Risk Estimation in the NCI Study of Thyroid Disease in Kazakhstan: Methods and Findings to Account for Complex Uncertainty in Radiation Dosimetry

• Activity Number: 303
• Type: Topic Contributed
• Date/Time: Tuesday, August 11, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Section on Teaching of Statistics in the Health Sciences
• Author(s): Deukwoo Kwon* and F. Owen Hoffman and Brian E. Moroz and Steven Simon
• Companies: University of Miami and Oak Ridge Center for Risk Analysis and National Cancer Institute and National Cancer Institute
• Keywords: Bayesian dose-response model ; Bayesian model averaging ; radiation epidemiology ; shared uncertainty ; two-dimensional Monte Carlo

Abstract

Most conventional risk analysis methods rely on a single best estimate of exposure per person which does not allow for adjustment for exposure-related uncertainty. Here, we propose a Bayesian model averaging method to properly quantify the relationship between radiation dose and disease outcomes by accounting for shared and unshared uncertainty in estimated dose. The exposure model used in this work is taken from a study of the risk of thyroid nodules among a cohort of 2,376 subjects following exposure to fallout resulting from nuclear testing in Kazakhstan. Our Bayesian risk analysis method utilizes multiple realizations of sets (vectors) of doses generated by a two-dimensional Monte Carlo simulation method that properly separates shared and unshared errors in dose estimation. We assessed the performance of our method through an extensive series of simulation tests and comparisons against conventional regression risk analysis methods. An evaluation of the dose-response using our method is presented for an epidemiological study of thyroid disease following radiation exposure.