Assessing excess lifetime risk for disease after radiation exposure

Area of Science:

• Radiation epidemiology
• Biostatistics
• Cancer research

Background:

• Radiation exposure is a known risk factor for various cancers, including thyroid cancer.
• Quantifying excess lifetime risk (ELR) is crucial for understanding radiation's long-term health impacts.
• Assessing ELR in populations with varying ages and exposure levels presents statistical challenges.

Purpose of the Study:

• To define and evaluate the excess lifetime risk (ELR) of disease following radiation exposure.
• To propose and compare two distinct statistical methods for estimating ELR.
• To facilitate risk comparison and hypothesis testing between different exposed groups.

Main Methods:

• Developed two methods: a standard maximum likelihood (ML) approach and a Monte-Carlo (MC) simulation approach.
• Both methods enable the construction of confidence intervals for ELR estimates.
• Both methods support hypothesis testing for comparing excess risks between groups.

Main Results:

• The ML approach is computationally simple but may be inaccurate if normality assumptions are violated.
• The MC approach is consistently reliable but computationally intensive.
• Both methods were applied to pooled data from five cohorts with external radiation exposure and subsequent thyroid cancer incidence.

Conclusions:

• The study provides robust statistical tools for quantifying radiation-induced excess lifetime risk.
• The choice between ML and MC methods depends on data characteristics and computational resources.
• Accurate ELR estimation is vital for public health and radiation protection policies.