What data sources do the maps use?
The Heart Maps draw on the following data sources:
- Hospital Admissions (2012-14): AIHW National Hospital Morbidity Database (Customised data request)
- CHD Mortality (2010-14): AIHW Mortality Over Regions and Time (MORT) books
- Obesity Rates (2014-15) Public Health Information Development Unit (2017) Social Health Atlas (risk factor modelled estimates)
- Smoking Rates (2014-15) Public Health Information Development Unit (2017) Social Health Atlas (risk factor modelled estimates)
- Indigenous CHD mortality rates (2010-2014) ABS Annual Cause of Death tables
- Indigenous obesity and smoking rates (2012-13) ABS Australian Aboriginal and Torres Strait Islander Health Survey: Updated Results, 2012–13
Heart maps FAQs
Australian Heart Maps
Australian Heart Maps
Where does the population risk factor data come from?
The population risk factor data are modelled estimates produced by the ABS in collaboration with the Public Health Information Development Unit at Torrens University (South Australia). This robust model is based on themeasured 2014-15 Australian Health Survey data in addition to local-level demographic, socioeconomic information that indicates the expected level of each health indicator in that area. Click here for more information on the modelled risk factor data.
Does the risk factor data relate to the admitted patients?
No, the risk factor data is not linked to the hospital admission or mortality rates.
What does age-standardised rate (ASR) mean?
The Heart Maps present both crude and age-standardised rates. Crude (also referred to as unadjusted) rates are calculated as the number of events in a given period, divided by the population of the region (Using the 2013 ABS ERP to derive the population denominator for the relevant years of interest (2012-13 to 2013-14). Crude rates allow comparison across regions of different population sizes, and can provide an indication of the health service needs of a region. Age-standardised rates (ASR) have been adjusted for the age distribution of the population. ASRs allow comparison across regions that have differences in both population size, and age distribution. If a region has a high ASR relative to another region, it means that there are factors other than age and population size that are driving the number of admissions or deaths, such as lifestyle risk factors like obesity and smoking.
How have the age-standardised rates been calculated?
The direct method has been used to calculate age-standardised rate (ASR) at the level of state/territory as well as a way to assess health inequalities across different sub-populations of interest including SES quintile, region of birth, indigenous status, and remoteness area.
The indirect method has been used to calculate age-standardised rates (ASRs) for all geographical areas within a state/territory. The denominators used to calculate the indirectly age-standardised and crude rates are the ABS Estimated Resident Populations (ERPs) for the relevant periods.
What is a standardised ratio?
A Standardised Ratio (SR) expresses the overall experience of a population in reference to a standard population. In the Australian Heart Maps, the Standardised Ratio indicates the relative risk of a given area (relative to the national average) for a specified event or risk factor. The maps use a convention of 100 for the national average.
A standardised ratio greater than 100 indicates the risk of heart attack for a given population is higher than would be expected if it had the same risk profile as the standard population. A standardised ratio of less than 100 indicates the risk of heart attack is lower than what would be expected compared to the standard population. For example, the Central Desert LGA has a standardised ratio of 1357 which is equivalent to over 13x the national average.
What are the geographical boundaries used?
There are 95 ‘regions’ in the 2017 release of the Australian Heart Maps (V1.1); including 86 Statistical Areas Level 4 (SA4) and for the first time, the 9 Statistical Areas Level 3 (SA3) regions in the ACT. These regions are based on the Australian Bureau of Statistics’ geographical structures and typically have between 30,000 to 130,000 people (SA3) and between 100,000 and 500,000 people (SA4). Two SA3s in the ACT (Fyshwick-Pialligo-Hume and Cotter-Namadgi) have much smaller populations but are forecast to grow significantly in the coming years. Rankings will differ from those in the 2016 release of the Maps due to the inclusion of the SA3 ACT regions.
Local Government Areas (LGA) boundaries are taken from the ABS (2012). We are unable to report rates for new LGAs (in NSW for example) until the digital boundaries for these new structures are available from the ABS.
What is a statistical area level 4 (SA4) and statistical area level 3 (SA3) region?
Statistical Areas Level 4 (SA4) are geographical structures created by the Australian Bureau of Statistics’ as part of their Australian Standard Geographical Classification . SA4s typically have between 100,000 and 500,000 people. In this release of the Australian Heart Maps (V1.1) there are 95 ‘regions’ - 86 Statistical Areas Level 4 and 9 Statistical Areas Level 3 (SA3) regions in the ACT. The SA3 regions are smaller, with approximately 30,000 to 130,000 people. However, two of the ACT’s SA3 regions have much smaller populations including: Fyshwick-Pialligo-Hume with a population of 1506 and Cotter-Namadgi with a population of 744 (in 2013).
These areas have seen significant growth since 2013 and are predicted to be major growth areas in the coming years. The data for these areas has been suppressed for mortality and hospital admissions, but not for the risk factors. Care should be taken in interpreting their rates of obesity and smoking.
How is a hospital ‘admission’ defined?
A hospital ‘admission’ (also known as a separation) is a completed episode of patient care in hospital, where the completion can be the discharge, death or transfer of the patient, or a change in the type of care (e.g. from acute to rehabilitation). The admission data excludes admissions where a patient has been transferred from another hospital. Hospital admission data includes admissions to private and public hospitals. The hospital admission data comprise separations by principal diagnosis only.
At the state and regional area (Statistical Area 4), the Heart Maps display separations for five heart diagnoses: Heart Attack, NSTEMI, STEMI, Unstable Angina and Heart Failure. The Heart Maps also present rates for All Heart admissions (which combines NSTEMI, STEMI, Unstable Angina, and Heart Failure).
Which conditions do ‘all heart’ admissions include?
Rate for “All Heart” admissions is the sum of rates for the four diagnosis groups: NSTEMI, STEMI, Unstable Angina, and Heart Failure. You can find more information on hospital codes in the Technical report.
Which hospital codes are included for each admission group?
Single events involving heart disease can result in more than one episode in hospital due to transfers and readmissions. For this reason, hospital records are likely to overestimate the number of new cases of heart disease and it is not possible, at the national level, to avoid such over-counting.
In order to reduce over-counting of cases for the Australian Heart Map project, records have been omitted in which the mode of admission is the result of a transfer between hospitals. Whilst such records would be useful in determining hospital burden, e.g. length of stay etc, excluding those records should provide a more reliable assessment of variation in rates of heart disease across different areas and sub-populations.
In particular, transfers will account for a high proportion of admissions for heart disease where procedures such as angiograms tend to be conducted in hospitals within major cities and inner regional areas. Therefore, exclusions of transfers should provide a less biased comparison of rates across areas with disparate geographies and/or remoteness.
What data has been excluded from the hospital admissions?
The following admission data have been excluded in the Heart Maps.
- Separations for which the care type was reported as Newborn with no qualified days, and records for hospital boarders and posthumous procurement have been excluded.
- The data exclude hospital transfers. As such, patients admitted to one hospital and transferred to another hospital will only be counted once. As a result, the admission rates likely underestimate the burden of these admissions on the health system.
- To protect confidentiality, hospital admission rates where the population is less than 1,000 or where there are fewer than five admissions during the two-year period (2012/13 and 2013/14) are not presented.
- Due to small numbers, hospital admissions for Rheumatic fever; for Aboriginal and Torres Strait Islander Peoples and by Country of birth are presented in chart format.
Is the hospital data based on hospital location or patient’s usual address?
The admission rates at a regional level are shown for patients according to their usual residential address. At a State/Territory level the data is based on admissions at a hospital level, and will reflect admissions where a person has travelled from another State or Territory.
How is disadvantage measured? What is the SEIFA index? How is it calculated?
SEIFA (Socio Economic Indexes for Areas) is the Australian Bureau of Statistics measure of disadvantage for areas. SEIFA is designed to convey social and economic conditions of an area using indictors of education, occupation, employment, income, families, and housing collected in the Census.
The Heart Maps use the 2011 SEIFA Index of Relative Socioeconomic Disadvantage (IRSD) which measures how disadvantaged an area is in relation to a base Australian score (1,000). For ease of use, the SEIFA Index for Relative Socioeconomic Disadvantage (IRSD) scores are presented into five groups (quintiles) each comprising 20% of areas and ranked by socioeconomic disadvantage (with Quintile 1 representing the most disadvantaged areas and Quintile 5 representing the least disadvantaged areas).
What do the SEIFA quintiles mean?
For ease of use, the SEIFA Index for Relative Socioeconomic Disadvantage (IRSD) scores are presented into five groups (quintiles) each comprising 20% of areas and ranked by socioeconomic disadvantage (with Quintile 1 representing the most disadvantaged areas and Quintile 5 representing the least disadvantaged areas). The IRSD SEIFA score provides an indication of relative area-level disadvantage. However, its use at a local government area can be limited, sometimes masking pockets within LGAs having greater disadvantage. Local level demographic data and context is important when interpreting SEIFA scores.
How has the avoidable hospitalisation data (due to disadvantage) been calculated?
The number of excess cases has been derived for each SEIFA quintile by calculating the difference between the observed and expected number of admissions (for each age/sex group). The excess for all quintiles is then summed to calculate the total excess; representing the number of admissions that would have been avoided if the rate for the least disadvantaged quintile applied to the other quintiles. This has been presented as a percentage of all admissions.
The standard population was taken as the whole of Australia (in 2012-13 to 2013-14) and quintiles of disadvantage were defined on this standard population. Rates within a small number of sub-populations were already lower than the least disadvantaged quintile. For example, the rate of STEMI admissions in older sub-groups within Western Australia was slightly better than the least disadvantaged quintile for Australia. The number of excess cases was set to zero and therefore interpreted as no excess risk over and above the least disadvantaged quintile for these sub-populations.
Why can’t I see admission data for some local government areas?
If the population for a local government area is less than 1,000 or there are fewer than five hospital admissions, the data is not displayed in order to protect patient confidentiality.
Why can’t I see mortality rates for some local government areas?
A mortality rate will not be available by AIHW where the population in any age group in an area is less than 30, or where there are fewer than 20 cause-specific deaths (excluding those with missing age at death) in an area, by sex. For more information, refer to the MORT books. For the Australian Heart Maps, unadjusted rates have been further supressed where adjusted rates are not available.
When will data be available for the new NSW local government boundaries?
Digital boundaries for the recent LGA changes have only just been released by the Australian Bureau of Statistics. The Australian Heart Maps are based on data from the 2011 Australian Standard Geographical System’s Non-ABS data structures. We are working to bring the new LGA structures to the 2018 Heart Map updates.
Why can’t I see admission rates For Aboriginal and Torres Strait Islander Peoples at a state/territory or local government level?
Hospitalisation and mortality rates for Aboriginal and Torres Strait Islander Peoples are only presented as charts at a national level, due to data limitations. The data is available for each of the heart admission categories and can be viewed by both genders.
We are working hard to bring regional-level hospitalisation rates for Aboriginal and Torres strait Islander Peoples to the Maps. Make sure you register to stay-up-to-date and we will let you know when new data is coming.
Why can’t I see variations in admission rates according to people’s cultural background?
The number of hospital admissions according to country of birth can only be presented as charts at a national level, owing to data limitations. Rates are available for each of the heart admission categories and can be viewed by both genders. The Heart Foundation is pursuing data on coronary heart disease mortality with the Australian Institute of Health and Welfare.
What are the country of birth categories?
The ABS use the Standard Australian Classification of Countries (SACC) (cat. no. 1269.0) when collecting, aggregating and disseminating Country of Birth data. The highest level of aggregation (major group) has been used here to report on admission rates by 9 groups. Australia and Oceania are combined into one group 'Australia and Oceania'.
How is “remoteness” defined?
The Remoteness Structure is a geographic classification designed by the ABS. The purpose of the Remoteness Structure is to divide Australia into broad geographic regions that share common characteristics of remoteness for statistical purposes. For more information refer to the ABS Remoteness Structure (ASGS 2011).
There are more people living in the capital cities so wont they have more hospital admissions and deaths?
Around two-thirds of Australia’s population live in major cities, with the remaining third residing in regional cities, towns and remote areas. So, in absolute numbers, there are more heart attack admissions in the major cities but the rate of admissions for heart attacks in the country is higher (after adjusting for the age of the population). Using admission rates allows us to ‘correct’ for population size so that we can compare areas in a more meaningful way.
Why do Victoria and South Australia have more detailed maps? And why do they look different?
The Heart Foundation first developed state-based maps for Victoria and South Australia using their respective state hospital data and state population risk factor data. The success of the Victorian and South Australian maps fuelled the development of the Australian Heart Maps to provide a national platform for comparing standardised data across Australia.
The SA and Victorian Heart Maps have a state-based focus whereas the Australian Heart Maps have a broader national focus, comparing rates to a national average. Rates in the SA and Victorian Maps cannot be compared to rates in the Australian Heart Maps because they use different data sources.
The national Heart Maps have a different “look and feel” because they are featured on a new software, called Tableau, which will (when configured) enable access on Tablets and Mobile devices. Tableau is increasingly being adopted by government departments and other agencies.
Why are the hospital admission rates in the Victorian and South Australian heart maps different than those seen in the national heart maps?
- The South Australian maps use a different denominator (persons aged 30+) to derive rates.
- Hospital transfers were excluded for the national Heart Maps to improve comparability across states/territories.
- The national maps are based on 2 years of hospital admissions data (2012/13 to 2013/14) and as such provide different estimates compared to the other two state-level maps which were developed based on more years of admissions data.
Admission rates in the separate state-based maps are different than that presented for Victoria and South Australia in these maps because hospital transfers have been excluded in the national maps.
The Victorian and South Australian maps have counted patients who have been transferred from another hospital as an episode. Transfers account for many heart disease admissions where patients in rural and remote areas are initially hospitalised, and then transferred to a major regional or metropolitan centre for procedures such as angiograms. Whilst such records are useful in determining hospital burden, they have been excluded here to provide a more reliable assessment of variation in rates across different areas and sub-populations.
Additionally, the Australian Heart Maps are based on a different dataset which was not used in any way to construct either of the two state level heart maps. The Victorian Heart Maps were developed independently and in collaboration with the Department of Health and Human Services and it therefore utilises data that has been extracted and put together by the Victorian Department of Health. As no attempt has been made to link datasets, there will be differences due to the manner in which the data are managed at the source.
The Vegemite and cheese sandwich is a treat only an Aussie could love. But can you guess how much of our daily salt… twitter.com/i/web/status/9…