Abstract

The mortality of a census population and a prospective cohort of men employed on an antimony smelter in the north east of England was followed up from 1961-1992. The workers studied were exposed to a variety of agents including antimony and its oxides, arsenic and arsenic oxides, sulphur dioxide, and polycyclic aromatic hydrocarbons. The regional mortality rates were used to calculate expected deaths and a group of zircon sand workers employed on the site were used as a comparison group. For the census population of men working on the smelter before 1961 a significant increase in deaths from lung cancer was found (32 observed v 14.7 expected, P < 0.001). A similar excess was seen among maintenance men (12 observed v 5.3 expected P = 0.016). No such excess was found in the cohort recruited after 1960 (5 observed v 9.2 expected, maintenance workers 3 observed v 2.8 expected). There was evidence of a minimum latency period of around 20 years between first exposure and death from lung cancer. No evidence was found for a correlation between length of time worked and mortality from lung cancer. The results show that an increased risk of lung cancer existed in the workers employed before 1961, but it was not possible to attribute this excess to any particular agent. Mortality analysed by five year calendar periods of first exposure show a lessening of effect after 1955. Although the power of the study is clearly less for more recent periods of exposure the absence of any excess in the population after 1960 is encouraging.