Abstract

STUDY OBJECTIVE--The aim was to determine whether female manual workers have higher mortality than other women. DESIGN--This was a retrospective cohort study in which mortality was compared with that of the general female population. Main outcome measures were standardised mortality ratio (SMR) and 95% confidence intervals (CI). SETTING--Reykjavík region. PARTICIPANTS--Participants were 18,878 women, the cumulated members of a pension fund for manual workers between 1970 and 1986. MAIN RESULTS--A healthy worker effect was observed in the total cohort. The study was then restricted to those who had contributed to the pension fund any time after reaching 20 years of age, and a 10 year latency period was instituted. When analysing subcohorts by duration of employment the standardised mortality ratios for all causes of death and all cancers increased with longer employment time up to 10 years. However, the ratios were low in the group with over 10 years of employment. Those who began contributing to the fund in 1977 or later had higher mortality than those who began earlier. There was an excess of lung and bladder cancer in the total cohort and in all the subcohorts except in the group with over 10 years' employment. Mortality from accidents and suicides was in excess in all the groups. CONCLUSIONS--Mortality is high among some groups of female manual workers. A deficit was found among those with the longest employment. Differences in mortality have widened in recent years. An excess of suicides shows that women in this group have, for some reason, less will to live than other women.