The Effects of Nuclear War

CALCULABLE EFFECTS: IONIZING RADIATION

A large body of scientific literature addresses itself to the issue of long-term effects from low levels of ionizing radiation, There has been an intensive study over the years of the health of the survivors of Hiroshima and Nagasaki, and of some of those who were subjected to radioactive fallout as a result of nuclear weapons testing. There has been considerable research into the question of how large a quantity of radioactive particles of various kinds are produced by nuclear weapon explosions. There is a body of theory regarding the effects of ionizing radiation on the human body. But there are also formidable uncertainties. New information is coming to light regarding some of the effects of past weapons testing, and there are unresolved scientific controversies over matters as basic as whether a small dose of radiation does more damage to the human body (or, from a statistical point of view, is more likely to do a given amount of damage to a human body) if it is absorbed during a brief period of time than if it is absorbed over a longer period. There are pertinent questions whose answers are only known to within a factor of 10.

Previous chapters have discussed the effects of very intensive ionizing radiation: 1,000 rems will almost certainly be lethal if absorbed within a matter of days; 450 rems will kill 50 percent of a healthy adult population, and a slightly higher percentage of the young, the old, and those without adequate medical care; 250 rems will cause acute radiation sickness, from which “recovery” is probable; and even lower doses may lower the body’s resistance to infectious diseases of various kinds. It is generally assumed that because of the rate at which fallout radiation decays, doses of this magnitude are likely to be received during the first so days after an attack if they are received at all. The preceding chapter, and appendix D, include calculations on the numbers of people who might die from radiation effects during the first 30 days after various kinds of nuclear attack.

However, doses of ionizing radiation that are too small or too slowly accumulated to produce prompt death or radiation sickness nevertheless have harmful effects in the long run. These effects can only be discussed statistically, for it appears that if a large population is exposed to a given (small) dose of radiation, some will suffer harmful effects while others will not. The larger the dose, the greater the percentage of the population that is harmed, and the greater the risk to any one individual.

There are a number of ways in which a nuclear attack would lead to radiation exposures which, although too low to cause death within the first 30 days, nevertheless pose an appreciable long-term hazard:

• Prompt radiation from the nuclear explosions could inflict sublethal doses on some survivors, especially if the weapons are small ones. Most of the radiation absorbed by survivors of the Hiroshima and Nagasaki attack was direct radiation. A substantial number of U.S. weapons have yields in the tens of kilotons, and might inflict radiation on people far enough away from the explosion to survive the blast effects. Few Soviet weapons are of such low yields and high-yield weapons are expected to kill those within radiation range by blast. A terrorist weapon would almost certainly inflict direct radiation on survivors. There is a particular area of uncertainty regarding the effects on humans of low levels of neutron radiation.
• Local fallout will inflict small doses of radiation on people who are on the fringe of “fallout zones,” or on people who are in fallout shelters in zones of heavier fallout. It is important to realize that even the best fallout shelters attenuate fallout rather than block it completely, and the whole theory of fallout shelters is to see to it that people who would, if unsheltered, receive a lethal dose would instead receive a sublethal dose. However, this sublethal dose will produce harmful long-term effects for some percentage of those exposed.
• After a period of time, local fallout radiation levels decay to the point where the area would be considered “safe,” and survivors in fallout shelters would emerge. Nevertheless, low levels of radiation would persist for some time— indeed, low levels of radiation have persisted for years at some sites of nuclear weapons tests. The question of safety here is a relative one. By the standards of peacetime, many such areas would be considered unsafe, because living in them would expose a population to a significant risk of long-term hazards— cancer, genetic damage, etc. However, in the aftermath of a nuclear attack, there may be few habitable areas that do not have a measurable (though low) level of additional radiation, and the survivors would simply have to accept the hazards.
• Some fallout is deposited in the troposphere, and then is brought down to Earth (largely by rain) over a period of weeks. Such fallout reaches areas quite far from the blast. While the doses inflicted would be relatively small, they would add to the risk.
• Some fallout is deposited in the stratosphere. It returns to Earth over a period of years (through the effects of gravity), and consequently only very long-lived radioactive isotopes pose a significant hazard. If the attacks are confined to the territory of the United States and the Soviet Union (and, for that matter, to Europe and China as well), then stratospheric fallout will be confined mostly to the Northern Hemisphere, and the region between 300 and 600 north latitude will receive the bulk of it.

In quantifying the radiation dose received by individuals, radiation from external and internal (ingested) sources must be distinguished. External radiation passes through the skin. Ingested radioactivity derives its effects from particular radioactive isotopes becoming concentrated in specific organs. For example, radioactive iodine (I-131), which may enter the body through breathing, eating, and drinking, is concentrated in the thyroid, and radioactive strontium (Sr 89 and Sr 90) is concentrated in bone.

An OTA contractor performed a series of calculations to estimate the magnitude of the long-term health hazards that would be created by the long-term, low-level radiation that each of the OTA cases might produce. The basic method was to calculate the total amount of radiation that all the survivors of each hypothetical nuclear attack might absorb during the 40 years following the attack, and then calculate the numbers of adverse health effects that this much radiation could be expected to produce. (Tables 12 and 13 present the risk factors used for these calculations.) The difficulties in such a procedure are formidable, and precise results are manifestly impossible to obtain.

The major uncertainties, which result in a wide range of answers, are the following:

• All of the uncertainties discussed in previous chapters about the size and nature of the attack, and the distribution of the population.
• How much of the population benefits from what degree of fallout sheltering? It has been noted that there is no necessary relation between civil defense plans and actual shelter received.
• How many people die in the immediate aftermath of the attack?
• Does radiation that is part of a low exposure or a very slow exposure do as much damage per rem absorbed as radiation received as part of a high and rapid exposure? One theory holds that, given time, the body can repair the damage done by radiation, and that hence the same dose spread over years does less damage than it would if received within a few days. Another theory is that radiation damages the body in ways that are essentially irreparable. The contractor calculated the effects both ways (DEF = 1 and DEF = 0.2), which accounts for some of the range in the answers.
• Is there a threshold dose below which radiation exposure does no harm at all? If there is, then the methodology used produces somewhat exaggerated results, since it attributes damage to radiation absorbed by people whose total dose is below the threshold.
• How to deal with the distribution of ages of the population at the time of the attack, since susceptibility to cancer, etc., from causes other than radiation varies with age.
• How great are the genetic effects from a given level of radiation? Extensive experimental results permit an approximate calculation of the number of mutations that would be produced, although one source notes that the doubling dose for genetic disorders might be anywhere from 20 to 200 rems. However, it is far more difficult to predict exactly how these mutations would manifest themselves in future generations.

The results of these calculations are summarized in table 14. (The full report of the contractor is available separately. ) The ranges result from the uncertainties noted above, and it is expected that the “actual” results if a war took place would be some distance from either extreme. It is observed that:

• Cancer deaths in the millions could be expected during the 40 years following a large nuclear attack, even if that attack avoided targets in population centers. These millions of deaths would, however, be far less than the immediate deaths caused by a large attack on a full range of targets.
• A large nuclear war could cause deaths in the low millions outside the combatant countries, although this would represent only a modest increase in the peacetime cancer death rate.
• These results might not apply if an attacker set out deliberately to create very high radiation levels.

Just as this study was going to press, the results of the new report of the Committee on the Biological Effects of Ionizing Radiations (“BEIR II”) of the National Academy of Sciences (NAS) became available. (The full report, entitled “The Effects on Populations of Exposure to Low Levels of Ionizing Radiations,” will be published by NAS during the second half of 1979. ) In general, the new report suggests a slightly narrower range of uncertainty than the OTA calculations, but generally confirms their assumptions. OTA used assumptions of cancer deaths per million person-rems which appear to be about 10 percent higher at the high end of the range and about 40 percent lower at the low-end of the range than the findings of the new BEIR report. OTA calculated genetic effects on the basis of a doubling dose of 20 to 200 rems, compared with a range of 50 to 250 rems suggested by the new BEIR report, which may mean that the OTA estimates are too high at the high end of the range. The new BEIR report also notes that the incidence of radiation-induced cancer would be higher for women than for men.

Table 12. -Assumed Effects of Radiation Exposures

Effect	Number per million person-remsa
Somatic effects†
Cancer deaths (DEF - 1)*	194.3b c
Cancer deaths (DEF = 0.2)	38.9d
Thyroid cancers	131.4e
Thyroid nodules	197.4e
Genetic effects†
Abortions due to chromosomal damage	21-210f
Other genetic effects	66-660f

† These effects are in addition to those expected from natural or background causes.
^*DEF = Dose effectiveness factor.
^a This assumes that total exposure is governing -that is, that 1 rem each to a million
people produce the same effects as 10 rems each to 100,000 people. ^b This fiqure is modified from values presented in table VI, 9-4 of the Nuclear Regulatory Commission Reactor Safety Study (WASH 1400). The rationale for the modification was that although the latent period for cancer induction used by WASH 1400 was deemed appropriate, there is insufficient evidence that the plateau periods are limited to 30 years. Using the latent periods in WASH 1400 and the re mining lifespan as the plateau period along with the population age distribution, the cancer risk coefficients in the WASH 1400 study were converted to those in this table.
^c See table 12 for the sources of these cancer deaths.
^d Arrived at by multiplying the DEF = 1 figure by 0.2.
^e Taken from table VI, 9-8 of the WASH 1400 report.
^f These figures were derived from tables VI, 9-11 and VI. 9-12 of the WASH 1400 report. The range is based on a range of possible doubling dose from 20 to 200 rems, as suggested by the BEIR report (Washington, D.C.: National Academy of Sciences, 1972), p. 53.

Table 13. -Assumed Sources of Cancer Deaths^a

Cancer type	Cancer deaths per million organ-rems
Leukemia	45.4
Lung	35.5
Digestive tract	27.1
Bone	11
Others	75.3

^a See footnotes a and b to table 12.

---