Existing Remedies and Their Limitations 5 Although injuries in the course of IRB-approved biomedical or behavioral research appear to be neither frequent nor severe, they do occur. When subjects are injured, what remedies are currently available to them? Do existing judicial or administrative remedies assure adequate compensation? The Commission has found that to a certain degree, some of the claims by injured subjects can be and are being met outside any formal legal structure. Many investigators and research institutions apparently provide emergency and short-term medical care to subjects injured in research. 1/ Such care is provided, often without charge to the subjects (or their health care insurers), as a matter of professional and institutional responsibility. For subjects with minor, short-lived injuries (who constitute a large proportion of injured subjects), the provision of free, on-the-spot medical care dispenses with any need for formal "compensation" for research injuries. This informal means of resolution is, however, neither universal nor comprehensive. On occasion, research injuries could have severe consequences for the subject's physical well-being and ability to earn a livelihood or perform family responsibilities; further, they could require continuing medical care of supervision over an extended period. In such circumstances, the injured subject may not be able to count on the largess of the research institution. Moreover, the research institution and investigator may find that the financial demands exceed the capacity of their research budgets or the discretionary funds available to them. This chapter examines several issues which arise when serious injuries do occur. What factors currently 1. Testimony of Edward Holmes, M. D., on behalf of the Association of American Medical Colleges, transcript of 4th meeting of the President's Commission (September 15, 1990) at 169-74. 89. Compensating for Research Injuries: Chapter 5 govern the ability of injured research subjects to secure medical care and financial redress for injuries sustained as a result of participation in research? To what degree are these factors, particularly as reflected in the legal rights of injured subjects, in accordance with the ethical claims of the subject and the moral obligations and responsibilities of the investigator and the wider society? What changes are necessary to give legal effect to the underlying ethical claims? 2/ _____________________ 2. These issues are addressed, at least in part, by a growing legal literature on compensation of injured research subject. Many of the seminal early treatments are discussed in Chapter Two of this Report, including articles by Ladimer, Dietrich, Bergen, Stickel and Calabresi. Other articles noteworthy include: Bernard R. Adams and Marilyn Shea-Stonum. Toward a Theory of Control of Medical Experimentation with Human Subjects: The Role of Compensation. 25 CASE W. RES L. REV. 604 (1975): Nancy E. Cahill. Compensation of Subjects Injured in Experimental Medicine Programs: The Ethical and Legal Considerations, 1 J. LEGAL MED. 110 (1979): Ciba Foundation Study Group, MEDICAL RESEARCH: CIVIL LIABILITY AND COMPENSATION FOR PERSONAL INJURY: A DISCUSSION PAPER. The Ciba Foundation. London (1980); Patricia J. Cooper. Compensation for Human Research Subjects: Reform Ahead of Its Time?. 2 J. LEGAL MED. 1 (1930); Bernard M. Dickens, Constructural Aspects of Human medical Experimentation. 25 U. OF TORONTO I. J. 406 [Fall 1975); J. B. Harmon, Compensation of Research Subjects for Adverse Effects, in U. S. Department of Health, Education and Welfare. ISSUES IN RESEARCH WITH HUMAN SUBJECTS, DHEW Publication No. 80-1858, U. S. Government Printing Office, Washington (1980); Seymour Perry, Compensation of Research Subjects for Adverse Effects, in ISSUES IN RESEARCH WITH HUMAN SUBJECTS: John A. Robertson., Compensating Injured Research Subjects: II. The Law, 6 HASTINGS CTR. REP. 29 (1976); Arthur Jay Silverstein, Compensating Those Injured Through Experimentation, 33 FED. BAR J. 322 (1974). See also the discussions of compensation for research injuries in Charles Fried, MEDICAL EXPERIMENTATION: PERSONAL INTEGRITY AND SOCIAL POLICY, American Elsevier Publishing Co., Inc., New York (1974) at 13-43, 165-72; George J. Annas, Leonard H. Glantz and Barbara F. Katz, Compensation for Harm: An Additional Protection for Human Subjects, in INFORMED CONSENT TO HUMAN EXPERIMENTATION: THE SUBJECT'S DILEMMA, Bailinger Publishing Co., Cambridge, Mass. (1977) at 257-277. Two additional bodies of legal literature address Compensating for Research Injuries: Chapter 5 90 Negligence Customarily, redress for personal injuries is provided in the legal system by the law of torts. For a subject to receive compensation through the courts for a research-related injury, the subject would have to establish legal liability for the injury on the part of one or more named defendants--typically, the investigator or the research institution. At present, such legal liability must be predicated on a showing of negligence. 3/ That is, unless the plaintiff-subject can prove the defendant was "at fault," he or she cannot legally recover from the injuries. __________________ questions similar to those rising in the research context. For pioneering discussions of nonfault insurance for adverse effects resulting from standard medical practice, see Clark C. Havighurst and Laurence R. Tancredi, 'Medical Adversity Insurance' - A No-Fault Approach to Medical Malpractice and Quality Assurance, 51 MILBANK MEMORIAL FUND Q. 125 (1973); Robert E. Keeton, Compensation for Medical Accidents, 121 U. PA. L. REV. 590 (1973). Articles discussing compensation for injuries associated with vaccine or immunization programs include Marc A. Franklin and Joseph E. Mais, Jr., Tort Law and Moss Immunization Programs: Lessons from the Polio and Flu Episodes, 65 CALIF. L. REV. 754 (1977); Irving Ladimer, Legal and Regulatory Perspectives in Mass Immunization Programs. 643 INSURANCE LAW JOURNAL 459 (1976); Leslie C. Ohta, Immunization Injuries: Proposed Compensatory Mechanisms - An Analysis 11 CONN. L. REV. 147 (1978); Bonnie L. Siber, Apportioning Liability in Mass Innoculations: A Comparison of Two Views and a Look at the Future, 6 N. Y. U. REV. OF LAW AND SOCIAL CHANGE 231 (1977). For comprehensive recent discussions of compensation issues in broader contests, see Eli P. Bernzweig, BY ACCIDENT NOT DESIGN: THE CASE FOR COMPREHENSIVE INJURY REPARATIONS, Praeger Publishers, New York (1981): Jeffrey O'Connell, ENDING INSULT TO INJURY; NO-FAULT INSURANCE FOR PRODUCTS AND SERVICES, Univ. of Illinois Press, Urbana, Ill. (1975). 3. Cases based solely on "mal-research" have not been reported. Virtually all decisions resulting in the award of damages to injured research subjects appear to be based in whole or part, upon proof that injuries resulted from participation in research to which the plaintiff (or individuals authorized to act on his or her behalf) did not grant legally effective informed consent. Such cases might be litigated on the basis of negligence or, in some jurisdictions, on the legal theories of trespass or battery. Alexander M. Capron, Informed Consent in 91. Compensating for Research Injuries: Chapter 5 was "at fault" he or she cannot legally recover for the injuries. Requirements for, and Obstacles to, Recovery. The traditional formula for the elements necessary to state a cause of action for negligence includes: (a) A duty or obligation, recognized by the law, that a person conform to a certain standard of conduct (" due care") for the protection of others against unreasonable risks. (b) A failure to conform to the standard required. (c) A reasonably close causal connection between the conduct and the resulting injury ("proximate cause"). __________________________- Catastrophic Disease Research and Treatment, 123 U. PA. L. REV. 340, 403-423 (1974). Thus far, there has been only a single reported appellate decision in a North American jurisdiction sustaining a damage award for a research injury. Halushka v. University of Saskatchewan, 52 W. W. R. 606 (Sask. Ct. App. 1965) (upholding jury verdict of $22,500 based on lack of adequate informed consent). In a second case, decided as this Report was in preparation, a New York jury awarded $2.9 million in damages to an individ-ual blinded by retrolental fibroplasia. The jury found that following his premature birth, the plaintiff had been enrolled in a randomized clinical trial of oxygen therapy without the consent of this parents. The jury awarded damages on the basis of both malpractice and lack of informed consent. Burton v. New York Hospital. An appeal is now pending. Informed consent issues have also been raised in ongoing litigation growing out of research on DES conducted at the University of Chicago in the early 1950s. Mink v. University of Chicago, 460 F. Supp. 713 (N. D. Ill. 1978). The first human use of an artificial heart, while perhaps not constituting "research with human subjects" in the technical sense of that term, also resulted in litigation over the adequacy of informed consent; in that case, plaintiff's challenge was rejected by the courts. Karp v. Cooley, 493 F. 2d 408 (5th Cir. 1974). Thus, when subjects are enrolled in research protocols without proper consent, and their participation results in injury, the injured subjects may seek relief in the courts on the basis of lack of informed consent. Compensating for Research Injuries: Chapter 5 92 (d) Actual loss or damage resulting to the interests of another. 4/ The standard of "due care" in human research conducted by a professionally qualified investigator would be that established by his or her peers. Thus, in order to prevail in a lawsuit against an investigator (or the research institution for which the investigator works and which is legally responsible for the investigator's acts), a subject must prove that the investigator departed from those standards (i.e., was "negligent") and as a result caused the subject's injury. Needless to say, it is often difficult to establish a "standard" of care for interventions that are (or contain components that are) by definition innovations from existing stan-dards. (The accepted norms for researchers, including consultation with peers and standards for judging the due care of a researcher on procedural matters.) In addi-tion, the injured subject would probably be barred from recovery if negligent conduct on his own part contributed to the injury or if he "assumed the risk" of injury in consenting to participate in the research. The obstacles in the way of recovery posed for any plaintiff by the requirements of tort law loom especial-ly large in the research context. Some--though not all--of these special difficulties rise because it is often hard to prove negligence in the setting of advanced scientific research. Of course, research scientists, like biomedical and behavioral scientists in standard practice (and like all other human beings), are not immune from simple carelessness. The lack of due care in such instances may be palpable and easy to prove. The very ethos of scientific investigation, however, requires that research be conducted precisely, and the available data on incidence of research injuries (and the virtual absence of court decisions awarding damages for negligently conducted research) suggest that errors careless enough to result in serious injury are very rare in biomedical and behavioral research. Considerably more subtle questions may be posed in ____________________ The remainder of this chapter places primary emphasis on means of redress available to injured subjects who would not be in a position to complain that their informed consent was totally absent. 4. William L. Prosser. HANDBOOK OF THE LAW OF TORTS, (4th ed.) West Publishing Co., St. Paul. Minn (1971) at 30, p. 143. 93. Compensating for Research Injuries: Chapter 5 particular cases regarding the decision to move from animal experimentation to research with human subjects, the design of the research protocol, and technical aspects of the actual conduct of the protocol. Scientific judgements must be made at many points along the way. When are such judgements negligent? Under traditional tort principles, the burden is on the injured subject to prove, by expert testimony, the "standard of care" owed by the investigator (and any reviewing bodies) to the research subjects. Marshalling such expert evidence may be an exceedingly difficult undertaking for the reserch subject. In addition to the highly technical nature of the issues involved, the subject may encounter difficulties in finding other investigators willing to testify on his behalf--a phenomenon parallel to the so-called "con-spiracy of silence" in medical malpractice litigation. 5/ The difficulties of establishing negligence in the research setting are further complicated by the legal requirements of "foreseeability." As stated by the late Dean William Prosser: Negligence...is conduct which falls below the standard established by law for the protection of others against unreasonable risk. It necessarily involves a foreseeable risk...If the defendant could not reasonably foresee any injury as the result of his act, or if his conduct was reasonable in the light of what he could anticipate, there is no negligence, and no liability. 6/ The application of this legal requirement of foreseeability to particular factual situations is often controversial, and legal scholars disagree even as to the guiding principles. Given the very nature of scientific research adventure into the unknown, the foreseeability doctrine introduces an unpredictable and confounding factor into any litigation concerning the conduct of research. Research is often carried out precisely because the range of consequences of a particular intervention and the probability of each outcome are unknown. The research is conducted in order to find out more about them. In these circumstances, what are the ________________________ 5. See e.g., Melvin M. Belli, An Ancient Therapy Still Applied: The Silent Medical Treatment, 1 VILLANOVA L. REV. 250 (1956); Richard M. Markus, Conspiracy of Silence, 14 CLEV.-MAR. L. REV. 520 (1965); Note Overcoming the 'Conspiracy of Silence: Statutory and Common Law Innovations, 45 MINN. L. REV. 1918 (1961). 6. Prosser, supra note 4, at 43, p. 250. Compensating for Research Injuries: Chapter 5 94 "foreseeable" risks, and what are the steps reasonably required to protect against them? Can that be known prior to carrying out the research? The intrinsic difficulty of such questions suggests the magni- FOR REFERENCE SEE (16bb10.gif)] 95. Compensating for Research Injuries: Chapter 5 Applying this analysis to the situation of the injured research subject, Professors Fleming and Sugarman find it "unnecessary to use the doctrine 'assumption of risk' to explain why well warned research subjects whose misfortune it is to suffer from a carefullly run exper-iment will fail to recover damages in a suit based on negligence." 9/ In such circumstances, there has simply been no negligence, and hence no basis for a recovery. A closely related question is whether a subject waives his legal rights to recovery in granting informed consent to participation in research, particularly since, in compliance with 45 C. F. R. 46.116(a)(6), subjects participating in research involving more than minimal risk must be provided "an explanation as to whether any compensation and an explanation as to whether any medical treatments are available if injury occurs..."The answer to this question is clear. Federal regulations explicitly preclude "any exculpatory language through which the subject...is made to waive or appear to waive any of the subject's legal rights, or releases or appears to release the investigator, the sponsor, the institution or its agents from liability for negligence." 10/ Even in the absence of such explicit regulatory language, the courts disfavor exculpatory clauses in medical contexts, 11/ and there are substantial policy arguments supporting this position. 12/ _____________________ 9. Id. at 23 10. 45 C. F. R. 46.116(1981). 11. See, e.g., Tunki v. Regents of the University of California, 32 CAL RPTR. 33, 383 P. 2d 441 (1963). 12. See e.g., Comment, Legal Implications of Psychological Research with Human Subjects, 1960 DUKE L. J. 265, 272. Although the subject may have known what the experimenter intended to do and may have been willing to have this done to him, still, the subject may not have fully realized the risks involved. Mere knowledge of the facts which create the risk of harm is not enough unless there is a true appreciation of the nature and extent of the risks; only then is the subject's assent given under circumstances which make it legally effective. The im-portant point here, however, is that, in ascertaining whether the subject truly appreciated the risk of harm involved, it is inevitable that considerations of policy, depending on the result desired, will consciously or unconsciously influence the court's decision. Compensating for Research Injuries: Chapter 5 96 Other difficulties encountered by injured research subjects seeking relief in the courts are common to all tort litigation. Foremost among these is the problem os causation. Proof of causation may be sophisticated anal-ysis of often subtle statistical variations in risks and outcomes. In negligence and other tort litigation, the burden is on the injured party both to come forward with evidence suggesting causation and ultimately to persuade the trier of fact that his or her injury was "proximately caused" by the defendant's breach of a standard of care. 13/ A plaintiff claiming negligence must also surmount hurdles posed in some situations and in some jurisdic-tions by doctrines of charitable immunity (when the research is conducted by a charitable institution) and sovereign immunity (when the research is conducted or, perhaps, supported by a governmental agency), and by statutes of limitations which, in some jurisdictions, effectively preclude recoveries for certain latent injuries. With the exception of the latent injury problem (which is discussed in the Appendix to this Report 14/), these doctrines are not peculiar to the research context and do not require detailed discussion here. Critique of Negligence as a Remedial Mechanism. Quite apart from the considerable difficulties of prevailing in a negligence action, a more fundamental question remains: as a matter of moral responsibility and public policy, ought proof of negligence to be required as a prerequisite to recovery for research injuries? This question lies near the heart of the Commission's discussion of ethical issues in Chapter Three: a few points are reiterated here to contrast the ethical arguments with the current state of the law. First, and most obvious, the negligence system is structured to preclude recovery for injuries not resulting from negligent conduct. It is widely recog-nized that "despite the exercise of the highest degree of care and skill by the medical investigator concerned, death or a personal injury which was quite unforeseen and indeed quite unforeseeable might be suffered" 15/ by a ___________________ 13. See pp. 138-139 infra 14. See pp. 140-141 infra: see also Appendix S to this Report. See also Annas et al., supra note 2 at 257-58, and Fleming and Sugarman, supra note 8, at 22-24. 15. Royal Commission on Civil Liability and Compensation 97. Compensating for Research Injuries: Chapter 5 research subject. In many settings, the obstacles to recovery may be--indeed, probably are--quite appropriate. There ought, after all, to be good reasons for the legal system to shift the costs of an activity from A to B. The difficulties in the way of such a shifting of costs simply reflect the protections established by society against the misapplication of this legal power. But if research subjects should, on ethical grounds, be compen-sated for their injuries, despite the lack of negligence by an investigator, the rules of negligence law will be patently inappropriate. Second, as pointed out in a recent article critical of the tort system's handling of the DES litigation. "[t]he doctrines of tort law were devised for straight-forward moral situations--somebody's misbehavior causes harm to somebody else--and for narratives that observe the three unities of classical drama: time, place, and action." 16/ The focus of traditional (negligence) tort doctrine or the immediate parties to an accident--the victim and the allegedly negligent defendant--fits the moral equation of research injuries only imperfectly. The immediate parties--subject and investigator--encom-pass only one aspect of the wider society's interest in biomedical and behavioral research. Much research is conducted with the encouragement or, indeed, financial support of the government itself. The fruits of research accrue to the society at large, in a form economists term "positive externalities." Thus, society as a whole has a direct stake in the conduct of scientific research. Even on the theories supportive of a "fault" component in tort litigation, the presence of strong societal exter-nalities suggests a modification in the usual calculus of fault, so that society will share in the financial burdens, as well as the benefits, of research activity. Thus, the traditional emphasis of negligence law on the "moral drama" of the defendant's behavior fails to capture (in economists' terms, to "internalize") the societal interest in the conduct of research. Other approaches not founded on negligence, could do so more successfully. A final point, different in character although carrying similar implications, derives from the nature of the relationship between investigator and subject in the for Personal Injury, REPORT, Her Majesty's Stationery Office, London (1978), Vol 1 at 1339 (hereinafter cited as ROYAL COMMISSION). _______________________ 16. Michael Kinsley, Fate and Lawsuits, 102 THE NEW REPUBLIC 20, 21, (June 14, 1980). Compensating for Research Injuries: Chapter 5 98 conduct of research. Ideally, the relationship is a collaborative one in which the subject knowingly accepts certain risks of physical injury on behalf of society and participates, together with the investigator, in the ongoing research. 17/ The bond between investigator and subject may, however, be rent asunder by the adversary system characteristic of tort litigation, most particu-larly when judgement in favor of a subject is dependent on proof of an investigator's wrongdoing. While the negligence system is often applauded for creating incentives for acting with care, the knowledge that an injury could result in a courtroom battle may also create incentives less worthy of support: a certain distance, lack of candor, and perhaps even reluctance to provide necessary immediate care when the provision of such care might alert a possibly litigious subject that something had gone amiss. Such incentives may be physically harm-ful to the subject; they are certainly destructive of the collaborative ideal held out for the relationship between investigator and subject. It is often argued that the existing system of recoveries for medical malpractice has poisoned relationships between doctors and patients. 18/ The Commission believes that the malpractice systems should not serve as the model for remedying injuries that arise in ethically conducted biomedical and behavioral research. Strict Liability Britain's Royal Commission on Civil Liability and Compensation for Personal Injury (the "Pearson Commis-sion") recently recommended the doctrine of strict liability in tort as an alternative to the negligence system for providing compensation to subjects injured in research. 19/ In the Commission's words, "any volunteer for medical research or clinical trials who suffers se-vere damage as a result should have a cause of action, on the basis of strict liability, against the authority ________________________ 17. See e.g., Talcott Parsons, Research with Human Subjects and the Professional Complex, 38 DAEDALUS 325-360 (1969); Renee C. Fox, EXPERIMENT PERILCUS, The Free Press, Glencoe, Ill. (1259). 18. See, e.g., David Mechanic, Some Social Aspects of the Medical Malpractice Dilemma, 6 DUKE L. J., 1179 (1975). 19. ROYAL COMMISSION, supra note 15 at 1341. 99. Compensating for Research Injuries: Chapter 5 to whom he has consented to make himself available." 20/ Strict liability has also been characterized by some commentators as an "appealing solution" to the problem of research injuries. 21/ An inquiry into whether strict liability provides an answer for subjects injured in research has two compon-ents. First, is strict liability a remedy now available to injured research subjects? Second, if it is not, ought it be? Roots in History and Policy. The doctrine of strict liability, which permits recovery without proof of negli-gence, has ancient roots in Anglo-American law. Indeed, in the early common law development of the law of torts, the focus was on causation; no inquiry was made into "fault" as that term is now employed. As the doctrine of negligence developed, largely replacing the earlier con-cepts, several pockets of strict liability remained. In more recent times, strict liability has reemerged as an increasingly prevalent theory of tort liability. Today, its principal applications are to activities viewed as "ultrahazardous" or a"abnormally dangerous" and to defective products (including, but not limited to, those products posing extraordinary danger). A variant of strict liability theory applies to pharmaceuticals and certain medical devices which are viewed as "unavoidably dangerous." 22/ Contemporary courts and scholarly commentators have sought to justify the expanding scope of applicability of strict liability on explicit policy grounds. In one landmark decision, the Chief Justice of the California Supreme Court explained: Even if there is no negligence, however, public policy demands that responsibility be fixed wherever it will most effectively reduce the hazards to life and health inherent in defective products that reach the market. It is evident that the manufacturer can anticipate some hazards and guard against the recurrence of others, as the public cannot. _______________________ 20. Id. 21. Note, Medical Experiment Insurance, 70 COLUM. L. REV. 965, 966 (1970) 22. Escola v. Coca Cola Bottling Co., 24 Cal. 2d 453, 150 P. 2d 436, 440-41 (1944). Compensating for Research Injuries: Chapter 5 100 Dean Prosser summarized the views of many contemporary scholars, stating that the courts, in applying strict liability to abnormally dangerous condi-tions or activities, have in effect recognized a new doctrine. that the defendant's enterprise, while it will be tolerated by the law, must pay its way. There is "a strong and growing tendency, where there is blame on neither side, to ask, in view of the exigencies of social justice, who can best bear the loss and hence to shift the loss by creating liability where there has been no fault." 23/ Applicability to Research Injuries. Whether strict liability doctrine applies to research with human sub-jects is, at present, an open question. No court has explicitly ruled on the issue in the context of contem-porary, scientifically designed and peer-reviewed re-search. The factors which govern the applicability of strict liability to new areas provide no unequivocal guidance. 24/ Three factors have been recognized as favoring application of the doctrine. Research activity is clearly valuable to the community. 25/ Risks of injury cannot be entirely eliminated by the exercise of reasonable care. 26/ In some sense, research is not a matter of "common usage," although this conclusion might be disputed. 27/ On the other hand, research is typical-ly not conducted in inapproprite settings. 28/ Probably the critical factors in determining the applicability of strict liability doctrine to research are whether research involves "a high degree of risk of some harm" 29/ and whether the "gravity" of the harm _______________________ 23. Prosser, supra note 4, at 75, p. 494 original citations omitted. 24. American Law Institute, RESTATEMENT (SECOND) OF TORTS, 519 ff (1970) 25. Id. at 520(f). 26. Id. at 520(c). 27. Id. at 520(d). 28. Id. at 520(e). 29. Id. at 520(a). 101 Compensating for Research Injuries: Chapter 5 which may result is likely to be "great" 30/ While the popular imagination might be tempted to answer these questions affirmatively, the evidence marshalled by the Commission strongly suggests that research conducted with IRB approval does not involve high risks of serious in-jury and, on empirical grounds, probably should not be viewed as an "abnormally dangerous activity." 31/ In addition to these factors specified by the Second Restatement of Tort, courts considering the applicability of strict liability to research with human subjects would be guided by existing legal precedents. An initially plausible argument can be constructed, on the basis of several early English and American cases, that physicians experiment "at their peril." Closer examination of these cases, however, indicates that they do not concern re-search in its modern sense and should not be viewed as governing precedents for the application of strict liability to contemporary research. The earliest such case, Slater v. Baker and Stapleton, C. B. 32/ was decided by an English court in 1767, prior to the advent of the legal doctrine of negligence. The decision upheld a judgement against an eminent surgeon who employed a novel means of treating a broken leg, with unfortunate results. The court noted that "many men very skillful in their profession have frequently acted out of the common way for the sake of trying experiments" and held that such behavior opened the way to liability in the event of injury: [T]his was the first experiment made with this new instrument; and if it was, it was a rash action, and he who acts rashly acts ignorantly ... 33/ Slater was followed a century later, by an American case, Carpenter v. Blake. 34/ There the court said: [W]hen the case is one as to which a system of treatment has been followed for a long time, 30. Id at 520(b). _________________________ 31. Se Chapter 4, infra. 32. 95 Eng. Rep. 550(1767). 33. Id. at 862-63. 34. Carpenter v. Blake, 60 Barb. 488(N.Y. Sup. Ct. 1871), rev'd on other grounds, 50 N.Y. 695(1872). Compensating for Research Injuries: Chapter 5 102 there should be no departure from it, unless the surgeon who does it is prepared to take the risk of establishing, by his success, the pro- priety and safety of his experiment. 35/ Taken together, the Slater and Carpenter cases appeared to establish the proposition that physicians depart from established modes of treatment at their peril, and that injuries resulting from experimental treatment would be actionable without further proof of negligence. Yet for several reasons it would probably be mistaken to apply this proposition to contemporary research. First, the conduct at issue in Slater and Carpenter was not scientifically designed research in the modern sense, but the "trying out" of an innovative therapy" or nonvalidated medical procedure," for which different norms may be appropriate. Second, contemporary requirements that Federally supported research pass scientific and IRB muster suggest that the gap between "human experimentation" and professional norms of prac-tice is narrower today than was true of the conduct condemned in the early legal cases. Third, and most important, the law itself has changed over the past century, pointing toward a more nuanced evaluation of the facts and circumstances surrounding the decision to employ innovative or experimental treatments. The beginnings of this change were already discernible in the nineteenth century. In an 1895 case, Jackson v. Burnham 36/ the court, while repeating the rule that "if a physician sees fit to experiment..he should do so at his peril 37/ construed the rule to require only that the physician "must be able, in the case of deleterious results, to satisfy the jury that he had reason for the faith that was in him, and justify his experiment by some reasonable theory 38/ By 1935, when the Michigan Supreme Court decided Fortner v. Koch, 39/ ______________________ 35. 60 Barb. at 524. 36. 20 Colo. 332, 39 P. 577(Colo. 1895). 37. Id. 39 P. at 58C. 38. Id at 580. 39. 272 Mich. 273, 281 N.W. 762(Mich, 1935). 103 Compensating for Research Injuries: Chapter 5 the courts recognized explicitly that "if the general practice of medicine and surgery is to progress, there must be a certain amount of experimentation carried on." 40/ The legal requirements were essentially, those of consent and reasonableness: "such experiments must be done with the knowledge and consent of the patient or those responsible for him, and must not vary too radically from the accepted method of procedure." The transition in legal standards, from "the physician experiments at his peril" standard of Slater and Carpenter, to the reasonableness criterion of Fortner v. Koch, paralleled a more general tendency in tort law during the nineteenth and early twentieth centuries: the emergency and solidification of negligence as the touch-stone of tort liability. This development was particu-larly evident in the field of medical malpractice liti-gation, which has steadfastly adhered to a standard of liability based upon professional negligence 42/ and which today provides perhaps the closest analogue to research "malpractice." Thus the proposition that research with human subjects is an "abnormally dangerous activity" subject to strict liability standards finds no clear support in the _____________________ 40. Id at 765. 41. Id. 42. One posible exception may be Helling v. Carey, 83 Wash. 2d 514, 519 P. 2d 961 (1974), in which the Washington Supreme Court found the defendant opthalmolo-gists negligent as a matter of law for failing to conduct a pressure test for glaucoma on a young patient, despite undisputed testimony that such testing was not heretofore required under the "standards of the profession." The court concluded that, in the circumstances presented, medical professionals should be held to a higher stan-dard. Another justice concurred in this result but argued that the court was imposing a standard of liability which "approached that of strict liability" (Concurring opinion of Utter. J.). Justice Utter urged that the court should explicitly recognize the applicability of strict liability doctrine to medical malpractice, in part in the belief that awarding compensation on the basis of strict liability rather than negligence would avoid unfairly 'imposing a stigma of moral blame upon doctors who...used all the precautions commonly prescribed by their profession in diagnosis and treatment." Id. at 986. See also Gates v. Jensen, 595 P. 2d 919 (Washington 1979). Compensating for Research Injuries: Chapter 5 104 empirical evidence or in legal precedent. While the possibility cannot be excluded that an innovative court might seek to apply strict liability to research injuries should a sympathetic case come before it (particularly in light of the steadily expanding scope of strict liability in recent decades), on the current state of the law the Commission concludes that injured subjects contemplating legal action would find little encouragement in the doctrine of strict liability. 43/ ___________________ 43. A second branch of strict liability doctrine, that applicable defective products, requires brief mention here. Under the doctrine of products liability, as formulated in 402A of the SECOND RESTATEMENT OF THE LAW OF TORTS, one who sells a "product in a defective condition unreasonably dangerous to the user or customer" may be strictly liable for physical harms caused thereby, without regard to whether the seller "has exercised all possible care in the preparation and sale of his pro-duct." The applicability of this doctrine to products which may be employed in research, particularly pharma-ceuticals and certain medical devices, is a matter of controversy. Comment k to 402A of the SECOND RESTATEMENT recognizes a category of "unavoidably dangerous pro-ducts," including drugs, "which, in the present state of human knowledge, are quite incapable of being made safe for their intended and ordinary use." Such products are not viewed as defective, and hence are outside the scope of 402A products liability, if they are "properly prepared, and accompanied by proper directions and warning." Nonetheless, particularly through their interpretation of the warning requirement, the courts have imposed rather expansive liability on drug manufacturers, often basing their decisions on policy grounds arguably applicable to biomedical research as well. In one much discussed case, concerning liability for a polio case arguably induced by a vaccine, the Fifth Circuit Court of appeals analyzed the issue this way: "Until Americans have a comprehensive scheme of social insurance, courts must resolve by a balancing process the head-on collision between the need for adequate recovery and viable enterprises....This balancing task should be approached with a realization that the basic consideration involved a determination of the most just allocation of the risk of loss between the members of the marketing chain."....Statistically predictable as are these rare cases of vaccine-induced polio, a strong argument can be advanced that the loss ought not lie where it falls (on the victim), but should be borne by the manufacturer as a foreseeable cost of doing business, and passed on to the public in the form 105 Compensating for Research Injuries: Chapter 5 Outside the main body of strict liability law which arose and expanded in scope primarily as a matter of judge-made law, instances have arisen in which the doctrine's scope has been expanded, or limited, by legis-lative action. Indeed, the very uncertainties about the doctrine's applicability to biomedical research also arose with respect to legal liability associated with accidents at nuclear power installations, and this uncer-tainty (and the possibility that courts in different states might resolve the issues differently) played a major role in prompting Congressional action on the Price-Anderson legislation, which in practical effect imposed a nationwide strict liability policy for major nuclear accidents. 44/ The Pearson Commission, which favored strict liability as a basis for compensating subjects injured in research in the United Kingdom, similarly recommended that strict liability be imposed by legislative enactment. Thus, the proper question becomes whether existing strict liability doctrine should be extended to encompass research with human subjects. Critique of Strict Liability. The question of extending strict liability to cover research with human subjects poses substantial policy issues, whose resolution depends in part on the alternatives to which ____________________ of price increases to his customers. Reyes v. Wyeth Laboratories, Inc., 496 F. 2d 1264 (5th Cir 1974) (quoting Helene Curtis industries v. Pruitt. 385 F. 2nd 841, 862 (5th Cir. 1967)). The Court thereby imposed liability for failure to warn of a less-than one-in-one-million risk of contracting polio from a live virus vaccine. A similar analysis might be employed in a research setting, suggesting that standards of liability might differ as between research employing drugs or medical devices and other modes of research. See also Fleming and Sugarman, supra note 8, at 29. 44. Pub. L. No. 85-256 21 Stat. 576, as amended, 2 U.S.C. 2110. In fact, the technique employed in the Price-Anderson Act to accomplish this result was not a straightforward declaration of a "strict liability" standard, but a complex waiver of certain defenses (including the absence of negligence) which might be available under state law. The reason for this rather roundabout approach was a Congressional hesitancy to "Federalize" the standard of tort liability, an area traditionally reserved to the states. See generally, Legislative Drafting Research Fund, Columbia University, ISSUES OF FINANCIAL PROTECTION IN NUCLEAR ACTIVITIES (1973), at 2-7 to 2-9. 106 FOR REFERENCE SEE (16bb11.gif) Compensating for Research Injuries: Chapter 5 107 strict liability is being compared. To be sure, the availability of a cause of action for strict liability would enhance the position of injured research subjects unable to prevail on negligence theories, a result many would consider more just than the existing system. On the other hand, a strict liability system would perpetu-ate many characteristics of courtroom tort litigation which have been widely criticized, not least by those in the medical profession. Litigation based on strict lia-bility principles would retain the adversary character of all tort litigation, pitting the interests of the injured subject against those of the investigator (even if not directly calling the conduct or competence of the inves-tigator into question). Financial recoveries would likely be sought, in the first instance, against the investigator or the research institution, although the principles justifying recovery point to the beneficiaries of research (i.e., the entire society) as the more appropriate source of redress. Like any courtroom litigation, liability as a legal remedy for research injuries is in no sense a panacea. Nonfault Approaches A number of administrative or insurance compensation mechanisms are already in place in the United States. While only a few small-scale, nongovernmental programs are specifically directed toward compensating subjects injured in research, several more broadly based social welfare programs provide incidental benefits to such persons. 45/ Government Compensation Programs. A multitude of Federal, state, and local government programs provide compensation of various types to the injured, medical care to the ill, and financial assistance to the needy. None of these programs is universal in coverage or com-prehensive in scope. Nor do initiatives for comprehen-sive national health care of guaranteed minimum incomres currently appear in prospect. Thus, injured research subjects are eligible for government assistance only insofar as they qualify under categories other than their status as research subjects. For some injured subjects, assistance available under existing governmental programs may be perfectly adequate. For Federal civilian employees participating __________________________ 45. Existing programs are discussed briefly infra at pp. 114-122 and are described more fully in papers prepared for the Commission. See Chapter 7 infra. 108 Compensating for Research Injuries: Chapter 5 in research "while in the performance of [their] duty," 46/ compensation is available in the event of injury through the Federal Employees' Compensation Act (F.E.C.A.). For elderly retirees injured in research, Social Security and Medicare may relieve much of the financial burden. For others, the only recourse may be to Federal and state welfare systems, for which they may or may not be eligible. As regards injured research subjects, the "social safety net" is highly permeable; many injured subjects may simply fall through. As part of its study of compensation for research injuries, the Commission examined the structure and operation of a number of currently functioning govern-mental compensation programs. One of the Commission's objectives was to determine the degree to which existing programs meet the needs of injured research subjects. In this regard, the Commission corresponded with over twenty Federal agencies and departments, and members of the Com-mission staff met with representatives of the Departments of Labor and of Health and Human Services, the Food and Drug administration, the Social Security Administration, and the Veterans Administration, to discuss the availa-bility of retrospective data and the feasibility of collecting prospective data reflecting the incidence of claims by injured research subjects and the benefits provided to them. None of the agencies contacted main-tained records in a form amenable to such analysis, and considerations of time and cost ruled out prospective studies by the Commission. Thus, deficiencies in data limited this portion of the Commission's study to the qualitative conclusions indicated above. Insurance Mechanisms. It is clear that existing insurance-in-force does provide some protection for subjects injured in research. The primary source of protection is general health insurance that may be carried by individual research subjects. Although such individual health insurance is neither universal nor comprehensive, it does provide a source of benefits to insured individuals for health costs associated with research injuries. 47/ Such insurance does not, _______________________ 46. F.E.C.A., 5 U.S.S. 8102(a) 47. While there have been occasional suggestions that injuries resulting from voluntary participation in research activity might be excluded from coverage under terms of existing health insurance policies, see John Robertson, Compensating Injured Research Subjects: II. The Law, 6 HASTINGS ctr rep. 29 (1976), it appears that Compensating for Research Injuries: Chapter 5 107 however, cover lost wages or other non-health-care associated with serious injuries. While general health care coverage for persons participating in research offers at least a partial answer for medical costs associated with research injuries, industry representatives were deeply skeptical of the practicability of marketing to potential research subjects policies specifically designed to cover the risks associated with participation in particular re-search projects. This skepticism was founded, in part, on the extraordinary difficulties and expense involved in efforts to secure private insurance to cover risks in a much-discussed bone marrow transplant case. 48/ More generally, the transaction costs and administrative burdens of writing coverage for individual participants in research were considered to pose insuperable problems in marketing such insurance. Insurance industry representatives were considerably more favorably disposed to a system which would provide blanket coverage for all research subjects at a particu-lar institution during a given period of time. A sub-stantial consensus emerged that such blanket policies could be written and premiums set on the basis of the overall research mix and past claims experience of a given institution, although specific high-risk protocols might require individualized underwriting examination. An institution-wide approach would entail substantially such statements result from a confusion between the use of insurance proceeds to support the research itself, which is typically precluded by insurance contracts, and the use of insurance proceeds to pay for standard medical care necessitated by an injury associated with participa-tion in research. The Commission has been unable to doc-ument instances of insurance company exclusions of the latter type, and the Commission's insurance expert, George Bernstein, testified that on the basis of his extensive inquiries with the insurance industry, no such exclusions currently exist or are contemplated. Trans-cript of the 6th meeting of the President's Commission ) January 9, 1981) at 187-88. ________________________________ 48. Letter from Albert L. Broseghini, Ph.D. (Director, Research Administration, The Children's Hospital Medical Center, Boston, Mass.) to Senator Edward M. Kennedy (December 2, 1974), reprinted in the HEW Secretary's Task Force on the Compensation of Injured Research Subjects, REPORT, U. S. Department of Health, Education, and Welfare, Washington (1977) (hereinafter cited as TASK FORCE REPORT) Appendix B at 123. 110 Compensating for Research Injuries: Chapter 5 lower transaction and administrative costs. (For example, the private insurance company that provided coverage for research injuries at the University of Washington as an "add-on" to the university's liability policy initially set the premium at $.50 per subject. 49/) Furthermore, rate-setting on an institution-wide basis was felt to provide sufficient grounding for under-writing decisions, with an initial break-in period during which insurers would be allowed some recovery of their costs in the event of excessive loss experience. Alternatively, insurance might be written separately for each research protocol. Theoretically, this would lead to individualized assessment of each project, there-by providing a means for the "market" in research to adjust the amount of research in light of an economic quantification of risk. The Commission is skeptical that such "fine tuning" has an adequate basis in scientific fact 50/ or that the additional costs and administrative burdens involved would be outweighed by any ability of the "market" to reduce the amount of truly risky research beyond the existing system of IRB review. Further, in-dustry representatives were dubious that such an approach would be administratively feasible. ____________________________ 49. The University of Washington has for several years been a self-insurer. The total amount paid in claims to injured subjects in the nearly 10 years it has had a formal compensation program averages less than $.5 per subject. This amount does not include the value of free medical services provided to subjects. See Chapter 7 infra. 50. Such a "market control" approach has been discussed by Guido Calabresi, Reflectiong on Medical Experimenta-tion in Humans, in EXPERIMENTATION WITH HUMAN SUBJECTS (ed. by Paul Freund). George Braziller, New York (1970), and Clark C. Havighurst, Mechanisms for Compensating Persons Injured in Human Experimentation, TASK FORCE REPORT, supra note 49, Appendix A. at 81-108, among others, Professor Harry Bostrom of the University of Uppsala in Sweden testified to the Commission that an analogous system is currently in effect in Sweden, Bostrom, On the Compensation for Injured Research Subjects in Sweden (1980); see Appendix K to this Report. However, the volume of research in Sweden is such that a single individual (Professor Bostrom) is able to review all insured protocols and to determine insurance premiums on the basis of comparative evaluation of risks associa-ted with them. The vastly greater scope of research conducted in the United States would, in the view of the Commission (and of representatives of the Compensating for Research Injuries: Chapter 5 109 The series of discussions between Commision staff and industry representatives culminated in testimony before the Commission by insurance industry representa-tives in September of 1980. In that testimony, the representatives stated "that we would like to participate with you in seeing that a reasonable compensation program for human subjects is developed" 51/ Stressing that the program should not be "unlimited and open ended," and that the development of a new insurance mechanism must proceed with moderation, the insurers reiterated their support for "the development of pilot programs and asses-sment of experience under those programs" as "the best route to develop an actual program for the compensation of human subjects." __________ insurance industry), render such a "ratings bureau" approach to American research totally impractical. Further, on the basis of a careful analysis of risk literature conducted for the Commission, a minutely graded ranking of risks would be impossible or grossly misleading. See Mary Harvey and Robert J. Levine, Risk of Injury Associated with Twenty Invasive Procedures Used in Human Experimentation and Assessment of Reliability of Risk Estimates (1980), Appendix E to this Report. 51. Testimony of Dennis R. Connolly, Senior Counsel to the American Insurance Council, transcript of 4th meeting of the President's Commission (September 15,1980) at 1.