The effect of occupational type of household on mortality in Japan.

The mortality of the Japanese population was examined focusing on the effect of household head occupation rather than individual occupations. The 'healthy worker effect' (HWE) which implies lower mortality observed in a working than in a non-working populations has been successively observed in comparative studies on mortality in different occupation populations. In this study we report quite similar results analysing the effect of household head occupation. We point out the possibility that HWE should have a component which is shared with the household members if the worker is a household head. We call this component the 'indirect effect'. The results obtained in this study are as follows. The employee household had the lowest mortality followed by self-employed, agriculture and 'others' households. The self-employed household also had a low standardized mortality ratio (SMR) except for diabetes in male. The SMR of the agriculture household was significantly high especially for those above 65 years of age and for suicide in both male and female of age 45 years old and over but low for cirrhosis of liver in male and tuberculosis in both male and female. The SMR of 'Others' household was the highest among the occupational type of households especially between 20-64 years. The similarity in SMR trend among male and female rather than in occupational type of household, despite the fact that males are more likely to be working than females, suggests the presence of indirect effect. Though it is difficult to distinguish direct and the indirect effects completely, we also discuss a method that enables us to distinguish the two effects more effectively.