Ching Choi, University of New South Wales
Rob Hyndman, Monash University
Len Smith, Australian National University

COAG has undertaken to close the gap in Indigenous life expectancy within a generation, but it is difficult to monitor progress, due to changes in the method for estimating life expectancy, and the lack of reliable information on Indigenous deaths. It is believed that virtually all Indigenous deaths are included in the ABS national deaths database, but they are not properly identified.

With support from OATSIH, AIHW and COAG we are assessing the feasibility of enhancing the national mortality database by incorporating information from other sources on whether the deceased was Indigenous. In the first stage, we tested the feasibility of improving identification using other records of Indigenous deaths. In the second stage, we are testing the feasibility of improving identification using records of the Indigenous population.

In the first stage, we linked 2001-2006 deaths in the National Death Index to Indigenous deaths in the national hospital inpatient database, the residential aged care database, and the national perinatal statistics database. We found about 12% more Indigenous deaths than were originally identified, and estimated life expectancy at 67 for males and 73 for females. The results varied considerably by state.

Having demonstrated its feasibility, the enhanced database is being brought up to date and extended by linking to all deaths.

Questions remain about the extent of false positives and false negatives. In the second stage, undertaken jointly with NTSCORP, a NSW Aboriginal organisation which maintains comprehensive records used to verify land claims and claims for benefits, we are currently testing the feasibility of validating Aboriginal identification in death records against this gold standard.