18
Casualties of War: The Short- and Long-Term Effects of the 1945 Atomic Bomb Attacks on Japan
© 2007 The College Board. All rights reserved. College Board and the acorn logo are registered trademarks of the College Board.
All other products and services may be trademarks of their respective owners. Visit the College Board on the Web: www.collegeboard.com.
18
It would not be ethical or feasible to randomize people to one of two groups and
force them either to be exposed or to be unexposed. Ethical considerations would pre-
vent investigators from subjecting people in the exposed group to potential harm.
When epidemiologists study the effect of potentially harmful substances, they have neither
controlled laboratory experiments nor randomized controlled trials at their disposal. Instead,
epidemiologists often use cohort studies or case-control studies as ways of assessing whether
a particular agent is associated with a disease. In cohort studies a group of people who are
exposed and a group of people who are unexposed to the agent are followed over time, and the
rate of disease occurrence in the two groups is compared. In case-control studies a group of
people with the disease of interest and a group of people without the disease are compared with
respect to the frequency of previous exposure to the agent. Interpretation of the results of such
studies is limited by the fact that the groups may not be comparable. People who are exposed
may differ from people who are not exposed in other factors that affect the risk of getting the
disease. For example, when comparing asbestos miners with people who are not asbestos miners
with respect to lung cancer occurrence, it is important to consider whether asbestos miners are
more likely to be smokers, as smoking is a risk factor for lung cancer.
When natural disasters or man-made catastrophes occur, epidemiologists are provided with
the opportunity to study situations that would be impossible to study under normal condi-
tions. When there is such a disaster, people can be observed and followed over time after
being exposed to potentially harmful substances. Such studies are often referred to as natural
experiments. In a natural experiment people are not randomized to exposure or nonexposure as
they would be in a randomized controlled trial. Nevertheless, in a natural experiment the usual
selection factors that complicate many studies of the association of exposure and illness may be
reduced or absent. Also, it is often possible to study the effect of the exposure on a wide range
of people under a wide range of exposure conditions.
The dropping of atomic bombs on Hiroshima and Nagasaki constitutes a natural experiment, result-
ing in the largest and longest prospective human epidemiologic study ever undertaken. Because
the entire population of the two cities was exposed, it was possible to study men, women and
children of all ages, as well as unborn children. Because different survivors experienced different
amounts of exposure (depending on where they were at the time of the blast), it was possible to
study the extent to which health effects were dependent on radiation dose. Although there was a
wide range of exposure among the survivors, the immediate mortality rate among those with the
highest exposures was high, so that it turned out that most survivors had been exposed to low
doses of radiation. Therefore, epidemiologic studies of the A-bomb survivors have been crucial to
understanding the effect of low doses of ionizing radiation such as may be encountered in occupa-
tional settings. Because exposure occurred at a single fixed point in time, it was possible to quan-
tify the latent period for various cancers. Also, most survivors’ radiation exposure was quite evenly
distributed throughout the body, so the effects of whole-body irradiation could be studied. This is