[Socioeconomic determinants of mortality for very old men in Austria]

Background: Although socioeconomic differences in the mortality of middle-aged persons is well-researched, little is known about socioeconomic inequalities in mortality in the very old population. Surveys and population census follow-up studies in this age group are associated with reliability and validity problems.

Objective: The aim of the study was to investigate, by linking statistical information from register data, to what extent differences in mortality exist among very old Austrian men by socioeconomic variables (e.g. education and income) and if these can be explained by differences in health status.

Material and methods: The data records on men aged 80-99 years counted in the Austrian register-based population census from 2011 were augmented by information from tax and social security data and merged with deaths in a 5-year follow-up period. Relative mortality risks were estimated by Cox regression. The health status was operationalized by the standardized level of care.

Results: Even in very old men significant differences in mortality exist depending on the socioeconomic status. A systematic effect was found for education and a poverty effect for income. When controlling not just for age but also for the standardized level of care, no socioeconomic disparities in mortality were observed. The higher mortality risk for less educated and low income very old men is thus a result of a poorer health status.

Conclusion: At ages over 80 years the health status is the essential influencing factor on mortality. Measures to reduce socioeconomic disparities in mortality must therefore focus on earlier phases in life, in order to reduce the extent of care dependency among socially disadvantaged groups.