Fallout exposures from US weapon tests:
health effects other than thyroid cancer

Before the Senate Appropriations Committee on
Labor, Health and Human Services and Education

October 1, 1997

Jan Beyea, Ph.D. (1)

I have asked Dr. Lawrence Mayer (2) to accompany me here today to answer any medical questions related to the work I will be discussing. For this hearing, I reviewed the NCI's exposure calculations based on the preliminary information published by NCI on its study, (3) as well as other uses of the databases relied upon by NCI. (4) In connection with litigation over radioiodine releases from the Hanford production facility, Dr. Mayer and I have also reviewed the scientific literature on the health effects associated with low-dose exposure to radioiodine, which has made us sensitive to disease risks other than thyroid cancer.

Since NCI in its preliminary reports has only discussed the possible connection between radioiodine exposure and thyroid cancer, I will focus most of my attention on thyroid nodules and autoimmune thyroid diseases. Based on my studies and those by Dr. Mayer, I have reached a number of conclusions:

  1. The "gummed paper" data, which NCI used to calibrate its estimates of radioiodine exposure, are probably free from the political pressures of the time and, therefore, can be used as the basis for unbiased exposure estimates.
  2. DOE's Health and Safety Lab (HASL, now EML), an institution that has always prided itself on its independence and integrity, carried out the measurements of fallout radioactivity on gummed paper at the time. Having been privy to EML practices as part of discovery in legal cases, (5) I have seen evidence of HASL scientists resisting orders to suppress information, finding ways to make the information public. There may be mistakes and limitations in the underlying gummed-paper data, but they are probably honest mistakes and limitations.

  3. Nevertheless, there are a number of reasons to expect that the current best estimates for exposure may be low.
  4. Generating historical exposure estimates requires making judgments, particularly in choosing values for parameters that enter the exposure model. It is difficult, even without the political pressure that radiological dose estimates have engendered, to pick an unbiased set of best-estimate parameter values. Thus, the choice of study members and the makeup of advisory committees can play a crucial role in a study's outcome, and hence can generate ferocious infighting. Usually, there are many free model parameters to specify in a dose reconstruction exercise. A member of the analytical team with a strong particular bias, or a member of an advisory committee who believes he or she knows the "correct" answer, not to mention the employers of the team, can selectively, even unconsciously, bias some of the results for non-scientific reasons. Without countervailing critiques of the choices made for important parameters, analysts may adjust parameters to satisfy advisory committees, supervisors, or dominant team members. Only if the advisory committees have a full representation of independent scientists with varying points of view can the choice of parameter values be kept from favoring the perceived goals of one political faction or agency. I have found some hints of this phenomenon in the early descriptions of the work provide by NCI. (6) Congressional investigators should scrutinize the makeup of the various advisory committees over time, the process by which study leaders and members were chosen, and whether or not independent scientists were appointed as members. Investigators should also examine if NCI was prepared to handle the heavy politicization that has existed in this country over radiation health effects. I have a very high degree of respect for NCI, based on personal experience and knowledge of its work. NCI has done many marvelous studies. It has a top notch, peer-review process for scientific projects, but it may not have been prepared for the hard-ball politics that was played with radiation health effects by other government agencies and influential figures. It would be a tragedy if public confidence in NCI were undermined because of this incident. Full and open disclosure at this point is the best way to protect the reputation of NCI. (7)

  5. The uncertainties in the dose estimates may have been understated in the counties with lower exposure.
  6. Published articles dealing with the underlying databases have pointed out certain inconsistencies in estimates made outside the highest dose regions. (8) The inconsistencies imply that the uncertainties in the lower dose regions are likely to be greater than stated in the NCI report. Thus, it may be unwise to dismiss consideration of medical actions in these areas, based solely on the dose ranges provided in the NCI report.

  7. The NCI reports published to date have ignored diseases other than thyroid cancer.
  8. Thyroid nodules, another known consequence of radiation exposure, have been neglected by NCI, even though such nodules require follow-up, and sometimes surgery. The authors of the thyroid disease study carried out around the Nevada Test Site were much more than "suggestive" when it came to induced neoplasms:

    "We conclude that in the cohort that was studied, an excess of between one and 12 neoplasms (0 to six malignancies) was probably caused by exposure to fallout radioiodines from nuclear weapons testing." (9)

    The NCI report is strangely silent on this consequence of radioiodine exposure.

    Of great concern is that physicians should be alerted to watch for signs and symptoms of lesser diseases than thyroid cancer, namely autoimmune hypothyroidism, mild thyroiditis, and possibly hyperthyroidism incident to Graves' disease. These diseases can be initiated by low to moderate radiation doses, the recent literature suggests, presenting themselves years after exposure. They are most likely caused by an autoimmune reaction in susceptible individuals. Once triggered, the body attacks its own thyroid cells, eventually causing clinical disease. Such effects were not identified in the A-bomb survivors by Nagataki et al. until 40 years after the bombing of Japan. (10) This finding suggests that these diseases should now be evident in the US population exposed to fallout doses of about 40 rads, assuming that the A-bomb situation is comparable. (11)

    The extensive data collected on people living around the 1986 Chernobyl disaster show that radioiodine exposure at surprisingly low doses between zero and 30 rads, mid-point = 15 rads leads to significant production of antithyroid antibodies. (12) The dose response stays fairly flat as dose increases up to a few hundred rads suggesting that a susceptible subgroup exists that is sensitive to low doses. Now, the presence of antibodies alone does not necessarily imply any diseases are yet present in this population, neither autoimmune thyroiditis, nor autoimmune hypothyroidism, nor autoimmune hyperthyroidism. However, the presence of elevated thyroid antibodies is a well-known risk factor for chronic thyroiditis and hypothyroidism. (13,14)

    Most of these data on autoimmune thyroid disease are new (15) and contradict some older published work. (16,17,18) The major study of the Nagasaki population wasn't published until 1994 (19) and the striking thyroid antibody study on the Chernobyl population wasn't published until 1996. (20) Although not yet appreciated or accepted by all scientists, the new information contained in these studies needs to be communicated to physicians and other health providers practicing in those counties where exposures may have exceeded an appropriate threshold, taken here to be 15 rads. Since exposure uncertainties may be as high as a factor of ten, these diseases could be appearing today in counties with average exposures as low as 1.5 rads. (21)

  9. I question if NCI has the full range of expertise necessary to consider these non-cancerous, autoimmune diseases in its contract with the Institute of Medicine.
  10. An important question is whether or not the contract NCI has developed with the Institute of Medicine will allow the Institute to consider this new literature in detail, as well as its importance for medical surveillance. Possibly, other divisions of NIH with expertise in autoimmune diseases, including immunology and endocrinology, should have a role in the follow-up study NCI has commissioned.

  11. A medical response program is in order for a significant fraction of the United States population exposed to fallout.
  12. Such a program should deliver medical information to physicians, perhaps through medical societies or continuing education. It should inform exposed persons of the early or pre-clinical symptoms of disease. It should provide medical surveillance in areas where the prior likelihood of disease is likely to be significant. Finally, in such areas, it should provide treatment for persons with a wide range of thyroid diseases who do not have adequate medical care due to insurance limitations.

    Obviously, a great deal of thought needs to be given to designing a proper medical surveillance program. However, based on the review of the literature and on discussions with Dr. Mayer, it seems clear that a carefully designed program would provide tremendous health benefits for those who do not already have, or do normally take advantage of, regular and thorough medical care. (22,23) The monitoring efforts required will generally not entail a large individual expense, although the total for the country will be significant. A surveillance program might consist solely of regular thyroid palpations and blood tests for thyroid dysfunction, including a "TSH" assay.

    Diseases that should be monitored:

    1. Thyroid cancer and its typical precursor, thyroid nodularity.
    2. As has been discussed by NCI, cancer risks are heavily weighted towards those exposed as children, particularly those exposed at the ages of 0-4 years. For external radiation, cancer has been confirmed in the literature down to an average dose of 9 rads, (24) and there is no evidence of a threshold to suggest cancer cannot be caused down to lower doses. Even if iodine-131 exposure were to be somewhat less of a risk factor for cancer compared to external radiation, this reduced risk would not change the fact that disease will result from the weapons fallout. It would only change the expected number of cases.

    3. Autoimmune thyroid diseases should also be monitored, particularly autoimmune hypothyroidism, autoimmune thyroiditis, as well as hyperthyroidism incident to Graves' disease. (25)

    The lowest dose in the literature shown to trigger the autoimmune thyroid process is found in a study of Chernobyl-exposed children. It is 15 rads. (26)

    An important goal of a medical surveillance program for autoimmune thyroid diseases, in addition to identifying any severe cases that have been missed, should be to identify mild and subclinical cases of hypothyroidism, which are difficult to detect and easily confused with non-disease conditions. Studies show that people with subclinical hypothyroidism benefit from drug treatment, leading to improved quality of life. (27) Paradoxically, severe cases are easily identified and easily treated with corrective medicines. Mild cases can go on for long periods of time, reducing the quality of life for people until symptoms get so severe that the need for treatment is recognized.

    Although it is now clear that those children who drank milk at the time of exposure are the key population to monitor for thyroid nodularity and cancer, the effects of exposure age on the risk of autoimmune disease is not known. This uncertainty complicates the picture for determining the scope of a medical surveillance program for radiation-induced autoimmune diseases.

    1. Information on the importance of the food chain for radiation exposures has been known for a long time.
    2. Government health physicists have known of the importance of the food pathway as a source of exposure since 1946. (28) They knew about the value of early intervention by, at least, 1958. (29) By the time of many, if not all, of the weapons tests, government officials knew or should have known that a mitigation strategy would prevent injury. Advisories to the public to avoid drinking fresh milk after weapons tests, particularly after rain, could have significantly reduced the expected number of thyroid cancer and nodules (and, as we now know, non-functioning thyroids). Such advisories could have been presented simply as a precaution, without having to admit to the public that any harm would necessarily fall upon them. Obviously, such a warning might have weakened public support for nuclear weapons testing and, therefore, would have had to be balanced against issues of national security. Somewhere, there may exist records of high level discussions about this difficult choice. Their content may be very revealing in helping to determine the degree of responsibility owed by the government to the public for the high exposures.

    3. It was not necessary to know the full outcome of the NCI study before making recommendations on medical surveillance.

      By 1982, when Congress asked NCI to undertake the fallout study, the connection between thyroid cancer and radioiodine was well known. At what point should NCI or DOE have taken steps to inform the medical community about the potential risks to their patients? Answering this question will, no doubt, be a key goal of congressional investigations.

    4. There may be other relevant studies languishing under bureaucratic confinement

      The fact that NCI has taken so long to report the essential public health message in this work raises questions about other studies that may be ongoing. (30) For instance, are political pressures slowing reports of the NCI's Chernobyl study? Will the Institute of Medicine have access to the preliminary results? Are there pending studies in other government agencies, such as DOE, that bear on the questions before the committee?

      Conclusion

      There are a number of factors that suggest a medical information, surveillance, and treatment program is in order for those exposed to weapons test fallout. These factors are 1) the magnitude of the projected radioiodine exposures, 2) the large uncertainties in the estimates, and 3), the recent findings that autoimmune thyroid diseases can be triggered at relatively low doses of radiation.

      I hope that the need for action not be forgotten as attention focuses on why there has been such a delay by government agencies in formulating a recommendation on medical surveillance.

      Acknowledgements

      I thank Dr. Mayer for advising me on the medical aspects of my statement and Joseph Wayman and David Beavers for collecting some of the historical documents relating to the discovery of the food exposure pathway for radioiodine.