DATA QUALITY AND AVAILABILITY

Almost all deaths in Australia are registered. However, Indigenous status is not always recorded, or recorded correctly. The incompleteness of Indigenous identification means that the number of deaths registered as Indigenous is an underestimate of the actual number of deaths which occur in the Indigenous population.


Coverage

The extent to which the identification of Indigenous Australians occurs in data collections is referred to as 'coverage' or 'completeness of coverage'. While there is incomplete coverage of Indigenous deaths in all state and territory registration systems, some jurisdictions have been assessed by the Australian Bureau of Statistics (ABS) as having a sufficient level of coverage to enable statistics on Aboriginal and Torres Strait Islander mortality to be produced. Table 9.1 presents the implied coverage of Indigenous deaths for each of the jurisdictions (except Tasmania and the Australian Capital Territory) for the period 2001-2005. Implied coverage is calculated by dividing the number of Indigenous deaths registered for the period 2001-2005 by the number of projected Indigenous deaths for the same period (ABS 2007c). The projected deaths are obtained from the low series of population projections in Experimental Estimates and Projections, Aboriginal and Torres Strait Islander Australians 1991 to 2009 (ABS 2004c).

9.1 INDIGENOUS DEATHS, implied coverage - 2001-2005
	Deaths registered as Indigenous	Projected Indigenous deaths	Implied coverage of Indigenous deaths(a)
States/Territories	no.	no.	%
New South Wales	2 479	5 469	45
Victoria	364	1 182	31
Queensland	2 822	5 430	52
South Australia	642	1 011	64
Western Australia	1 860	2 658	70
Tasmania	123	(b). .	(b). .
Northern Territory	2 229	2 430	92
Australian Capital Territory	36	(b). .	(b). .
Australia(c)	10 564	18 973	56
. . not applicable
(a) Calculated as the ratio of deaths registered to projected Indigenous deaths.
(b) Not calculated due to small numbers.
(c) Includes Other Territories.
Source: ABS 2007c

The jurisdictions assessed as having a sufficient level of coverage are Queensland, Western Australia, South Australia and the Northern Territory. These data have been combined for 2001-2005 to enable an analysis of Indigenous mortality. The less than complete coverage of Indigenous deaths in these four jurisdictions means the aggregate analyses presented in this chapter, which compare Indigenous and non-Indigenous mortality statistics, will underestimate the actual mortality experience of Indigenous people.

Longer term mortality trends discussed in this chapter are based on an analysis of data from three jurisdictions - Western Australia, South Australia and the Northern Territory - the only jurisdictions with 15 years of reasonable coverage of Indigenous deaths registrations. Indigenous mortality rates have been compared with the mortality rates of 'other Australians' (which include deaths of non-Indigenous people and deaths for which Indigenous status was not stated). This is due to a late inclusion of a 'not stated' category of Indigenous status in 1998, before which 'not stated' responses were included with non-Indigenous deaths.


Year of registration or year of occurrence

Deaths can be analysed by year of occurrence of death or by year of registration of death. While the majority of deaths are registered in the year they occur, some of those registered in a given year occurred in previous years, and some which occurred in one year are not registered until subsequent years. Delays in registration can occur when deaths are subject to the findings of a coroner and are more common when the death occurs in a remote area. Late registrations are more common among deaths of Indigenous than non-Indigenous people. For example, in Queensland, Western Australia, South Australia and the Northern Territory for deaths of non-Indigenous Australians that occurred in 2004, 95% were registered in 2004 while 5% were registered in 2005. For Indigenous deaths, the corresponding figures were 88% in 2004 and 12% in 2005. The proportion of deaths that occurred and were registered in the same year has remained relatively stable over the period 1991-2004 for both Indigenous and other Australians (table 9.2).

While late registrations are more common among Indigenous Australians, table 9.2 shows that there is little difference between the number of deaths registered in a given year and the number of deaths that occurred in the same year for both Indigenous and other Australians. This is because for each year, the number of deaths that are not registered in the year they occur are offset by deaths that occurred in previous years but were registered late. This indicates that analysis of mortality data using year of registration of death will produce similar results to analysis using year of occurrence of death for the period of interest.

The analyses of deaths reported in this chapter are based on year of registration of death for the period 2001-2005 (1991 to 2005 for trends analysis). Year of registration of death was used so that deaths for the most recent year of data available can be included in the analysis and to be consistent with the publication of mortality statistics by the ABS. It should be noted that the 2005 edition of this publication used year of occurrence of death for all years of analysis except for the latest year of available data for which year of registration of death was used. Rates published in this report may therefore differ slightly from those published in the previous edition for comparable years of data.

9.2 PROPORTION OF DEATHS THAT WERE REGISTERED IN THE YEAR THEY OCCURRED(a) - 1991-2004
	Indigenous	Other
	%	%
1991	84.5	95.3
1992	85.3	94.7
1993	85.9	95.0
1994	84.7	95.4
1995	86.8	94.3
1996	80.4	95.4
1997	84.3	95.1
1998	85.1	95.0
1999	83.3	95.0
2000	84.7	94.9
2001	84.2	94.4
2002	85.8	95.0
2003	86.2	95.3
2004	88.3	95.1
(a) Data for Qld, WA, SA and NT combined. Deaths are based on year of registration of death.
Source: AIHW National Mortality Database

Cause of death statistics in this chapter are based on the tenth revision of the International Classification of Diseases (ICD-10). Mortality coding using ICD-10 was introduced in Australia for deaths registered from 1 January 1997. All rates and ratios derived in this chapter are calculated using the ABS 2001 Census-based experimental Indigenous population projections (low series).

Previous Page

Next Page