You need JavaScript to view this

Radiation effects on cancer mortality among A-bomb survivors, 1950-72. Comparison of some statistical models and analysis based on the additive logit model

Journal Article:

Abstract

Various statistical models designed to determine the effects of radiation dose on mortality of atomic bomb survivors in Hiroshima and Nagasaki from specific cancers were evaluated on the basis of a basic k(age) x c(dose) x 2 contingency table. From the aspects of application and fits of different models, analysis based on the additive logit model was applied to the mortality experience of this population during the 22year period from 1 Oct. 1950 to 31 Dec. 1972. The advantages and disadvantages of the additive logit model were demonstrated. Leukemia mortality showed a sharp rise with an increase in dose. The dose response relationship suggests a possible curvature or a log linear model, particularly if the dose estimated to be more than 600 rad were set arbitrarily at 600 rad, since the average dose in the 200+ rad group would then change from 434 to 350 rad. In the 22year period from 1950 to 1972, a high mortality risk due to radiation was observed in survivors with doses of 200 rad and over for all cancers except leukemia. On the other hand, during the latest period from 1965 to 1972 a significant risk was noted also for stomach and breast cancers.  More>>
Authors:
Otake, M [1] 
  1. Hiroshima Univ. (Japan). Faculty of Science
Publication Date:
Dec 01, 1976
Product Type:
Journal Article
Reference Number:
AIX-08-332454; EDB-78-011207
Resource Relation:
Journal Name: J. Radiat. Res.; (Japan); Journal Volume: 17:4
Subject:
63 RADIATION, THERMAL, AND OTHER ENVIRON. POLLUTANT EFFECTS ON LIVING ORGS. AND BIOL. MAT.; A-BOMB SURVIVORS; DELAYED RADIATION EFFECTS; CARCINOMAS; RADIOINDUCTION; LEUKEMIA; MORTALITY; STATISTICAL MODELS; DOSE-RESPONSE RELATIONSHIPS; HIROSHIMA; NAGASAKI; RBE; STATISTICS; ASIA; BIOLOGICAL EFFECTS; BIOLOGICAL RADIATION EFFECTS; DISEASES; HEMIC DISEASES; JAPAN; MATHEMATICAL MODELS; MATHEMATICS; NEOPLASMS; RADIATION EFFECTS; 560151* - Radiation Effects on Animals- Man
OSTI ID:
7083139
Country of Origin:
Japan
Language:
English
Other Identifying Numbers:
Journal ID: CODEN: JRARA
Submitting Site:
INIS
Size:
Pages: 262-321
Announcement Date:

Journal Article:

Citation Formats

Otake, M. Radiation effects on cancer mortality among A-bomb survivors, 1950-72. Comparison of some statistical models and analysis based on the additive logit model. Japan: N. p., 1976. Web.
Otake, M. Radiation effects on cancer mortality among A-bomb survivors, 1950-72. Comparison of some statistical models and analysis based on the additive logit model. Japan.
Otake, M. 1976. "Radiation effects on cancer mortality among A-bomb survivors, 1950-72. Comparison of some statistical models and analysis based on the additive logit model." Japan.
@misc{etde_7083139,
title = {Radiation effects on cancer mortality among A-bomb survivors, 1950-72. Comparison of some statistical models and analysis based on the additive logit model}
author = {Otake, M}
abstractNote = {Various statistical models designed to determine the effects of radiation dose on mortality of atomic bomb survivors in Hiroshima and Nagasaki from specific cancers were evaluated on the basis of a basic k(age) x c(dose) x 2 contingency table. From the aspects of application and fits of different models, analysis based on the additive logit model was applied to the mortality experience of this population during the 22year period from 1 Oct. 1950 to 31 Dec. 1972. The advantages and disadvantages of the additive logit model were demonstrated. Leukemia mortality showed a sharp rise with an increase in dose. The dose response relationship suggests a possible curvature or a log linear model, particularly if the dose estimated to be more than 600 rad were set arbitrarily at 600 rad, since the average dose in the 200+ rad group would then change from 434 to 350 rad. In the 22year period from 1950 to 1972, a high mortality risk due to radiation was observed in survivors with doses of 200 rad and over for all cancers except leukemia. On the other hand, during the latest period from 1965 to 1972 a significant risk was noted also for stomach and breast cancers. Survivors who were 9 year old or less at the time of the bomb and who were exposed to high doses of 200+ rad appeared to show a high mortality risk for all cancers except leukemia, although the number of observed deaths is yet small. A number of interesting areas are discussed from the statistical and epidemiological standpoints, i.e., the numerical comparison of risks in various models, the general evaluation of cancer mortality by the additive logit model, the dose response relationship, the relative risk in the high dose group, the time period of radiation induced cancer mortality, the difference of dose response between Hiroshima and Nagasaki and the relative biological effectiveness of neutrons.}
journal = {J. Radiat. Res.; (Japan)}
volume = {17:4}
journal type = {AC}
place = {Japan}
year = {1976}
month = {Dec}
}