Uncertainty in estimating probability of causation in a cross-sectional study: joint effects ofradiation and hepatitis-C virus on chronic liver disease

Abstract

Exposure to other risk factors is an important consideration in assessing the role played byradiation in producing disease. A cross-sectional study of atomic-bomb survivors suggestedan interaction between whole-body radiation exposure and chronic hepatitis-Cviral (HCV) infection in the etiology of chronic liver disease (chronic hepatitisand cirrhosis), but did not allow determination of the joint-effect mechanism.Different estimates of probability of causation (POC) conditional on HCV statusresulted from additive and multiplicative models. We therefore estimated the riskfor radiation conditional on HCV status using a more general, mixture modelthat does not require choosing between additivity or multiplicativity, or decidingwhether there is interaction, in the face of the large uncertainty. The results supportthe conclusion that POC increases with radiation dose in persons without HCVinfection, but are inconclusive regarding individuals with HCV infection, the lowerconfidence bound on estimated POC for radiation with HCV infection being zeroover the entire dose range. Although the mixture model may not reflect the truejoint-effect mechanism, it avoids restrictive model assumptions that cannot bevalidated using the available data yet have a profound influence on estimatedPOC. These considerations apply more generally, given that the additive andmultiplicative models are often used in POC related work. We therefore consider that anempirical approach may be preferable to assuming a specific mechanistic model forestimating POC in epidemiological studies where the joint-effect mechanism is indoubt.