Linking census and death data reveals large difference in death rates by education

4 October 2019

Researchers from ANU, together with partners from the Australian Bureau of Statistics and the Australian Institute of Health and Welfare, are among the first to analyse linked census and death registration data. They found that death rates were higher among those with lower levels of education, and these inequalities were greater than previously thought.

The study found that Australian men aged 25-84 with low education, i.e. those who did not finish high school or have another qualification, had a death rate more than double that of men with high education, i.e. those with a university degree. For women of the same age, death rates were 60% higher among those with low compared to high education.

Inequalities differed substantially by age and sex. Differences were largest among men aged 25-44 years, where death rates were almost four times higher among people with low compared to high education, and smallest in older women (65-84 years), where death rates were around 40% higher among people with low compared to high education.

Associate Professor Rosemary Korda from the Research School of Population Health led the study and is keen to point out that low education itself does not directly ‘cause’ death.

“Education is usually an indication of your socioeconomic position. It often determines your income, and access to other resources, such as healthcare, needed for good health,” says Korda

The ability to link education and death registration data is a significant step forward for Australia. Not only does it revolutionise the way we assess health inequality, it also brings us in line with other countries already doing this, including Canada and a number of European countries.

“Historically the Australian Bureau of Statistics and the Australian Institute of Health and Welfare have assessed inequality in mortality based on the area in which a person lives, such as their postcode, grouping people into areas from most disadvantaged to most advantaged” says Korda. “This approach has served us well, however, it misses much of the socioeconomic variation, and as a result, tends to underestimate inequality.”

The ability to measure inequalities at an individual, rather than an area level, opens up a number of opportunities for Australia to improve the population’s health.

“What is really exciting about this information is what we do with it next,” says Korda.

“We often talk about Australia as being a healthy nation, and by world standards we are, but this says nothing about the distribution of health. If you have a lot of inequality within the population, it indicates that there is room for improvement. If you can achieve low death rates in one group, then why not another?”

The study found, for example, that the death rate in Australians aged 25-84 years would have hypothetically been 30% lower if there were no education-related inequalities in mortality. This would be equivalent to around 27,000 fewer deaths each year.

Addressing socioeconomic inequalities in mortality is a priority in Australia. Aside from being inherently unjust, differences in death rates are also potentially avoidable. Better measurement of inequalities has the potential to improve the way we monitor population health, formulate policy and target resources to improve the health of all Australians.


* This research is published in International Journal of Epidemiology, and you can download a copy of the paper here.

** This research was undertaken through a partnership grant awarded by the National Health and Medical Research Council of Australia. Researchers from ANU partnered with the Australian Bureau of Statistics, Australian Institute of Health and Welfare and the National Heart Foundation of Australia. Participating institutions are the University of Adelaide, University of New South Wales and University of Melbourne.