Geographical variation in disease: diabetes, cardiovascular and chronic kidney disease

The Australian Bureau of Statistics (ABS) 2011–12 National Health Measures Survey (NHMS) is the biomedical survey component of the 2011–12 Australian Health Survey (AHS). Participants were drawn from the combined respondent sample of the National Health Survey (NHS) and the National Nutrition and Physical Activity Survey (NNPAS). The AHS collected information from a large sample of households about health-related issues, including health status, risk factors, health-related actions, socioeconomic circumstances, nutrition and physical activity, as well as self-reported long-term health conditions and biomedical information on chronic diseases and nutrient status.

The NHMS collected urinary and blood samples from over 11,000 voluntary respondents with children aged 5–11 years asked to provide a urine sample only.

The prevalence of chronic diseases, such as CKD and diabetes, are most accurately estimated using measured (or biomedical) data, in the form of markers found during blood and urine testing, rather than using self-reported data alone. For example, results from the AHS indicate that only 11% of the survey respondents who showed biomedical signs of CKD self-reported that they had the condition and for every 4 adults with diagnosed diabetes, there was 1 who was undiagnosed.

While the reliability of disease estimates is improved by using biomedical data, it should be noted that the AHS has limitations, which may result in the underestimation of the prevalence of these conditions:

• People living in non-private dwellings, including people living in institutional care facilities such as hospitals and aged care facilities, were not included in the survey.
• Those living in Very remote areas and in discrete Aboriginal and Torres Strait Islander communities were not included in the survey.
• For self-reported data items, some respondents may not have known, or were not able to accurately report, their health status.
• The response rate of the biomedical component of the survey was lower than the AHS. Overall participation in the NHMS was 37.1% among those aged 5 and over and 37.6% among those aged 12 and over with 34.7% providing a urine sample and 36.9% providing a blood sample (participants 12 years and over only) (ABS 2013).

Estimates for states and territories are direct survey estimates extracted from the 2011–12 NHMS. As unit record files in the NHMS do not include information at PHN or PHA level, the ABS produced modelled estimates for these smaller geographical areas. For explanatory notes on modelled estimates, see Technical notes.

For more information on the NHMS.

For more information on the AHS.

Conducted every 3 years, the Australian Bureau of Statistics (ABS) National Health Survey (NHS) provides information on the health status of the Australian population; including their use of health services and facilities, and health-related aspects of their lifestyle. The ABS NHS in 2017–18 is the latest survey. The NHS collects self-reported data on whether a respondent had 1 or more long-term health conditions; that is, conditions that lasted, or were expected to last, 6 months or more. A few exceptions have applied to health conditions not meeting the duration or currency criteria such as stroke, angina, heart attack, heart failure or epilepsy. These data are used in this report to estimate the prevalence of selected conditions. The survey had 20,700 respondents.

Consider the following limitations when interpreting data from the 2017–18 NHS:

• Much of the data is self-reported and therefore relies heavily on respondents knowing and providing accurate information.
• The survey is household-based and does not include information from people living in nursing homes or other institutions.
• The survey excludes residents of Very remote areas and discrete Aboriginal and Torres Strait Islander communities. This is unlikely to affect national estimates, but impacts prevalence estimates by small areas where a large proportion of the population live remotely (e.g., 20% of the population of the Northern Territory live in very remote areas).

For further information on the NHS.

In 2017–18, the National Health Survey (NHS) and the Survey of Income and Housing collected a number of items that were common to both. A pooled data set has been created from these 2 surveys and contains information on smoking behaviour from 44,904 individuals aged 15 and over. It allows derivation of more robust estimates of smoking status, particularly for smaller geographic areas.

When interpreting data from the pooled data set, the same limitations of the 2017–18 NHS apply.

For further information on Smoker Status.

The Australian Bureau of Statistics (ABS) 2016 Census of Population and Housing was Australia’s 17th, 5-yearly national census with more than 95% of Australians participating. In addition to key demographic characteristics, the Census collected information about housing tenure, education, labour force participation, occupations and industries, marital status and family size.

The Census provides information for the country as a whole, as well as for small geographic areas and small population groups.

Household, dwelling and population characteristics were extracted for inclusion in the socioeconomic profile, complementing the health risk profile of an area, which can be compared with state and territory and national profiles. Together, these provide a comprehensive picture of disease health outcomes and behaviours within a social and economic context.

For more on the 2016 Census data quality statement

For more information on the independent review of the data quality statement of the 2016 Census.

Residential aged care data are available from the AIHW National Aged Care Data Clearinghouse (the Data Clearinghouse) and primarily sourced from the Department of Health.

The Data Clearinghouse is located at AIHW to provide aged care data to a range of stakeholders including policy makers, researchers, service providers and general consumers. The Data Clearinghouse includes a wide range of data on aged care providers, services, places, residents, care recipients and payments.

The majority of these data are derived from administrative data collections designed to support payment of subsidies to service providers, and include data on the numbers of clients, their age and Indigenous status. The residential aged care collection is used for this project.

Residential aged care provides care and support to senior Australians, who can no longer live at home and need ongoing help with everyday tasks or health care. Residential care is provided on a permanent or respite basis to any eligible person.

The services provided through residential care include:

• help with day-to-day tasks (such as cleaning, cooking, laundry)
• personal care (such as bathing, dressing, grooming, going to the toilet)
• clinical care (such as wound care and medication administration) under the supervision of a registered nurse
• other care services.

For people who need almost complete assistance with most activities of daily living, residential care can provide 24-hour care.

This project used June 2018 data on the permanent aged care residents aged 70 and over as additional information on the local population at risk living permanently in institutions.

At 30 June 2018, there were 2,695 residential aged care services, run by 886 operational residential aged care providers, an increase of 0.9% from the previous year.

For the data quality statement of the Data Clearinghouse.

The cause of death data used in this report are sourced from the Registrar of Births, Deaths and Marriages (RBDM) in each state and territory and the National Coronial Information System. The data were coded by the Australian Bureau of Statistics (ABS) using the 10th revision of the International Classification of Diseases and Related Health Problems (ICD-10). Registration of deaths is the responsibility of the state and territory RBDM, with the Australian Coordinating Registry managing and coordinating the registry on their behalf. AIHW holds these records in the National Mortality Database (NMD).

A medical practitioner or a coroner is required to certify the cause of death and to report all medical conditions that directly contributed to the death and in some cases, the circumstances that led to the death:

• underlying cause of death—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
• associated causes of death—all causes listed on the death certificate, other than the underlying cause of death. They include the immediate cause, any intervening causes and conditions that contributed to the death but were not related to the disease or conditions causing the death.

In addition to information about causes of death, the NMD captures other characteristics of the person, such as sex, age at death, area of usual residence and Indigenous status.

Analysis of cause of death was based on:

• the year of registration of death over a 5-year period between 2013 and 2017
• area of usual residence.

Deaths registered in 2015 and earlier are based on the final version of cause of death data. Deaths registered in 2016 are based on the revised version. Deaths registered in 2017 are based on the preliminary version. Revised and preliminary versions are subject to further revision by the ABS.

Note: These data have not been adjusted for Victorian additional death registrations in 2019. For more detail please refer to Technical note: Victorian additional registrations and time series adjustments in Causes of death, Australia, 2019 (ABS Cat. no. 3303.0).

The data quality statements underpinning the AIHW NMD can be found in the following ABS publications:

• ABS quality declaration summary for Deaths, Australia
• ABS quality declaration summary for Causes of death, Australia

For more information on the AIHW National Mortality Database.

The National Hospital Morbidity Database (NHMD) is a compilation of episode-level records from admitted patient morbidity data collection systems in Australian hospitals.

The data are based on the Admitted Patient Care National Minimum Data Set (APC NMDS) and include demographic, administrative and length of stay data; as well as information on patient diagnoses, procedures and external causes of injury and poisoning.

The purpose of the APC NMDS is to collect information about care provided to admitted patients in Australian hospitals.

The scope of the APC NMDS is episodes of care for admitted patients in all public and private acute and psychiatric hospitals, freestanding day hospital facilities and alcohol and drug treatment centres in Australia. Hospitals operated by the Australian Defence Force, corrections authorities and in Australia’s offshore territories are not in scope but some are included.

Reporting to the NHMD occurs at the end of a person’s admitted episode of care (separation) using the clinical documentation for that episode. Separations—referred to in the map and data workbooks as hospitalisations—are reported to the NHMD in accordance with the requirements of the APC NMDS. A separation can be a total hospital stay (from admission to discharge, transfer or death) or a portion of a hospital stay beginning or ending in a change of type of care (for example, from acute care to rehabilitation). The NHMD does not include episodes of non-admitted patient care provided in outpatient clinics or emergency departments. Patients in these settings may be admitted subsequently, with the care provided to them as admitted patients being included in the NHMD.

The Geographical variations product reports data both for a single year (2017–18) and as an aggregation of multiple years (2015–16 to 2017–18) in order to compare the smallest geographical areas Population Health Area (PHA). Both single and combined year data are reported at the national, state/territory and Primary Health Network (PHN) levels.

The APC NMDS requires the principal diagnosis and any additional diagnoses to be reported according to the ninth and 10th revisions of the International Statistical Classification of Diseases and Related Health Problems, 10th Edition, Australian Modification (ICD-10-AM) and associated Australian Coding Standards.

Principal diagnosis is the diagnosis established after study and chiefly responsible for occasioning the patient’s hospitalisation.

Additional diagnoses are conditions or complaints, either coexisting with the principal diagnosis or arising during the episode of admitted patient care (hospitalisation), episode of residential care or attendance at a health-care establishment that requires the provision of care. Multiple diagnoses may be recorded during a single episode of care.

Hospitalisations for cardiovascular disease and where dialysis was undertaken are reported where they were recorded as the principal diagnosis. Type 2 diabetes, chronic kidney disease and acute kidney injury hospitalisations are reported where they were recorded as any diagnosis (principal or additional diagnosis).

The comparability of the coded diagnosis data can be affected by variations in the quality of the coding, and can be influenced by state-specific coding standards. Further information on the quality and comparability of the coded data at the state and territory level can be found in the following AIHW publications:

• Admitted patient care 2015–16: Australian hospital statistics
  (AIHW Health services series no.75. Cat.no. HSE 185)
• Admitted patient care 2016–17: Australian hospital statistics
  (AIHW Health services series no.84. Cat.no. HSE 201)
• Admitted patient care 2017–18: Australian hospital statistics
  (AIHW Health services series no.90. Cat.no. HSE 225).

Data on geographical location are collected based on the area of usual residence of patients in the NHMD. These data are specified in the NMDS as state/territory of residence and Statistical Area Level 2 (SA2). Missing SA2 information explains why not all hospitalisation records could be assigned a PHN or PHA.

For the NHMD data quality statements for 2015–16, 2016–17 and 2017–18, see, respectively:

• Admitted Patient Care 2015-16
• Admitted Patient Care 2016-17
• Admitted Patient Care 2017-18

For more information on the NHMD.

Established in 1987, the National Diabetes Services Scheme (NDSS) is an initiative of the Australian Government, administered with the assistance of Diabetes Australia (DA). The NDSS subsidises the supply of diabetes-related services and products—such as pens and needles to administer insulin, blood glucose test strips and insulin pump consumables—to people registered with the scheme. A diagnosis of diabetes by a health professional is required in order to register with, and purchase products through, the NDSS. The NDSS registrant data provides demographic information as well as basic clinical details, such as diabetes type, while the sales data provides information on insulin use.

Not all people with diabetes register with the NDSS. However, given the incentives, it likely represents a high proportion of people with diagnosed diabetes in Australia.

As an administrative data set there are a number of data quality issues that should be noted in the context of this project:

• Prior to 2002–03 diabetes was classified as either insulin-dependent diabetes mellitus (IDDM) or non-insulin-dependent diabetes mellitus (NIDDM), with the IDDM records classified as type 1 diabetes and the NIDDM as type 2 diabetes. The changes in classification of diabetes type on the NDSS may underestimate the number of registrants with type 2 diabetes who also needed insulin to manage their diabetes.
• Diabetes type is classified by a health practitioner at the time of NDSS registration; however, the recorded type might not always be correct as the symptoms of type 1 and type 2 diabetes may be similar. To correct for the misdiagnosis of diabetes type (such as late onset type 1 diabetes), an algorithm proposed by Harding et al. (2014) was applied to reclassify registrants recorded as having type 2 diabetes to type 1 if they were:

• less than 30 years of age at diagnosis; and
• were using insulin; and
• the time between date of diagnosis (or registration date if the date of diagnosis was missing); and
• first insulin use was less than 1 year.

This may result in a small proportion of registrants with a true type 2 diagnosis being reclassified to type 1 diabetes and will result in discrepancies with published data from the NDSS.

• A large proportion of early NDSS registrations (prior to 2000) had a lower rate of completion for diagnosis date. However, the proportion of registrants with complete diagnosis date information has improved over time. Around 15% of new registrants had missing diagnosis date information in 2018 compared to 20% or over prior to 2014.
• There is limited coverage of the NDSS in remote areas. NDSS Access Points assist in delivering support services and products to people with diabetes in all states and territories. These Access Points may be limited in rural Australia and unavailable in remote communities, with other programs being available in these areas to assist with the purchase of diabetes-related products. As a result, the coverage of the NDSS may be lower in remote and very remote areas or across states and territories with large remote communities (AIHW 2017).
• The geographical information on NDSS registrants is based on the postcode of residence and is collected at the time of registration. To date, the extent to which the postcode information is updated over time is unknown, which might have an impact on the accuracy of the geographical distribution of type 2 diabetes in Australia.
• Interpretation of the results for type 2 diabetes for small geographic areas such as PHAs should be done with caution as a number of factors can affect the prevalence of type 2 diabetes at this level. These include the changes in postcode of residence information over time and the quality of the concordances for converting postcode data to PHA data for some PHAs. AIHW’s internal assessment showed that over half of the postcodes (52%) poorly corresponded to the SA2 geography, leading to potential inaccuracy for 62% of the PHAs.

Despite these limitations, the NDSS is a valuable and unique national source of information about people diagnosed with diabetes.

NDSS-NDI linked data used for this project are based on the 2018 reference year.

For more details, see the 2018 quality statement of the National (insulin-treated) Diabetes Register which lists the NDSS limitations.

The National Death Index (NDI), is a database housed at the AIHW, which contains records of all deaths occurring in Australia since 1980. The data are obtained from the RBDM in each state and territory. The NDI is linked to the NDSS to flag which registrants may be deceased and the date of their death to derive diabetes prevalence estimates at any point in time. The year of reference for this project was 2018.

For the data quality statement underpinning the NDI.

Estimates available from the Australian Bureau of Statistics (ABS) for the estimated resident population (ERP) have been used to define the denominator and standard populations for all analyses in this product. ERP counts by SA2 from the 2016 ASGS were used to derive the population denominators for the PHN and PHA geographical levels. The following approach for specific data sets was used in this project:

• Denominators for the AHS were derived from the weighted population aged 18 and over. For more information about the denominators for direct estimates, see Australian Health Survey: user’s guide 2011–13 (ABS 2013a). For explanatory notes on modelled estimates, see Technical notes.
• Denominators for the NHS were derived from the weighted population aged 18 and over. For more information about the denominators for direct estimates, see National Health Survey: user’s guide (ABS 2019). For explanatory notes on modelled estimates, see Technical notes.
• Denominators for the NHMD were derived from the ERP as at 30 June. For each financial year, December population counts were calculated, based on the averaged ERP at 30 June for each calendar year. Where the denominator for 3 financial years (2015–16, 2016–17 and 2017–18) was used, the sum of the December ERPs for each of the 3 years was divided by 3 to obtain the annual average population estimate for the denominator, and reported as annualised population. The rate calculation was based on the 3-year combined population denominator.
• Denominators for the NMD were derived from the ERP as at 30 June. Where the denominator for 5 calendar years (2013–2017) was used, the sum of the ERPs at June for each of the 5 years was divided by 5 to obtain the annual average population estimate for the denominator and reported as mean population per year. The rate calculation was based on the 5-year combined population denominator.
• Denominators used for the socioeconomic profile from 2016 Census data are summarised below with the corresponding characteristics:

• occupied private dwellings (internet access and overcrowding)
• usual resident population (socioeconomic quintile, educational attainment among people aged 25–74)
• usual resident population speaking at home language other than English (self-assessed proficiency if spoken English)
• usual resident in the labour force aged 15 to 64 (unemployment).

Estimates of the Aboriginal and Torres Strait Islander Australians in 2016 by small areas are based on the ABS Estimated Aboriginal and Torres Strait Islander Resident Population after adjusting for net census population undercount (ABS 2018d). Population estimates by Primary Health Network were derived using the correspondence files between SA2 area and Primary Health Network belong to a different geographical system than the ASGS. For data quality statement of the ABS Estimated Aboriginal and Torres Strait Islander Resident Population, see ABS Quality Declaration.