Mortality By Different Dimensions of Stratification – a Comparison of Education, Class, Status and Income with Finish Register Data
We use a 11% random sample from the Finish population in 1990 with an 80% oversampling of deaths. Our data includes information on education, occupation; individual taxable income, household taxable income, household disposable income; and mortality in a follow-up from 1991 to 2000. We do the same analysis ten years later with 2000 as baseline year and a follow-up from 2001-2010. We conduct separate analysis for men and women using Cox regression survival models to calculate relative risks of dying (hazard ratios) taking into account outmigration as a censoring event.
From the unique quantity and quality of the data we expect significant results on the relation between multiple dimensions of social status and mortality. This will concern gender differences, differences between income measures, age differences (35-59 and 60-85), differences between the 1990s and the 2000s, and differences to Swedish results by Torssander and Erikson. Our results will provide important new evidence on the best measurement of social status for studies of health inequalities, in particular with regard to the comparative question whether relevant dimensions of social status differ between gender, age groups, countries or periods. The question of measurement is crucial for exploring the mechanisms from social status to health. Our results will shed light on the relative importance of several suggested mechanisms and will reveal rather universal risk factors and those that differ between groups and contexts.