Beck and Co-Workers Explained:

Red Bone Marrow and Male Breast Doses for a Cohort of Atomic Veterans.

Reference: Beck, H.L., Till, J.E., Grogan, H.A., Aanenson, J.W., Mohler, J.H., Mohler, S.S. and Voilleque, P.G. (2017) Red Bone Marrow and Male Breast Doses for a Cohort of Atomic Veterans, Radiation Research, 187, pp. 221-228.


What was the research question?

Prior to this study, scientists had estimated the doses received by U.S. nuclear test veterans in an American government programme called the Nuclear Test Personnel Review (NTPR) in the 1970s and 1980s.

The NTPR was required by law to favour veterans who were seeking compensation relating to their possible exposure. This meant that when relevant information needed was not available and a range of dose estimates were possible, the scientists chose the highest estimate. For example, there was a ship for which it was known that the highest risk of exposure was in the engine room. The NTPR did not have film-badge data or a list of duties for the crew members and so assumed the worst-case scenario, i.e. that every crew member had worked for 16 hours a day in the engine room.

The research team who performed this study described the doses assigned by the NTPR as being “high-sided”. The authors argued that more accurate dose estimations were required so that epidemiologists could investigate the health effects of ionising radiation based on the dose received.

In 2005 the American Department of Defence began to release individual veteran service records, i.e. a record becomes declassified 62 years after the veteran leaves the military. Environmental radiation readings from the nuclear tests have also been declassified over the same period. The authors of this paper wanted to use this information to make more accurate dose estimates.

How was the scientific problem approached?

The researchers began by reviewing the data from the film badges that had been worn by some (25%) of the veterans selected for this study. These badges contain photographic film which darkens in response to radiation, but can also be darkened by the hot and humid conditions in which the veterans performed their military service. So, the researchers discarded dosimetry data where they had evidence for it coming from a damaged film badge.

The majority of veterans however had not been given a film badge. To account for this, the NTPR made the assumption that if there was badge data for one member of a military unit, then the other members of that unit will have received the same dose. For this study, the availability of service records enabled the researchers to compare both the duties and the activities of badged and non-badged veterans and reconstruct the circumstances in which exposure occurred, thereby providing a more nuanced estimate of dose.

What did the research involve?

The researchers recruited veterans who had participated in at least one of eight series of nuclear tests that had been performed at various locations in the Pacific Ocean and in the U.S. states of Nevada and New Mexico between 1945 and 1963. These tests had the codenames Upshot-Knothole, Plumbbob, Crossroads, Greenhouse, Castle, Redwing, Hardtack I and Trinity.

Approximately 114,000 U.S. veterans served in these operations in the army, navy, marines and air force. The researchers randomly selected 1% of this total and also, every veteran that had either leukaemia or male breast cancer to form a total of 1,982 study participants. Due to practical constraints, the researchers chose to perform reconstructions only on veterans who had been assigned doses greater than 5 mGy by the NTPR.

When performing reconstructions, the researchers found that the service records contained varying levels of detail about the duties and the activities of veterans and so they divided them into three groups.

Group A: Veterans whose activities and duties were well known and hence for whom it was relatively straightforward to reconstruct the circumstances of their exposure.

Group B: Veterans whose activities and duties were less well known but there is sufficient information in their records to infer the probability of exposure.

For example, some naval servicemen were required to board other ships which had been intentionally contaminated with radioactive materials to perform clean-up operations. However, it was not standard practice in the U.S. Navy to record the names of the members of boarding parties and so some ships did not have this information. Despite this, based on their rank and duties the researchers could estimate the probability that a given naval veteran was included in a boarding party.

Group C: Veterans for whom little is known about their activities and duties. These veterans were assigned the average dose for activities that they could have participated in, however these estimates have the largest margins of error.

In addition, a small number of the study participants had also made compensation claims and in these cases, there was also interview data to supplement the information given in their service records.

To assist the epidemiologists, the scientists estimated for each veteran a red bone marrow dose and a male breast dose.

What did they find?

The doses received by the veterans were lower than determined by the NTPR in all groups and the average bone marrow dose was 5.9 mGy. Two thirds of the veterans were estimated to have received a bone marrow dose < 5 mGy and 4 individuals had a bone marrow dose > 50 mGy. The doses to the male breast were estimated to be approximately 20% higher than the bone marrow doses.

How did the researchers interpret their basic results?

It was not unexpected to the researchers that the estimates that they produced were lower (on average 36%) than those produced by the NTPR, because the latter were high-sided for compensation purposes. The authors regard these updated estimates to be more accurate because they have improved the methods for making them. However, the authors caution that for some of the veterans their estimates will still contain uncertainties due to there being insufficient detail in their service records. The authors stated that they intend to use these new estimates in future epidemiology studies.

Who did this research?

This work was done by a team of American researchers from the following private companies: Risk Assessment Corporation, Scientific Consulting Incorporated, Cascade Scientific Incorporated, Bridger Scientific Incorporated and MJP Risk Assessment Incorporated.

This research was funded by the U.S. government agencies, the National Cancer Institute and the Department of Energy, and the Vanderbilt-Ingram Cancer Centre, an American treatment and research centre.


Supporting Resources



Key Messages

An improved method, based on using dosimeter film badge data in combination with individual service records and environmental readings, has been developed to estimate the doses received by U.S. nuclear test veterans.

The ability of researchers to historically reconstruct the circumstances of exposure depended upon the level of detail about the activities and duties of each veteran in their service record. This varied and so the new estimates have varying degrees of uncertainty.

The estimates for the doses received based on this new method are generally 36% lower than those previously obtained.

Two thirds of the veterans received bone marrow doses < 5 mGy, the average bone marrow dose was 5.9 mGy and only four individuals had a dose > 50 mGy.

The male breast doses were on average 20% greater than the bone marrow doses.

Links to the research paper

This is a peer-reviewed study meaning that other scientists have reviewed this work before the authors published it in the journal Radiation Research in 2017.

This article is a sequel to one by Till and co-workers that was published in
it in the journal Radiation Research in 2014.

Please let us know what you think. Did you find this useful?

Click on a star to rate it!

We are sorry that this page was not useful for you!

Tell us how we can improve this?