Technical notes

Data source: Australian Institute of Health and Welfare National Mortality Database

Cause of Death Unit Record File data are provided to the Australian Institute of Health and Welfare (AIHW) by the Registrars of Births, Deaths and Marriages and the National Coronial Information System (managed by the Victorian Department of Justice and Community Safety) and include cause of death coded by the Australian Bureau of Statistics (ABS). The data are maintained by the AIHW in the National Mortality Database.

Analysis was performed by year of registration of death. Deaths registered in 2020 and earlier are based on the final version of cause of death data; deaths registered in 2021 are based on the revised version; deaths registered in 2022 and 2023 are based on the preliminary version. Revised and preliminary versions are subject to further revision by the Australian Bureau of Statistics.

Causes of death are based on underlying causes of death and classified using an AIHW-modified version of Becker R, Sivli J, Ma Fat, L'Hours A, Laurenti R. 2006. A method for deriving leading causes of death. Bulletin of the World Health Organization 84: 297–304.

For more information on the AIHW National Mortality Database, see Deaths data at AIHW.

The data quality statements underpinning the AIHW National Mortality Database can be found in the following ABS publications:

Data source: Australian Bureau of Statistics 2022 National Health Survey

About the data and factors influencing disease estimates

The 2022 National Health Survey (NHS) was conducted by the ABS from January 2022 to April 2023. Data was collected from approximately 13,100 households around Australia. The survey collected a range of information about the health of Australians including the prevalence of long-term health conditions, health risk factors, demographic and socioeconomic characteristics and self-reported health status.

For further information on how the NHS data are collected, see ‘How the data is collected’ in the ABS 2022 National Health Survey methodology.

While the accuracy of estimates may vary across conditions self-reported to the NHS, these data importantly enable us to examine the co-occurrence of a wide range of long-term health conditions across the Australian population to produce estimates of multimorbidity. This is not possible using separate data sources.

The following sections provide more information on factors influencing disease estimates.

ABS 2022 NHS long-term health conditions: definitions and factors influencing disease estimates

A long-term health condition is defined in the NHS as a condition which was current at the time of interview and had lasted, or was expected to last, 6 months or more.

Information on long-term health conditions was collected using:

  • condition-specific modules to capture detailed information on a selection of conditions associated with substantial health impacts: asthma, cancer, cardiovascular disease and diabetes.
  • questions where respondents were prompted to review lists of conditions and identify each condition they had.
  • an open-ended question to capture any other conditions not already captured.

The self-reported nature of the NHS data relies on survey respondents providing accurate information. Conditions that are not specifically prompted for, that are undiagnosed or asymptomatic in early stages are likely to be under-reported. Potential overdiagnosis of conditions may also affect results based on self-reported data.

Furthermore, an individual’s tendency to self-report a condition can differ based on characteristics such as their age and cultural background and will influence results. Whether a condition is self-reported may also be influenced by characteristics of the condition for example, whether it is episodic (such as migraine) or persistent in nature (such as diabetes).

NHS prevalence estimates based on self-reported information may therefore differ to estimates based on diagnostic surveys or surveys collecting biomedical samples (such as blood and urine) for testing.

Scope of the ABS 2022 NHS and influence on disease estimates

The ABS 2022 NHS is a community-based survey and does not include information from people living in residential aged care facilities, hospitals or prisons. This will exclude people likely to experience certain long-term health conditions such, as dementia, and may underestimate the prevalence of multimorbidity.

The NHS does not capture residents of Very remote areas and discrete First Nations communities. While this is unlikely to affect national estimates, it is not possible to report estimates of multimorbidity among First Nations people using the NHS.

For further information, see ‘Scope’ in the ABS 2022 National Health Survey methodology. For further information about chronic conditions among First Nations people, see National Aboriginal and Torres Strait Islander Survey 2022-23.

Conditions included in NHS multimorbidity estimates and how they are counted

A list of 72 selected long-term health conditions was used in analysis of ABS 2022 NHS data (Table 1). The 72 conditions are a subset of chronic conditions from the Australian Burden of Disease Study (ABDS) disease list (AIHW 2021) that could be reasonably identified in the 2022 NHS survey data.

The ABDS disease list was used as the conditions have been assessed to be of substantial burden to at least one age group or sex, or to be of significant policy interest. The list includes conditions commonly diagnosed among younger people, such as attention deficit hyperactivity disorder (ADHD), as well as conditions more common among older people such as deafness and hearing loss. This supports the analysis of multimorbidity among people of all ages.

The severity or impact of the 72 conditions may vary depending on the condition, condition severity or stage of progression. For example, asthma may be self-reported where it is mild, moderate or severe although it is not possible to distinguish these cases in the NHS data. Where possible, conditions that have been corrected, such as vision conditions corrected with glasses, are excluded from analysis. It was not possible to exclude hearing conditions corrected with hearing aids.

For more information on the selection and classification of diseases see, Australian Burden of Disease Study: Methods and supplementary material 2018.

Determining multimorbidity

The 72 conditions are counted individually to determine multimorbidity (whether a person has 2 or more conditions). For example, mental health conditions such as depression, anxiety and drug and alcohol use disorders are counted individually so that an individual with more than one of these long-term health conditions is considered to have multimorbidity.

Condition group multimorbidity

For analysis by condition group, groupings are based on the Australian burden of disease condition groups.

The 72 conditions are counted individually to determine multimorbidity in analysis by condition group so that people with 2 conditions in the same group are counted as having multimorbidity.

Derived variables used in analysis of NHS data

Estimating prevalence and statistical significance

Crude and age standardised estimates

Unadjusted (crude) weighted prevalence estimates provide important information on the actual level of multimorbidity in the population under study. Crude proportions are reported unless otherwise stated.

Statistically significant differences in age-specific proportions are determined by comparing the crude proportions and their standard errors.

Age-standardisation is used to remove the influence of age when comparing populations with different age structures. Age-standardised proportions are used to inform whether comparisons between population groups (such as men and women of all ages) are statistically significantly different.

The age-standardised proportions in this report have been directly age-standardised to the 2001 Australian standard population. For age-standardised estimates, see Data tables.

Relative standard error, margin of error and confidence intervals

The relative standard error (RSE) of an estimate is a measure of the error likely to have occurred due to sampling. The RSEs of the estimates were calculated using the standard errors (SEs):

RSE% = (SE/estimate) * 100

The margin of error (MoE) at the 95% confidence level for each estimate was calculated using 1.96 as the critical value:

MoE = 1.96 * SE

The MoE was then used to calculate the 95% confidence interval (CI) around each estimate:

95% Confidence Interval = estimate ± MoE

The 95% CI is a range of values determined by the variability in data, within which there is a 95% chance that the confidence interval will contain the true value of the population quantity being estimated.