Abstract

A nested case-control study of lung cancer among men exposed to ink mist in newspaper production with rotary letterpress technology is presented. It is based within a historical cohort of 9232 printing workers in Manchester (1949-63). Men who operated newspaper rotary letterpress machines had a lung cancer standardised mortality ratio (SMR) of 179 (95% confidence interval (95% CI) 144-218) when compared with rates for England and Wales for the follow up period 1950-83. When adjustment was made for the higher rates in the local area, the SMR was reduced to 122 (95% CI 98-148). The nested case control study was based on 110 lung cancer cases (1949-86) and 316 matched controls. Duration of work in a rotary letterpress machine room was positively associated with risk of lung cancer (chi 2 linear trend = 3.30, p = 0.07); mean with 30 or more years duration of exposure had a risk of 1.73 (95% CI 0.94-3.17), relative to those with less than 20 years of exposure. Adjustment for period of first exposure in a machine room reduced the strength of the positive duration effect. The magnitude of the SMRs found in the cohort study could be explained by confounding with smoking. The duration effect seen in the case-control study, however, suggests that there may be a real effect of exposure to letterpress ink mists. This is biologically plausible, as benzo[a]pyrene, a known human carcinogen, has been found in appreciable concentrations in the atmosphere of rotary letterpress machine rooms.