EH-93-1 Job Factors, Radiation, and Cancer Mortality at ORNL
             OFFICE OF EPIDEMIOLOGY AND HEALTH SURVEILLANCE
----------------------------------------------------------------------
                           HEALTH BULLETIN
----------------------------------------------------------------------
U.S. DEPARTMENT OF ENERGY                             WASHINGTON, D.C.
Issue 93-1                                               February 1993
----------------------------------------------------------------------

            Job Factors, Radiation, and Cancer Mortality at
         Oak Ridge National Laboratory:  Followup Through 1984


A study published in the March 20, 1991, issue of the Journal of
the American Medical Association (JAMA 265:1397-1402) examined the
causes of death among white men who worked at the Oak Ridge
National Laboratory (ORNL) between 1943 and 1972.  The authors
compared death rates for the calendar period 1943 to 1984 among
workers exposed externally to ionizing radiation with those among
workers who had no recorded exposure.  The cancer death rate
increased with total radiation dose, but this relationship was not
observed for any specific type of cancer.  The death rate for
leukemia among ORNL workers was higher than that for U.S. white
men, but this excess was not related to radiation exposure.

Additional analyses of death rates among the same ORNL workers,
will be published in the February 1993 issue of the American
Journal of Industrial Medicine.  These analyses focus on other
factors that might influence death rates to see whether accounting
for these factors would change the statistical relationship between
cancer and radiation exposure.  These include:  length of
employment in specific job categories, factors related to time of
hire (for example, men hired during World War II may have been less
healthy than men drafted for military service, or may have had
exposure to higher levels of radiation than men hired later in
time), and exposures to substances such as beryllium, lead, and
mercury.

The study population consisted of 8,318 white men.  One thousand
four hundred and ninety of these men died between 1943 and 1984. 
Death rates for cancer were calculated for men employed in 15 broad
job categories that were assumed to have similar occupational
exposures.  The death rate for cancer rose with length of
employment in six of the categories:  radioisotopes production,
chemical operations, physics, engineering (construction),
engineering (hazards), and "unknown" jobs.  Further statistical
analyses suggested that measured external radiation exposure might
account for this finding in two of the job categories
(radioisotopes production and chemical operations).  Accounting for
potential exposure to lead, mercury, or beryllium did not change
the relationship between cancer and radiation exposure.   The
cancer death rate among workers hired during World War II was
similar to that among men hired later.

There were several limitations in the study methods that should be
considered in interpreting the results.  Most important, results
were shown only for all types of cancer combined.  The term
"cancer" comprises many diseases that have different causes.  Broad
job categories were used as a substitute for exposure.  Individual
exposure levels for specific substances like beryllium, lead, and
mercury were not determined.  Each worker had an average of 1.5
jobs during his career, thus, any individual worker could have been
included in more than one category.  One of the most important risk
factors associated with cancer (lung cancer, in particular),
cigarette smoking, was not considered.  Because about half of all
deaths from cancer are lung cancer, it would be important to
consider smoking as a factor.

This study is one of many that has looked at the mortality
experience of ORNL workers, and radiation exposed workers in
general. 

      Epidemiologic Note:

      In evaluating whether there is a cause and effect relationship
      between an exposure (or risk factor) and a disease, researchers
      carefully consider the following questions: 

      How strong is the relationship between the disease and the exposure?

      The larger the difference between the rate of disease in an exposed
      population compared to that in an unexposed population, the more
      likely the disease is caused by the exposure. 

      Was the relationship statistically significant?  If there were more
      cases of the disease than would be expected due to chance alone, the
      relationship is more likely to be causal. 

      Does the rate of disease increase consistently with increasing level
      of exposure?  If the relationship is a causal one, we might expect
      to see higher rates of disease among people who had higher levels of
      exposure or who were exposed for a longer time. 

      Did the exposure occur a sufficient amount of time before the
      disease developed to have been able to cause the disease?  Some
      diseases, like certain forms of cancer, take many years to develop,
      and to cause the disease the exposure must occur many years earlier.
      
      Do similar studies of different populations having the same exposure
      show similar results?  If the same relationship is seen in many
      studies, then it is likely that the relationship is causal.  If the
      results of studies are different, then it is difficult to make any
      interpretations regarding cause. 
 
      Are the results consistent with what we know about human biology and
      disease natural history?  If the results are consistent, then it is
      more likely that the relationship is causal. 

      Finally, researchers consider the design and limitations of the
      study.  Are there any weaknesses in the methodology that may
      influence the conclusions?  Have other factors that may also be
      related to the disease of interest been considered?  How accurate
      are the measurements of the exposure and of the occurrence of the
      disease?



      This Health Bulletin is one in a series of routine publications
      issued by the Office of Health to share data from health studies
      throughout the DOE complex.  The authors conclusions do not
      necessarily reflect those of the Department.  For more information
      contact:  Dr. Terry L. Thomas, Director, Health Coordination and
      Communication Division, Office of Epidemiology and Health
      Surveillance, U.S. Department of Energy, Washington, D.C.  20585;
      Telephone (301) 903-5328.

.