A new report by the School for Social and Policy Research at Charles Darwin University has implications for the Federal Government’s commitment to close the gap in Indigenous life expectancy.
One of the report authors and demographer Professor Tony Barnes said that apart from the Northern Territory, “we don’t actually know what that gap is, and given the Government’s commitment to close this gap, measuring it should be dealt with as a national priority”.
The report, commissioned by the National Advisory Group on Aboriginal and Torres Strait Islander Health Information and Data (NAGATSIHID), found that two estimates of Indigenous life expectancy for 1996-2001 published by experts in this area were “too unstable to be reliably used for estimating Indigenous mortality” and could “seriously mislead policy makers”.
According to Professor Barnes, confusion and uncertainty has existed among governments and policy makers for the past year since two estimates of Indigenous life expectancy, one made by the Australian Bureau of Statistics (ABS) and the other developed by the University of Queensland, were published.
“These estimates differed by as much as five years, providing contradictory messages on the magnitude of the life expectancy gap,” he said.
“Using data on the Northern Territory’s Indigenous population and deaths, the most accurate in the country, the report explored the reasons for these differences and found that estimates of life expectancy can change by as much as 10 years merely by making small changes in the data sets used in the analyses.”
The report highlights the complicated challenges the ABS faces when estimating Indigenous life expectancy. It contends that the methods currently proposed for use with poor-quality Indigenous data do not work well. Instead of overcoming the data problems, the methods make them worse.
The report has recommended that in the short term the ABS use a simpler approach to derive official life expectancy estimates for Indigenous Australians. This approach would see the ABS revert to using what is called a “standard method”, after first correcting data for expected deficiencies.
“We didn’t expect the standard method to be so robust against a range of data quality errors,” Professor Barnes said. “But the more we tested the methods with the real and simulated data, the clearer it became that this is a superior method of estimating Indigenous life expectancy.
“In the longer term, there is no substitute for continuing the hard work with the ABS and the State and Territory Registrars of Births Deaths and Marriages to systematically improve the quality of Indigenous population and vital statistics data,” he said.