Modeling Effects of Radiation Exposure on Disease Outcomes with Radiation-Influenced Risk Factors (RFs), using Adult Health Study (AHS) Data

• Activity Number: 233
• Type: Topic Contributed
• Date/Time: Tuesday, July 31, 2007 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Kenneth Kopecky*+ and Lori A. Williams
• Companies: Fred Hutchinson Cancer Research Center and University of Washington
• Keywords: Radiation epidemiology ; Time-dependent confounding ; Cardiovascular mortality ; Radiation Effects Research Foundation

Abstract

The AHS is a longitudinal study of radiation effects in survivors of the A-bombings of Hiroshima and Nagasaki. The cohort, initially about 20,000 survivors, has been offered medical examinations since 1958, with about 17,000 examined at least once, averaging 11 exams in 23 exam cycles to 2004. Linking AHS data to disease outcomes may allow testing if apparently radiogenic disease risks arise in part from radiation-related changes in disease RFs. For example risks of hypertension (HT) and of death from heart disease and stroke (together cardiovascular morality, CVM) increase with radiation dose. Since HT is a RF for CVM mortality, radiation's apparent effect on CVM might be due in part to radiation-related HT. Methods for estimating the radiation effect on disease risk that account for time-dependent confounding using RF data from the AHS will be discussed with the example of CVM and HT.