Dementia in Australia

The number of people living with dementia was estimated for this report based on the methodology used in the previous AIHW 2012 Dementia in Australia report (AIHW 2012), with some data source adjustments.

For people aged 60 and over 

Prevalence rates for those aged 60 and over were derived from the Alzheimer’s Disease International 2015 report. At the time of writing, this report provided the most up-to-date global estimates of dementia prevalence, with published prevalence rates based on a systematic review of dementia prevalence literature globally (Alzheimer’s Disease International 2015). Age-specific (and where possible sex-specific) estimates were presented by world regions (for example, Australasia, North America, and Western Europe) for people aged 60 and over. Prevalence rates for Australasia alone were based on just three outdated regional studies and are not as reliable as those of other regions. To overcome this challenge, prevalence rates presented here were derived from a combination of the Australasian, North American and Western European prevalence rates. North America and Western European rates were based on many more recent studies, and the estimated dementia prevalence rates for these two regions were most similar to Australasia.

Sex-specific estimates

Since sex-specific rates for Australasia were not available in the Alzheimer’s Disease International 2015 report, sex-specific rates for Australasia were calculated by averaging sex rate-ratios from Western Europe and North America and applying them to the Australian sex-specific rates. Age and sex-specific rates for Australia were then calculated by averaging rates from Australasia, Western Europe, and North America, giving each region the same weight.

People aged under 60

Prevalence rates for people aged under 60 were based on a recent large-scale Australian study by Withall et al. (2014), as the Alzheimer’s Disease International 2015 report does not present prevalence rates for people aged under 60. Withall et al. used a methodology consistent with the 2003 UK study by Harvey et al. that was used to estimate dementia prevalence in the AIHW 2012 Dementia in Australia report. Since the Withall et al. study does not report age-specific prevalence rates by sex, we applied sex rate-ratios from Harvey et al. (2003) to derive sex-specific rates.

Calculation of age and sex-specific rates

The newly calculated sex and age-specific rates for people aged 60 and over, and sex-specific rates for people aged under 60 for Australia were applied to ABS Australian population estimates to estimate the total number of people with dementia in Australia overall and by geographic areas, including remoteness, jurisdiction, socioeconomic area, Primary Health Network (PHN) and statistical area 2 (SA2). While the prevalence estimates are synthetic and have no potential for being disclosive in smaller areas, cell sizes smaller than 5 estimated people have been suppressed due to unreliability.

Dementia prevalence projections

As well as providing estimates of dementia prevalence based on the ABS Australian population estimates (ERPs), projections of dementia prevalence are also reported. Where data is reported as a projection, these estimates are derived using a different population estimate from the ABS (ABS Series B or medium series population projections (ABS 2023)). As these two population estimates are different, there may be differences where projected data is reported for the current or previous years.

References

ABS (Australian Bureau of Statistics) (2023) Population Projections, ABS, Australian Government, accessed 17 November 2025.

AIHW (Australian Institute of Health and Welfare) (2012) Dementia in Australia 2012, AIHW, Australian Government, accessed 17 August 2022.

Alzheimer’s Disease International (2015) World Alzheimer report 2015: the global impact of dementia: an analysis of prevalence, incidence, cost and trends, ADI, accessed 17 August 2022.

Harvey RJ, Skelton-Robinson M & Rossor MN (2003) 'The prevalence and causes of dementia in people under the age of 65 years', Journal of Neurology, Neurosurgery and Psychiatry, 74(9):1206–9, doi:10.1136/jnnp.74.9.1206.

Withall A, Draper B, Seeher K & Brodaty H (2014) 'The prevalence and causes of younger onset dementia in Eastern Sydney, Australia', International Psychogeriatrics, 26(12):1955–1965, doi:10.1017/S1041610214001835.

WHO (World Health Organisation) (2012) Dementia a public health priority, WHO, accessed 18 August 2022.

The number of people estimated to be living with dementia by place of residence (living in community or living in cared accommodation) was calculated using several different data sources. This report uses the definition of ‘cared accommodation’ from the 2022 ABS Survey of Disability Ageing and Carers (SDAC). People living in cared accommodation includes people who are a resident in hospitals, nursing homes, aged care hostels, retirement villages, psychiatric institutions, and other 'homes' such as group homes.

To calculate the number of people with dementia living in cared accommodation, the number of people with dementia who were living in residential aged care homes between 1 July 2021 and 30 June 2022 by sex and 5 year age groups from the Aged Care Funding Instrument was inflated using data from the 2022 Survey of Disability Ageing and Carers to obtain the number of people with dementia living in cared accommodation (and not just in residential aged care homes). According to the 2022 Survey of Disability Ageing and Carers, 3.3% of people with dementia living in cared accommodation were not living in a residential aged care home but living in another type of cared accommodation setting.

The number of people with dementia living in the community was calculated by subtracting the estimates of people living in cared accommodation from the total number of Australians estimated to be living with dementia in 2022.

This method suggests that in 2022, 66% of Australians with dementia are living in the community and 34% are living in cared accommodation. These proportions are consistent with findings from an Alzheimer’s Disease International survey in high-income countries, which indicated that 30% of people with dementia lived in ‘care homes’ (WHO 2012).

References

WHO (World Health Organisation) (2012) Dementia a public health priority, WHO, accessed 18 August 2022.

Due to data quality issues, the AIHW provides a minimum estimate of the total number of people aged 15 and over who provide consistent care for a person with dementia who is living in the community. It excludes people who provide formal assistance (on a paid basis, usually associated with an organisation) to a person with dementia, as well as people who provide ongoing care to a family member or friend with dementia living in residential aged care.

The minimum number of unpaid carers of people with dementia was estimated by applying the rate of carers of people with dementia from the SDAC 2022 to the estimated residential population aged 15 and over. However, the SDAC will be an underestimate as only carers who lived in households in the SDAC sample were included. In addition, the SDAC also underestimates the number of people with dementia living in the community.

As there is no single source of data to report total expenditure due to dementia, several different data sources of varying quality were used to estimate health and aged care expenditure attributable to dementia. It must be noted that estimates may also be subject to revisions in methodology and comparisons of such estimates over time may not be possible, such as the improved health care expenditure estimates for 2022–23 based on more detailed costs data and improved coding (refer to Health expenditure section for more details).

Direct spending on dementia-specific programs, packages and services was sourced from the Australian Government Department of Health, Disability and Ageing. Health care expenditure was estimated from the AIHW Disease Expenditure Database (AIHW 2025), noting a revised methodology has been used for 2022–23 estimates.

Due to data limitations, the dementia expenditure estimates presented in this report do not include expenditure for:

• specialised mental health care services
• state and territory government expenditure on aged care
• private aged care services (both home care and supported residential services and facilities)
• government payments to support people with dementia (such as payments made under the National Disability Insurance Scheme) and payments to support carers of people with dementia who are prevented from undertaking substantial paid employment due to their caring role
• indirect expenditure – such as travel costs for patients, the social and economic burden on carers and family, and lost wages and productivity.

Aged care expenditure

Direct Australian Government expenditure on aged care programs are sourced from the 2022–23 Report on the Operation of the Aged Care Act 1997 and Report on Government Services 2024. Non-government-expenditure (for example, by individuals, private health insurers and other non-government sources) is however, included in health expenditure estimates in relation to hospital services, out-of-hospital medical services and prescription medications.

Aged care expenditure is allocated to dementia using the proportion of care delivered within programs to clients with dementia diagnoses and supplements. For the purposes of this report, spending on community-based respite care for people with dementia (which is part of the Commonwealth Home Support Program) is shown separately.

Due to limited data availability relating to diagnoses managed through some aged care programs, the 2022–23 Home Care Packages Program data and the 2022–23 dementia and cognition supplement for home care data were used to estimate dementia spending on the Commonwealth Home Support Program, Home Care Packages Program and the Department of Veterans’ (DVA) Community Nursing Program:

• Commonwealth Home Support Program dementia spending was estimated using the proportion of days in 2022–23 that clients received the dementia and cognition supplement in Home Care Packages Program levels 1 and 2.
• Home Care Packages Program dementia spending was estimated using the proportion of days in 2022–23 that clients received the dementia and cognition supplement (levels 1 to 4).
• DVA Community Nursing Program dementia spending was estimated using the proportion of days in 2022–23 that clients with a DVA entitlement received the dementia and cognition supplement in Home Care Packages Program levels 2 and 3.

Data from the 2022–23 National Screening and Assessment Form (NSAF) were used to allocate spending on residential and community-based respite care, the Transition Care Program, aged care needs assessments, and Veterans Home Care Program:

• Residential respite care expenditure for dementia was estimated by determining the proportion of NSAF approvals for residential respite care that were associated with a dementia diagnosis, for clients not currently living in permanent residential aged care.
• Community based respite care expenditure for dementia was estimated according to the proportion of NSAF assessments where the client was receiving respite care (informal, community and mixed) that were associated with a diagnosis of dementia, among clients not currently living in permanent residential aged care and not approved for residential respite care.
• Transition Care Program expenditure for dementia was estimated by determining the proportion of approvals for transition care that were associated with a dementia diagnosis, for clients not currently living in permanent residential aged care.
• Aged care needs assessment expenditure for dementia was estimated according to the proportion of aged care needs assessments where dementia was recorded as a health condition impacting the person’s care needs (see, National Aged Care Data Clearinghouse (Table 2), for a full list of codes used).
• Veterans Home Care Program expenditure for dementia was estimated according to the proportion of clients with a DVA entitlement who are diagnosed with dementia.

Data from the 2020–21 Aged Care Funding Instrument (ACFI) was used to allocate spending on the National Aboriginal and Torres Strait Islander Flexible Aged Care Program in 2022–23:

• National Aboriginal and Torres Strait Islander Flexible Aged Care Program expenditure on dementia was estimated by determining the proportion of First Nations clients with a dementia diagnosis from the 2020–21 ACFI data (see Table 3 for a full list of codes used).

Estimating expenditure in residential aged care homes directly attributable to dementia

To determine what portion of total funding for a permanent resident in an aged care home is directly related to dementia, information on the resident’s comorbidities is required. While some information on health conditions is collected within the ACFI, these data do not include a complete list of comorbidities, nor do they indicate the relative severity of these conditions. It is therefore not possible to use data collected through the ACFI to separate the cost attributable solely to dementia from the total cost of caring for people living with dementia in residential aged care homes. To allow the estimation of the costs due to dementia, data from the 2018 Australian Bureau of Statistics Survey of Disability Ageing and Carers (SDAC) were used to supplement ACFI data. This approach is consistent with the approach taken for the previous Dementia in Australia report (AIHW 2012) and is outlined below.

Data from the 2018 SDAC relating to people living in residential aged care homes were used to estimate the differences in care needs and funding between people with and without dementia. SDAC questions relating to need for assistance were mapped to related ACFI questions, such that an estimated ACFI score was created for each SDAC respondent living in residential aged care homes.

Health conditions recorded in the SDAC were allocated across eight categories, grouped according to similarities in the likely need for assistance for the condition. For example, arthritis was grouped with hip damage from injury in the group ‘Conditions affecting mobility’. The groups were defined by the ABS categorisations within the SDAC, and include dementia and Alzheimer disease, stroke, conditions affecting mobility, mental health, other cardiovascular disease, hearing loss, Parkinson disease, and other conditions. Each group was only counted once, which means that an individual with multiple conditions within a group is treated the same as an individual with one condition in the group.

A regression model was fitted to the data using the estimated ACFI scores as the dependent variable, and all possible combinations of the eight condition groups (more than 200) as the independent variables. The resultant model had 187 degrees of freedom, an F value of 14.40 (Pr < 0.0001) and an adjusted R2 of 0.29. From this model, a predicted ACFI score was generated for each combination of condition groups, which provided an average ACFI score and level of funding for each combination of condition groups within the model.

Comparisons were then made between combinations of conditions with dementia and without dementia, to quantify the impact of dementia on predicted ACFI scores and associated levels of funding. For example, the predicted ACFI score for a resident with dementia, stroke and mobility problems was compared to that of a resident with just stroke and mobility. The higher the ACFI score for a resident, the more complex their care needs and the more funding they receive. The average predicted ACFI score of a resident with dementia was 125, compared to 95 for those without dementia. This translates to 24% of costs for residents with dementia allocated directly to dementia.

These results were applied to 2020–21 data for ACFI residents, with age and sex accounted for, and used to estimate the proportion of Australian Government funding for permanent residents in residential aged care that was attributable to dementia. This 2020–21 proportion was applied to 2022–23 expenditure data to estimate residential aged care expenditure due to dementia in 2022–23.

Health expenditure

Health care expenditure estimates for 2022–23 were sourced from the AIHW Disease Expenditure Database using their revised methodology (AIHW 2025). The AIHW Disease Expenditure Database takes data from Australia’s National Health Accounts, that forms the base of reporting for the AIHW Health expenditure Australia report series and further examines the data to understand more about the people receiving care and the diseases and conditions being managed. The database contains spending estimates for 17 Australian Burden of Disease Study (ABDS) groups and the 220 conditions within those groups and presents spending by area of expenditure, age group and sex.

The revised methodology in the AIHW Disease Expenditure Database includes updates to the MBS and PBS mapping files to allocate expenditure to specific conditions, improved capture of some in-hospital MBS costs, and use of the 12th edition of the ICD-10-AM to code conditions for hospital patients. The estimates presented in this report should therefore not be compared with previous versions.

Refer to AIHW Disease Expenditure Database methodology for detailed information on the methods used to derive data in the AIHW Disease Expenditure Database.

Estimating expenditure for hospitalisations where a dementia diagnosis was recorded

The approach for estimating expenditure on admitted patients with dementia in this report is similar to what has been used in previous reports, but uses more detailed cost data and includes updates to the MBS and PBS mapping files, improved capture of some in-hospital MBS costs and use of the 12th edition of the ICD-10-AM to code conditions for hospital patients.

In public hospitals, a cost for every separation was estimated using the AIHW Hospital Morbidity Costing Model (HMCM). The HMCM estimates acute hospital admitted patient costs by apportioning the total admitted patient expenditure to individual episodes of hospitalisation with an adjustment for the resource intensity of treatment for the specific episode. Separations with a care type of 7.3 – Unqualified newborns, 9 – posthumous organ procurement or 10 – hospital were excluded from the costing analysis as they are not indicative of admitted patient activity.

Resource intensity and cost for each separation was estimated using the Independent Health and Aged Care Pricing Authority (IHACPA)’s National Weighted Activity Unit/National Efficient Price (NWAU/NEP) calculator, on the basis of principal diagnosis and Diagnostic Related Group (DRG), additional diagnoses, episode complexity, procedures, age, sex, and Indigenous status.

A Diagnosis Related Group (DRG) is a patient classification system that groups hospital in-patients with similar diagnoses and treatments into groups that require a similar amount of hospital resources.

In private hospitals, admitted patient expenditure for dementia is estimated based on charges for separations in the Private Hospital Data Bureau (PHDB) data, and allocated to private hospital separations in the National Hospital Morbidity Database based on principal diagnosis, DRG, and state. This data collection includes all costs except for medical charges. Medical charges are allocated to separations using the MBS items recorded for the separation, and the average in-hospital fee charged for each MBS item by state.

Allocation of total expenditure for a separation to additional diagnoses was based on modelling the estimated separation cost and diagnoses record for a patient. A regression model was used to estimate the fraction of each public hospital separation cost that is due to each condition being treated to account for the impact of comorbidities on costs, and more accurately reflect the expenditure for each condition.

The excess expenditure for each principal diagnosis due to comorbidities was modelled with a log-linear regression model that estimated expenditure for each principal diagnosis (grouped by condition reported in the Australian Burden of Disease Study), with indicators of additional diagnoses as independent variables. The estimated coefficients of the models quantify the impact of additional diagnoses on expected expenditure; that is, the extent to which the charge associated with a given separation for a given principal diagnosis is expected to increase in the presence of additional diagnoses. The results from the regression model were used to estimate the predicted proportion of expenditure associated with each diagnosis within each separation in the hospital data.

For further information on the methods used to produce estimates in the AIHW Disease Expenditure Database, refer to Health system spending on disease and injury in Australia, 2023–24: Overview of analysis and methodology.

Reference

AIHW (Australian Institute of Health and Welfare) (2012) Dementia in Australia 2012, AIHW, Australian Government, accessed 17 August 2022.

AIHW (Australian Institute of Health and Welfare) (2025) Health system spending on disease and injury in Australia, 2023–24, AIHW, Australian Government, accessed 29 October 2025.