Mortality 1967-1977 of industrial workers and ex-workers from the British steel industry: further analyses

Further analyses have been carried out on mortality from selected causes (all causes; non-malignant respiratory disease; lung cancer; stomach cancer; ischaemic heart disease) in a large-scale though methodologically suspect cohort of 86 548 steel workers in Britain, of whom 94% were followed up over a 10-year period 1967-1977. These analyses extend those reported previously by Jacobsen et al (1982), where the study cohort and its reliability have been described in greater detail. The objectives have been to evaluate, insofar as the data allow, whether steel workers in general are at special risk of mortality from non-malignant respiratory disease; and to identify what subgroups, if any, may be at special risk from these and other causes.Expected deaths were based on the general population death rates, taking account of age-group, calendar year of follow-up, and regional differences within Britain. Results are reported by age-group and by region.Associated analyses of mortality in relation to occupation during 1971/73 have been carried out on a subgroup of 16 120 men over a shorter (43-month) follow-up period. These are men who had worked in the industry for at least 10 years; they are likely to be unrepresentative of steel workers generally in some other respects also. Results are reported by various combinations of age, occupation and region, and include’results from regression modelling of the standardised death rates.The overall (all-cause) SMR was 78.5, based on 9299 deaths.The SMR for non-malignant respiratory disease was 72, based on 614 deaths which occurred almost exclusively at ages 45-74. The study therefore gives no evidence that British steel workers in general have during the 1970s experienced excessive death rates from bronchitis, emphysema and asthma relative to the general population.In the subgroup analysed for occupational differences, the SMR for non-malignant respiratory disease was lower, at 52. Against this low background level, more detailed analyses identified suggestively higher SMRs in the West Midlands region. The SMR of 94 in blastfurnace or sinter operatives was also higher than average, but the difference was not statistically significant.The overall lung cancer SMR was 89.5, with some variations by region, and pointers that any excess occurred in younger men. Jacobsen et al had indicated excess lung cancer mortality among blastfurnace and sinter men. The present analyses found a statistically significant twofold relative risk in blastfurnace/sintermen compared with the other occupations studied, and indicated that there may be regional variations in these risks. Although the subgroup studied for occupational differences may be unrepresentative, this finding underlines a possible problem. The overall SMR for ischaemic heart disease (IHD) was 80, based on 3331 deaths.Results showed a clear and consistent trend in SMR with age, the highest SMRs being in the youngest age groups. (The SMR at ages less than 45 was almost exactly 100 whereas mortality of an industrial cohort is expected to be less than in the general population.) Limited occupational results indicated that, again,blastfurnace/sinter operatives were at greater risk than most other steel workers, but the differences were not statistically significant, and the comparison varied by region. Some’Other suggestivevpatterns’by: occupation were identified.The overall SMR for stomach cancer was 83. Analyses by occupation were based on 18 deaths only, giving very little information from which to draw conclusions. There was however no suggestion that blastfurnace workers were at special risk from stomach cancer.In conclusion, the results provide a strong reassurance that there has not been an ‘epidemic’ of non-malignant respiratory disease through the 1970s among workers in the British steel industry, while at the same time providing information on possible problems relating to lung cancer and IHD which is helpful in guiding further research. Extension of the mortality follow-up of some of the present study may be one useful part of that further research.