Liability Exposure When Offspring are Injured Because of Their Parents' Participation in Clinical Trials
Ellen Wright Clayton
The exclusion of fertile men and women, and more particularly pregnant women, from clinical trials is often justified on the basis that fetuses must be protected and that if fetuses are injured, the potential liability will be too great. Both of these arguments are fundamentally flawed. Looking first just from the child's perspective, the earlier the teratogenic or mutagenic effects of drugs are detected, the fewer children will be exposed. From the perspective of investigators and particularly of the manufacturers whose interventions are being tested, the risk of incurring liability during the early stages of drug investigation is actually quite small whereas the potential for substantial liability is much greater once a fetotoxic drug enters widespread use.
Does the desire of the federal government to include women in clinical trials preempt claims alleging the children were injured as a result? A threshold question is whether children and their parents can bring tort claims at all because permitting such claims works counter to the federal purpose. The argument here is that if there is a threat of liability, efforts to include women in clinical trials will fail because potential parents will not enroll in and third parties will not include potential parents in clinical trials. The majority's dictum on Johnson Controls that the goal of Title VII of ensuring equal access to the work place could preempt tort claims brought by children allegedly injured by their mothers' exposure to lead in the work place would appear to support this line of reasoning.1 Yet when directly confronted with defining the scope of preemption in 1992, the Supreme Court made clear that it will not find preemption unless
there is a direct conflict between state and federal law or unless Congress clearly intended to occupy the field completely.2 These conditions are only rarely met. In the area of drug research, for example, no appellate court has found that the extensive federal regulation of the information contained in package inserts and the Physicians' Desk Reference totally preempts state tort claims alleging inadequate warning.3 Further, there is no evidence that Congress intends to eliminate claims by later-born children in order to ensure that potential parents be included in clinical trials, so it is unlikely that courts will find that these tort claims are preempted.
Identifying potential claimants and their claims. Assuming that tort claims are not preempted, defining the types of claims that can be brought following the ingestion of a fetotoxic drug becomes important because the potential claims vary widely and lead to very different measures of damages. The two most important factors in this analysis are whether the child was born alive and what the parents would have done had they known about the risk. Claims can be brought on behalf of children only if they were born alive.4 When children are born alive and when their parents allege that they would not have taken the drug had they known about the risk of fetotoxic effects, children can assert claims for prenatal injury, saying that they could otherwise have been born healthy. In recent years, most courts have looked favorably on prenatal injury claims and on occasion have awarded very large amounts of damages, commensurate with the child's special needs over the course of a lifetime.5 By contrast, children are said to be seeking damages for "wrongful life" when they are born alive but their parents say that more information would have led them to avoid childbearing. These claims have almost universally been rejected by courts and legislatures.6 Interestingly, one court permitted two children who were born with fetal hydantoin syndrome to recover some damages for wrongful life when their mother alleged that she had expressed concern about taking Dilantin during pregnancy to her providers, who inappropriately reassured her that there was little risk.7
The nature of parents' claims differs depending on whether the child was born alive or not and on whether the parents would have avoided the drug or put off procreation. When the children were live-born and the parents say they would not have taken the drug, the parents may recover the medical expenses not covered by the child's claims. In most cases, however, the parents will not receive damages for their own emotional pain and suffering because they are seen as bystanders to their child's injury. Even in the jurisdictions that are most sympathetic to parents' claims for emotional injury, parents must allege that they were aware at the time of the child's injury that the defendant's actions were responsible, conditions that can probably never be met for injuries caused by fetotoxic drugs.8
If the parents have a live child but assert that they would have avoided childbearing had they known more, their claim is one for "wrongful birth." Courts have generally permitted at least some aspects of the parents' claims, although some legislatures have limited these causes of action.9 These claims, when permitted, can also support large damage awards.
When a child is stillborn as a result of negligence, some jurisdictions permit the beneficiaries of the fetus' estate—usually the parents—to bring an action for wrongful death,10 although some states require that the fetus be viable at the time the lethal injury occurred.11 Many other states, however, refuse to permit such claims at all.12 Even when these claims are permitted, however, the amount of damages is usually relatively small.
Children's claims against their parents. The next question is who can be sued. Whether children injured by drugs can sue the parent who took them turns, first of all, on the law of parent-child immunity in the state where the family resides. States generally have begun to limit this sort of immunity: some never had it or got rid of it altogether; others have declared that only some decisions by parents may give rise to liability. Yet a few states retain complete immunity.13 Three states have specifically addressed children's claims that they were injured by their mothers' behavior during pregnancy. Two states permitted the children's claims,14 including one case in which the child alleged that he was injured when his mother took tetracycline while she was pregnant. One state denied a child a cause of action,15 saying that allowing the claim would intrude too deeply into the lives of pregnant women.
In the states that have struck down parent-child tort immunity in whole or in part, the next issue is what sort of parental behavior will give rise to liability. At the least, parents must be negligent, that is, their choices must be ones that other reasonable people would not have made, before they can be held liable. The factors that should enter into that calculus include the general deference we give to patients in choosing their treatment,16 the seriousness of the parent's medical problem, and the availability of nonfetotoxic alternatives. Parents' choices would probably be found acceptable in most situations. It would probably be reasonable, for example, to choose chemotherapy with powerful mutagens for treatment of cancer. It might be less reasonable, by contrast, for a woman of childbearing years to use ACE inhibitors which are known teratogens to treat hypertension when there are other nonfetotoxic options available.
If the parent's decision to take the research drug was not negligent, the final argument that might be made by a child is that the parent was unreasonable or negligent in deciding to have the child. Such "wrongful life" claims by children are highly disfavored in general, as was discussed above, but they are particularly unlikely to succeed when made against parents.17
Impact of one parent's negligence on the other parent's claims. This issue has most commonly been litigated in the context of wrongful death, but the analysis is applicable as well to claims for wrongful birth or the ancillary claims to prenatal injury. In general, a potential beneficiary may not recover damages if his or her negligence caused the wrongful death, but the negligence of one beneficiary does not preclude recovery by other beneficiaries.18 One court recently held that the father of a child stillborn because of the mother's negligence could not recover damages because the mother might indirectly benefit.19 This imputation of the negligence of one parent to the other, however, has been criticized as a ''senseless survival of a discarded concept of marital unity."20 Thus, if one parent negligently takes a fetotoxic drug in a jurisdiction that permits wrongful death actions when children are stillborn, the other parent may well be able to recover damages.
Claims against third parties. Although children may be able to sue their parents for taking part in a clinical trial, the public debate about the desirability of including fertile and pregnant individuals in clinical trials is much more heavily influenced by the fear that injured children will sue third parties, including the researcher/physician who provided the drug, the institution in which the research occurred, and the drug manufacturer. The basic rules of negligence and strict liability are discussed at length in this volume's paper by Flannery and Greenberg.21 Claims on behalf of children, however, present some special issues since they did not directly consent to the intervention. For purposes of this discussion, we can group the possible bases of liability into two groups—one for lack of warning of possible fetotoxic effects, based in the law of informed consent or duty to warn in strict liability; and the other for problems with the intervention itself or the way it was administered, based in the rules governing defective products or negligence.
Cases in which parents were not warned are in some ways easier. If the parent says that she was not told of the potential fetotoxic effects of the drug, then the child and parents may be able to pursue claims against researchers and the health care institutions in which they work for prenatal injury and economic/emotional injury or wrongful life and wrongful birth, depending on whether the parent would have avoided taking the drug or decided against childbearing. Claims against manufacturers, however, may be barred by the learned intermediary doctrine.
Courts have not confronted a case in which the parents say that they knew at the time they took the drug about its potential effects on their unborn child, even though this situation is likely to occur in research because of the stringent federal requirements for disclosure. In this setting, although the law is by no means clear, the distinction between the two types of claims is critical. If the parents were adequately informed, the child who was injured prenatally probably cannot be heard to complain that he or she would not have agreed to the
exposures. Fetuses or preconceptuses cannot be warned, and as Professor Merton demonstrated, parents can consent to some types of nontherapeutic research for children who are already born.22 Prospective parents should have even greater latitude to "choose" nontherapeutic research on their unborn children, particularly when the protocol offers potential benefit to the parent who is enrolled. There may, however, be settings in which third parties are not entitled to rely on even the fully informed consent of prospective parents to immunize them from later liability if the protocol poses very serious risks to the unborn child while offering little benefit to the subject adult or to adults in general.
A different problem arises when the intervention was provided negligently or was defective. In that setting, as was mentioned by Justice White in Johnson Controls,23 many cases hold that parents cannot anticipatorily release their children's potential claims.24 If even documents clearly designated as releases do not preclude children's claims, then a parent's signature on a document for informed consent to research involving a potentially fetotoxic drug would not bar the injured child's suit. One might argue that cases rejecting anticipatory releases, which often deal with potential liability for recreational activities, do not apply to the more complex decisions of a potential parent to take a drug that may help him or her at the risk of harm to future children. This argument may not prevail because the rule barring parental releases of children's claims is based in part on the desire to avoid conflicts of interest between parents and children, which is exactly the problem posed by parental use of potentially fetotoxic drugs. Finally, even if parents can release some claims, releases are strictly construed and therefore must be clearly written, especially where they purport to waive claims for future negligence.
Another potential barrier confronting children in this setting is the argument that the parent's consent acts as a superseding cause that shields the researcher or manufacturer from liability. The line of reasoning should be rejected on the grounds that the parent's agreement to use the drug is a clearly foreseeable result of the efforts to test it and that permitting this defense would vitiate the ban on parental releases.25
Does it matter whether the researcher works for the government? The states vary widely in whether they allow physicians who work for governmental institutions and the institutions themselves to be sued. Some states say that provision of health care is not protected because it is not inherently governmental or because the state does not exercise control over the provider,26 while others waive immunity to the extent of available insurance coverage.27 Some states, however, retain immunity for some providers and government institutions.28 Under the Federal Tort Claims Act, physicians who are working within the scope of their employment for the federal government may not be sued in their
individual capacities. The federal government can be sued for some of its employees' torts, including medical malpractice and lack of informed consent, but not for battery.29
May children sue the federal government if they were injured before birth by their parent's participation in research while enlisted in the armed services? In Feres v. United States,30 the Supreme Court held that an injured serviceman cannot bring a tort action against the government for claims incident to service. Subsequently, a number of cases held that claims on behalf of fetuses harmed by negligent prenatal care of women who were members of the military were barred by the Feres doctrine.31 These courts reasoned that medical care was incident to service and that if the service member cannot sue, the child cannot sue for any derivative injuries.32 Recent cases, however, have ruled to the contrary, challenging both of these lines of argument. First, some cases have held that the children injured by negligent prenatal care to a service member are suing for their own injuries and not asserting simply derivative claims.33 Second, a recent and potentially more far-reaching case has challenged the notion that all injuries resulting from medical care are "incident to service." A service woman was allowed to pursue a claim for injuries sustained as a result of her voluntary participation in a military-run blood donor drive on the grounds that her donation did not implicate sensitive military matters and was more civilian in nature.34 Thus, while children injured as a result of their parents' participation as members of armed services in clinical research are still unlikely to succeed in their claims against the federal government, these recent trends raise the possibility that their claims soon will be more successful.
The risk of liability for injuries incurred during research is low. Looked at simply from the perspectives of legal doctrine and of the extensive disclosure that typically occurs in the research setting, then, there will rarely be any basis for successful lawsuits by children or particularly by parents. In addition, as a practical matter, there are a very small number of cases alleging injuries incurred during research. Perhaps subjects do not sue because they feel that they somehow "assumed the risk," even though the law no longer accepts this notion as a defense to tort claims. Perhaps the small number of suits is part of the phenomenon that the overwhelming majority of people injured by medical malpractice do not pursue claims.35 But the result of all these theoretical and practical limitations is that the risk that researchers, institutions, or manufacturers will be held liable for large amounts of money for injuries children suffer as a result of their parents' participation in clinical trials is actually quite low.
The potential for liability is much greater if efforts are not made to detect fetotoxic effects. Here it is useful to compare the situation of a
manufacturer of a drug that was in widespread use while its teratogenic effects were unknown but knowable with that of a manufacturer of a known teratogen. In the former case, many children may be injured and their lawyers may argue that the manufacturer had a duty to find out about these effects. As Flannery and Greenberg have suggested, this argument is likely to meet with increasing success, particularly in light of growing pressure to include fertile people in trials; and where it is accepted, it could lay the basis for tremendous liability. Physicians, too, would face potential liability if they did not warn their patients about the lack of data or failed to notice early reports in the literature. Both manufacturers and physicians would also face the prospect of having issues of causation decided in the crucible of the courtroom, where the scientific method is but one factor in the determination.
By contrast, the manufacturer of a known teratogen can transfer the risk of liability to the health care provider under the learned intermediary doctrine by promulgating a sufficiently stringent warning about the fetotoxic effects.36 The provider, in turn, can avoid liability by choosing another nonfetotoxic alternative. Where an appropriate alternative is not available, the provider can decrease or eliminate his or her exposure by telling patients about the potential risks and allowing them to share or shoulder the responsibility.
Finding out about fetotoxicity averts harm to children. Even though the children injured by thalidomide and DES have not often succeeded in their efforts to obtain compensation from drug manufacturers,37 they were nonetheless harmed because their mothers took these drugs. The legal system may shift some of the costs of injuries away from the injured parties, but the uncompensated harms do not go away. It is also possible that children are currently being injured by their parents' ingestion of drugs whose fetotoxic effects have not yet been detected. Finally, it is clear that having this information alters behavior in ways that minimize children's exposure—physicians usually will not prescribe fetotoxic drugs and fertile individuals usually will not choose to take them. It may be that most fetotoxic effects will not be detected in clinical trials. But unless the drug will never be used in fertile individuals, the answer to this dilemma is not to exclude such people from clinical trials but rather to broaden the scope of inquiry to require animal studies of mutagenicity and teratogenicity prior to testing or at least marketing for human use and to implement truly effective methods of long-term surveillance that begin during clinical trials.
In closing, since information obtained during research will benefit future children, proposals to limit or ban liability for children injured while the data are being collected seem unjust unless support were otherwise provided for their medical and other needs. Simply to ban claims would mean that the children injured by research and their families would bear all the costs while potential
parents and other children receive all the benefits. In the end, the costs of potential liability to children injured as a result of their parents' participation in clinical trials may simply be ones that must be borne as costs of ensuring that new products are fully tested before they are brought to market.
I would like to thank Carrie Genova for her excellent research assistance and Jay Clayton and Nicholas S. Zeppos for their helpful comments on earlier drafts.