A note on sex-specific mortality and social class in Sydney, 1979 to 1983.

Simple indicators such as education, income, occupation, and employment status are often used to adjust for social class in studies of morbidity or mortality. The effect of social class on such outcomes can be different for men and women. In population-based studies, residential area may be the only means of assessing social class. This study investigated sex differences in the association between mortality between 1979 and 1983 and several area indices of social class--percentage with a degree, percentage professional, percentage blue collar, percentage unemployed, percentage with annual income greater than $40,000, and the composite indicator constructed by the Australian Bureau of Statistics. Thirty-seven Sydney statistical local areas were ranked for men and women separately according to the indicators and standardised mortality ratios. Correlations between the indicators and mortality were all negative and stronger for men than for women. The composite indicator and percentage unemployed correlated most strongly with mortality. There was a moderate correlation between the remaining indicators and mortality for men, but only a weak correlation for women. The results suggest that the use of simple area indicators may be inadequate to classify the social class of women, and that the unemployment rate in those over 35 years of age may be a reasonable alternative to the composite indicator for the purpose of social class adjustment in mortality studies.