Report to the Congress from the Presidential Commission on Catastrophic Nuclear Accidents

Standards and Procedures for Latent Illnesses

I.   Introduction

    This chapter discusses the Commission's recommendations for establishing standards and procedures for deciding and paying claims for latent illnesses caused by a catastrophic nuclear accident.1 For the most part, the latent illnesses discussed in this chapter are claims for cancer induction and benign thyroid nodules alleged to have been caused by exposure to radiation as a result of a nuclear incident, but delayed in manifestation for some years or decades.2 The focus is on claims by the offsite public.3

    The principal problem that the Commission found with respect to properly compensating such claims under present law is the difficulty of proving or defending such claims on the issue of causation in fact. The Commission sought to minimize that problem in the context of the catastrophic nuclear accident and the profusion of claims to which such an accident would give rise. Consequently, the Commission recommends that the best available proxy for direct proof of such causation (which is not presently possible for latent illnesses that might have been induced by doses in the 10 to 100 rad range) be utilized. At present, the best available proxy for this purpose appears to be “probability of causation” (PC), that is, the attribution of group risk to group members based on relative strength of association between radiation dose and disease induction.

II.   Summary, Conclusions and Recommendations

    The central problem with claims for latent cancer induction by radiation exposure arising from a nuclear accident is proof of causation in fact. Cancers that could be radiogenic also occur spontaneously or due to exposure to other carcinogens. There is no way to distinguish directly (as opposed to inferentially) among possible causes. A related problem is reconstructing the dose actually received long after the event. Under the Commission's recommendations, such issues would be essentially eliminated with respect to claims for medical monitoring (cancer and thyroid nodule screening) in that such screening would be provided without cost based on increased risk at a given estimated dose or above. Difficulties in reconstructing dose received would be ameliorated by the use of state-of-the-art dose reconstruction techniques (chromosomal aberrations and external thyroid scans). Causation issues would be minimized with respect to diagnosed cancers by adopting the best techniques then available, which today would be PC and proportionate recovery (at least for the purpose of offering settlements) and the payment of damages for cancers that meet the corresponding eligibility criteria (as recommended by the court-appointed PSA and adopted by the court), all in accordance with the recommendations set forth in Chapter 2 and below.

    The best technique currently available for resolving causation in fact issues appears to be the so-called4 PC approach. By PC, the Commission means the assignment of group risk to individual cancers that actually appear. Better techniques may be developed in the future; the Commission assumes that the PSA will avail itself of the best information and methodology available at the time. Nonetheless, to process the claims from a major nuclear accident, some mechanism will be needed for attributing causation based on readily computed factors such as level of exposure, age, sex, the type of cancer, and the latency period. Hence, in the absence of a better method, the Commission has employed PC methodology in the examples provided in Appendix H and suggests how awards might correspond to dose received. Similarly, the Commission uses the estimated increase in group risk at lower dose levels to determine eligibility for medical monitoring without charge.

    Proportionate recovery would allow more people to be paid than strict adherence to traditional levels of proof It would involve paying a reduced award for cases that do not meet the eligibility criteria for a full award (as more fully explained in the examples discussed in Appendix H) but still exceed some minimum threshold. The advantage of proportionate recovery for cancers that have actually appeared is that it can, depending on the choices made by Congress and the courts, avoid any shortfall in the aggregate amount actually paid for cancer claims (which, in the absence of some method such as PC, might arise because of the difficulty of proving causation) as compared to the estimated value of excess cancers in the affected population.5

    Increased risk of incurring disease as a basis for payment of claimants shortly after the event rather than if and when a disease develops is not a concept that the Commission finds, on balance, appealing, as more fully explained in Chapter 3. Paying on the basis of increased risk implies preclusion of future claims if the disease develops. It also produces relatively small awards because they are discounted by present worth techniques and may involve very small values of increased risk. The basic reason for this conclusion is the “windfall/shortfall” problem: those that never get cancer receive a windfall, those who do develop cancer receive much less than their costs.

    The Commission is aware of a proposal that claimants be permitted to elect a small award now in lieu of a chance at full recovery later, on the theory that they would either purchase “cancer insurance” or self-insure. This proposal could have a chance of becoming workable, in the Commission's view, only if a schedule of cancer insurance policy limits corresponding to significant levels of exposure were adopted, if choosing the cancer insurance route precluded a future claim, and if the claimants were provided such policies rather than cash. Increased risk of incurring disease is more logically used as an eligibility criterion for benefits that are provided in kind. Thus, as already noted, the Commission has adopted the increased risk approach in the examples given in Appendix H but only with respect to eligibility for free medical monitoring. The increased risk also qualifies the individual for emotional distress counseling, as explained in Chapter 3.

    The Commission would stress that, following a nuclear accident, offsite surveys and calculations should be performed to quantify whole body and organ doses to the public. All relevant pathways should be surveyed and/or calculated. These include, but are not limited to, direct radiation shine, plume immersion, ground deposition and internal dose through ingestion and breathing.

    Information gathered from or about individuals who may have had significant radiation doses to any organ (including whole body) should include:

    1. Blood sample within 1 month of exposure.

    2. Thyroid analysis within 1 month of exposure. (A simple geiger counter placed on the neck of the subject over the thyroid gland is sufficient.)

    3. Signed statement of location during the accident.

    Individuals shall receive appropriate compensation for their time providing this information.

    In summary, therefore, the Commission recommends that a system be established that pays the full award where PC methodology indicates that it is more likely than not that a particular illness occurred as a result of the accident. Further, it is recommended that a level be established on the other end of the scale where it is extremely unlikely that an illness results from the accident, for which no payment is made. For those whose probability falls between the above two limits, there should be some level of proportional recovery related to the likelihood of causation. However, claimants who wish to rely on traditional notions of tort causation may submit their cases to binding arbitration or adjudication.

    While the Commission recommends this system for paying latent injury claims, it also recognizes that in this situation, Congress may feel it necessary to pay for any cancer or other illness following a major nuclear accident because exposed persons will believe their cancer was caused by the event whether, in fact, it was or was not

III.   Radiation Exposure and Latent Health Effects

    It is clear from scientific observations that exposure to ionizing radiation at high enough levels can result in health consequences that manifest themselves later in life. The latency period can be from a few years to several decades. The dominant latent consequence is cancer, including leukemia and tumors at specific body sites, e.g., lung, thyroid, etc. Because they so dominate the latent health risks, cancer morbidity and mortality will be the main focus of the discussion that follows.6

    The cancers caused by a radiation exposure will show up as an increase in the normal occurrence rate of various cancers and they will not be distinguishable from those of other causes. The normal occurrence rate of cancer mortalities in the U.S. was about 193 deaths per 100,000 persons per year in 1985.7 Cancer accounted for 22 percent of the deaths in that year. It has been estimated that the rate of cancer fatalities might increase by about 10 percent in the population exposed by a very severe nuclear accident.8 This increase might begin about 5 years following the accident, and persist for about 30 years9

    The estimates for the radiation caused cancers are obtained by calculating the estimated dose to each member of the public and summing over the relevant population. This gives a population dose in units of person-Gray (pGy) or person-rad (prad).10 The expected number of cancers is obtained by dividing the population dose by the number of person-rad on the average required to produce a cancer11 As suggested in the preceding footnote, because of the use of a more or less linear dose-response hypothesis, a significant fraction of the calculated cancers are assumed to result from doses of less than .1 Gy (10 rad) to the populations at large distances. Recent studies suggest that 10 rads increase the risk of cancer to the average population by about 3.7 percent.12 Since the dose response in this range of exposure is taken to be roughly linear, this means that a dose of .01 Gy (I rad) would result in an increase of cancer risk of less than one-half of 1 percent in the exposed population.13 Likewise, a dose of 1 Gy (100 rads) would increase the cancer risk by about 37 percent.

    A second group of latent effects of considerable concern to the public are genetic effects. These effects, of course, manifest themselves in future generations of the exposed persons. Such effects have been observed in animal populations, including the fruit fly (drosophila) and mice. However, careful studies of the Japanese atomic bomb survivors have shown no statistically significant increase in observed genetic effects in the descendants of that exposed group.14 This leads to the conclusion that the human population is significantly less sensitive to radiation caused mutations than was inferred from studies of mice. The data are consistent with man's sensitivity being as much as a factor of about 3 to 10 less than that of the mouse.15 Thus, the dose required to double the natural incidence rate of genetic defects (the “doubling dose”), which is on the order of 1 Gy (100 rads) for mice, maybe about 3 Gy (300 rads) for humans. However, because of the considerable uncertainty in the data, the BEIR V report conservatively concludes only that “the value of 100 rad represents an approximate lower 95 percent confidence limit for the human doubling dose.”16 In any event, then, genetic effects seem to be quite small. For example, although the approximately 40,000 Japanese atomic bomb survivors who received more than 0.005 Gy (0.5 rads) showed an excess of cancer mortalities of about 1,700 over the about 13,000 expected, this same group has shown no observable genetic effects.17

    However, a recent article in Nucleonics Week18 reports that some British health officials have characterized studies of childhood leukemias in the offspring of employees at British Nuclear Fuels Limited (BNFL) Sellafield reprocessing plant19 as the first indication in the world of a link between occupational exposure and leukemia in the next generation. In this regard, according to reports, the principal author of the studies, Professor Martin Gardner of the Medical Research Council's Environmental Epidemiological Unit at Southampton University, points out that the study does not establish cause and effect, and that further research would be needed to confirm the suggested link. In a forward to the study, the director of the Imperial Cancer Research Fund's Cancer Epidemiology Unit at Oxford was reported to have pointed out the inconsistency with the Japanese data:

    The only other relevant human data available are on the 7,400 children of the Japanese men who survived the atomic bomb explosions, and these show no hint of an increased risk of leukemia in the offspring. [Yet) the average exposure to external ionizing radiation of the Japanese men was four times higher than that of the Sellafield workers.

    The case of latent effects on the thyroid gland must be mentioned. One of the more volatile radioactive fission products is iodine-131. The thyroid gland has a great propensity for removing iodine from the blood. Thus, persons who ingest iodine-131, either from direct inhalation or by eating food contaminated with it, may be subject to significant thyroid doses, depending upon the amount ingested. Very large doses can destroy the gland's function and lead to acute effects; however, smaller doses may produce only latent effects. Most effects will occur as growths on the thyroid gland called nodules, some fraction of which (about 10 percent)20 will be malignant. The usual treatment for this malignant condition is to remove the gland surgically and prescribe medication to supply the hormone produced by the gland. The prognosis for the patient is excellent, except for about 3 percent of all nodules (about one-third of the malignant ones) where a malignancy has spread to other parts of the body before the treatment.21 Although latent thyroid disorders would be expected to produce very little life shortening in the population, they could affect substantial numbers of people, and, in severe cases, double the normal occurrence rate of this disease. In addition to the impact on the individuals affected, this would represent a significant economic obligation for health care to the exposed population. The normal incidence rate of thyroid cancer among the U.S. population is about 80 cases per 100,000 persons in the population per year. Calculations suggest that, following a very severe nuclear accident, the incidence rate among the exposed population may double.22

IV.   Legal Problems

    A.   In General.

    The Commission considered the traditional problems with processing tort claims for latent illnesses, whether cancers allegedly caused by radiation exposure or cancers (or some other latent illness) caused by another toxic agent. These problems are discussed in Appendix D. Suffice it to say here that (1) the issue of whether claims are barred by a statute of limitations that runs from the event of exposure to radiation was resolved for substantial nuclear accidents in the 1988 amendments to Price-Anderson, (2) the problem of whether to make payments shortly after the accident because of the possibility of latent illnesses emerging later (in the form of increased risk or emotional distress claims) has been addressed in Chapter 3, and (3) issues of when to assemble funds are less pressing now that the 1988 amendments have greatly reduced the prospect that the court having jurisdiction will be administering only a limited fund.

    B.   The Causation-in-Fact Issue and Proposed Solutions.

    The most difficult problem that the Commission faced was that of dealing with the issue of causation of latent illnesses, including cancers. As was pointed out above, the central difficulty with radiogenic cancer claims is that many cancers occur in the absence of exposures to ionizing radiation in excess of that from natural background and medical treatment and therapy. Even when excess cancers are estimated to increase in the exposed population as a result of a nuclear accident, it is impossible to distinguish the excess, accident-related cancers from the spontaneous ones. In a very real sense, one does not know whom to pay. The Commission, therefore, sought to identify the “next best” approach, since attaining the “best” solution, compensating only those whose cancers or other latent illnesses were caused by the accident, is not currently possible.

    If the option of paying for increased risk of getting cancer (which suffers from the windfall/shortfall vice) is left aside, the remaining choices include: Option A, relaxing traditional notions of proof of causation and paying something to everyone who gets cancer; Option B, retaining and rigorously applying traditional standards, which would result in paying few, if any, claims; and Option C, adopting some proxy for direct proof of causation, such as imputing group risk to individuals who actually develop cancer and paying those claims where the association between radiation exposure and a particular cancer is the strongest (or at least at some minimum level), with the option, where a strong association is required for a “full” award, of also paying lesser amounts on those claims with a somewhat weaker association.

    Sub-options to Options A and C also exist; some or all benefits could be scheduled, based on a survey of actual jury awards in the area for similar claims, with or without discounting to account for increased certainty of recovery; certain relief, such as medical monitoring or treatment could also be scheduled, with the schedule in that instance perhaps being based on a survey of actual medical costs in the area; and some remedies, such as cancer or thyroid nodule screening, could be provided in kind (and thus at “wholesale” costs), on the theory that a given level of funding could thus provide greater benefits for more people.

    Given that about one-third of the U.S. population get some form of cancer during their lifetimes, Option A (paying for all cancers) would be very expensive unless individual benefits were capped at a level that would come nowhere near paying the costs of medical treatment for many forms of cancer. In light of the Congressional experience with black lung compensation, such a well-intentioned but overly inclusive approach to resolving the causation problem may not be acceptable. Moreover, Option A would do violence to the policy of Price-Anderson (dating from the 1975 amendments), which has been to internalize the costs of nuclear power through advance and retrospective premiums payable by the industry. That is, the internalization goal would be furthered by a compensation system that would require nuclear power licensees and their ratepayers to pay the costs of those cancers that it caused, but it would be violated if costs were imposed on them for treatment of cancers that would have occurred even if the postulated accident had not.

    Option B (traditional causation) would not improve the present situation and would not be responsive to the charge given this Commission by the Congress. Under traditional burden of proof requirements, the claimant would have to demonstrate causation by a preponderance of the evidence. To the extent that expert opinion evidence was adduced to satisfy this burden, the opinion might well have to be “to a reasonable medical certainty.” In the Commission's view, insistence upon these standards would lead to rejection of many deserving claims.

    For these reasons, the Commission recommends some version of Option C (payment based on strength of association). In the Commission's examples in Appendix H, three variations are discussed. The first would pay the full amount for any diagnosed cancer where the probability of causation (PC)23 is .5 or greater, and a declining amount down to a cutoff of PC = .2, at which compensation would be 20 percent of the full award, determined in accordance with Chapter 3.

    The second variation would pay the full amount for any diagnosed cancer where the PC is .5 or greater, and a declining amount down to a PC of .2, at which compensation would be 30 percent of a full award.

    The third variation, which is most like Option A, above, would simply pay a benefit to anyone in the affected area with a diagnosed cancer whose radiation exposure indicated a PC of 20 percent or greater. Congress might elect to make this a full award determined in accordance with Chapter 3, or a fixed dollar amount, or reimbursement for actual medical expenses.

    Under all three variations, cancer and thyroid nodule screening would be provided to those whose exposure was such as to indicate a significantly increased risk of those illnesses, such as 10 percent.

    C.   Alternative Approaches Considered.

    Dissatisfaction with traditional causation in fact analysis and with the wide variation in results reached by the courts has led to some conventions being proposed to narrow the problem. Such conventions can be used to screen claims, rejecting the most tenuous-ones at the threshold, or as a guide for the parties' experts or court-appointed experts, or as a burden-shifting (or even binding) norm to decide claims.24

    1. Probability of Causation

      One such convention is attribution of causation (often, but somewhat misleadingly, called probability of causation) based on standardized characteristics, such as age at time of exposure, personal habits such as smoking cigarettes, level of exposure, type of cancer, latency period, and age at manifestation of disease.25 This so-called probability of causation approach can be based on tables prepared by the National Institutes of Health (NIH) of the Department of Health and Human Services at the direction of the Congress.26 The PC tables were prepared under a directive contained in unrelated legislation,27 with the apparent intent that subsequent legislation would be enacted mandating their use in certain cases involving radiation-exposed veterans and perhaps others, such as “downwinders” — residents of areas downwind of atmospheric nuclear explosive device testing — or uranium miners.28 A critique of the PC approach has been presented to the Commission29 and was made part of the record of Congressional hearings on earlier versions of the bills that led to the 1988 amendments.30 However, a more receptive view of the usefulness of PC methods was also provided by a former critic .31

      As noted at the outset of this discussion, there are a number of ways to use PC tables. They can be used to screen claims, not necessarily to make awards. That is, they can be used to dismiss claims at the threshold that do not have a minimum PC value. Such claims as survive summary dismissal can be decided along traditional, “battle-of-the experts” lines. Alternatively, the PC values can be used not just to screen claims at the threshold, but employed by experts in forming their opinions. PC could be used as a basis for settlement offers or Claim Master awards. PC can also be used to create rebuttable presumptions of causation, or even irrebuttable ones. However, unless the probability is reduced below a preponderance of the evidence (i.e., 50 percent) the use of the tables as dispositive would mean essentially no awards for low-level exposures and common cancers, because most of the PC values are well below 50 percent even for exposures in the tens of rads range.

    2. Proportionate Recovery

      Another response to the individual causation problem that has been proposed is proportionate recovery. That is, where proof falls short of a preponderance of the evidence, but the increased risk brought about by exposure to a particular toxin exceeds some threshold value (e.g., the toxin increased the inherent risk by 50 percent), recovery is allowed in proportion to the increased risk. This approach is often associated with the probability of causation concept, as in the case of the BNFL scheme discussed later, but could be applied directly, based on prevailing expert opinion in a given case.

    3. Increased Risk of Incurring Disease

      As noted earlier, another way to approach the problem is to deal with it shortly after exposure, rather than waiting for the disease to emerge, if ever it does, and to pay everyone exposed above some trigger level of radiation exposure for the increased risk of morbidity and mortality as a result of the exposure, taking into account the increased risk and all of the other relevant factors in calculating awards, such as past and projected earning power, life expectancy, medical monitoring and care, and so on.

      Such an approach, assuming that partial recovery now forecloses claims for the balance later if the disease does emerge. has the virtues of finality, of dealing with the potential latent claims in the same time frame as the current claims, of permitting the latent claims class to access the fund on the same footing as the current claims, and of resolving factual questions while the evidence is still fresh. The increased risk approach did not commend itself to the Commission because it would lead to (a) windfalls to those who never incur disease (because of the tendency of triers of fact to be generous to victims or potential victims and because of the conservatism thought to inhere in population risk estimators), (b) failure to cover the full costs of those that do incur disease, and (c) imposition of retroactive liability on insurers.32 A novel approach that was brought to the Commission's attention, and briefly referred to above (in Chapter 3), has been suggested by Professor Frank B. Cross of the University of Texas.33 Briefly, it involves a “cancer insurance” approach to proportionate recovery. In cases where there has been some unlawful release of toxic pollutants, the author argues for proportionate recovery and an administrative small claims process to provide (a) potency standards (i.e., how much does a given level of exposure add to the risk of disease) that would function as an eligibility threshold, (b) a schedule of benefits that would equate to an insurance premium for the cost (apparently medical costs and lost earnings, at least) of a future cancer discounted by the risk factor and present-worth techniques, and (c) claims processing, which might be an elective alternative (i.e., two-tier system) to a tort claim in court. Cross leaves open the possibility that successful claimants could also recover for emotional distress, medical surveillance and, in appropriate cases, punitive damages.

      Cross acknowledges but discounts the “windfall-shortfall” criticism of compensation for increased risk.34 Under his scheme, payment for increased risk now will preclude a future claim if and when the cancer arises. The proposal would require the system administrators to be coolly objective and not inflate either risk or benefits, while implying that the benefits might be fixed (as in classic workers' compensation) at relatively modest levels in exchange for the certainty of recovery. Cross assumes that defendants will not be held liable unless they violate applicable safety standards, though experience suggests that fact finders have been instructed that they can ignore compliance with such standards in determining liability.35

    D.  Solutions to the Causation Problem Adopted Elsewhere.

    During its deliberations, the Commission considered various approaches to compensating radiogenic cancer claims that have been adopted in other contexts.

    1. Atomic Veterans

      On May 20, 1988, the Radiation-Exposed Veterans Compensation Act of 1988 was approved by the President.36 Subsection 312(c)(1), added by the Veteran's Act, creates a presumption that certain diseases37 that become manifest in a radiation exposed veteran within a specified presumption period shall be considered to have been incurred in or aggravated during the veteran's service on active duty, notwithstanding that there was no record evidence of the disease during active service, thus making the veteran or his or her survivors eligible for certain benefits.

      The legislative history of the Veteran's Act indicates considerable controversy as to whether scientific evidence supported the proposed legislation. It was argued by the Veterans' Administration, as discussed in Appendix E, that the bill would pay over 32,000 claims in order to ensure payment of about 10 valid (but indistinguishable) ones. In signing the bill (H.R. 1811), the President pointed out that:

        Enactment of this legislation does not represent a judgment that service-related radiation exposure of veterans covered by the Act in fact caused any disease, nor does it represent endorsement of a principle of permitting veterans to receive benefits funded through veterans programs which bear no relationship to their former military service.

        Instead, the Act gives due recognition for the unusual service rendered by Americans who participated in military activities involving exposure to radiation generated by the detonation of atomic explosives. The Nation is grateful for their special service, and enactment of H.R. 1811 makes clear the Nation's continuing concern for their welfare.38

    2. Ontario Industrial Disease Standard Panel

      In 1984, the Province of Ontario, Canada, created a statutory Industrial Disease Standards Panel to advise the provincial Worker's Compensation Board on industrial disease matters, with particular reference to issues of causation.39

      One feature of the Ontario system that is worth emphasizing here is that the statutory standards panel is politically appointed and diverse in composition, giving it certain perceived advantages over a pure science panel yet it retains the ability to make use of the consensus scientific view of a separate specialist panel in arriving at -its recommendations, and has always done so to date.

      A fuller discussion of this approach is found in Appendix F. The Commission's recommendations draw upon some of the features of the Ontario system including the use of scientific panels.

    3. British Nuclear Fuels /UKAEA

      In the United Kingdom, British Nuclear Fuels plc (formerly BNFL) and the United Kingdom Atomic Energy Authority (UKAEA) have entered into a Compensation Agreement with their respective unions covering cancer mortality and morbidity among radiation workers. The agreement provides a voluntary alternative to litigation for determining whether the disease or death should be deemed to have been linked to employment related radiation exposure. The parties have agreed to use a PC approach, under which the essential facts are stipulated and the causation issue can be determined by computation from PC tables or by expert panels using the PC tables as a guide. When there is a sufficient likelihood that a cancer or cancer death should be attributed to radiation exposure in the course of employment, partial or full compensation is paid, depending upon the degree of likelihood (an example of proportionate recovery).

      About 1,200 cancer deaths among radiation workers employed by BNFL and its predecessors were known at the end of 1987; by the end of 1988, nearly all of them had been processed under the agreement and awards made in 17 cases. The range of awards in these 17 cases was from about $30,000 to about $200,000.40

      A fuller discussion is found in Appendix 0. The Commission drew upon this approach in its recommendations and in the discussion of exemplary cases in Appendix H.

    4. Medical Monitoring

      It seems likely that, following any nuclear power plant accident that might involve substantial releases of radiation (orders of magnitude in excess of normal operating releases), claims for medical monitoring of those who might have been substantially exposed will be made. Given the Three Mile Island Unit 2 accident experience, which did not involve major radiation releases, even lesser accidents will give rise to such claims. These are not claims for injury or even for the increased likelihood of future disease; rather, they are claims for reimbursement for or provision of medical services that a reasonably prudent individual would incur at intervals over a number of years given the fact of the radiation exposure, and the expenses involved are ones that would not have been incurred but for the radiation exposure. Accordingly, one of the basics of post-accident claims management would seem to be the creation of a fund (or prepaid medical monitoring service) to serve this function;41 the reasonable costs of such services should be considered part of the legal liability for a nuclear accident, and should be (and are understood to be) covered by required nuclear liability insurance. The threshold to receive medical monitoring need not be at the same level as the threshold for compensation; it could and should be somewhat lower because of the uncertainties involved.

V.   Recommended Approach for Dealing with Latent Health Effects Claims

    The major underpinnings of the Commission's recommended approach for dealing with problems of dose reconstruction and proof of causation in fact for latent health effects is discussed here and in detail in Appendix H.

    With respect to those who have developed latent health effects from radiation exposures due to the accident, the major philosophy that has been adopted is to put in place a mechanism for compensating the affected individuals quickly and efficiently after the effects emerge. Doing so requires lifetime tracking and medical monitoring of those at some significant risk of latent health effects. The goal should be to get the largest fraction of the available funds possible into the hands of the legitimate claimants by minimizing both the costs of administering the process and incentives to abuse the system for financial gain.

      A.   Key Elements of the Proposed Process.

      An estimated increase in cancers resulting from doses received in the course of the nuclear accident implies that at least some offsite areas were contaminated or persons in such areas were otherwise exposed42 at a significant level. Just as with any process that requires making distinctions, there is a need for an initial trigger point. As will be discussed a bit more fully below, there win be additional thresholds; who should receive a medical examination and have a blood sample taken, who should receive long-term medical monitoring, and who, having developed symptoms that could have been caused by the accident, should be considered for compensation.

      The proposed process depends heavily upon two things. First, it assumes that a threshold dose can be defined, below which the chance that radiation caused the health effect is so small that it can be dismissed as a cause. Second, it assumes that a method for reliably reconstructing the dose received will be available.

      1. Threshold Dose Issue

        In this chapter the Commission has not attempted to define this (health effects) threshold dose. However, the recent report of the BEIR V Committee shows that, based upon the best scientific evidence currently available, a 0.1 Gy (10 rad) dose to a typical population would cause about a 3.7 percent increase in the normal cancer mortality rate in that group. The Committee further suggested that to a good approximation, the dose response would be linear in this dose range. Thus, a dose of .01 Gy (1 rad) would cause a less than one-half of 1 percent increase (0.37 percent) and a dose of 1 Gy (100 rad) would cause an increase of about 37 percent. The cut-off for long-term medical monitoring (which might be termed the “monitoring threshold”) could lie somewhere in the 10 to 20 rad range. The monitoring threshold would be somewhat higher than the threshold for invoking the system of registration, etc., (which might be termed the “system threshold”). Because the understanding of the nature of radiation risks is constantly changing, it seems reasonable to make a decision about the magnitude of this quantity after the accident, using the best available information at that time. Since the latent effects would not begin to appear for several years or more, deferring decision in this respect would not delay payment of latent injury claims.

      2. Dose Estimation Issue

        There are several ways that a person's dose might be determined, for example, by knowing where the person was during and after the accident, by observing acute clinical effects43 following the exposure, by direct external measurement of radioactive iodine present in the thyroid;44 and by noting chromosomal aberrations following the exposure.45 This latter method seems to offer considerable potential because, in well-trained hands, it is currently capable of measuring doses as low as 0.1 to 0.2 Gy (10-20 rads).46 At higher doses (about 1 Gy [100 rad] and above), other clinical effects would be increasingly expected, which would be useful in assessing the dose. Since information on these acute effects would be required for settling acute effects claims, they would be part of the patient's medical record. Finally, from measurements of the radiation fields during the course of the accident and the location of the claimant during the accident, it should be possible to reconstruct his or her dose in many cases. For doses below about .1 Gy (10 rads), this combination of registry information and reconstruction of exposure could be used to estimate the dose. Depending upon where the threshold is set, the chromosomal aberration technique might be capable of measuring the full range of interest.

        The method has one serious problem in that the measurement of the chromosomal aberrations has not been automated — at present it can only be done by eye, using a microscope, so this measurement is time consuming and expensive. Considerable work is currently underway to develop an automated method of doing this measurement.47 Current techniques do permit processing a blood sample and preserving it so that it could be counted later, if needed. However, the number of aberrations in a person's blood do decrease with time following an exposure, so the blood sample should be obtained fairly soon (1 to 2 months) after the exposure.48

        The Commission recognizes that there may be considerable (and understandable) demand on the part of those whose blood samples are taken to have them processed immediately, so that they can learn whether they personally received a significant dose. At present, if very large numbers of samples are taken, processing all of them in a short time would be impractical from the standpoint of personnel resources since, as described above, the process for reviewing such samples is not automated. If there were an overriding need for automation of blood sample processing, this situation could change.

        It would be highly desirable that a blood sample be taken for dose reconstruction even if the individual has no present intention of filing a claim. It would also be desirable to take samples from everyone in an even potentially affected area within a certain radius of the accident site. In any event, reconstruction of doses would be necessary to process latent illness claims so that, as noted in Chapter 2, a blood sample should be required to preserve a claim, subject to exceptions for good cause. All reasonable measures should be taken to assure and facilitate participation, such as payment of a small stipend and travel allowance, comparable to what is paid in the federal district courts for jury duty.

        These samples would have to be processed and stored. This is the only special requirement for the latent health effects claims at the time of registering all, and initially providing medical examinations for at least some, of the affected population.

        This examination should also look for any obvious signs of acute radiation syndrome. It should take a blood count and measure the thyroid gland for any signs of iodine-131 that might have been ingested. The Commission recommends that a panel of competent physicians and other appropriate experts be assembled, perhaps by legislation directing NIH to develop the protocol for this physical exam. This medical examination protocol panel should be formed now and develop a written procedure, with provisions for periodic review and update, that could be made available to physicians at the scene.

      B.   Types of Claims.

      Before proceeding into the details of the proposed process, the Commission considered the types of medical claims that will be made: claims for medical monitoring, for latent cancers (including malignant thyroid nodules), for benign thyroid nodules, and for genetic effects.

      1. Latent Cancer Claims and Medical Monitoring

        Reasonable people are likely to differ considerably on how to assess the value of a cancer claim. As more is learned about the process by which radiation induces cancer, better judgements should be possible. Since there will be a latent period of some years after the accident before cancer claims are expected, there ought to be time to form the proposed expert panel. The panel could assemble the then-current information and propose a fair method for settling claims.

        The Commission felt a responsibility to illustrate how the proposed method might work given the tools for mass tort claims processing presently available, and consistent with the present level of knowledge for radiogenic cancers. This is done in Appendix H. What follows is a brief summary.

        To simplify the processing of large numbers of claims, the best technique now available seems to be PC, which might be coupled with proportionate recovery. PC essentially means attributing an increased risk in a population to members of that population.49 The first step is determining group risk. If the fatal cancers estimated to occur at a given radiation dose are divided by the total number of fatal cancers (excess and normal occurrence), the result is the “probability” (in a group average sense) that a given cancer was caused by the radiation exposure. For example, the normal occurrence in a group of 100,000 U.S. males50 is 20,510 fatal cancers. According to BEIR V,51 a dose of 2.6 Sv (260 rem) would produce 20,000 excess cancers in such a group. The PC that an “average” cancer was caused by such a dose is:

        PC =    40,510 - 20,510   
        = 0.49

        Figures in Appendix H plot the values of PC versus equivalent dose for various groups.

        “All cancer” PC values for entire populations are not very exact for assessing cancer causation in individual cases. Some cancers are more strongly associated with radiation exposure than others. Cancers that appear at a certain age or after exposure at a certain age may be more strongly associated with radiation exposure than those appearing earlier or later, or appearing after exposure of older persons. Thus, ideally, values that can be applied to individuals and particular cancers would be used, if available. The BEIR V report contains the basic information to update the relations for groups of varying age (at exposure and, in some cases, age at manifestation of disease or for certain minimum latency periods) and sex for a variety of different cancers given in the NIH Radioepidemiological Tables.52

      C.   Proposed Procedure.

      The proposed procedure begins by defining two PC values, PC(A) and PC(B). The higher of these, PC(A), defines a level above which a cancer that has actually appeared is likely to have been caused by the radiation released in the course of the accident. If PC(A) is to serve as a proxy for proof of causation, claimants at this level or above would consequently receive full compensation in accordance with Chapter 3 ($V). In the range between PC(A) and PC(B), the cause is less likely to have been the exposure to radiation, so the proposed award could be the calculated value of PC (or some other fraction corresponding to a given PC value53) for the dose received by the particular individual times the amount of fun compensation determined in accordance with Chapter 3 (all pecuniary losses including medical expenses and lost income plus a scheduled amount for pain and suffering and any other nonpecuniary loss, $V).

      Instead of applying the proportionate recovery fraction to the amount determined in accordance with the principles of Chapter 3, the Congress might elect to apply this fraction to a flat amount, a substitute value for $V. In the discussion that follows, the Commission suggests one method that Congress might consider in determining the amount of this substitute $V.

      The Commission's interpretation of the Act is that the exposed population should be compensated for its loss. Thus society should be reimbursed for the costs associated with all of the excess cancers attributed to the accident. This could be expressed by multiplying the number of excess cancers times the average cost of cancers ($Av). Cancer risk values such as those appearing in BEIR V with an analysis of population doses could be used to determine the number of excess cancers, thus defining the cost to society for the latent cancer “pool” as follows:

      (Excess cancers) x ($Av) = “pool”$.

      The Congress (or the courts) could then assign values for PC(A), PC(B) and $V that it felt would most fairly distribute these “pool” funds to the injured parties.

      The value of equivalent dose for a particular claimant would in most cases54 be determined from a blood sample taken shortly after the accident. Then the appropriate version of a PC table for a particular cancer based on the applicable age at exposure (and, in some cases, age at manifestation of disease), latency period, and sex would be used to obtain a value for PC.55

      The Commission is not aware of any logic for precisely calculating the values of PC(A) and PC(B). These are clearly matters to be decided by the Congress, which might leave some or all of them to the court under appropriate guidelines.

      The Commission intends that claims would be paid for both fatal and nonfatal cancers. However, all the data used from the BEIR V report in Appendix H are for cancer fatalities. Since some cancers are cured and hence do not lead to a cancer fatality, the number of cancers observed will be larger than the number of cancer fatalities. If, however, the fraction of radiogenic cancers cured is the same as the fraction from other causes that am cured, as is now believed,56 then the probability of causation will not be changed. For example, suppose that, on the average, one-third of cancers of all origins were cured. Then, the number of excess cancers observed in the population of interest following a dose of I Sv (100 rem) (see Table H.2 in Appendix H) would become 1.5 x 7,700 = 11,550. The number of cancers observed in the unexposed population would also increase by the nonfatal ones, i.e., by a factor of 1.5, so their number would be 1.5 x 20,5 10 = 30,765. Thus,

      PC =    42,315 - 30,765   
      = 0.27

      and is unchanged from the PC for cancer fatalities, as indicated in Table H.2 in Appendix H.

      In addition to cancer claims, the proposed system might have to deal with a large number of benign thyroid nodule claims. Medical monitoring and a PC method for processing such claims are discussed in Appendix H. PC values and the corresponding percentages of a full award are shown in Tables H.3, H.4, and H.5 of Appendix H.

      As indicated in several places in the Report, those analytical techniques most widely accepted at the time of the accident should be utilized in dealing with latent illness issues. In this report, the PC method, which currently appears to be the best technique available, was used to illustrate the manner in which a large number of latent illness claims might best be handled equitably.


      1. Act 170(1)(3)(C), 42 U.S.C. 22100)(3)(C). Although the Act uses the term “latent injuries” here and elsewhere (see Appendix C. it is clear from the legislative history that what is meant is latent disease or illness, and the latter terminology is used throughout this report. A brief explanation of how Price-Anderson and state law interact with respect to latent illness claims and a summary of what Price-Anderson already provides with respect to such claims is set forth in Appendix C.

      2. Genetic effects and birth defects are also briefly discussed. See Section III of this chapter.

      3. Tort claims for cancer induction on behalf of workers at the site of, and employed in connection with, a licensed or contract activity might also be brought; the Price-Anderson law excludes claims by such onsite workers that arise under workers' compensation laws.

      4. See infra. n. 49.

      5. This estimated value is arrived at by multiplying the estimated number of excess cancers by the chosen value of a full award.

      6. This chapter also briefly discusses a second group of potential latent effects. genetic effects. There is no separate discussion of either cancers or birth defects arising from in utero exposure of the fetus. The rate of cancer incidence for fetal exposure is a difference of sensitivity or dose response, not a difference in kind. Clinically observable birth defects would become manifest during the same time frame as “early” claims, so they are not latent in the sense of effects that arise after a hiatus of several years.

      7. The World Almanac and Book of Facts (New York., Pharos Books, 1987), p.771. About one-third of the U.S. population incurs some form of cancer at some time during their lives; about half of these are fatal. Ibid.. p.88.

      8. Nuclear Regulatory Commission, Reactor Safety Study - An Assessment of Accident Risk in U.S. Commercial Nuclear Power Plants, WASH - 1400. NUREG 75/014, 1975, p. 85 [hereinafter WASH 1400].

      9. Ibid.

      10. “Rad” is a unit of dose of ionizing radiation to body tissues in terms of the energy absorbed per unit mass of the tissue; one rad is the dose corresponding to the absorption of 100 ergs per gram of tissue. 10 CFR 20.4(b). The Gray (Gy) is the dose corresponding to the absorption of one joule per kilogram of tissue. I Gy equals 100 rad.

      11. When this generally accepted method, which is entirely appropriate for standard-setting, is used to estimate health effects in a population it should be remembered that no distinction is made between population doses composed mainly of very small doses to many people and population doses composed of larger doses to a few. In the former case, the number of excess cancers may be overestimated if there is, contrary to the basically linear assumption used by the standard-setting bodies, a safe threshold below which biological repair occurs, as it does with other insults to the body.

      12. Committee on the Biological Effects of Ionizing Radiations, Health Effects of Exposure to Low Levels of Ionizing Radiation - BEIR V (Washington. D.C.: National Academy Press, 1990), Table 4.2. p. 172 [hereinafter BEIR V Report].

      13. The increase in cancer estimates in the BEIR V Report, as compared to earlier studies that estimated lower rates of health effects, is attributable to its use of a relative risk model (whereas most earlier studies used an absolute risk model), and also to its use of revised dosimetry for the Japanese bombing victims. The absolute risk model assumes that a given dose adds an amount to the underlying risk that is independent of the magnitude of the underlying risk. In the relative risk model, one obtains the increase in risk by multiplying the underlying risk by appropriate factors. Thus, in the relative risk model, the increase in risk is proportional to the underlying risk, while in the absolute risk model, the exposure increases the risk by an additive factor that is independent of the underlying risk.

      14. BEIR V Report. P.65.

      15. Ibid.

      16. BEIR V Report, P. 125.

      17. BEIR V Report, P. 182.

      18. Marshall, “Children's Leukemia Link Posited for Fathers Working at Sellafield,” Nucleonics Week, XXXI, No. 8 (Feb. 22. 1990). p. 1.

      19. Gardner “Results of case-control study of leukemia and lymphoma among young people near Sellafield nuclear plant in West Cumbria.” British Medical Journal, February 17, 1990, CCC, p.423. Gardner, “Methods and basic data of case-control study of leukemia and lymphoma among young people near Sellafield nuclear plant in West Cumbria.” British Medical Journal, February 17, 1990, CCC, p. 429.

      20. 63 American Journal of Medicine, pp. 967. 972.

      21. Early detection of thyroid abnormalities is therefore important. Annual thyroid examinations would help minimize the number of malignant nodules that metastasize before detection.

      22. Nuclear Regulatory Commission, Reactor Safety Study - An Assessment of Accident Risk in U.S. Commercial Nuclear Power Plants, WASH-1400, NUREG 75/014 (1975), p. 85 (hereinafter WASH-1400).

      23. See infra, n. 30 and text.

      24. National Institutes of Health, Report of the National Institutes of Health Working Group to Develop Radioepidemiological Tables, NIH Publication 85-2748 (1985). p. 79 (hereinafter NIH Pub. 85-2748) (disclaiming the use of the radioepidemiological tables as other than a guide).

      25. See discussions of Atomic Veterans Legislation and BNFL/UKAEA agreements with their unions in Appendices E and G, respectively, for more comprehensive listings of relevant factors.

      26. NIH Pub. 85-2748. See supra, n. 24 for cancers covered in the tables.

      27. The Orphan Drug Act, 7, 42 U.S.C. 241.

      28. See Radiation Exposure Compensation Act of 1979, Joint hearings before the Subcommittee on Health and Scientific Research of the Senate Committee on Labor and Human Resources and the Senate Committee on the Judiciary, 96th Cong., 2d sess., 1980; and Radiation Exposure Compensation Act of 1981, hearing before the Senate Committee on Labor and Human Resources, Part 2, 97th Cong., 2d sess., 1982. On June 5, 1990, the House of Representatives approved H.R.2372 the Radiation Exposure Compensation Act. This legislation would set up a $100 million fund to compensate victims of radiation-induced illnesses caused by the nuclear weapons testing program and uranium miners who were employed in mines in Colorado, New Mexico, Arizona, and Utah. Similar legislation has been introduced in the Senate. 136 Congressional Record H 3145.

      29. Dr. Joseph Fiksel. Cimflex Teknowledge, Inc., Meeting Transcript. Nov. 30.1989. pp. 111-122.

      30. See Fiksel and Cox, “A Critical Review of the Probability of Causation Method,” The Price-Anderson Law; Six Reports on Price-Anderson Issues (Farmington. Conn.: ANI/MAELU. 1985).

      31. Robert Willmore. Attorney, Arent, Fox, Kintner, Plotkin and Kahn, Meeting Transcript. Feb. 7, 1990, p. 328.

      32. The theory here would be that premiums for guaranteed cost insurance were set based on the reasonable and sealed expectation that the “bodily injury, sickness or disease, or death resulting therefrom” covered by the approved form of nuclear liability insurance applies to actual instances of disease, not increased risk of incurring disease. It is less clear that this theory argues against compensating emotional distress that results in impaired function and associated economic losses, such as the expense of medical treatment and lost wages due to inability to work, although insurers might argue whether emotional distress in the absence of present physical harm is covered under the above policy wording.

      33. Cross, Environmentally Induced Cancer and the Law (Westport, CT: Greenwood Press, 1989).

      34. His main point is that the present system undercompensates on average, so one should not be unduly concerned if the proposed system does this also; as to windfalls, he seems to think that providing all those who are eligible enough money to buy modest amounts of “cancer insurance” is the answer, because only those who develop cancer would receive the benefit but all at risk would pool their risk. While this is appealing, it does seem to overlook the “spontaneous” rate of cancers, which implies that a great many policies would pay benefits to those whose cancers were not caused by the toxic pollutant

      35. See Silkwood v. Kerr-McGee Corp., 485 F. Supp. at 566.599 (1979). (Appendix; jury instruction 11.)

      36. Radiation Exposed Veterans Act of 1988, amending 38 U.S.C. 312.

      37. See Appendix E.

      38. 24 Weekly Compilation of Presidential Documents 641, May 23, 1988.

      39. Workers' Compensation Amendment Act of 1984. No. 2, Ch. 58. See Appendix F to this report

      40. The range has been reported as from 16,924 to 120,633 pounds sterling. See Appendix G.

      41. There might be two basic types of medical monitoring: (1) complete physicals stressing cancer detection and (2) for those who have ingested radioactive iodine but have no other significant exposure, periodic thyroid exams could suffice. Thyroid exams are quite straightforward and should involve only a brief office visit, with correspondingly lower costs than a complete physical.

      42. That is by inhalation, ingestion, or direct radiation emitted from a cloud of radioactive material that passes overhead without deposition of particulate matter.

      43. That is, the symptoms of acute radiation syndrome or radiation sickness, which include headaches, abnormal sensations of taste or smell, and decreases in blood pressure and white blood cells. Those symptoms would be expected to occur with individual doses in excess of I Gy (100 rad).

      44. This is a simple matter of holding a Geiger tube to the skin of the neck over the thyroid gland to detect radioiodine.

      45. Of course, if there is reason to expect uptake of radioactive material, then use of whole body counters and bioassay techniques might also be indicated.

      46. Bender, et. al., “Current Status of Cytogenetic Procedures to Detect and Quantify Previous Exposures to Radiation.” Mutation Research, CXCVI (1988). p.103.

      47. Ibid.

      48. Ibid.

      49. A more correct term might be “assigned share” of group risk, as suggested by the Oversight Committee on Radioepidemiological Tables of the National Research Council in its report. Oversight Committee on Radioepidemiological Tables of the National Research Council. Assigned Share for Radiation as a Cause of Cancer (Review of Radioepidemiological Tables Assigning Probabilities of Causation). (Washington, D.C.: National Academy of Sciences, 1984). As pointed out at pages 2 and 3 (and explained in Chapter II) of that report, “... the Committee uses the term 'assigned share' rather than the more familiar 'probability of causation.' for technical reasons .... The latter is in some ways a misnomer.... the expression 'assigned share' is more appropriate ... because the quantities being computed are not probabilities in the usual sense and are truly properties of the group to which a person belongs, but in practice are assigned to the person for purposes of compensation.”

        Precisely because the term “probability of causation” is the more familiar one, the Commission has chosen to use it, (1) trusting that readers will bear in mind that it is technically a misnomer, and (2) intending that it be read as having a specialized meaning in this usage, viz: “assigned share.”

      50. Data for both males and females are shown in Appendix H.

      51. The Commission's discussion and the examples in Appendix H rely heavily on information obtained from the recently released BEIR V report. That report concludes that the risks of cancer from a radiation exposure may be three or more times higher than stated in previous reports. This conclusion has, not as yet, received widespread acceptance. While the Commission used BEIR V in the sample calculations because its conclusions seem to be in the conservative direction, the Commission does not comment on its validity.

      52. See supra, n. 24.

      53. Compare Examples 1 and 2 in Appendix H.

      54. In instances where the only dose was iodine-131 to the thyroid, the dose might be determined by an external measurement as explained in Appendix H.

      55. Again, in the manner of the NIH Radioepidemiological Tables, as updated using then-current information regarding risk coefficients and so on.

      56. Since radiogenic cancers are indistinguishable from other cancers, it follows that cure rates do not differ.

Chapter Three « Index » Appendix A