Executive Summary


National Cancer Institute
Study Estimating Thyroid Doses of I-131 Received by Americans From Nevada Atmospheric Nuclear Bomb Test


Public Law 97-414, in part, directs the Secretary of Health and Human Services to "conduct scientific research and prepare analyses necessary to develop valid and credible methods to estimate the thyroid doses of Iodine-131 (131I) that are received by individuals from nuclear bomb fallout (and) to develop valid and credible assessments of the exposure to Iodine-131 that the American people received from the Nevada atmospheric nuclear bomb tests."

The National Cancer Institute was asked to respond to this mandate, and the present report was prepared for that purpose. The full study report, to be available as soon as possible, provides estimates of human exposure to and thyroid radiation doses from iodine-131 resulting from individual nuclear tests conducted at the Nevada Test Site (NTS).

Ninety nuclear tests released almost 99% of the total iodine-131 entering the atmosphere from the bomb tests conducted at the NTS. These ninety tests released about 150 million curies of iodine-131, mainly in the years 1952, 1953, 1955, and 1957. Some radioiodine was deposited everywhere in the United States, with the highest deposits immediately downwind of the NTS. The lowest deposits were on the west coast, upwind of the NTS. In the eastern part of the country, most of the deposited iodine-131 was associated with rain, while in the more arid west, dry deposition (where particles settle on the ground) prevailed. Because iodine-131 decays with an 8-day half-life, exposure to the released iodine-131 occurred primarily during the first two months following a test.

Estimating Exposure

Historical measurements of the amounts of radioactivity deposited and of a daily rainfall were used as the basis for the dose calculations whenever feasible. These historical measurements consisted of a simple collection of daily fallout on sticky paper (i.e., gummed film) made at the time of and during several days following most of the tests. The number and location of the monitoring stations across the United States varied with time but never exceeded 100. The collected fallout was measured daily for the amount of gross beta radioactivity present. The monitoring system was intended to determine where and when fallout occurred, but did not measure specific radionuclides. In other words, the system did not measure individually the amounts of different kinds of radioactivity, such as iodine-131, strontium-90, and cesium-137.

Reanalyses of these data together with the use of mathematical modeling, and the incorporation of precipitation data for each county during the time fallout clouds were over the United States, permitted estimates of iodine-131 deposition in each county for each day following each test. This reanalysis included: 1) the assessment of the collection efficiency of the gummed film for fallout collection; 2) the assessment of the efficiency of the radioactivity counting equipment, which varied from test series to test series; 3) accounting for the loss of volatile radionuclides during sample processing at the time of the original measurements; and 4) the use of more recently declassified and published characterization of the distribution and quantity of radionuclides in the fallout cloud produced by each test.

Measurements of the amount of radioactivity deposited were not available for 3 tests conducted in 1951, and for 6 tests conducted between 1962 and 1970. The latter six tests are thought to have led possibly to significant depositions of iodine-131 in the U.S. For these nine tests, atmospheric dispersion and deposition models were used to estimate the amount of iodine-131 deposited by county.

Regional data on consumption of pasture grasses by cows and on the transfer to milk of iodine-131 deposited on pasture grasses were used to estimate concentrations of iodine-131 in milk fresh from cows. These concentrations, together with milk distribution patterns in the 1950s, were used to estimate local concentrations of iodine-131 in the cows' milk available for human consumption throughout the country. (Milk consumed immediately after milking a family cow would have a higher concentration if iodine-131 than does milk processed and then consumed days after a cow was milked.) Finally, milk consumption rates, based upon diet surveys, were used to estimate the amounts of iodine-131 ingested by age group and by gender. The transfer of iodine-131 to people through the other exposure routes was similarly analyzed.

The overall average thyroid dose to the approximately 160 million people in the country during the 1950s was 2 rads. The uncertainty in this per capita dose is estimated to be a factor of 2, that is, the per capita dose may have been as small as 1 rad or as large as 4 rads, but 2 rads is the best estimate. The study also demonstrated that there were large variations in the thyroid dose received by subcategories of individuals. The primary factors contributing to this variation are county of residence, age at the time of exposure, and milk consumption patterns.

Scope of Study

The legislation called for the development of methods to estimate iodine-131 exposure to the American people, to assess thyroid doses from iodine-131 received by individuals across the country from the Nevada tests, and to assess the risk for thyroid cancer from these exposures. This study fulfills the first two of these requirements; other studies have and are fulfilling the third. The complete study report includes estimates of the cumulative average iodine-131 dose, by age and sex, to the thyroid for representative persons in each county after each test during the period when the nuclear tests were conducted in Nevada. Estimates of thyroid doses have been made for persons by age, sex, and source and quantity of milk consumed because milk was the source of most of the iodine-131 exposure for most people. Uncertainty is associated with the dose estimates developed by the study because the estimates are based on a small number of radiation measurements made at the time of the tests and the study authors had to rely heavily on mathematical models to develop the estimates.

Estimating Individual Exposures


The importance of geographical location can be seen in Figure 1, which shows the overall per capita doses by county. In general, the highest per capita thyroid doses, in the range of 9 to 16 rads, were obtained in counties of western states located east and north of the NTS, such as Colorado, Idaho, Montana, South Dakota, and Utah. In many counties on or near the west coast, the border with Mexico, and parts of Texas and Florida, the per capita thyroid doses were lowest, in the range of less than 0.1 to 0.5 rad. By comparison, the average individual in the United States receives a thyroid dose of about 0.1 rad each year from exposure to cosmic rays and naturally occurring radioactivity, with relatively large variations from one location to another.

The counties with the highest estimated average doses are listed in Table 1. Individuals living in these five western counties were estimated to have a cumulative average dose of 12 to 16 rads. These were Meagher County, Montana, and Custer, Gem, Blaine, and Lemhi Counties in Idaho. The table lists another 20 counties, mostly in Montana, where cumulative individual doses were estimated to be in the range of 9 to 12 rads.

It should be noted that the exposure ranges for the counties in Table 2 and other ranges merge into one another, especially considering the uncertainties associated with all of these estimates. There are no sharp dividing lines between these ranges.


The thyroid doses to individuals at a particular location were strongly dependent upon age at the time of exposure. Thyroid dose estimates for young children are uniformly higher than those for adults, assuming that individuals in particular geographic areas consumed milk from the same source at average rates for their age group. For any particular test, the thyroid doses for children between 3 months and 5 years of age exceeded the average per capita thyroid dose following that test by a factor of about 3 to 7 because of greater milk consumption and their smaller thyroid.

The date of birth and geographic residence of individuals also are strong determinants of the cumulative dose received from all tests. The variation in cumulative thyroid doses to individuals born at different times, each of whom lived in a single county and consumed cows' milk from local sources at average rates, is illustrated in Table 2. This can be considered a dose table for six typical families located in the identified counties throughout the testing period. The factors affecting the doses to parents are approximately independent of birth dates up to 1930; doses to adult men and women born prior to this time were nearly the same. Thyroid doses to children born about six months prior to three major test series (1952, 1953, and 1957) were substantially higher in general than the adult doses. The thyroid doses to teenagers would have been intermediate between those to small children and to adults. The last column shows doses to children born in 1958, which is the year when the last test series (but not the last individual tests) in the atmosphere took place at the NTS. Cumulative thyroid doses to most of the children born in later years are estimated to be less than 0.1 rad.

Diet, Particularly Milk Consumption

For most people, the major exposure route was the ingestion of cows' milk contaminated as the result of iodine-131 deposited on pasture grasses; other exposure routes such as the inhalation of contaminated air and the ingestion of contaminated leafy vegetables, goats' milk, cottage cheese, and eggs also were considered. For individuals within a particular age range, milk consumption can vary substantially. For example, surveys have shown that 10% to 20% of children between ages 1 and 5 do not consume cows' milk. Their doses were only about one tenth of those received by children who consumed fresh cows' milk at average rates for their age. Conversely, the milk consumption of 5% to 10% of individuals in the same age range was two to three times greater than the average and their thyroid doses were therefore proportionally larger. The type of milk consumed also is important. It is estimated that at that time about 20,000 individuals in the U.S. population consumed goats' milk. Thyroid doses to those individuals could have been 10 to 20 times greater than those to other residents of the same county who were the same age and sex and drank the same amount of cows' milk. Goats' milk concentrates iodine-131 more than cows' milk.

The foregoing examples illustrate that the thyroid dose received by any particular individual depends on his/her source of milk and dietary habits and thus may differ considerably from the group dose estimates. Furthermore, the person's total thyroid dose from all tests depends upon place of residence and age at the time of each test. Because of the very large number of variations in residence location, age, and dietary habits, it is not feasible to provide estimates of cumulative doses for individuals. However, detailed information is provided in the full report so that individual cumulative doses can be estimated based upon personal residence and dietary history.

Uncertainties and Model Validation

There are large uncertainties in the estimated thyroid doses given in the report because it is impossible to know all the information needed to determine exact doses. These uncertainties were assessed in two ways. First, calculated concentrations of iodine-131 were compared with the few historical measurements of iodine-131 in people and the environment that are available. Second, the uncertainties in the historical daily deposition data and in each of the factors used to estimate the transfer of iodine-131 to people's thyroids through the various exposure routes yielded an estimate of the total uncertainty. The uncertainty in the thyroid dose estimated for an individual is greater than the uncertainty in the overall average thyroid dose to the entire United States population. In general, the uncertainty of the thyroid dose from NTS iodine-131 for representative individuals is about a factor of 3, e.g., if the thyroid dose estimate for an individual is 3 rads, it will likely lie between 1 and 9 rads.

The results obtained from the mathematical models used in this study were compared with any data collected at the time of the tests in order to compare the findings of the modeling with those of the actual data collection. The comparisons also provide an estimate of the uncertainty attached to the calculated doses. As a result of these comparisons, a relatively good agreement was found between actual data and predictions made by the mathematical models. For example, independent analysis of urine samples volunteered by soldiers at Army bases throughout the United States following one of the test series showed iodine-131 dose levels consistent with doses predicted. However, it should be noted that the comparison between measured and predicted values required the use of several assumptions, and there is no guarantee that the samples measured were representative of county averages.

Information To Be Included in the Full Report

Thyroid dose estimates are given for representative individuals in specified age groups residing in each county of the contiguous United States. The report also contains extensive tables of information organized by test and by county so that individual radiation doses to the thyroid from iodine-131 can be estimated based upon personal residence and dietary histories. Thyroid doses from iodine-131 were estimated for 13 age categories, including the fetus, with adults subdivided by gender, in 3,071 counties of the contiguous United States, and for all periods of exposure. There are four consumption scenarios calculated for each category. The report's maps, tables, and formulas will allow local governments and other organizations to calculate dose estimates for individuals falling in these categories in their geographic region.

back to Report Contents