Lack of concordance between reported lung-cancer risk levels and occupation-specific diesel-exhaust exposure

Some epidemiologic studies of occupational groups show associations between surrogates of diesel-exhaust exposure and increased lung cancer, but causality has not been established. For several occupational groups, we compared information on the reported lung-cancer risk with estimated diesel-exhaust concentrations. Although none of the epidemiologic studies had concurrent measurements of diesel-exhaust concentrations, such data are available from more contemporary studies. Measurements of particle concentrations yield three, overlapping "order-of-magnitude" groups: Truck drivers, dock workers, railroad workers (excluding shop workers, hostlers), 5-100. mu g/m(3). Bus garage workers, railroad shop workers, and hostlers, 50-700 mu g/m(3). Underground miners, 500-2000 mu g/m(3). Bhatia ei ai. (1998) conducted a meta-analysis of lung-cancer risk it with occupational exposure to diesel exhaust and found an overall relative risk (RR) value of 1.33, with a range of 1.11 to 1.49 in the subanalysis by occupation. If diesel exhaust were causally increasing lung-cancer risk by 50% for low-exposure occupations (for example, truck drivers, RR = 1.49), then the lung-cancer risk in a more heavily exposed population (for example, railroad shop workers) should be much higher however, the shop workers experienced an RR of around 1.0 (Crump, 1999; HEI, 1999). Similarly, the added lung-cancer risk for bus garage workers (RR = 1.24) is hail that of truck drivers, but diesel-exhaust concentrations were considerably higher for garage workers. There is an approximately two orders of mag nitude difference in potential diesel-exhaust exposure, yet, the epidemiologic relative risks cluster in a narrow range. Such a lack of concordance between reported lung-cancer risk and estimated exposure argues against a causal role for diesel exhaust in the epidemiologic associations.