Inequalities in women's health: looking at mortality differentials using an alternative approach.

Data obtained from follow up of the 1971 census sample in the Office of Population Censuses and Surveys longitudinal study of England and Wales were used to look at women's mortality differentials at ages 15-59. Women were grouped by combining information on marital state, own occupation, husband's occupation (if married), economic activity, and indicators of household wealth (housing tenure and access to a car). Large groups were found with considerable differences in mortality. High mortality was associated with working in manual occupations and living in rented housing with no car in the household. In contrast, low mortality was associated with non-manual occupations and living in owner occupied housing with a car. Among married housewives and single women these extreme groups contributed 44% of expected deaths, the disadvantaged group experiencing death rates two and a half times that of the advantaged group. Smaller differences were found among married women with an occupational class. These findings are further evidence of the "health divide" in England and Wales and show that accurately to reflect the relation between a woman's life circumstances and mortality it is necessary to utilise other measures than those based solely on occupation.