Sensitivity Analysis of Median Lifetime on Radiation Risks Estimates for Cancer and Circulatory Disease amongst Never-Smokers

*Lori J Chappell, USRA Division of Space Life Sciences 
Francis A Cucinotta, NASA Lyndon B. Johnson Space Center 

Keywords: transfer models, multiplicative, additive

Radiation risks are estimated in a competing risk formalism where age or time after exposure estimates of increased risks for cancer and circulatory diseases are folded with a probability to survive to a given age. The survival function and disease rates change with calendar year, gender, smoking status and other demographic variables. An outstanding problem in risk estimation is the method of risk transfer between exposed populations and a second population where risks are to be estimated. Approaches used to transfer risks are based on the multiplicative risk transfer models where risks are proportional to background disease rates or the additive risk transfer model where risks are independent of background rates. In addition, a mixture model is often considered where the multiplicative and additive transfer assumptions are given weighted contributions. We studied the influence of the survival and disease characteristics from the population of interest on the risk of exposure induced cancer and circulatory disease. Emphasis was placed on discovering how estimated risks would compare between the average U.S. population and a healthier U.S. population that would have a longer lifespan due to factors such as refraining from tobacco use leading to reduced cancer and circulatory disease risks for never-smokers (NS) compared to the average U.S. population. Risks are estimated to be reduced between 30% and 60% dependent on model assumptions. Lung cancer is the major contributor to the reduction for NS, with additional contributions from cancers of the stomach, liver, bladder, oral cavity, esophagus, and leukemia. Results for circulatory disease risks are influenced by model assumptions at older ages (>80 y) to a greater extent than cancer. Risk estimates could be improved with a greater understanding of risk transfer models and elucidating the role of radiation on the various stages of disease formation.