A STUDY OF MORTALITY AMONG 14730 MALE WORKERS IN 12 NORWEGIAN FERROALLOY PLANTS - COHORT CHARACTERISTICS AND THE MAIN CAUSES OF DEATH

Objectives - Concern about the health hazards of exposure to workers i n the ferroalloy industry has initiated this historical cohort study. The aim was to examine the mortality pattern among male employees in 1 2 Norwegian ferroalloy plants. Methods - All men employed for at least six months who started their first employment during 1933-91 were eli gible for the cohort. Deaths observed during 1962-90 were compared wit h expected figures calculated from national mortalities. Internal comp arisons of rates were performed by Poisson regression analysis. The fi nal cohort comprised 14 730 male employees who were observed for 288 8 86 person-years. Results - Mortality from all causes of death was slig htly increased (3390 deaths, standardised mortality ratio (SMR) 1.08, 95% confidence interval (95% CI) 1.04-1.11). Regression analysis of to tal mortality showed a significant negative trend for the rate ratios with increasing duration of employment. An increased mortality was fou nd among employees in urban plants compared with employees in rural pl ants (rate ratio (RR) 1.21, 95% CI 1.13-1.29). Excess deaths from canc er (SMR 1.11) and sudden death (SMR 1.47) were found among employees w ith at least three years of employment. Mortality from accidents, pois onings, and violence was increased among all employees (SMR 1.28). Exc ess deaths from this cause were however only found for the time after the end of employment in this industry and not during employment (SMR 0.90). Conclusions - The increased mortality from cancer and sudden de ath could be related to work exposures, at least in sub-groups, and th ese results warrant further studies. The excess deaths from accidents, poisonings, and violence were probably not related to work exposures. The mortality results for short term workers and other information in dicate that systematic errors contribute to the increased overall mort ality.