Indigenous status data
Although the identification of Indigenous Australians in deaths data is incomplete in all state and territory registration systems, 5 jurisdictions (New South Wales, Queensland, Western Australia, South Australia, and the Northern Territory) have been assessed by the ABS and AIHW as having adequate identification from at least 2001 onwards (AIHW 2014).
As a result, this report presents trends data for Indigenous Australians from 2001–02 to 2016–17. Mortality data for these 5 jurisdictions should not be assumed to represent the experience in other jurisdictions. Data for these 5 jurisdictions over-represent Indigenous populations in less urbanised and more remote locations.
Late and revised registration of Indigenous deaths
Issues on deaths in Western Australia that were incorrectly recorded as Aboriginal and/or Torres Strait Islander deaths between 2007 and 2009, and some Indigenous deaths that were late registrations in Queensland in 2010 are detailed in previous reports in this series.
Adjustment of injury deaths
The extent of under-identification of Indigenous deaths in death registrations has been estimated in the ABS Census Data Enhancement Indigenous Mortality Quality Study, by linking 2011 Census data with deaths registered from 10 August 2011 to 27 September 2012 (ABS 2013b). The method described in this report has been used as a basis for adjusting for under-identification of Indigenous deaths in some reports.
Indigenous injury deaths in this report are as reported, and have not been adjusted for under-identification for 2 main reasons:
- The coverage estimates are for deaths from all causes. Injury deaths differ from most deaths in the way data are collected, which might affect the number of deaths recorded as Indigenous—most deaths are certified by a doctor, while the great majority of injury deaths are reported by police to a coroner. No adjustment factors are specific to coroner-certified deaths (or injury deaths).
- Comparable adjustment factors are not available for years before 2011–12, due to differences in the method used. This report covers 1999–00 to 2016–17, for which coverage of Indigenous deaths is likely to vary.
Population data and the calculation of rates
General population
Rates were calculated using, as the denominator, the estimated resident population as at 31 December in the relevant year (for example, 31 December 2016 for 2016–17 data). Where possible, the final release of estimated resident population was used.
Indigenous population
Rates of injury death of Aboriginal and Torres Strait Islander people are provided in this report for 2001–02 to 2016–17, using data from 5 jurisdictions (New South Wales, Queensland, Western Australia, South Australia, and the Northern Territory). Data were selected on the basis of place of usual residence.
The assessments of the quality of identification of Indigenous status are affected by restrictions that jurisdictions place on what is included in the data. The assessments are subject to review, and some recent AIHW reports include New South Wales data from 1999 onwards (AIHW 2014).
For non-Indigenous Australians, population denominators were derived by subtracting the Aboriginal and Torres Strait Islander population based on the 2011 Census (ABS 2014) from the total Australian estimated resident population (of the states and territories eligible for inclusion), as at 31 December of the relevant year. Current standard practice in AIHW reports is to omit cases where Indigenous status was not stated or unknown.
Rates and change in rates
Directly age-standardised rates were calculated using the Australian population in 2001 as the standard (ABS 2002). Estimated trends in age-standardised rates were reported as average annual percentage changes, obtained using negative binomial regression modelling, performed in Stata.
Quantifying variability
The data presented in this report are subject to 2 types of statistical error—non-random and random (a third type of statistical error, sampling error, does not apply in this report, because none of the data sources used involved probability sampling).
Non-random error
Some level of non-random error is to be expected in administrative data collections, such as the NMD on which this report relies. For example, non-random error could occur if the approach to assigning cause codes to deaths were to differ systematically between jurisdictions or over time. Systems are in place to encourage uniform data collection, and coding and scrutiny of data during analysis include checking for patterns that might reflect non-random error. But some non-random error remains.
Random error
The values presented in the report are subject to random error, or variation. Variation is relatively large when the case count is small (especially if less than about 10), and small enough to be unimportant in most circumstances when the case count is larger (that is, more than a few tens of cases).
Some of the topics for which results are reported compare groups that vary widely in case count, largely due to differences in population size (for example, the population of New South Wales is more than 30 times as large as the Northern Territory population, and the population of Major cities is nearly 90 times that of Very remote areas). In this situation, year-to-year changes in counts or rates for the smaller-population groups might be subject to large random variation. There is potential to misinterpret such fluctuations as meaningful rises or falls in occurrence.
Classification of remoteness area
Remoteness area’ in this report refers to the place of usual residence of the person who died. The remoteness areas for 1999–00 to 2009–10 were specified according to the ABS ASGC, while remoteness areas for 2009–10 to 2016–17 were specified according to the ABS ASGS.
Australian Standard Geographical Classification
Australia can be divided into several regions based on their distance from urban centres. This is considered to determine the range and types of services available. In this report, remoteness area refers to the place of usual residence of the person who died, assigned on the basis of the reported statistical local area (SLA) of residence.
Remoteness categories were defined based on the Accessibility/Remoteness Index of Australia (ARIA). According to this method, remoteness is an index applicable to any point in Australia, based on road distance from urban centres of 5 sizes. The reported areas are defined as the following ranges of the index:
- Major cities (for example, Sydney, Geelong, Gold Coast): ARIA index 0–0.2
- Inner regional (for example, Hobart, Ballarat, Coffs Harbour): ARIA index 0.21–2.4
- Outer regional (for example, Darwin, Cairns, Coonabarabran): ARIA index 2.41–5.92
- Remote (for example, Alice Springs, Broome, Strahan): ARIA index of 5.93–10.53
- Very remote (for example, Coober Pedy, Longreach, Exmouth): ARIA index more than 10.53.
Most SLAs lie entirely within 1 of the 5 areas. If this was so for all SLAs, then each record could simply be assigned to the area in which its SLA lies. But some SLAs overlap 2 or more of the areas. Records with these SLAs were assigned to remoteness areas in proportion to the area-specific distribution of the resident population of the SLA, according to the 2006 Census. Each record in the set having a particular SLA code was randomly assigned to 1 of the remoteness areas present in it, in proportion to the resident population of that SLA.
Australian Statistical Geography Standard
The ASGS is a hierarchical classification system of geographical regions and consists of interrelated structures. The ASGS brings all the regions for which the ABS publishes statistics within a single framework, and has been used by the ABS to collect and disseminate geographically classified statistics from 1 July 2011. It provides a common framework of statistical geography, and enables the production of statistics that are comparable and can be spatially integrated.
Australian Statistical Geography Standard (ASGS) volume 1—main structure and greater capital city statistical areas (ABS 2010b) is the first in a series of volumes that detail the various structures and regions of the ASGS. Its purpose is to outline the conceptual basis of the regions of the main structure and of the greater capital city statistical areas, and their relationship to each other. This product contains several elements, including the ASGS manual, maps, codes, and names and the digital boundaries current for the ASGS Edition 2011 (date of effect 1 July 2011).
The digital boundaries for Volume 1 of the ASGS are the spatial units for the main structure and the Greater Capital City Statistical Areas. These spatial units are:
- Mesh Blocks
- Statistical Area Level 1 (SA1)
- Statistical Area Level 2 (SA2)
- Statistical Area Level 3 (SA3)
- Statistical Area Level 4 (SA4)
- Greater Capital City Statistical Areas
- State and Territory.
Each case is allocated to 1 of 5 remoteness areas on the basis of the place of usual residence of the person who died, according to Statistical Area Level 2 (SA2). Most SA2s lie entirely within 1 of the 5 areas. If this was so for all SA2s, then each record could simply be assigned to the area in which its SA2 lies. But some SA2s overlap 2 or more of the areas. Records with these SA2s were assigned to remoteness areas in proportion to the area-specific distribution of the resident population of the SA2, according to the 2011 Census. For death registrations, each record in the set having a particular SA2 code was assigned to 1 of the areas probabilistically, in proportion to the resident population of that SA2. The resulting values are integers. A SA2 to remoteness area map can be found at the ABS website (ABS 2012b).
Socioeconomic area
Data on socioeconomic groups are defined using the ABS’s SEIFA 2011 (ABS 2013a).
The SEIFA 2011 data are generated by the ABS using a combination of 2011 Census data, such as income, education, health problems/disability, access to internet, occupation/ unemployment, wealth and living conditions, dwellings without motor vehicles, rent paid, mortgage repayments, and dwelling size.
Composite scores are averaged across all people living in areas and defined for areas based on the Census collection districts. But they are also compiled for higher levels of aggregation. The SEIFAs are described in detail.
The SEIFA Index of Relative Socio-economic Disadvantage is one of the ABS’s SEIFA indexes. The relative disadvantage scores indicate the collective socioeconomic level of the people living in an area, with reference to the situation and standards applying in the wider community at a given point in time. A relatively disadvantaged area is likely to have a high proportion of relatively disadvantaged people. But such an area is also likely to contain people who are not disadvantaged, as well as people who are relatively advantaged.
Mortality rates by socioeconomic area were generated by the AIHW using the Index of Relative Socio-economic Disadvantage scores for the SA2 of usual residence of the person who died.
The ‘1—lowest’ group represents the areas containing the 20% of the national population with the most disadvantage, and the ‘5—highest’ group represents the areas containing the 20% of the national population with the least disadvantage. These groups do not necessarily represent 20% of the population in each state or territory.
The following labels for each socioeconomic group have been used throughout this report:
| Label | Socioeconomic area |
|---|---|
| 1—lowest | Most disadvantaged |
| 2 | Second most disadvantaged |
| 3 | Middle |
| 4 | Second least disadvantaged |
| 5—highest | Least disadvantaged |