Technical notes

Deaths data

Information about deaths is collected on death certificates and certified by either a medical practitioner or a coroner. Registration of deaths is compulsory in Australia and is the responsibility of each state and territory Registrar of Births, Deaths and Marriages (RBDM) under jurisdiction-specific legislation. Additional information about coroner-certified deaths is maintained by the National Coronial Information System (NCIS). On behalf these agencies (RBDM & NCIS), deaths data are assembled, coded and published by the Australian Bureau of Statistics. Causes of death are coded by the ABS to the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10). The Australian Coordinating Registry undertakes the coordination and management of the data on behalf of the RBDMs, and supplies the AIHW with the Cause of Death Unit Record File. The data are maintained by the AIHW in the National Mortality Database (NMD).

For more information about Australian mortality data, including scope and coverage of the collection and a quality declaration, please refer to Deaths, Australia and Causes of death, Australia available from the ABS website.

The data used in this report was extracted from the NMD. The NMD comprises two sets of causes:

  • record data – which has causes of death that have been coded to an ICD-10 code based on the standard international coding rules. These data contain the underlying cause and the associated causes (that is, all conditions that were not the underlying cause).
  • entity data – which has the causes of death in the order and location that they were recorded on the Medical Certificate of Cause of Death. These data contain information about the location (Part I or Part II) on the Medical Certificate of Cause of Death. The location of the cause can be used to identify whether the cause was in the chain of events leading directly to death (Part I) or whether it was a cause that significantly contributed to the death (Part II).

For more information on how deaths data is derived, see Where do death statistics come from? in What do Australians die from?

Year of occurrence and year of registration

Trends may be presented by year of occurrence of death or year of registration of death.

Using year of occurrence of death is common when the exact time period of the death is important (for example, seasonal deaths) however the latest data available underestimates the occurrence of recent deaths due to a lag in registration.

For this reason, year of registration of death is often used to allow the latest year of data to be compared to previous years.

In both cases the latest year of data are coded with preliminary causes of death information and may underestimate causes of death that are usually certified by a coroner (for example, external causes of death including suicide).

Unless otherwise specified, deaths statistics presented here are based on year of registration of death. Previously data were presented by the reference year in which the death was received and processed by the Australian Bureau of Statistics. As a result, historical data in this report may not equal previously published data for these years. For more detail, refer to the ABS Data release: Presentation of mortality data in Causes of death, Australia, methodology.

For more information on how deaths are registered, coded and updated, see Deaths data.

Cause of death terminology

Death statistics are often based on the underlying cause of death only – that is, the disease or injury that initiated the train of events leading directly to death, or the circumstances of the accident or violence that produced the fatal injury. Analysis of the underlying cause of death is important because it points to where interventions can be targeted.

Multiple cause of death statistics are based on all the causes, conditions and health events listed on the Medical Certificate of Cause of Death (MCCD). Using the information supplied on the MCCD, causes involved in death can be considered as direct (health events that arise from the underlying cause) or contributory (conditions that significantly contributed to the death but were not in the chain of events leading to death). From a public health perspective, understanding the different roles played by common causes of death highlights the extent of their involvement in causing death. This can inform different ways to target prevention strategies in addition to what is known about the underlying causes of death.

For more information on how these causes are derived, see Where do death statistics come from? in What do Australians die from?

Further descriptions of cause of death terminology are available in the Life expectancy & deaths glossary.

Classifying causes of death

Leading causes of death are determined by grouping specific causes of death and counting the number of deaths assigned to each cause group. Over 14,000 specific causes of illness, injury and death are presented in the International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10). Grouping causes of death in a meaningful way is a useful measure of population health. It is of most value when making comparisons over time or between population groups. Changes in the pattern of causes of death can result from changes in behaviours, exposures to disease or injury, and social and environmental circumstances, as well as from data coding practices.

Cause lists for different purposes

There is no standard method for grouping causes to identify leading underlying causes of death, however, the AIHW follows the recommendations of the World Health Organization (WHO) (Becker et al. 2006) with minor modifications to suit the Australian context. This grouping is a mix of ICD chapters, blocks and specific diseases to maximise information, separate out ill-defined causes and highlight health priority areas. The leading underlying causes of death presented in this report are classified using an AIHW-modified version of Becker et al (2006).

 For the analyses of multiple causes of death, the ICD-10 codes for each cause (underlying, direct and contributory) in each death were mapped to the specifically developed multiple cause of death cause list (see Cause list). The cause list was designed to incorporate common acute direct causes and risk-factor related causes. Due to different purposes, these cause groupings can differ to the modified Becker cause list which is designed to elicit the leading underlying causes of death in a manner that lends to international comparability. The purposes and examples of differences between the classification for identifying the leading underlying causes and leading multiple causes is described below.

Differences in cause of death analytical groupings

Methods for identifying the leading underlying causes of death are well established and internationally consistent. The AIHW for example, uses a classification (suggested by Becker and colleagues) to identify the leading underlying causes of death. Aside of minor modifications by the AIHW to enable reporting relevant to the Australian public health context, this approach largely facilitates international comparisons of causes of death. An example AIHW modification is asthma. Given the public health importance of asthma, the codes for identifying deaths due to asthma are specifically extracted from the broader Becker grouping of chronic respiratory conditions.

Using a multiple cause approach to identify causes of death provides a more complete picture of the health status of Australians. However, these methods require a finely detailed cause list to identify, in addition to the underlying causes, conditions that commonly occur as consequences of the underlying cause (direct causes) or which are coexisting or contextual causes (contributory causes). Identifying the direct and contributory causes highlights a broader range of conditions which can be used to

  • emphasise the involvement of potentially preventable complications of the underlying cause, such as sepsis, to inform strategies to minimise their occurrence.
  • show the contribution of preventable causes and modifiable risk factor-related conditions, such as hypertension, to provide additional focus for prevention strategies.

It is important to note that the ‘Becker’ classification is specifically designed to identify the leading underlying causes of death. A more detailed grouping of conditions is required to identify common direct and contributory causes.

Some conditions, for example, sepsis, hypertension and pneumonitis, are not considered valid underlying causes of death. Theoretically, there is another (underlying) condition that led to sepsis or pneumonitis. Therefore, in assessing only the underlying cause, there is little need to draw out these sorts of conditions.

For assessment of the multiple causes, a detailed cause list was developed to facilitate identification of the common direct and contributory causes. Lower respiratory infections, pneumonitis and acute renal failure are common conditions that arise as consequences of disease and injuries, while hypertension (a preventable condition) and chronic kidney disease, commonly occur as significant contributors to death.

For example, influenza and lower respiratory infections were each extracted from the broader Becker grouping of Influenza and pneumonia to identify common direct causes. Similarly, acute renal failure and chronic kidney disease were extracted from the broader Becker grouping of diseases of the urinary system to identify these as direct and contributory causes, respectively.

The multiple cause of death cause list has undergone some modification since reporting What do Australians die from? This has resulted in greater consistency between the leading underlying cause and the multiple cause reporting. Some examples of changes in the classification of health conditions include refining the cause codes for:

  • Cerebrovascular diseases from G45, I60–I69 to I60–I69
  • Diabetes to include the codes related to renal complications (previously aligned to chronic kidney disease)
  • Dementia to align with Deaths in Australia reporting (by excluding G31)
  • Falls to now include X59.0.

A new cause has been included — Other degenerative neurological diseases defined by the ICD10 codes F02, G31, G32.

Cause list: Cause group and cause name, International Classification of Disease (ICD-10) code inclusions for Multiple Cause of Death analysis

Calculating and interpreting life expectancy by SEIFA

Life expectancy estimates by socioeconomic area presented in this report are based on life tables by Socio-Economic Indexes for Areas (SEIFA) developed by the ABS using similar methods to life tables by SEIFA produced by the ABS. While the ABS methods combine 3 years of deaths to produce life tables, the AIHW estimates use single year age-specific death rates to produce an annual life expectancy estimate.

The AIHW has presented life expectancy estimates by socioeconomic area by population-based quintiles (distributing areas of socio-economic disadvantage into 5 groups). Estimates published by the ABS in Life expectancy by Socio-Economic Indexes for Areas (SEIFA), 2011-2024 are presented by SEIFA deciles (distributing into 10 groups; ABS 2026). The quintiles range from Quintile 1, representing the most disadvantaged areas to Quintile 5, representing the least disadvantaged areas. 

Abbreviations and symbols

Abbreviation or symbol

In full or meaning

ABS

Australian Bureau of Statistics

AIHW

Australian Institute of Health and Welfare

COPD

Chronic obstructive pulmonary disease

HALE

Health-adjusted life expectancy

ICD

International Statistical Classification of Diseases and Related Health Problems

ICD-10

International Statistical Classification of Diseases and Related Health Problems, 10th Revision

OECD

Organisation for Economic Co-operation and Development

PYLL

Potential years of life lost

WHO

World Health Organization

%

Per cent


Glossary