp. v. [leash (722 P.2d 410 [solo. 1986]), an oral contraceptive case in which the court held that the trial court properly submitted the case to the jury on a design defect theory of strict liability, despite FDA approval of the drug. The court reversed the case and sent it back to the trial court, however, because the facts entitled the defendant to a comment k instruction on unavoidable risk of harm, and the trial court had not given such an instruction. The case was ultimately settled for $800,000 (Bureau of National Affairs, 1987).
PRODUCTS LIABILITY AND CONTRACEPTIVE DEVELOPMENT 1 3 1 three states, however, have surprised manufacturers by holding that the warning must be given to the user of the contraceptive (MacDonald v. Ortho Pharmaceutical Corp., 394 Mass. 131, 475 N.E.2d 65 [19851~; Odgers v. Ortho Pharmaceutical Corp., 609 F. Supp. 867 [E.D. Mich. 19851; Stephens v. G.D. Searle & Co., 602 F. Supp. 379 [E.D. Mich. 19851; Lukaszewicz v. Ortho Pharmaceutical Corp., 510 F. Supp. 961 [E.D. Wisc. 19811~. Courts in Massachusetts and Michigan reasoned that oral contraceptives share many of the characteristics of over-the-counter drugs, and that the users do not rely on the advice of physicians in the same way they do with other prescription drugs. The court in Wisconsin reached the same result by a different route: a federal district court held that, under Wisconsin law, violation of a regulation designed to protect the class of persons of which the plaintiff is a member is negligence per se. The court then held that, since the FDA requires warnings to be furnished to the users of oral contraceptives in the form of package inserts (21 C.F.R. §310.501) and this is a regulation designed to protect the user, the manufacturer had a duty under Wisconsin law to warn the user herself (Lukaszewicz v. Ortho Pharmaceutical Corp., 510 F. Supp. 961 LE.D. Wisc. 1981]). The scope of the manufacturer's duty has been stated as a duty to warn the medical profession of untoward effects that the manufacturer [mows, or has reason to know, are inherent in the use of a drug. This duty requires the manufacturer to keep abreast of developments that could possibly require additional or modified warnings (McEwen v. Ortho Pharmaceutical Corp., 270 Ore. 375, 528 P.2d 522 [19741~. Most courts that have considered the question of the adequacy of a warning have held that compliance with FDA regulations does not, of itself, necessarily constitute an adequate warning; FDA compliance is only to be considered along with all other evidence of whether, in the circumstances, the warning was adequate (McEwen v. Ortho Pharmaceutical Corp.; MacDonald v. Ortho Pharmaceutical Corps. The question of the adequacy of the warning is generally decided by the jury, provided the plaintiff has introduced some substantial evidence of inadequacy. However, many judges have held warnings adequate as a matter of law, thereby preventing juries from considering the issues and resulting in judgments for the defendants.5 In other cases, there has been sufficient evidence to go ~ the jury, sly the case of Eiser v. Feldman (SO7 N.Y.S.2d 386 [App. Div. 1986]), the plaintiff took Ortho- Novum pills and suffered visual impairment; in Cobb v. Syntex Laboratories, Inc. (444 So.2d 203 [La. App. 1984]), the plaintiff took Nonnyl 1+50 and suffered a stroke; in Reeder v. Hammond (125 Mich. App. 223, 336 N.Vi.2d 3 [1983]), the plaintiff took Ovral while she was pregnant and gave birth to a retarded child; in Spinden v. Johnson & Johnson (177 N.J. Super. 605, 427 A.2d 597 [1981]), the plaintiff took Ortho-Novum and suffered thrombophlebitis and pulmonary embolism; in Goodson v. Searle Laboratories (471 F.Supp. 546 [D. Conn. 1978]), the plaintiff took Demulen 21 and suffered a stroke; in Dunkin v. Syntex laboratories, Inc. (443 F.Supp. 121 [W.D. Tenn. 1977]), the plaintiff took Norinyl 1+80 and suffered a stroke; and in Cambers v. G.D. Searle & Co. (441 F. Supp. 377 [D. Md. 1975]), aff'd, 567 F.2d 269 [4th Cir. 1977]), the plaintiff took Enovid E and suffered a stroke.
132 DEVELOPING NEW CONTRACEPTIVES but the jury has found the warning adequate.6 Nonetheless, a number of cases have found warnings to be inadequate and have held the manufacturer liable: fine plaintiff took Nonnyl and Ortho-Novum pills in 1966 and 1967. In 1968 she suffered hemorrhages in both eyes and became blind in her right eye. ~ 1966 and 1967 the manufacturer knew or should have known of British studies and U.S. animal studies indicating that injuries such as those suffered by the plaintiff were occurring (McEwen v. Ortho Pharmaceutical Corp.~. A plaintiff began taking Ortho-Novum 2 in 1967 and suffered a stroke in 1971. In 1970 and 1971, the manufacturer knew or should have known of the increased risks presented by oral contraceptives with higher estrogen levels and should have informed physicians of this higher risk (Brochu v. Ortho Pharmaceutical Corp., 642 F.2d 652 [1st Cir. 19811~. A plaintiff began taking Norlestrin in December 1975 and died of a stroke 19 days later. The warning was found inadequate because it did not (1) advise of the availability of lower estrogen formula pills that were associated with reduced risk of clotting disorders, (2) advise physicians to inquire about any family history of strokes, (3) advise that a test be administered to detect hypercoagulability before the drug was administered, and (4) advise that women with type A blood are more likely than others to suffer clotting disorders (May v. Parke Davis and Co., 142 Mich. App. 404, 370 N.W.2d 371 [198519. A plaintiff began taking Ortho-Novum pills in 1973 and suffered a stroke in 1976. The manufacturer was held liable because it did not use the word stroke in its warnings and in other information directed to the user; the manufacturer did warn of risks of blood clotting (MacDonald v. Ortho Pharmaceutical Corp.~. -A plaintiff took Ortho-Novum 1/80 from 1972 until 1976. In 1976 she experienced extremely high blood pressure, hemolytic uremic syndrome, and total kidney failure. Both of her kidneys were removed. The manufacturer did not adequately warn physicians that use of Ortho-Novum 1/80 might cause hemolytic uremic syndrome, malignant hypertension, or acute kidney failure (Wooderson v. Ortho Pharmaceutical Corp., 235 Kan. 387, 681 P.2d 1038, cert. denied, 469 U.S. 965 [19841~. This is one of the very few oral contraceptive cases in which punitive damages were awarded and the award was upheld on appeal. A plaintiff died in 1979 from a massive pulmonary embolism after Eking Ovral-21. The defendant's warning was found inadequate because it did not stress the risk of thromboembolic disorders resulting from prolonged immobilization or suggest tests for blood coagulation factors, especially in women with type A blood (Taylor v. Wyeth Laboratories, Inc., 139 Mich. App. 389, 362 N.W.2d 293 t19841~. 6In the case of Jordan v. Ortho Pharmaceutical Corp. (696 S.W.2d 228 [Text App. 1985]), the plaintiff took Ortho-Novum and developed liver tumors; and in Lawson v. G.D. Searle & Co. (64 B1.2d 543, 356 N.E.2d 779 [1976]), the plaintiff took Enovid and died from multiple pulmonary emboli.
PRODUCTS [lABl~Y AND CONTRACEPTIVE DEVELOPMENT ~ 33 Even if the warning is found to be inadequate, the plaintiff must show that the inadequacy was the proximate cause of her injury. Put another way, if the defendant can show that the plaintiff's injury would have occurred anyway, even if the warning had read the way the plaintiff contends it should have read, then the defendant is not liable. Several cases have absolved manufacturers of liability on this ground.7 Other Contraceptives Only three products liability actions have been reported involving manufacturers of contraceptives other than oral contraceptives or IUDs. The central issue in two cases concerned whether the injury was caused by the product. The third case concerned the determination of recoverable darnages. In Baroldy v. Ortho Pharmaceutical Corp. (157 Ariz. 574, 760 P.2d 574 t19881), a diaphragm manufacturer appealed a $1.5 million award granted to a plaintiff who alleged injury due to toxic shock syndrome (TSS) on a theory of failure to warn. The plaintiff had worn a diaphragm for "extended periods of time" over the course of three days until she developed TSS symptoms. At the time of her use, the manufacturer's patient information booklet stated that it was safe to leave the diaphragm in place for 24 hours and that failure to remove it was not cause for concern. The experts for both parties presented conflicting medical testimony regarding causation, which was resolved in favor of the plaintiff by the jury. The manufacturers argued on appeal that the plaintiff' s evidence of causation at the trial was not based on a generally accepted scientific principle. The appeals court disagreed and decided that there was sufficient evidence presented by both parties in the form of expert testimony and publications for the issue to have been resolved by the jury. The court also upheld the admission, for limited purposes, of subsequent revisions to the manufacturer' s patient information booklet; a letter to physicians; and articles, reports, and medical records published subsequent to the plaintiffs injury concerning Me possibility of a connection between TSS and extended use of the diaphragm. Only one products liability case against ~ spermicide manufacturer has been reported by the sources surveyed, Wells v. Ortho Pharmaceutical Corp. (615 F. Supp 262 [N.D. Ga. 1985], aff'd, 788 F.2d 741 [11th Cir. 1986], cert. denied, 93 L.Ed.2d 386 [19861), and liability was imposed on the manufacturer. In Wells, a woman used Ortho-Gynol contraceptive jelly (the active ingredient of which is 7In Seley v. G.D. Searle & Co. (67 Ohio St.2d 192, 423 N.E.2d 831 [1981]), the plaintiff did not inform her prescribing physician that she had suffered from toxemia during her first pregnancy; in Lawson v. G.D. SearIe & Co. (64 I11.2d 543, 356 N.E.2d 779 [ 1976]), the plaintiff was predisposed to blood clots because she was overweight and because of her parity (she had had five children); and in Vaughn v. G.D. Searle & Co. (272 Ore. 367, 536 P.2d 1247 [1975], cert. denied, 423 U.S. 1054 [ 1976]), the plaintiff did not inform her treating physicians of premonitory symptoms of a stroke.
]34 DEVELOPING NEW CO=~CE~~ES octoxynol-9) before conception and for four months thereafter. Her child was born with no left arm, a partial left shoulder, and a deformed right hand. The mother sued the spermicide's manufacturer, the case was tried before a judge without a jury, and damages in the amount of $5,151,030 were awarded. After presentations of scientific evidence on the issue of causation by each party's expert witnesses, the court found that the defendant's expert witnesses were either biased (several were employed by the manufacturer and others had previously expressed views on the subject), unqualified (had no advanced degrees or specialized training in the relevant disciplines), or did not appear to be credible on cross-examination. None of the defendant's experts had examined the child, and the court seemed to be unfavorably impressed by the omission. Under Georgia law, the plaintiff had only to show that the defendant's product caused her injury by a reasonable degree of medical certainty, and the court found that the plaintiff had met this burden of proof. Having established causation, the court had no difficulty in finding that the defendant had negligently breached its duty to warn users of the possibility of birth defects. The result in this case, which surprised many members of the scientific community, runs contrary to generally accepted scientific opinion on the issue of causation of birth defects by spermicides. Specifically, two independent advisory review panels commissioned by the FDA had evaluated the safety of octoxynol-9. The FDA's Advisory Review Panel on OTC Contraceptives and Other Vaginal Drug Products in 1978 found octoxynol-9 to be safe (U.S. Department of Health and Human Services, 1980~; the FDA's Advisory Committee on Fertility and Maternal Health Drugs in 1983 found no relationship between the sperrnicide and teratogenic effects and thus found no reason to change the spermicide's labeling (FDA, 1983~. In addition, one coauthor of a scientific study relied on by the plaintiffs in proving causation Dick et al., 1981) reassessed the original study and concluded that the link between a mother's exposure to a spermicide prior to conception and certain birth defects was unsupported by the evidence (Watkins, 1986~. Another coauthor questioned the wisdom of publishing the original study in the "present litigious environment" and expressed concern that the article was used as " 'proof' of a causal relationship" without regard to the qualifications and reservations contained in the report (Holmes, 1986:3095; 1987~. (See Bracken, 1987, 1985, for other studies pointing to a lack of association between birth defects and spermicides.) It seems clear in this particular lawsuit that the presentation of the evidence by the injured user's expert witnesses was simply more persuasive to the judge than the presentation of evidence by the manufacturer's expert witnesses. In J.P.M. and B.M. v. Schmid Laboratories, Inc. (178 NJ. Super. 122, 428 A.2d 515 [19811), a husband and wife brought a products liability action against a condom manufacturer on the theories of strict liability, negligence, and warranty. In this case the plaintiffs alleged that the condom was defective and resulted in the birth of healthy twin daughters. The couple sought compensatory damages resulting from the birth of their daughters, including the costs of rearing and
PRODUCTS [IABl~Y~D CO=~CEPT~E DEVE~PME ~135 educating the children. The appeals court specifically disallowed the specific damage claim for the costs of raising and educating the children. The case was subsequently dismissed and, according to a representative of the manufacturer, no payment was made to the plaintiffs (Kaiser, 1989~. PRODUCTS LIABILITY INSURANCE AND CONTRACEPTIVE DEVELOPMENT Given the liability and risk inherent in manufacturing and selling all products, contraceptive manufacturers, like manufacturers of any product, seek to protect themselves financially against products liability claims through some form of insurance. Some insurance industry experts consider 1984 and 1985 the worst period in the history of the U.S. insurance industry; in many respects it could be called the worst period for contraceptive products liability insurance in the United States as well. That period saw some contraceptive manufacturers' liability insurance Mtes more than triple; for certain contraceptive products, liability insurance was totally unavailable; courts approved multimillion dollar awards against a manufacturer of a spermicide and against manufacturers of oral contraceptives and IUDs; at least two pharmaceutical companies withdrew FDA- approved contraceptives from the market; and at least one large pharmaceutical company, which had traditionally been involved in contraceptive development, discontinued research and development of new methods. Causes of the Insurance Crisis The reasons for the recent problems of increasing price and lack of availability of comprehensive general liability insuranc~which covers products and a range of other liabilities of corporations, municipalities, and nonprofit organizations are multiple and complex (the reader is referred to the glossary for explanations of some of the technical terms in this section). At least three factors appear to have contributed. First and most fundamental is the underlying upward trend in both the frequency of claims (including products liability and medical malpractice) and the size of awards (Hensler et al., 1987; U.S. Department of Justice, 1986~. Second, adjustment to these trends has been exacerbated by interest rates and by the insurance underwriting cycle, in which "soft" insurance markets marked by low premium rates and widespread availability of insurance tend to be followed by "hard" insurance markets in which there are abrupt and dramatic increases in insurance premium rates. The availability of specific Apes of insurance coverage becomes much more limited in these circumstances. Third, increased unpredictability of future insurance costs results from the unpredictability of tort law in "long tailed" lines of insurance in which liability extends for many years. These factors apply to the liability of developers and manufacturers of contraceptives as well as to medical providers.
36 DEVE~Pl~G NEW CO=^CE~IVES Some have contended that the insurance crisis was fabricated by the insurance industry. In 1988 the attorneys general's offices in 19 states filed antitrust lawsuits against major insurance companies, charging that they conspired to limit the availability of commercial general liability insurance (Reske, 1989~. Others, pointing to the competitive structure of the insurance industry, have argued that the crisis was caused by the unpredictability of the tort law system and other factors (e.g., Clarke et al., 1988; Priest, 1987; Winter, 1988~. It will be years before this issue is resolved; the committee does not take a position in this controversy. One often-cited factor in the rising cost of liability insurance over the past decade has been the rising size of awards and settlements paid to plaintiffs and litigation expenses (Danzon, 1988~. One measure of trends in total costs of liability insurance is incurred losses. Losses incurred on general liability policies rose steadily from about $3 billion in 1975 to $8 billion in 1983 and increased dramatically to $14 billion in 1985 (U.S. Department of Justice, 19863. Harrington (1988) has estimated that the rate of growth of insurance losses incurred between 1984 and 1986 was over 40 percent per year. The cyclical patterns in insurance markets in the early 1980s have been more extreme than in previous cycles, particularly the swing to a hard market for commercial general liability insurance in 1984-1985. A major distinguishing feature during this period has been increased uncertainty in predicting future liability costs. Two factors are largely responsible: (1) unanticipated trends in tort law, especially the unpredictable size of awards, including awards for punitive damages and (2) judicial interpretations of insurance contracts (albeit in cases other than products liability cases) that have threatened the ability of insurers to contractually limit their exposure. (For environmental liability, see, for example, Jackson Township Municipal Utilities Authority v. Hartford Accident and Indemnity Co., 186 N.J. Super. 156, 451 A.2d 990 [19821; for directors' and officers' liability, see, for example, Federal Insurance Co. v. Oak Industries, Sec. Law Rep. [CCH] pare. 92,519 [S.D. Cal. 19861.) By the nature of the insurance product, the price is set before the costs are known. The ability to form reliable predictions about future costs is therefore critical to the willingness of insurers to write a particular coverage and to the price at which they will write it. The liability system is unpredictable because different courts and juries arrive at different findings with respect to the facts and the appropriate amount of compensatory and punitive damages in similar cases relating to the same product. Unpredictability is particularly severe when liability extends many years after the product is manufactured, introduced into the market, and used. Unpredictability tends to create a risk that cannot be readily diversified through standard insurance mechanisms. Insurers require higher prices for bearing undiversifiable risks. This climate of unpredictability applies to comprehensive general liability
PRODUCTS [lability ID CO=^CEPT~E DEVE~PME ~ 37 insurance, which includes products liability. In addition, certain Apes of insurance have been particularly adversely affected. Contraceptive liability tends to be a high-risk category because contraceptives are used by so many women over long periods of time, can create risks that may be latent for years after the product has been discontinued, and can cause severe injuries-birth defects and loss of fertility that tend to result in very large awards. Cost and Availability of Liability Insurance for Contraceptives Contraceptive developers and manufacturers found it difficult to obtain liability insurance coverage during the mid-1980s. This problem was an acute manifestation of the industry-wide price increases and lack of availability of liability insurance. For some pharmaceutical companies and private nonprofit research organizations, the cost of liability insurance coverage more than doubled in a period of only one or two years. For example, liability insurance costs for Family Health International (1985 to 1987) and the Population Council (1982 to 1984) more than doubled during a two-year period (~llaway, 1988; Lynch, 1988~. From 1984 to 1988, Stone Research and Development Corporation gradually decreased its development work on new IUDs, while it continued to observe the IUD liability situation in other companies (Lewis, 1988~. According to an FH! representative, the future availability of products liability insurance coverage at PHI is uncertain (Lynch, 1988~. For some manufacturers of IUDs, liability insurance was unavailable at any price during the mid-l98Os (G.D. Searle & Co., press release, January 31, 1986~. Sharply increased costs of liability and liability insurance can have substantial disruptive effects on the supply of established contraceptives and on the development of new ones. The effects of unanticipated liability costs for products that have already been marketed can be severe. New liability costs associated with products previously marketed cannot be passed on to current consumers of current products. These costs must be paid out of current profits, and thus they erode the flow of funds available to fund research and development. The effects of dramatic increases in insurance costs for new products can also be severe. Such costs may ultunately be passed on to consumers through higher product prices, but this pass-through is not immediate, particularly when prices for contraceptives are set by public agencies. Nonprofit research organizations involved in contraceptive development are particularly adversely affected; because they do not market products directly to the public, they do not have the power to raise prices, nor do they receive immediate increases in funding to accommodate substantially higher insurance premiums In general, if a manufacturer cannot charge a price for a product that is sufficient to cover all costs, including expected liability costs, the product will be withdrawn from the market. In He contraceptive area, G.D. Searle & Co.
|38 DEVELOPING NEW CO~RACE~IVES withdrew the Copper-7 IUD from the market in 1986 because the "escalating cost of defending unwarranted product litigation makes continuing in the IUD business in the U.S. no longer economically feasible" (G.D. Searle & Co., press release, January 31, 1986~. The company stated that the costs of successfully defending four jury trials exceeded $1.5 million, while the total annual sales for the Copper- 7 amounted to only $1 1 million. Almost 500 cases were still pending when Searle withdrew the product from the market, and several hundred more cases have been filed against Searle since the withdrawal of this product. In the unprecedented hard market situation of the mid-1980s, most insurance companies that previously provided liability insurance for special classes of products such as contraceptives were no longer willing to do so. As the market again softens, it is reasonable to expect that insurance coverage for contraceptives will become more readily available, conditioned on a thorough risk assessment of the product by the insurance company's engineers and scientists, and possibly with limits on the dollar amount of coverage and on other terms of the contract. To some extent, what is reported as a problem of availability of liability insurance for contraceptives may at bottom be a problem of affordability. In theory, there should be some price at which the insurance market would be willing to write coverage for any risk. However, in some cases this price may be unacceptably high to contraceptive developers and manufacturers, given their budgets and the prices that they can charge for their products. Responses to the Crises In response to the liability insurance crisis, a number of adjustments were made in the insurance market. The problems of cost and availability have been resolved in part by a contraction in the market for risk spreading: policy holders are retaining a larger share of the risk through higher deductibles and upper limits of coverage. There is increased use of innovative insurance mechanisms, including self-insurance, captive insurance companies, and risk retention groups, which provide alternatives to traditional insurance and offer advantages to certain types of policy holders. The 1986 amendments to the Product Liability Risk Retention Act have expanded the range of options available and substantially reduced the regulatory costs of using these alternatives. The contraction in the market for risk spreading seems surprising at first, since the increase in defendants' potential exposure to liability might normally be expected to increase their demand to spread risk through the purchase of insurance. The "solutions" that are being adopted in insurance markets are less than ideal to the extent that they leave the policy holder-or the claimant- bearing more real risk. They do nothing to alleviate the underlying problem of high and unpredictable liability exposure. However, greater risk retention and use of self-insurance may be a preferred second-best solution for some policy holders, given the costs of liability insurance in the face of the unpredictability of future liability costs.
PRODUCTS [lABl~Y ID CO=~CEPT~E DEVE~PME ~139 Responses by Insurers In some situations, insurers have reduced their risk by shifting from the occurrence policy form to the claims-made form. An occurrence liability policy covers claims, no matter when they are made, as long as the injury occurred during the period in which the policy was in effect. A claims-made liability insurance policy covers claims filed within the policy period, for injuries that occurred during the policy period or within the retroactive period defined by the policy. The price of future coverage for claims that may be filed after the policy period but arising from injuries that occurred within the policy period is not guaranteed, but the insurer usually at least guarantees that "tail" coverage will be available. Thus, a claims-made policy shifts the uncertainty of the future liability costs from the insurer to the insured (Danzon, 1985~. The claims-made approach was widely adopted for medical malpractice following the difficulties associated with medical malpractice insurance in the mid-1970s, but it has so far not been widely adopted for other commercial classes. Other responses by the insurance companies have been to increase the deductible, increase the premiums, or no longer offer liability coverage for what they consider risky products. Responses by Policy Holders The simplest response to high price or lack of availability of liability insurance- short of ceasing to manufacture the product is self-insurance. The manufacturer sets aside a reserve fund for contingent liabilities; if liability costs turn out to be larger than anticipated, so that the fund is inadequate, any shortfall must be paid out of the manufacturer's other funds or future profits. A manufacturer who self- insures must therefore anticipate liabilities with some degree of accuracy in order to set the price of the product at a level sufficient to finance an adequate reserve fund. To achieve a less extreme form of risk retention, the manufacturer can self- insure for the lower levels of risk and buy coverage only for costs above some very high threshold, provided such "excess" coverage is available. Representatives of each of the contraceptive manufacturers and developers who spoke to the committee stated that their organizations were unable to obtain, or found it very difficult to obtain, liability insurance and consequently were resorting to some forth of self-insurance. Some large pharmaceutical companies are now self-insured for their contraceptive products, for example, G.D. Searle & Co.; Syntex Laboratories, Inc.; Parke-Davis & Co.; and Ortho Pharmaceutical Corporation. Another form of self-insurance is for a business firm or group of firms to create their own "captive" insurance company. They enjoy regulatory advantages not available to domestic insurers and tax advantages not available to self-insurers. The Planned Parenthood Federation of America and ALZA Corporation (manufacturer of the Progestasert IUD) use captive companies.
140 DEVELOPING NEW CO=~CE~~ES A third alternative is the formation of a risk retention group. The federal Product Liability Risk Retention Act of 1981 permits business firms and nonprofit entities that are engaged in similar activities or share similar risks to form a risk retention group solely for the purpose of insuring members of the group. The group may provide product liability and other types of liability insurance. The 1986 amendments extended the scope of the act to cover all commercial liability risks except worker's compensation, and it eased the regulatory requirements on such groups. Essentially, the act preempts state insurance regulation except in the state in which the group is chartered, thereby eliminating the need to be licensed in every state in which a member of the group is located. Several states have enacted special enabling statutes to facilitate the formation of these groups. Captives, risk retention groups, and other forms of self-insurance share the common feature that more risk is retained by the insured than in traditional commercial insurance. Although there may be some savings to the insured, if these quasi-insurance mechanisms are to be viable in the long run, they must set aside reserves to cover the cost of claims and litigation expenses. Thus, such solutions do not reduce the real risk or cost of tort liability that is faced by contraceptive developers and manufacturers. Responses by State Legislatures In some states, legislatures have acted to require increased data reporting by insurance companies and to regulate insurance rates (Danzon, 1988~. Such measures do not address the fundamental causes of the problems of cost and availability of liability insurance and could in fact prove to be cumbersome and counterproductive in such a diverse area of insurance as commercial general liability. Even if rate regulation were feasible, evidence in other areas indicates that it tends to exacerbate the problem of lack of availablility of insurance. For example, when malpractice insurance rates were regulated in many states in response to the malpractice insurance crisis of the mid-1970s, at rates below levels considered adequate by insurers, a massive withdrawal of commercial insurers from that market took place (Danzon, 1985~. In states in which malpractice rates remain heavily regulated, coverage is still available only through mutual insurance companies and joint underwriting associations, whereas coverage is readily available from commercial carriers in unregulated states. Rate regulation is not a solution that appropriately addresses the underlying problems of unpredictability. So far, products liability insurance, including liability insurance for contraceptives, has generally not been subject to rate regulation by state insurance departments. Thus, direct regulation of rates cannot explain the lack of availability of liability insurance for some contraceptives in the mid-1980s.
PRODUCTS [IABlL~Y AND CONTRACEPTIVE DEVELOPMENT i4 1 CONCLUSIONS AND RECOMMENDATIONS From the evidence available to the committee, we conclude that recent products liability litigation and the impact of that litigation on the cost and availability of liability insurance have contributed significantly to the climate of disincentives for the development of contraceptive products. Two aspects of the litigation are especially significant in the context of contraceptives. The first is the unpredictable nature of litigation, which results in part from the absence of stable and uniform national products liability rules and in part from the often erratic character of the litigation system. The second is that, although manufacturers may introduce evidence of compliance with FDA regulations in a contraceptive products liability lawsuit, this evidence is given no special status in most states, such as entitling the manufacturer to a presumption that it acted with due care. Because of the length of time necessary for development of a new contraceptive product and the costs of development, manufacturers, in considering whether to remain in the contraceptive field, are likely to give the prospects of extensive litigation special importance. Without changes in the products liability rules and procedures, it appears likely that even fewer firms will allocate even fewer resources to contraceptive research and development. The committee makes no recommendations for changes in liability insurance mechanisms. However, concern over the cost and availability of liability insurance is one of the reasons for the recommendations with respect to products liability for contraceptives. The impact of products liability litigation on contraceptive research and development is a matter of great concern. As noted in Chapter 2, continued contraceptive research and development by U.S. firms is important to the health and welfare of people in the United States and in other, especially less developed, countries. The committee believes that the products liability rules can be changed to remove most of their undue negative consequences for contraceptive development without increasing the health risks of contraceptive users. The committee concludes that an aspect of a contraceptive drug or device that complies with the requirements of federal food and drug law should not be determined to be a defect or a breach of warranty under state law; that the manufacturer of that contraceptive product should not be held negligent for complying with FDA- approved designs or warnings; and therefore that the manufacturer of a specific contraceptive drug or device should not be the source of compensation to someone injured by that aspect of the particular contraceptive drug or device. Three qualifications are in order. First, the committee has not made a comprehensive study of the products liability system, but has limited its investigation to that system as it affects contraceptive products. It is possible that the committee's proposal could have applicability to all FDA-regulated drugs and medical devices, but our investigation did not go beyond contraceptive drugs and
142 DEVELOPING NEW CO=RACE~IVES devices. Second, the committee's proposal addresses only the safety, or deterrent, function of products liability law; the important issue of providing adequate compensation to persons injured by defective products is part of the much broader question of the adequacy of existing private and social insurance mechanisms. There are significant gaps and overlaps in the network of insurance programs, and it appears that if different systems of social and medical compensation or insurance existed, the liability rules would be implemented differently. It goes beyond the scope of this report, however, to recommend reforms in these programs or to suggest alternative programs. However, we note that, for purposes of providing compensation, the tort system is much more costly and less equitable than alternate private and social insurance mechanisms. Third, we assume that the FDA will continue to apply high standards in its review of the safety and effectiveness of contraceptives, and that changes made in its requirements for contraceptives (whether those recommended in this report or others) will be fully justified by an appropriate weighing of risks and benefits and are in accordance with statutory mandates. On the basis of these considerations, the committee therefore recommends that Congress enact a federal products liability statute that gives contraceptive manufacturers credit for approval of contraceptive drugs and devices (and their labeling) by the FDA. The first part of this recommendation is straightforward: the enactment of a federal products liability statute is intended to deal with the unpredictability and uncertainty caused by requiring manufacturers of nationally marketed contraceptive products to face the possibility of 50 different state liability rules. Of course, a single products liability statute for contraceptives will not completely solve the problem of a diversity of standards in trial courts, both state and federal, across the country. Such a statute, however, would reduce the problem, especially if Congress amended the statute when necessary to assert national uniformity on important points. The key contribution that a federal statute would make, however, is not so much the reduction of diversity at any given time, as the reduction in unpredictability over time. A system governed by Congress, the Supreme Court, and the 12 federal courts of appeals will produce fewer unpredictable doctrinal trends than one governed by 50 legislatures and state supreme courts. The second part of the recommendation is intended to address the fact that contraceptive products, as drugs or medical devices, are regulated by a national agency charged with the responsibility of weighing their risks and benefits and having the scientific expertise to execute this charge. Pharmaceuticals and medical devices are unique among products in the United States in the degree to which quality is regulated before they are released into the market. Given that a system of premarketing reviews exists, the necessity for liability as a quality control mechanism is greatly reduced. When the FDA has considered the relevant
PRODUCTS [lABluTY AND CONTRACEPTIVE DEVELOPMENT 143 health and safety data on a contraceptive product, has approved the product, and has required warnings and instructions to accompany the product, it is sound national policy to make this approval available to manufacturers as a defense and not to penalize them for something they could not have known at an earlier point. In the remainder of this chapter, we describe and explain the sort of statute the committee recommends and discuss the effect such a statute would have on several of the cases discussed earlier in this chapter. Because the statute would interact with postmarketing surveillance efforts, our recommendation would be more compelling if formal postmarketing surveillance studies were generally required. Possible Elements of a Statute The extent of the credit that the committee proposes be given for approval of a contraceptive drug or device by the FDA is that, as a general matter, there be no liability for design defect or inadequate warning if the FDA has reviewed and approved the contraceptive product or the warning and has addressed the characteristics of the product that caused the plaintiff's injury. The defense should not be available if the manufacturer withheld relevant information from the FDA in the approval process or if inflation developed after approval was not reviewed by the FDA for the purpose of determining whether the product or its labeling should be changed. The committee suggests that the proposed statute have five major provisions. First, if it is established that the injury-causing aspect was in compliance with all applicable requirements of the FDA at the time the contraceptive drug or device was made or sold, then a manufacturer or seller of a contraceptive drug or device would not be liable under any of the relevant legal theories (misrepresentation, negligence, warranty, or strict liability) for any injury related to design; nor would it be liable for a failure to provide an adequate warning or instruction regarding any danger associated with its use; nor would it be liable if the FDA had not asserted that the contraceptive drug or device was not in compliance. Second, a determination by the FDA that an aspect of a contraceptive drug or device complies with the requirements of federal law or with FDA requirements would be considered conclusive evidence of such compliance. Third, the defense would not be available if a claimant is able to establish that the manufacturer should have made design modifications or given different or additional warnings or instructions. Specifically, the defense would not be available if the manufacturer or seller knew or should have known of studies showing an increased risk of harm from the contraceptive drug or device and if consideration of these studies would have led to the conclusion that, without design modification or different warnings, there was an increased likelihood of
|= DEVELOPING NEW CO==CE"~ES serious injury occurring to the claimant or persons sharing the claimant's medical characteristics. Fourth, the defense would continue to be available to contraceptive manufacturers for action after initial FDA review and approval. That is, if a contraceptive manufacturer or seller complies with the FDA directions on the basis of information developed after initial FDA review and approval, then the contraceptive drug or device could not be found defective with respect to any aspects in compliance with the FDA. Fifth, the defense would not be available if a claimant establishes that the FDA was not informed of dangers regarding the contraceptive drug or device that were known to the manufacturer or seller but not to the FDA and that the claimant's injury is attributable to such dangers. The FDA Defense In operation, the FDA defense provided by He proposed statute would work as follows. If a lawsuit is brought alleging injury caused by a contraceptive, the manufacturer of the contraceptive product would have the opportunity to demonstrate that the alleged injury-causing aspect of the product had been reviewed by the FDA and that the agency had not asserted that, by reason of that aspect, the product was adulterated or misbranded or otherwise not fully in compliance with the laws and regulations administered by the FDA. The FDA review may have occurred in connection with an application for Remarketing approval (for example, an NDA or PMA), in connection with a §510 submission, or otherwise. If the FDA had not reviewed the alleged injury-causing aspect, then the defense would not be available. This limitation would ensure that state law remains applicable to aspects of contraceptives that have not been subjected to actual FDA review. If the manufacturer establishes the defense, and the plaintiff does nothing more, the manufacturer would have presented a complete defense, and the lawsuit could be decided before trial by pretrial motion. If, however, the plaintiff seeks to maintain that the manufacturer or seller should have made design modifications or given different warnings, then the burden is on Be plaintiff to show by a preponderance of the evidence that: (1) reports or studies showing an increased risk of harm were available to the manufacturer, (2) these reports or studies were scientifically valid or had received acceptability in the relevant scientific community, and (3) a review of these reports or studies should have led a reasonable manufacturer to conclude that design modifications or different or additional warnings or instructions were necessary to avoid He increased likelihood of serious injury to the plaintiff or persons sharing the plaintiff's medical characteristics. If He substance of the new reports or studies proffered by the
PRODUCTS [UBlL~Y AND CONTRACEPTIVE DEVELOPMENT 145 plaintiff were considered by the FDA and the manufacturer had complied with the FDA's directions as a result of that new information, then the manufacturer would be found to have acted reasonably and would prevail. Effect on Products Liability Cases What would be the effect of the proposed FDA defense on the specific products liability cases discussed in the chapter? For the most part, the answers must be speculative because of the limited information available: it is difficult to assess the quality of the plaintiff's evidence from an appellate opinion. Moreover, if the FDA defense had been in effect in each case, the plaintiff might have been able to overcome it with other evidence that was available but not adduced at trial. With this qualification in mind, we examine several cases in light of the FDA defense. In MacDonald v. Ortho Pharmaceutical Corp. (394 Mass.131,475 N.E.2d 65, cert. denied, 106 S.Ct. 250 [1985]), the court held that the warning regarding possible adverse effects of oral contraceptives must be given to the user and that the adequacy of the warning was a question for the jury, even though it was in compliance with FDA regulations. The plaintiff contended that the warning was inadequate because it did not contain the word stroke. This case would almost certainly have been decided in the manufacturer's favor under the FDA defense, because both parties agreed that the warning complied with FDA regulations. In Ortho Pharmaceutical Corp. v. Heath (722 P.2d 410 [solo. 1986]), the court held that the plaintiff was entitled to a jury instruction regarding liability for design defects even though the drug had been approved by the FDA. This part of the Heath case would be changed by the FDA defense; under the evidence presented, there could be no finding of a design defect. In McEwen v. Ortho Pharmaceutical Corp. (528 P.2d 522 tOre. 1974]), the court held that in 1966 and 1967 when the plaintiff took the defendant's drug, there existed British studies and U.S. animal studies showing that injuries such as the one the plaintiff suffered were caused by oral contraceptives. These studies should have caused the defendant, as a reasonable manufacturer, to change its warning. This case would probably have been decided the same way under the FDA defense; that is, it appears that the plaintiff's evidence would have been sufficient (under the provision that the manufacturer or seller should have made design modifications or given different warnings) to overcome the defense. In Wells v. Ortho Pharmaceutical Corp. (615 F.Supp 262 EN.D. Ga. 1985], aff'd, 788 F.2d 741 [11th Cir. 1986], cert. denied 93 L.Ed.2d 386 [1986]), the court found the defendant liable for failure to warn of possible teratogenic effects from its spermicide despite FDA approval of the warning and despite the fact that two FDA expert panels had reviewed the evidence of causation and had concluded
146 DEVELOPING NEW CO=RACE"IVES there was no basis for changing the warning. The proposed statute would change the result of this case; compliance with FDA directions would have been a complete defense. CONCLUSION The operation of the legal system in this country makes it difficult to make precise forecasts of the extent to which enactment of such a statute would change the perception of the risk of liability, and therefore what contraceptive developers and insurance underwriters will do. As a first step the proposed statute is important, but we recognize that it would take several years before its impact could be evaluated, and modifications may be needed in the future. We believe that it is important to preserve tort liability for contraceptive products with a few exceptions. The committee did not find that more extreme approaches, such as those proposed by others examining specialized drug development, are called for in the case of contraceptives. For example, an Institute of Medicine (1985) committee studied the problems of vaccine development in the United States; its report proposes a series of options for dealing with problems arising from vaccine-related injury, which includes the option that vaccine-related liability be taken out of the tort law system. We do not believe such action is appropriate with respect to contraceptives. That said, the committee recognizes that, if the proposed FDA defense is enacted and no changes took place in the pace of contraceptive development, other steps might be needed. The committee believes that the proposed statute constitutes a modest reform and is by no means a radical proposal. It is our belief that a change in the products liability law would change the climate of disincentives for the development of contraceptive products, without compromising the safety of contraceptive use.
