DISCLAIMER The following is a staff memorandum or other working document prepared for the members of the Advisory Committee on Human Radiation Experiments. It should not be construed as representing the final conclusions of fact or interpretation of the issues. All staff memoranda are subject to revision based on further information and analysis. For conclusions and recommendations of the Advisory Committee, readers are advised to consult the Final Report to be published in 1995. TAB G þþþDRAFT þ FOR DISCUSSION PURPOSES ONLYþþþ MEMORANDUM TO: Members of the Advisory Committee on Human Radiation Experiments FROM: Advisory Committee Staff DATE: December 12, 1994 RE: Discussion Memo on Proposed Remedies Guidelines [This memo was prepared by the staff in order to help facilitate a Committee discussion on how to address the issue of what remedial guidelines may be appropriate where wrongs or harms have occurred. This draft is not intended to represent the views of the Committee or any Committee Member. Rather, it provides one possible approach for determining potential remedies for harms caused by government sponsored human radiation experiments conducted in the past.1 [Prospective preventive measures that may be implemented to prevent similar harms in the future -- including heightened penalties for violations of current regulations and greater openness in government -- will be addressed separately. ]] INTRODUCTION In proposing remedies for victims of radiation experiments, the Advisory Committee is both guided and circumscribed by its Charter, which directs the Committee to make recommendations regarding "human radiation experiments." The Charter defines "human radiation experiments" as (1) "Experiments on individuals involving intentional exposure to ionizing radiation," and (2) "Experiments involving intentional environmental releases of radiation that (A) were designed to test human health effects of ionizing radiation; or (B) were designed to test the extent of human exposure to ionizing radiation." The Charter also directs the Committee to provide information and recommendations on thirteen specific atmospheric releases of radiation that do not appear to meet the definition of "human radiation experiments" provided by the Charter. The Committee recognizes that many individuals may have been exposed to radiation outside of these contexts; however, the Committee must limit its recommendations to the matters under its purview. Several challenges exist in proposing remedies for those individuals within the Committee's mandate. For example, the passage of time has, in many cases, made reconstructing a detailed record 1 of events difficult. In addition, many of the victims have long since died. Moreover, in recommending remedies for the many experiments which were conducted, the Committee must assess the government's behavior in light of both contemporary mores and the ethical and scientific standards that prevailed at the time they were conducted. (The conduct of the government includes all of the agents and contractors that acted on the government's behalf.) Thus, the Committee must determine whether an experiment that complied with prevailing standards would be less susceptible of a remedy than one that did not, even if neither experiment would meet today's ethical criteria. The need to make retrospective moral judgments necessarily complicates any remedies analysis. Finally, in making recommendations concerning available remedies, the Committee is aware of the need to carefully distinguish between the various types of experiments. JUDGING THE PAST: POSSIBLE CRITERIA FOR RETROSPECTIVE MORAL JUDGMENT Before making specific recommendations regarding appropriate remedies, the Committee must make a foundational decision regarding the criteria it will use in making retrospective moral judgments about past experiments. It must be noted that the Charter grants wide latitude in this regard, directing the Committee to evaluate whether "the experiments' design and administration adequately met the ethical and scientific standards, including standards of informed consent, that prevailed at the time of the experiments and that exist today." There are several approaches that could be used in evaluating past experiments, which are discussed in separate manuscripts and memorandum. However, given the language of the Charter, any conclusion regarding ethics should take into account both past and present ethical standards. The ethical model adopted by the Committee may have implications for its decisions regarding the application of guidelines for remedies. LESSONS FROM THE PAST In order to best evaluate appropriate remedies for the victims of experiments that the Committee has been studying, it is useful to review how Congress, the courts, and previous Commissions have approached the issues of both compensating injured research subjects, and compensating victims of exposure to ionizing radiation. These sources can inform the Committee of the degree to which remedies may already be available and also of types of remedies that may be particularly appropriate. The Committee will review the statutory schemes available for veterans and others who have been exposed to ionizing radiation from nuclear weapons tests, as well as statutory, legal, and analytical conclusions that have been reached concerning the compensation of research subjects, such as the findings and recommendations of the 1982 President's Commission for the Study of Ethical Problems in Biomedical and Behavioral Research regarding the need for and means of establishing a program to compensate injured research subjects. 2 GUIDELINES FOR REMEDIAL MEASURES The following guidelines are meant to assist policymakers in determining appropriate remedies for individuals exposed to ionizing radiation in the context of an experiment. The guidelines attempt to address several major themes. The first concerns the overall policy considerations and limitations facing the Committee. The Advisory Committee is guided by its Charter, which limits its purview to "human radiation experiments" as defined therein. Accordingly, the Committee's guidelines for remedial measures are addressed to deal only with the types of experiments outlined in the Charter. These guidelines do not address every situation in which the government has caused persons to be exposed to ionizing radiation. The Committee notes that Congress has enacted legislation to provide compensation to certain groups of individuals exposed to radiation in the course of production, testing, and deployment of atomic weapons -- e.g., uranium miners, persons living downwind of the Nevada Test Site, and civilians and veterans who participated in atomic bomb tests. However, most of these individuals do not fall within the Committee's Charter. To the extent that there is a subset of individuals within those groups who may also have been involved in human radiation experiments, the Committee's guidelines may suggest that additional remedial measures are appropriate. Nor does the Charter include most workers at government owned atomic weapons production facilities, who may have been exposed to radiation, by accident or otherwise, in the course of their work. Some of these individuals may be entitled to remedies established by statute (e.g., occupational requirements) or imposed by the courts. The Committee will then offer a set of "methodological guidelines" for assessing the type of remedy that would be most appropriate for a given type of experiment, based on the type of harm, the extent of government culpability (including subject selection and informed consent), and considerations of causation. Finally, the Committee will address the need for effective implementation of any type of remedy scheme that is put in place. The ultimate responsibility for providing remedies of any form lies with the policymakers in Congress and the Executive branch. The Committee is aware of the dissatisfaction that has been expressed with respect to some of the existing compensation statutes, and believes, accordingly, that any program that is responsive to the harms engendered by government sponsored human radiation experiments must be clearly crafted and objectively administered so that all persons who are entitled to a remedy can receive it with the minimum of difficulty. Guideline 1:The Committee's Remedial Guidelines are Designed to Address Remedies only for those Subjects Covered by Its Charter. The Committee's Charter articulates two general categories of human radiation experiments: (1) "Experiments on individuals involving intentional exposure to ionizing radiation"; and (2) "Experiments involving intentional environmental releases of radiation that (A) were designed to test 3 human health effects of ionizing radiation; or (B) were designed to test the extent of human exposure to ionizing radiation." Guideline 2:Possible Remedies that could be Considered under the Committee's Guidelines include Access to Information, Health Insurance, Life Insurance, Medical Evaluation, Medical Monitoring, Monetary Compensation, and Official Apology. The kind of remedial measure available to individuals who may have been harmed by human radiation experiments (or their survivors, where applicable) will depend largely on the nature of the harm and the recognition of some degree of government culpability. (Although, as a general proposition, the government should not be liable where it is not culpable, the Committee will consider whether a form of strict liability is appropriate for persons who are injured in the course of serving as a research subject. The question of ensuring that remedies are available to research subjects who are harmed in the course of such research, regardless of fault, has been debated and contemplated for many years, with no firm policy currently in place.) The remedy should be molded to fit the specific harm. In some cases, more than one remedy may be appropriate.  Access to Information. In many situations, experimental subjects have difficulty obtaining information pertaining to the particular event and their role in it. Such existing information could serve to answer questions and alleviate concerns about what took place. Thus, an important remedy, particularly for individuals who fear that they are at risk of developing a disease at some time in the future, will be the ability to gain access to records that bear on their experience.  Health Insurance. Guaranteed health care is an appropriate remedy for persons who have permanent health problems as a result of their being the subject of an experiment. Such care is provided, for example, to veterans who are injured in the course of their military service. Although many medical institutions provide medical treatment for research subjects who are injured, such treatment is often only short term and is not guaranteed to all such subjects, nor does it necessarily extend to persons who move away from the hospital at which they were injured. One uncertainty concerning health insurance (and life insurance as well) is how broadly it should apply--i.e., would it cover all medical conditions of the recipient, or only those that are the result of the experiment? The VA, for example, generally provides health care only for service connected disorders.  Life Insurance. A life insurance guarantee would serve to provide a death benefit to the survivors of an experimental subject. This form of remedy would be appropriate for people 4 who are still alive but might die "prematurely" because of their involvement in research. Life insurance could also be an additional benefit for injured persons in cases when the government's conduct is more culpable. As with health insurance, the question arises as to whether such coverage applies only to any cause of death, or only to death attributable to the radiation exposure.  Medical Evaluation. Oftentimes, persons who know that they were the subject of an experiment do not know whether they have been injured or are at greater risk of developing a disease. This is particularly true where the dose was low, and the probability of developing a disease is low. For such persons, the most appropriate remedy may be the opportunity to receive an independent, expert medical evaluation to determine what, if any, risks they face. Such a one-time evaluation would be based on an assessment of the experiment, which includes all available information on the experiment, most importantly the dose, along with a medical examination. This type of remedy will often serve to alleviate fears that persons may have by assuring them that the chance of disease is minimal, if it exists at all. It could also include some form of counseling for emotional distress and anxiety. This remedy is not without its risks. Many of the problems associated with medical monitoring could also apply to medical assessments -- e.g., false positives and negatives, diagnosis of unrelated, non-radiogenic disease, etc. (See discussion below.)  Medical Monitoring. Medical monitoring may be an appropriate remedy for persons who have been exposed to doses for which there is a reasonable probability of developing a radiogenic disorder. This remedy would include period "check-ups" in order to detect disease at an early stage for more effective treatment, and should also include health insurance in the event that the subject does develop a disease. (The Committee's Charter explicitly charges the Committee to consider whether "medical follow-up" is "required to protect the health of individuals who were subjects of a human radiation experiment.") The Committee recognizes that there are potential hazards associated with medical monitoring (as well as medical assessment). For example, there are only a limited number of conditions for which there is scientific evidence that early diagnosis prolongs life -- e.g., cervical and breast cancer, but not lung cancer.2 [Although screen-detected cases do tend to live longer than clinically-diagnosed cases, this apparent improvement is often the result of "lead-time" bias (advancing the date of diagnosis rather than delaying the date of death) and "length" bias (the tendency for screening to preferentially detect the less aggressive tumors). ] If early detection does not in fact prolong life or contribute to one's quality of life, then its value may be outweighed by other problems. For example, it can increase the personal anxiety of the patient if they are told at an early stage that they are likely 5 to develop a disease for which there is no cure. In addition, such information could result in burdens imposed by others, such as stigmatization by family and friends. Medical and life insurance may be cancelled or premiums raised to allow for a pre-existing condition. It may be difficult to get or keep a job because the employer is concerned about their health insurance or workers compensation premiums. On the other hand, people might welcome the opportunity to plan their life in accordance with their medical condition and might prefer that they be given the choice to know, regardless of the consequences. Medical monitoring could also produce false negatives, which can have serious ramifications. Monitoring must be recurrent if it is to be effective. However, there are great debates about the optimal frequency of screening. A single cross-sectional survey of an exposed population is unlikely to be of much use, either scientifically or for the benefit of the individuals. Parallel to the problem of false negatives is the problem of false positives, which any screening method can also be expected to produce. Eventually, these cases should be weeded out, but usually not without further unnecessary and possibly risky tests, diagnostic radiation, biopsies, or other procedures, and not without considerable anxiety to the patients. The second problem is that in any population, some diseases are bound to be detected, which may have nothing to do with exposure to radiation. This might argue for a tightly focused screen for diseases that have a plausible causal connection with exposure, but even there, the radiogenic cases may be only a small proportion of those detected. The inclusion of diseases in a medical monitoring program is likely to be seen as conferring legitimacy to the public's belief that their conditions are causally related to their exposures. Certainly, most exposed populations tend to ascribe a wide range of conditions to their exposure, without recognizing the high frequency of many of these conditions in the general population. Poorly designed epidemiologic studies can feed this perception by overestimating the true prevalence of various subjective complaints, some of which may not represent any real pathology.  Monetary Compensation. Monetary compensation, alone or as an adjunct to other remedies, is appropriate usually for the most egregious harms, or when the government's conduct is most culpable. In the legal context, such compensation generally occurs in cases involving death, bodily injury, or the loss of earning potential. In addition, punitive awards can serve as a form of retribution. Monetary compensation can be disbursed in a lump sum payment or through regularly scheduled (disability-type) payments.  Official Apology. In certain situations, experimental subjects (or their survivors) may be entitled to official recognition of their involvement, and sacrifice, in the experiment. Whether or not some other form of remedy has already been or should now be provided, it may be vitally appropriate that the government formally apologize to the subjects or provide some other significant symbol of national regret. 6 Guideline 3:Remedial Measures Must be Based on an Analysis of the Government's Culpability and the Nature of the Harm to the Individual. Consideration of possible remedial measures for subjects of human radiation experiments should be based on a number of key factors. The two foremost are the degree of government culpability (Guideline 4) and the nature of the harm to the individual (Guideline 5). Different combinations of these two elements may result in different forms of redress. The interrelationship between the two can be illustrated in the form of a grid, with a vertical axis representing increasing degrees of government culpability and a horizontal axis representing various types of harm to the individual. (See proposed grid at Attachment A.) Several additional factors must be considered in order to properly determine appropriate remedies. The first is causation, of which there must be some element connecting the government's conduct to the resulting harm: i.e., in the case of all harms except lack of adequate consent, there must be a recognized causal relationship (even if presumptively established) between the radiation exposure and the resulting injury (Guideline 6). Another factor is informed consent. While a potential harm in itself, whether or not adequate consent was obtained will often inform upon the nature and degree of government culpability (Guideline 4). Similarly, the process by which subjects were selected for a particular experiment may also be a factor in assessing government culpability; this issue will generally arise in connection with the consent issue (Guideline 4). Guideline 4:Culpability Varies on a Scale from None to Knowing. The Committee may recognize five degrees of government culpability -- knowing, reckless, negligent, strict liability, and none. Although these categories are rooted in legal terminology, they are not intended to have an exclusively legal meaning.3 [See section 2.02 of the Model Penal Code for descriptions of the various terms employed here; see generally Paul H. Robinson and Jane A. Grall, "Element Analysis in Defining Criminal Liability: The Model Penal Code and Beyond," 35 Stanford Law Review 681 (1983). Note however, that in the civil law context, the distinctions between negligence, recklessness, and inadvertence are themselves unclear, and appear to refer more to quality than kind. The distinction seems to have been generally rejected under the common law due to "the extreme difficulty in of classification [and] the almost complete impossibility of drawing any satisfactory lines of demarcation," although it is still used in certain statutes. William L. Prosser, Law of Torts 182 (1971). ] Rather, they are an attempt to delineate gradations of culpability through their common sense meanings. As noted above, final determinations regarding culpability will depend on the Committee's consideration of "retrospective moral judgments" -- e.g., the Committee could decide that conduct that it considers to be in accordance with standard practices of its time should be presumed not to be highly culpable, and thus probably not warranting remedies involving monetary compensation. (This issue will be discussed in an accompanying memo linking remedies and ethics.) 7 Assessing culpability will generally require analyzing the various components of the government's conduct. In particular, the extent to which the government adequately informed the subject and obtained his or her consent may be a significant factor -- e.g., did the government negligently, or knowingly, fail to get consent? Or, if consent was properly obtained, does it mitigate the government's liability, especially if it contains a liability waiver? Moreover, the government's expectations and obligations regarding consent may differ depending on the nature of the experiment -- i.e., a one-time intentional releases over a large population raises different issues than in a biomedical experiment. Existence or absence of consent can also factor into how one judges the severity of other harms -- e.g., injury to someone who gave consent may result in a different remedy than if the same injury occurred in the absence of consent.) An additional component of the government's conduct is the extent to which abuses in the subject selection process may be indicative of culpability. Selection of certain subject populations for particular experiments -- e.g., mentally retarded students, prisoners, indigent hospital patients -- raises the issue of whether these groups were deliberately exploited by the government because of their perceived vulnerabilities. For example, an exploited group may be less cognizant of the informed consent process and therefore less able to respond in a meaningful manner. The Committee must determine whether this issue constitutes a form of remediable harm that should be included in the current calculus, or whether the issue should be given separate consideration.  Knowing. "Knowing" applies to situations in which the government acts with full knowledge that its action is wrong and will almost certainly cause damage, but it nonetheless believes that such action is required. An example might be if the government deliberately administered a harmful dose of a toxin to someone for the sole purpose of seeing what biological effect it would have, without telling the subject what it was doing or why it was doing so. (The same example could be considered reckless if the dose was not considered to be harmful.)  Reckless. Recklessness applies in situations in which the government is aware that its conduct might be wrong and might cause injury, but it nonetheless acts without taking necessary precautions. Whereas knowing means "wilful" conduct, reckless means "careless" conduct. However, for both knowing and reckless, the government is aware of the harmful consequences that may result, and is therefore presumably both blameworthy and deterrable for its actions. An example of reckless conduct might be where a government agency continues to fund a research protocol unchanged after learning that subjects were suffering unanticipated injuries.  Negligent. The government can be considered negligent when it fails to recognize injurious conduct that it should have recognized. The Model Penal Code, for example, states that negligence applies when the actor's failure to recognize a risk "involves a gross deviation from the standard of care that a reasonable person would observe in the actor's situation." An example might be where the government funded a research program without thoroughly 8 reviewing available information that would have shown injurious deficiencies in the design of the program.  Strict Liability. Finally, it may be appropriate to hold the government responsible even when it has not done anything wrong. Even if the government took every precaution and followed all established requirements, a judgment could be made that the subjects should nonetheless be compensated in light of the harm suffered, particularly if the government's conduct would not meet current standards. (Such a no-fault approach to injury has been advocated in the biomedical community for treating research subjects who are injured in the course of medical research.) Guideline 5:The Types of Harms that could result from Human Radiation Experiments Range from None to Death. The harms associated with radiation experiments fall into three general categories. The first involve physical injuries, such as acute injury from which there is full recovery (including radiation sickness -- nausea, vomiting, anorexia, infection -- skin burns), serious disease that does not result in death (including flash blindness, cataracts, mental radiation from in-utero exposure, and various non-lethal cancers), and premature death (fatal cancer or infection). The second involves emotional distress that may result when a person believes that he or she is at an increased risk of developing a disease or passing on a genetically transmissible injury to his or her offspring from a radiation experiment. The third are so-called "dignitary" injuries, which involve the violation of a person's autonomy by failing to obtain consent or to inform the person adequately of the risks. (This category includes the outcome where the only "harm" is the wrong of failing to obtain consent the absence of any material harm or risk of harm.) These harms are presented as general categories. Their location on the axis is not necessarily an indication of rank order; nor does physical injury necessarily mean greater harm. Rather, individual harm must be assessed on a continuum where there are no clear distinctions or there are several different harms of varying severity. Guideline 6:Determination of Harm will Require some Consideration of Causation; Presumptive Causation will be Appropriate for Certain Types of Experiments. Perhaps the most difficult aspect of determining injury, particularly for latent illnesses, is establishing a causal link between the exposure to radiation and the subsequent disease. This link must be considered at both the population level and at the individual level. At the population level, there must be at least some acceptable scientific evidence or biological plausibility that the disease could have been caused by the exposure. For many diseases, there is controversy as to the strength of this evidence, and this uncertainty needs to be factored into the recommendations for remedies. If a causal association is not recognized at the population 10 level, then there is no need to proceed with consideration of individuals' claims for that disease. At the individual level, criteria are needed for judging whether it is likely that the disease was caused by the claimant's exposure rather than by other factors. In civil litigation, this judgment is traditionally made on the "balance of probabilities," i.e., a probability of causation (PC) of greater than 50% (but most legislative remedies have adopted a lower threshold). Others, including the Presidential Commission on Catastrophic Nuclear Accidents, have advocated that the magnitude of the award be proportional to the PC (perhaps with some minimal threshold to discourage frivolous claims and minimize administrative costs). One of the key factors entering into the PC calculation is the individual's radiation dose. (Other factors vary from disease to disease, but might include age, latency, and other exposures.) The requirement to provide a "dose reconstruction" has proved to be cumbersome at best and a serious barrier to many legitimate claims at worst. Determining the dose can be a difficult task, especially when the records of the original exposure either were not maintained or no longer exist. Accordingly, the Committee recommends that remedies not require dose reconstruction to prove the cause of the disease. Rather, for certain types of experiments, causation should be presumed based on what is known about the conditions of the experiment as a whole and epidemiologic evidence about the likely effects of that exposure. Thus, judgements about causality would be rolled into the criteria for evaluating outcomes in relation to particular experiments. This is not to say, however, that any degree of exposure would constitute an acceptable presumption. For example, everyone in the world was exposed to fallout from nuclear weapons testing and is at slightly elevated risk of cancer as a result. However, the excess risk for most people is minuscule in comparison with the background risk from other causes. Thus, the Radiation Exposure Compensation Act recognizes only those persons who have certain minimal lengths of residence at particular times in particular downwind counties as having been presumed to have a sufficiently elevated risk to merit compensation. (The accompanying table at Attachment B attempts to illustrate the causality issue.) Guideline 7:The Government's Relationship with a Contracting Party May Affect Government Culpability. Government culpability may depend on the interrelationship between the government and the contracting party actually conducting the experiment. In many cases, the government was the sponsoring entity, while an independent contractor performed the experiments. In general, the government is responsible for the actions of its agents (to the extent that the agent is acting on the government's behalf), such that the government is culpable to the same extent that the agent is culpable. 12 However, with respect to past actions, the Committee may want to consider whether government culpability could be mitigated in situations where an agent has acted in a way that the government could not have reasonably expected and protected against. If, for example, the government required that its contractor obtain informed consent and take all reasonable precautions, but the contractor nonetheless violated these requirements and injured a research subject, the government might not be culpable to the same degree as the agent, but might be culpable to a lesser degree. The Committee may also want to consider the reverse situation, where the agent is not deemed culpable because he or she acted with the standards of the time, but the government is still held accountable because the conduct is deemed to violate current standards (e.g., strict liability). Guideline 8:Absent Evidence of Lost Earnings or Premature Death, Remedies will generally Inure only to the Subject, and not to the Relatives or Estate of Deceased. The Committee must consider whether the remedies it proposes will be available only to experimental subjects or whether they should pass on to their relatives if the subjects are dead. Survivorship is generally not an issue when the remedy itself involves any form of medical support, which can only be of use to persons who may themselves be injured or at risk (except in the case of in-utero or genetically transmissible diseases). However, more difficult questions arise as to whether monetary remedies should survive the death of an experimental subject. Such remedies may survive where the experiment itself caused the death or lost earning potential of a subject and the government was culpable, which is a common measure in tort law for such damages. Where the experiment did not have these effects, but the subject has since died, monetary compensation would generally not seem to be indicated, except perhaps in situations where the government sought to keep experiments secret precisely in order to avoid legal liability.[ Another option might be to distribute awards for these types of harms to relatives of the entire class of subjects or some other appropriate repository. ] Official apologies could apply to relatives as well as subjects, and would most likely apply to dignitary harms -- i.e., lack of informed consent or risk of latent disease. Guideline 9:Measures Must be Taken to Insure that Remedies Proposed by Government Policymakers are Properly Implemented. As noted above, the purpose of this chapter is to provide guidelines that the government policymakers in Congress and the Executive branch can use to determine what, if any, remedies should be available to the various experimental subjects covered by the Committee's Charter. The Committee believes that the government must ensure that any remedy it offers is properly and effectively implemented, so that eligible recipients are not unduly burdened in obtaining that to which they are 13 entitled. Although the Committee will not make definitive findings regarding how the government should implement the proposed remedies, it will review potential methods of effective implementation. To do so, the Committee will examine current compensation schemes for both radiation, non-radiation, and research-related injuries, and may make findings regarding which schemes serve as the best models for effective administration of the Committee's proposed remedies. 14