Improving Measurement of Inequalities in Working Age Mortality and Life Expectancy using an Individual-Level Socio-Economic Index: Insights from Australia

Tim Adair , University of Melbourne
Han Li, University of Melbourne
Sergey Timonin, Australian National University
Andrew Cairns, Heriot-Watt University
Vladimir Canudas-Romo, Australian National University

Conventional analyses of socio-economic inequalities in mortality in many high-income countries are constrained by reliance on area-level indices or isolated socio-economic indicators (i.e. education). This study introduces a Socio-Economic Index of individuals to more accurately capture the magnitude of inequalities in working age mortality (25-64 years) and life expectancy at age 25 in Australia, using the 2016 Census data linked to deaths occurring in 2016-19. The Socio-Economic Index was developed from several Census variables using Multiple Correspondence Analysis - a dimensionality reduction method for categorical variables similar to Principal Components Analysis. The Index was calculated for each of 10-year age groups and converted to percentiles. Age-standardised death rates (25-64) in 2016-19 showed a very strong gradient with Index percentiles. The death rate ratio of the 1st (p1) to 100th (p100) percentile was 18.1 for males and 13.7 for females, while for p10 to p90 was 5.5 and 4.1, respectively. Death rates were particularly strongly related with the Index at younger ages. Life expectancy at age 25 differed by 20.3 years for males and 15.2 years for females between p1 and p100, and by 9.4 years for males and 5.1 years for females between p10 and p90. The new Socio-Economic Index reveals much wider inequalities in Australian working age mortality and life expectancy than previous studies. It highlights high death rates for the lowest few percentiles and advanced longevity in the highest percentiles, and demonstrates how multiple socio-economic characteristics influence longevity. Results will also be contrasted with inequalities in European countries.

See extended abstract

 Presented in Session 78. Assessing and Improving the Quality of Mortality Data