CONCLUSION
The quality of information about the Aboriginal and Torres Strait Islander population is dependant on the reliable completion of the Indigenous status question. Where people with Aboriginal and/or Torres Strait Islander origin are not recorded as such in datasets, their information will not contribute to the understanding of Aboriginal and Torres Strait Islander people's outcomes. The greater the number of Aboriginal and Torres Strait Islander people who are not recorded as such in a dataset, the greater the risk that outcomes being reported are not fully representative of Aboriginal and Torres Strait Islander people.
Through linking the 2011 Census with death registrations the Indigenous status recorded on the death registration is enhanced, as more certainty is provided about the total number of Aboriginal and Torres Strait Islander deaths, mortality rates and causes of death. These are all key pieces of information to help understand where health policy needs to be directed to improve Aboriginal and Torres Strait Islander people's life expectancy, as well as to more reliably measure whether improvements in life expectancy are being made over time.
This Information Paper provides valuable insights to enhance understanding of the quality of Aboriginal and Torres Strait Islander mortality data, and confirms that some under-identification of Aboriginal and Torres Strait islander people remains in the death registrations data. Whilst Registries of Births, Deaths and Marriages across all jurisdictions continue to work independently and with the ABS to improve data quality, linking death registration data to improve the quality of Indigenous status information offers the most efficient, direct and immediate method to improve the reliability of mortality data. The Indigenous Mortality Project has contributed to more reliable Aboriginal and Torres Strait Islander life expectancy tables through the examination of differences in the reporting of Indigenous status across the datasets, and the application of adjustment factors to deaths in order to produce the best estimates of life expectancy.