Cognitive performance and lifetime occupational exposures in a regional Australian population: a distributed lag mixtures approach

Examination of exposures over a lifetime of employment and work-related disease has largely focused on the effects of single agents. In reality, workers are exposed to mixtures of agents. In a community cohort of 1621 older participants (Hunter Community Study) with a mean age of 68.0 years (SD 6.8 yr), cognitive performance was assessed using the Audio Recorded Cognitive Screen, and lifetime occupational histories used to retrospectively estimate yearly exposures to a range of chemicals from the Finnish job-exposure matrix. This study utilised Bayesian Kernel Machine Regression - Distributed Lag Models to investigate potential time windows of increased susceptibility to cognitive effects following retirement. Furthermore, the research explored exposure-response relationships between multiple chemicals, with adjustments for covariates identified from a directed acyclic graph. Analysis identified a window of susceptibility from welding fume exposures 17 years after commencement of work at age 18. The negative effects from lead (Pb) at levels below current exposure standards were identified. Bivariate interactions from co-exposure of carbon monoxide to lead, toluene and trichloroethylene were identified, the magnitude of which may not be statistically meaningful. This study identified effects on cognition post-retirement associated with occupational exposures to a mixture of chemicals encountered throughout the participants' working life.