Abstract
A survey of 425 death certificates of radiologists dying between the ages of 35 and 74 during the years 1948 to 1961 reveals a statistically highly significant excess of deaths from leukemia, multiple myeloma, and aplastic anemia. That this excess is due to radiation exposure (or to some factor acting in a similar manner), rather than to an artifact of diagnosis is suggested by the absence of deaths ascribed to chronic lymphatic leukemia.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
H. C. March, Radiology 43, 275 (1944); Am. J. Med. Sci. 220, 282 (1950); ibid. 242, 137 (1961).
H. C. March, Am. J. Med. Sci. 220, 282 (1950)
H. C. March, Am. J. Med. Sci. 242, 137 (1961)
L. I. Dublin and M. Spiegelman, J. Am. Med. Assoc. 137, 1519 (1948).
E. B. Lewis, Science 125, 965 (1957); R. Seltser and P. E. Sartwell, Am. J. Public Health 49, 1610 (1959).
R. Seltser and P. E. Sartwell, Am. J. Public Health 49, 1610 (1959).
W. M. Court brown and R. Doll. Leukemia and Aplastic Anemia in Patients Irradiated for Ankylosing Spondylitis, Medical Research Council (Brit.) Special Rept. Series No. 295 (H. M. Stationery Office, London, 1957).
E. E. Pochin, Brit. Med. J. II, 1545 (1960).
Directory of Medical Specialists (Marquis, Chicago, 1950–1960).
When year of birth was not stated, it was readily found in the case of every male radiologist by reference to editions of the American Medical Directory (American Medical Association, Chicago).
Dates of births were punched to the nearest half-year for the decreased group. For the living population it was necessary to assume that a birth was equally likely to have occurred in the first as in the second half of a year. This accounts for the presence of “half-individuals” (Table 1). Any errors due to wrong assignment to sex are likely to be negligible since only 2.8 percent of the names of Board-certified radiologists in the 1950 and 1960 editions of the Directory of Medical Specialist are estimated to be of females.
Death rates were computed by dividing the number of deaths in each 5-year age group (obtained from Vital Statistics of the United States, U.S. Govt. Printing Office. Washington, D.C. 1948–59) by the latest mid-year census estimates of the living population of U.S. white males (obtained from Current Population Reports, Series P-25,Nos. 98, 246, and 265,U.S. Bureau of the Census,Washington D.C.). TheNational Vital Statistics Division of the U.S. Public Health Service kindly provided the following unpublished data: deaths of U.S. white males (by 5-year age groups) from leukemia for 1960–1961, from multiple myeloma for 1949–1961, from aplastic anemia for 1950 and for 1953–1959; and deaths of U.S. white males (by 10-year age groups) from lymphatic leukemia for 1951 and 1956.
A third method was used, but yielded no additional names. This involved checking the death notices in unpublished annual bulletins of the American College of Radiology, the majority of whose members are Board-certified radiologists. It remains possible that a few deaths occurred that were not detected by any of the three methods.
In two cases, death certificates were unobtainable because the place of death could not be traced; in the third case, death occurred outside the U.S. Among the 425 death certificates obtained, four were for deaths occurring outside the continental U.S.; in none of these four cases, however, was death ascribed to causes under discussion in this report.
Manual of the International Statistical Classification of Disease. Injuries and Causes of Death. Fifth, Sixth and Seventh Revisions. (World Health Organization, Geneva, 1938, 1948, 1955).
In this computation the terms of an 8ç14 matrix of white male death rates (eight 5-year age groups for 14 years of the study period) were identically multiplied by the terms of a similar matrix of the numbers of living radiologists, and then summed over all 112 products. Death rates were calculated for leukemia for the years 1948–1961 inclusive and for multiple myeloma for the years 1949–1961, inclusive. For all other diseases shown in Table 2, the necessary data were lacking for 1948 and for 1960–1961. For these two time periods, death rates for such diseases were assumed to be the same as the corresponding rates for 1949 and for 1959, respectively. The over-all systematic error introduced by using such substituted rates in computing the expected number of deaths of radiologists is believed to be, at the most, a 3 percent under-or over-estimation of the true expected numbers.
The value of the mortality ratio for deaths from all causes in the present study population is 0.8. This agrees well with Dublin and Spiegelman’s (2 ) findings for the years 1938–1942, that the class of full-time medical specialists, and the subclasses, thereof, including radiologists, each enjoyed lower death rates from all causes of death combined than did the class of all physicians (or of all white males). (See also reference 3 ). For the present study population, it should be remembered that the value of the mortality ratio for any given cause of death is expected to be less than 1.0 if the radiologists were to enjoy as low a mortality rate for that cause as they do for all causes of death combined.
A. G. Gilliam, Blood 8, 698 (1953).
In this computation the 1949 rates were applied to the years 1948–1949, the 1951 rates to the years 1950–1954 and the 1956 rates to the years 1955–1961.
B. MacMahon and D. W. Clarke, Blood 11, 871 (1956).
For four additional deaths in the study population, the death certificates reported leukemia as a contributory rather than as the underlying or main cause of death; among these four, one was listed as lymphatic (chronicity unspecified) and the other as CLL. With these four deaths added to the 12 valid deaths, the ratio stands at one death with mention of CLL to 15 deaths without mention of this type. For still another death in the study population, the Journal death notice listed the cause as “myelogenous leukemia”; however, a certified copy of the death certificate failed to mention leukemia and ascribed death in this case to another cause.
An analysis (in preparation) of the distribution of deaths in the study population from leukemia, multiple myeloma, and aplastic anemia by age at death and year of death shows (i) that the observed number of deaths in each age group tends to exceed the expected number to the same proportionate extent that the observed total number of deaths in all age groups exceeds the corresponding expected total number, and (ii) that there is no tendency for the death rates for leukemia and multiple myeloma to decline during the 14-year period; however, for aplastic anemia, all four observed deaths occurred between 1948 and 1953, inclusive.
W. Pohl [Med. Klin. Munich 55, 1839 (1960)] reports three cases of multiple myeloma among German medical technicians with long histories of occupational exposure to radiation: however, the statistical significance of such a finding cannot be assessed because no information was provided on composition, in terms of size and age, of the living population of technicians from which the there cases were drawn.
I. Snapper, L. B. Turner, H. L. Moscovitz, Multiple Myeloma (Grune and Stratton New York, 1953).
B. MacMahon and D. W. Clarke, J. Chronic Diseases 4, 508 (1956).
R. Heyssel, A. B. Brill, L. A. Woodbury, E. T. Nishimura, T. Ghose, T. Hoshino, M. Yamasaki, Technical Report 02-59, ABCC (National Academy of Sciences—National Research Council Washington, D.C., 1959).
Deaths from Multiple Myeloma (Plasmocytoma) in Selected Countries by Sex and Age. Epidemiology and Vital Statistics Rept. No. 8 (World Health Organization, Geneva, 1955), p. 24.
In this computation it was assumed that for each 5-year age group in which leukemia deaths occurred in the Court Brown and Doll study, the ratio of leukemia deaths to multiple myeloma deaths would be the same as the corresponding ratios for British males for the years 1950–1952, the only years for which published data (see reference 24 ) are available.
M. Faber, in Transaction of the 6th Congress of the European Society of Haematology (Karger New York, 1957), p. S–211; A. Stewart, W. Pennybacker, R. Barber, Brit. Med. J. II, 882 (1962).
I thank R. Giesen, M. Hershey and S. Hillyard and especially my wife for technical assistance. I thank F. Lawler for writing the programs used to calculate the age composition of, and the expected numbers of deaths in, the study population. This work was supported in part by an institutional grant from the American Cancer Society (IN-39).
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer Science+Business Media New York
About this chapter
Cite this chapter
Lewis, E.B. (2004). Leukemia, Multiple Myeloma, and Aplastic Anemia in American Radiologists. In: Lipshitz, H.D. (eds) Genes, Development and Cancer. Springer, Boston, MA. https://doi.org/10.1007/978-1-4419-8981-9_28
Download citation
DOI: https://doi.org/10.1007/978-1-4419-8981-9_28
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4613-4746-0
Online ISBN: 978-1-4419-8981-9
eBook Packages: Springer Book Archive