Australian Institute of Health and Welfare (2022) Health system spending per case of disease and for certain risk factors, AIHW, Australian Government, accessed 07 July 2022.
Australian Institute of Health and Welfare. (2022). Health system spending per case of disease and for certain risk factors. Retrieved from https://www.aihw.gov.au/reports/health-welfare-expenditure/health-system-spending-per-case-of-disease
Health system spending per case of disease and for certain risk factors. Australian Institute of Health and Welfare, 05 April 2022, https://www.aihw.gov.au/reports/health-welfare-expenditure/health-system-spending-per-case-of-disease
Australian Institute of Health and Welfare. Health system spending per case of disease and for certain risk factors [Internet]. Canberra: Australian Institute of Health and Welfare, 2022 [cited 2022 Jul. 7]. Available from: https://www.aihw.gov.au/reports/health-welfare-expenditure/health-system-spending-per-case-of-disease
Australian Institute of Health and Welfare (AIHW) 2022, Health system spending per case of disease and for certain risk factors, viewed 7 July 2022, https://www.aihw.gov.au/reports/health-welfare-expenditure/health-system-spending-per-case-of-disease
Get citations as an Endnote file:
PDF | 1.4Mb
The Australian Burden of Disease Study (ABDS) aims to measure the human cost of disease in terms of the burden experienced by individuals and society through quality of life, and years of life lost. This does not however, give a measure of the number of people afflicted by a condition, and their contribution to total spending.
Prevalence is a measure of disease that indicates the proportion of a population with a disease over a specific period. The term ‘case’ is used here to refer to the number of prevalent cases, which is the total number of cases of a disease existing in a population over that period (period prevalence). By estimating the health system spending per prevalent case for a particular disease or condition, we can better understand how health system spending is distributed across the various diseases and conditions that Australian’s suffer. Among other things, this is useful for evaluating policy interventions, treatments and programs.
The term spending is used here, rather than cost, because the data does not cover the full cost (financial and non-financial) experienced by the individual, their family or the health system. It doesn’t cover, for example, the traumatic impacts and other strains experienced by those caring for the disease sufferer. In many cases, spending and actual cost are also not always completely aligned. The actual cost of the treatments and materials used or the staffing requirements for the care of an individual patient in a hospital, for example, is not always directly reflected in how hospital spending occurs at the state and territory and Commonwealth levels or through private funding mechanisms like private health insurance.
The ABDS uses prevalence measures as an input to estimate the burden of disease. As the same classification of diseases is used across both studies, this analysis uses the AIHW’s disease expenditure database and derived prevalence estimates from the ABDS to estimate the spending per prevalent case of disease. The measure of prevalence used for this analysis is period prevalence (number of people with a disease over a given time frame) for the year 2018.
For hepatitis B and hepatitis C, the ABDS was the source of prevalence for acute cases only. For chronic cases, the Kirby Institute was the source of prevalence for hepatitis C and the Doherty Institute was the source of prevalence for hepatitis B. The prevalence data used in the calculations for hepatitis C was the number of cases in the community at the end of 2017 while for hepatitis B, it was the number of cases in the community at the beginning of 2019.
Burden estimates in the ABDS are derived from the prevalence of multiple sequela of a condition, with health loss durations applied. Sequela (or health states) are the outcomes of conditions and often pathways through which burden is experienced. Health loss is the poor health and decreased quality of life due to a condition, or an outcome of it. For example immobilisation for a broken collar bone, or recovery during chemotherapy. The duration of health loss is the total time that individuals experience a particular outcome. For example, an individual with laryngeal cancer would have a particular health loss associated with the duration of their initial diagnosis and primary therapy, and then further health loss if they undergo a laryngectomy.
A review of the ABDS prevalence estimates was undertaken and concordances developed between the two studies, to identify which condition and sequela combinations would have individuals being counted across two or more sequelae, leading to double counting at the person level. Sequela were in scope if the people being captured in it were not likely to also be captured in another sequela for a condition. For example, ‘pelvic inflammatory disease due to chlamydia’ was excluded, as people with this health state are also captured in the ‘chlamydial infection’ health state. Once these sequelae are excluded, prevalence estimates can be calculated. Prevalence for infertility as a sequela of conditions (such as ‘infertility due to polycystic ovarian syndrome’) was allocated to the infertility condition prevalence, as the overall prevalence for this condition is distributed to various other conditions as a sequela during the ABDS estimation process.
The method for calculating ‘point prevalence’ differed depending on whether the condition was acute (such as an infection), or chronic (such as coronary heart disease). Burden estimates for acute conditions in the ABDS database are calculated using the duration of health loss. This health loss duration needs to be removed to calculate the number of people or events i.e., prevalent cases. For example, the number of people with tuberculosis multiplied by the average duration of health loss (8 months) is used to calculate the point prevalence. To calculate the number of prevalent cases for most acute conditions, the point prevalence estimate divided by the duration will give the number of people or cases. The ABDS point prevalence estimates were used directly for non-acute conditions since point prevalence was assumed to be equivalent to the one–year period prevalence for chronic conditions, where there are no sequela with multiple durations applied.
For some conditions there are multiple health loss durations for the components of a health state, generally due to varying levels of severity. Where this was the case, prevalence estimates were calculated directly from source datasets, and includes:
Burden estimates for cancer are complex, as burden can be experienced over many years. They include estimates of diagnosis and primary therapy, controlled phase, metastatic phase, terminal phase, and for some cancers, the ongoing impact of various treatments such as mastectomy or stoma. In this structure, patients who die from cancer (terminal phase) are assumed to have experienced metastases, while those who do not die are not recorded as being metastatic (and are counted in diagnosis and primary therapy). Therefore, to calculate the number of individuals undergoing treatment for cancer in a year, the estimates for diagnosis and primary therapy as well as terminal phase are used in the prevalence calculations for this analysis.
Burden estimates for chronic kidney disease are based on a variety of symptoms experienced during severe disease (stage 3, 4, and end stage), and do not include those with lower severity disease. To calculate the total prevalence of chronic kidney disease, diagnosed chronic kidney disease at each stage was used instead.
All of these calculations are undertaken before quality-of-life measures are applied to conditions. See Australian Burden of Disease Study: methods and supplementary material 2018 for more information on the calculation of the ABDS point prevalence estimates and the key data sources for prevalence. Estimates reported in the ABDS and included in this analysis are modelled and should be interpreted as such.
Total disease expenditure estimates were calculated at the national level for each condition by sex and divided by the number of prevalent cases to estimate the spending per prevalent case. For example, in 2018–19 the total spending on depressive disorders was $2.1 billion, and the estimated number of prevalent cases was 828,785, giving a spending per prevalent case in that year of $2,549.
Spending per prevalent case estimates are of high importance for health economic modelling in Australia. It is the first time that a spending per prevalent case for almost all burden of disease conditions has been estimated using total health system expenditure, rather than a cost of illness approach. The data generated can be used to populate proportional multistate life table models, and various other health economic models.
In this study, conditions with a prevalence count less than or equal to 4 for either males, females or total persons are not included in the analysis. In addition, benign and uncertain brain tumours, haemophilus influenzae type-b and all burden of disease conditions that contain ‘other’, ‘unknown’ or ‘uncertain’ in their description have all been excluded from the analysis due to the difficulty in obtaining reliable prevalence estimates for these conditions.
Spending estimates for Hepatitis C include new treatments that are expensive, and cure historic cases of Hepatitis C. These historic cases are not captured in burden estimates, as only acute episodes are included. An estimate of the number of chronic cases for Hepatitis C is available from the Kirby Institute and has been added to the acute cases to derive spending per case for this condition.
Expenditure estimates do not include costs born outside of the health system, such as management of dementia in residential aged care facilities.
We'd love to know any feedback that you have about the AIHW website, its contents or reports.
The browser you are using to browse this website is outdated and some features may not display properly or be accessible to you. Please use a more recent browser for the best user experience.