Application of the Australian Bureau of Statistics Socio-Economic Indexes for Areas in cardiovascular disease research: a scoping review identifying implications for research
Hannah Beks
A , Sandra M. Walsh B , Sarah Wood A , Suzanne Clayden A C , Laura Alston A D , Neil T. Coffee A and Vincent L. Versace A *
A
B
C
D
Abstract
To scope how the Australian Bureau of Statistics Socio-Economic Indexes for Areas (SEIFA) has been applied to measure socio-economic status (SES) in peer-reviewed cardiovascular disease (CVD) research.
The Joanna Briggs Institute’s scoping review methodology was used.
The search retrieved 2788 unique citations, and 49 studies were included. Studies were heterogeneous in their approach to analysis using SEIFA. Not all studies provided information as to what version was used and how SEIFA was applied in analysis. Spatial unit of analysis varied between studies, with participant postcode most frequently applied. Study quality varied.
The use of SEIFA in Australian CVD peer-reviewed research is widespread, with variations in the application of SEIFA to measure SES as an exposure. There is a need to improve the reporting of how SEIFA is applied in the methods sections of research papers for greater transparency and to ensure accurate interpretation of CVD research.
Keywords: Australia, Australian Bureau of Statistics, cardiovascular disease, health policy, SEIFA, social determinants of health, socio-economic factor, Socio-Economic Indexes for Areas, scoping review.
Introduction
Socio-economic status (SES) is important to understand the relationship between chronic disease and risk factors. Globally, the association of SES and the risk of multimorbidity, non-communicable disease, frailty and disability is well established.1,2 Cardiovascular disease (CVD) is one of the most prominent causes of death globally, with 17.9 million people estimated to have died from CVD in 2019,3 which increased to 20.5 million people in 2021.4 A socio-economic gradient is well established for CVD morbidity and mortality outcomes, and it includes lower levels of educational attainment, income level and employment status.5,6 For example, residing in a disadvantaged neighbourhood has been associated with a higher incidence of coronary heart disease5 and a higher risk of CVD.7 In Australia, national data indicate that mortality rates associated with CVD are higher in populations residing in the lowest socio-economic areas compared with populations residing in the highest socio-economic areas.8
CVD research examining and understanding the role of SES is central to developing targeted preventative interventions, and informing health service planning and policy. Research has generally applied two approaches to classifying SES in health research.9,10 These include classification at an individual level using income, education or occupation data; and at an area level using a range of existing socio-economic information.10 In Australian health research undertaken with specific communities (Aboriginal and Torres Strait Islander peoples), individual-level approaches have been preferred by researchers to ensure that findings are more relevant to the research setting.9 For analysis of population-level data, area-level approaches are more commonly applied in a defined geographical area.11 The annual reporting of Australian health data (such as the Australian Institute of Health and Welfare Australia’s Health Report) uses the Socio-Economic Indexes for Areas (SEIFA) Index of Relative Socio-Economic Disadvantage (IRSD) as an area-level measure to report on socio-economic characteristics of health risk factors, chronic disease, mortality and morbidity.10 Further information is provided in Supplementary File S1.
General guidance from the Australian Bureau of Statistics (ABS) around applying SEIFA states that the indexes indicate the average socio-economic characteristics of populations in an area, and are best interpreted as ordinal measures.12–14 Researchers have critiqued the use of SEIFA as a composite measure in health research because it does not account for the heterogeneity of individual indicators of SES in a geographical area.15 International scholars have supported the use of area-level approaches using spatial measures of CVD risk and socio-economic disadvantage, to implement targeted public health interventions to communities at higher risk for CVDs.5 Examining how SEIFA has been applied in CVD research and at what spatial unit, is important to identify any opportunities to improve the consistency in application. To our knowledge, a review of Australian CVD peer-reviewed research examining the application of SEIFA, has not been conducted, indicating a research gap. No similar proposed reviews were identified.
The review question was:
How has the ABS SEIFA been applied to examine SES in peer-reviewed CVD research undertaken in Australia?
Objectives included:
Methods
The Joanna Briggs Institute’s (JBI) scoping review methodology was used.16 The Preferred Reporting Items for Systematic Reviews and Meta-analysis extension for Scoping Reviews (PRISMA-ScR)17 checklist was reported against (Supplementary File S2). Methods were specified in advance (Open Science Framework https://doi.org/10.17605/OSF.IO/M9WD7). The JBI three-step search process guided the development of the search strategy.18 Searches were developed for databases: Ovid MEDLINE, CINAHL Complete (EBSCOhost), APA PsycInfo (EBSCOhost), and Embase (Elsevier) (Supplementary File S3). Google Scholar was also searched using keywords.
Inclusion and exclusion criteria
Studies were screened according to the inclusion and exclusion criteria (Table 1). Peer-reviewed CVD research in which SES was examined as a primary or secondary exposure measure using the ABS SEIFA indexes, was included. Searches were limited from 1 January 2013 to capture research published since the release of SEIFA 2011 (March 2013).
| Inclusion criteria | Exclusion criteria | ||
|---|---|---|---|
| Population | Australian populations with CVD. Cardiovascular diseases were classified using the International Classification of Diseases (11 – diseases of the circulatory system) 19 and defined as disorders of the heart and blood vessels, and included coronary heart disease, hypertension, cerebrovascular disease, congenital heart disease, rheumatic heart disease, peripheral arterial disease, deep vein thrombosis and pulmonary embolism. 3 | Non-Australian populations, Australian populations without CVD. | |
| Concept | Peer-reviewed experimental or observational research, including, but not limited to, cross-sectional studies, cohort studies, randomised controlled trials, which used the ABS SEIFA indexes 12 to examine SES as a primary or secondary exposure measure. | Editorials, protocols, opinion-based pieces, and non-peer-reviewed literature (e.g. government reports) and research using non-ABS measures of SES and studies that examine SES as a confounder, not an exposure. | |
| Context | Studies published in English since 1 January 2013 to capture studies published since the release of SEIFA 2011. | Studies published prior to 1 January 2013 |
Study selection and data extraction
Citations were imported into Covidence (Veritas Health Innovation, Melbourne, Australia). Titles and abstracts were screened independently by two reviewers. Full text review and data extraction was then undertaken. Findings were synthesised using a descriptive approach.16 A quality assessment of included studies was undertaken using the JBI critical appraisal tools. To our knowledge, no validated quality assessment tool currently exists for ecological studies, therefore the JBI critical appraisal tool for analytical cross-sectional studies was applied where appropriate (an approach used by other reviews examining ecological studies).20
Results
Of the 95 citations eligible for full text screening, 49 studies were included. No additional citations were identified through a review of references (Fig. 1). Reasons for excluding studies are provided in Supplementary File S4. A narrative synthesis of studies and implications of studies is presented in Supplementary File S5.
Characteristics of included studies
Of the 49 included studies (Table 2), 10 studies had a national focus and two studies focused on populations residing in two or more Australian states or territories. The remaining studies had a geographical focus at a state or territory level: the highest proportion focused on populations in Victoria (n = 12), followed by Queensland (n = 10), Tasmania (n = 5), New South Wales (n = 4), Western Australia (n = 3), South Australia (n = 2) and Australian Capital Territory (n = 1).
| Citation | State/Territory/Australia | Research aim (study period) | Study design type (methods) | Outcome measure(s) (exposure measure(s)) | Participant sample | Aboriginal and Torres Strait Islander peoples focus (Yes/No) | Findings | Study limitations | Implications | |
|---|---|---|---|---|---|---|---|---|---|---|
| Adair and Lopez (2021) 22 | Australia | To assess whether mortality inequalities among specific non-communicable diseases in Australians aged 35–74 years, widened during 2006–2016 (2006–2016) | Ecological study (data linkage) | Premature mortality (SES; remoteness) | Not reported for CVD | No | CVD mortality inequalities were mostly wider than for all non-communicable diseases. The Q1:Q5 premature mortality ratio was 2.28 for male participants and 2.67 for female participants. | Due to different versions of SES measured using IRSAD for 2006–2010 (SLA) and 2011–2016 (SA2), an analysis of long-term trends was not possible. | Widening inequalities in premature mortality rates attributed to non-communicable disease, including CVD, is a public health issue requiring immediate policy responses with a focus on socio-economic disadvantage. | |
| Astley et al. (2020) 23 | South Australia | To determine the impact of cardiac rehabilitation attendance on cardiovascular re-admission, morbidity, and mortality between 2013 and 2015 (2013–2015) | Retrospective cohort study (data linkage) | Re-admissions; morbidity; mortality (SES) | 49,909 eligible patient separations | No | Of 49,909 eligible separations, 30.2% were referred to cardiac rehabilitation with an attendance rate of 28.4%. Referred/declined patients were older, more likely to be female, with more heart failure and arrhythmia admissions and higher socio-economic disadvantage (median IRSAD: 950.1 vs 960.4, P < 0.001). | Given that administrative datasets were used, data may be incomplete with the risk of systematic bias in reporting of outcomes. | System and program considerations for future cardiac rehabilitation programs are identified. | |
| Atkins et al. (2013) 24 | Western Australia | To characterise admissions for an atherothrombotic event in the major arterial territories in men and women aged 35–84 years to tertiary, non-tertiary metropolitan, and rural hospitals in Western Australia during 2007 (2007) | Retrospective cohort study (data linkage) | Sociodemographic features; clinical features; hospital type (SES) | 11,670 index admissions | No | Comparisons of socio-economic disadvantage identified that for those admitted to rural hospitals, more than one-third were in the most disadvantaged quintile, compared with one-fifth admitted to any metropolitan hospital. | Given that administrative datasets were used, analysis was limited by the variables requested. Furthermore, given that SEIFA scores were based on the residential address on hospital admission, this could potentially overestimate or underestimate the level of disadvantage if the address was not the usual location for the person. | Due to these significant differences, findings have implications for atherothrombotic healthcare provision and the generalisation of research findings from studies conducted exclusively in the tertiary metropolitan hospitals. | |
| Baker et al. (2017) 25 | New South Wales | To present a method for comparing temporal trends in disease outcomes between multiple diseases and examine the effect of residual shared latent factors over time after adjusting for known factors (2001–2006) | Case study (Bayesian spatiotemporal method) | Hospitalisation rates | 13,866 cases (coronary artery disease); 6401 cases (chronic obstructive pulmonary disease); 5150 cases (chronic heart failure); 4869 cases (type 2 diabetes mellitus); 804 cases (hypertension) | No | It was identified that the choice of model depends upon the application. SES was substantively associated with hospitalisation rates, which differed for each disease. | Socio-economic information was available only for 1 year of the study. Information regarding other factors was not available at a spatial or temporal level. | Selecting the appropriate joint disease model enables the examination of temporal patterns and spatial factors for each disease. | |
| Biswas et al. (2019) 26 | Victoria | To examine whether there is an association between SES and baseline risk profile, clinical outcomes and use of secondary prevention therapy in patients undergoing PCI for ST-elevated myocardial infarction (2005–2015) | Prospective cohort study (data linkage) | Major adverse cardiovascular events; 12-month mortality; 30-day mortality, and 30-day major adverse cardiovascular events; secondary prevention medications; smoking status; post-PCI (SES) | 5665 patients | No | Patients residing in a lower socio-economic area were more likely to have diabetes mellitus, be smokers, and present to a non-PCI-capable hospital (all P ≤ 0.01). The median time to reperfusion was slightly higher in lower SES groups (211 [144–337] vs 193 [145–285] min, P < 0.001). Twelve months following PCI, lower SES patients had higher rates of ongoing smoking and lower use of guideline-recommended secondary prevention therapy (both P < 0.01). Despite these differences, SES group was not found to be an independent predictor of 12-month major adverse cardiovascular events. | As a registry-based study, all possible confounders were unable to be accounted for. Individual SES was not accounted for, rather based on the patient’s residential area. Other factors not included in the IRSD score, may also affect patient health, and were not accounted for. | There were baseline differences in lower SES patients who had more comorbidities and slightly longer reperfusion times, however, clinical outcomes following a PCI were similar across SES groups. | |
| Busija et al. (2013) 27 | Victoria | To estimate the proportion of stroke patients who take part in clinical research studies and to compare demographic and clinical profiles of research participants and non-participants (2004–2011) | Cross-sectional study (administrative dataset) | Participation in clinical research (patient characteristics; re-admissions; SES) | 5235 patients | No | 10.7% of patients took part in at least one of the 33 clinical research studies during the study period. High SES (OR = 0.74, 95% CI: 0.59–0.93) were associated with lower odds of research participation. | Patients included were from a single metropolitan teaching hospital. | Stroke patients who take part in clinical research are not representative of the ‘typical’ patient admitted to a stroke unit, which has implications for interpretation of research findings reported in stroke literature. | |
| Carter et al. (2019) 28 | Australia | To project the long-term impacts of Australian CVD deaths in 2003 on labour force participation and the present value of lifetime income forgone (2003) | Economic modelling study (data linkage) | Labour force participation; present value of lifetime income forgone (SES) | 18,450 premature deaths | No | Premature deaths due to CVD in 2003 accounted for 51,659 working years and $A2.69 billion in present value of lifetime income forgone when modelled to 2030 (95%CI: $2.63–$2.75 billion). Deaths occurring in individuals residing in the most socio-economically disadvantaged areas at the time of death had a disproportionately large impact on the total present value of lifetime income loss. | SEIFA index used as a proxy for an individual’s SES. | The magnitude of costs identified indicates the need for investments in effective healthcare interventions to provide positive economic returns. | |
| Cheng et al. (2020) 29 | Queensland | To examine short-term effects of winter temperature on the risk of myocardial infarction and explore spatial associations of winter hospitalisations with temperature and SES (2005–2015) | Ecological study using time series and spatial analysis (data linkage) | Myocardial infarction hospitalisation; daily temperature (SES) | 4978 hospitalisations | No | At the city level, each 1°C drop in temperature below a threshold of 15.6°C was associated with an RR of 1.016 (95%CI: 1.008–1.024) for myocardial infarction hospitalisations on the same day. Winter myocardial infarction incidence rates varied spatially in Brisbane, with a higher incidence rate in areas with lower socio-economic levels (RR: 0.900, 95%CI: 0.886–0.914 for each decile increase in IRSAD). | Study was undertaken in one metropolitan centre, limiting generalisability of findings. A stratified analysis of patient disease history was not undertaken. | Findings support that short-term winter temperature drops were associated with an elevated risk of myocardial infarction hospitalisations in a subtropical region with a mild winter, and the need for particular attention for persons residing in socio-economically disadvantaged areas. | |
| Chew et al. (2016) 30 | Australia | To explore geographic, socio-economic, health service and disease indicators associated with variation in angiography rates across Australia (2011) | Ecological study (data linkage) | Rates of acute coronary syndrome, angiography, revascularisation, and mortality (SES; health workforce indicators; rurality) | Not reported | No (Aboriginal and Torres Strait Islander status included as a variable) | Socio-economic disadvantage and remoteness were correlated with disease burden, acute coronary syndrome incidence and mortality, but not with angiography rate. | A linear relationship between variables was assumed. A small sample was used. | Variation in rates of coronary angiography, not related to clinical need, occurs across Australia. | |
| Close et al. (2014) 31 | New South Wales | To investigate the relationship between heart failure outcomes and SES (1998–2002) | Ecological study (data linkage) | Hospitalisation rates; mortality rates (SES) | Not reported | No | Rates of heart failure hospitalisations per local government area were inversely correlated with level of SES. | Study population from a single metropolitan centre, therefore findings may be limited in generalisability. | Higher rates of heart failure hospitalisations were identified for persons residing in areas that were socio-economically disadvantaged and indicate the need for targeted strategies. | |
| Dawson et al. (2022) 32 | Victoria | To assess whether there are disparities in incidence rates, care, and outcomes for patients with chest pain attended by emergency medical services according to SES (2015–2019) | Cohort study (data linkage) | Clinical outcomes; quality of care (SES) | 240,466 patients | No (Aboriginal and Torres Strait Islander status included as a variable) | Age-standardised incidence of chest pain was higher for patients residing in lower SES areas (lowest SES quintile 1595 vs highest SES quintile 760 per 100,000 person-years; P < 0.001). Patients of lower SES were less likely to attend metropolitan, private, or revascularisation-capable hospitals and had greater comorbidities. In multivariable models adjusted for clinical characteristics and final diagnosis, lower SES quintiles were associated with increased mortality risks and re-admission. | A proportion of ambulance cases could not be linked to hospital admissions and SES data were not available for all patients. Individual SES was not examined. | Lower SES was associated with a higher incidence of chest pain presentations to emergency and differences in care and outcomes, which indicate socio-economic disparities. | |
| Gutman et al. (2019) 33 | Victoria | To determine whether traditional markers of disadvantage [female sex, low SES (SES), and remoteness] are associated with lower prescription of evidence-based therapy and higher mortality in patients with moderate–severe heart failure with reduced ejection fraction (2005–2016) | Cohort study (trial dataset) | Mortality; evidence-based therapy delivery (SES; remoteness) | 452 patients | No | No difference in overall survival based on sex (HR = 1.19, 95%CI: 0.74–1.92) was identified. Higher SES or inner-city residence did not have an overall survival benefit. | Analysis of an observational trial therefore subject to confounders. | Delivery of care and likelihood of death were comparable between the sexes, SES groups, and persons residing in rural vs metropolitan areas. | |
| Hanigan et al. (2017) 34 | Australian Capital Territory | To explore the impact of the scale of spatial aggregation when describing the spatial distribution of selected hospital admissions for CVD and examine associations of socio-economic disadvantage (2011–2013) | Ecological study (data linkage) | Hospitalisation (SES) | 1365 admissions (myocardial infarction); 10,441 (CVD) | No | Relationships observed differed between the two types of spatial units. SA1-level exposure–response curve for rates against the disadvantage index extended in a linear fashion above the midrange level, whereas the SA2 level suggested a curvilinear form with no evidence that rates increased with higher disadvantage beyond the midrange. | Limitations of not being able to analyse individual-level data given the use of administrative datasets. | Results support findings from other research that identified that disadvantage increases the risk of CVD. The scale of analysis does influence the understanding of geographical patterns of socio-economic disadvantage and CVD morbidity. | |
| Hastings et al. (2022) 35 | Australia | To project new-onset CVD and related health economic outcomes in Australia by SES from 2021 to 2030 (2011–2012, 2017–2018 and 2020). | Retrospective cohort and prospective population economic modelling (data linkage) | New-onset cardiovascular events; SES | 3299 participants | No | Modelling showed that 8.4% of people in the most disadvantaged quintile were at high risk of CVD, compared with 3.7% in the least disadvantaged quintile. | Use of administrative datasets may not reflect current distribution of cardiovascular risk and use of an area-based measurement of SES. | There is a need to implement primary prevention interventions to reduce cardiovascular health inequity | |
| Huynh et al. (2018) 36 | Tasmania | To determine whether regional markers of SES were associated with days at home after discharge from hospital (2009–2012) | Ecological cohort study (data linkage) | Days at home; 30- and 90-day re-admission; mortality; re-admissions; days to first re-admission (SES; rurality) | 1391 patients | No | Included patients had a median of 352 days at home [IQR, 167–361]. All four SES indexes (i.e. IRSAD, IRSD, IEO, IER) and the remoteness index (P < 0.001) were adversely associated with days at home, independent of other clinical and non-clinical factors. | Use of administrative dataset and possibility of missing data, data only from public hospitals, and did not account for individual measures of SES. | Residential SES is associated with adverse outcomes in heart failure patients and requires targeted strategies. | |
| Hyun et al. (2018) 37 | Australia | To examine the influence of SES on in-hospital care, and clinical events for patients presenting with an acute coronary syndrome to public hospitals in Australia (2009) | Cohort study (data linkage) | In-hospital care; clinical events (SES) | 9238 patients | No | Following adjustments for patient characteristics, there were no differences in the odds of receiving coronary angiogram, revascularisation, prescription of recommended medication, or referral to cardiac rehabilitation across SES groups (P = 0.06, 0.69, 0.89 and 0.79, respectively). The most disadvantaged group were 37% more likely to have a major adverse cardiovascular event than the least disadvantaged group (OR (95%CI): 1.37 (1.1–1.71), P = 0.02) driven by incidence of in-hospital heart failure. | Based on observational data of which the follow-up data were self-reported. SES measure did not capture individual SES. | Gaps in delivery of care do not differ according to a patient’s SES. The likelihood of death is also comparable between SES groups. | |
| Jacobs et al. (2018) 38 | Australia | To assess the extent to which SES contributes to geographic disparity in CVD mortality (2009–2012) | Ecological study (data linkage) | CVD mortality (SES; rurality) | 180,530 deaths | No | After allowing for the mediating effect of SES, female participants living in inner regional areas and male participants living in remote/very remote areas had the greatest CVD mortality rates (mortality rate ratio (MMR): 1.12, 95%CI: 1.07–1.17; MRR: 1.15, 95%CI: 1.05–1.25, respectively) compared with those in major cities. | Possible that results may be influenced by misclassification of cause of death in administrative datasets, and use of an area-level measure of SES makes it difficult to account for other confounding factors. | SES explained a substantial proportion of the association between where a person resides and CVD mortality rates; however, remoteness has an effect above and beyond SES for subpopulations. Focusing on both socio-economic disadvantage and accessibility to reduce CVD mortality in regional and remote Australia is imperative. | |
| Jahan et al. (2022) 39 | Australia | To examine the care of patients with comorbid coronary heart disease and depression in general practice and explore the use of antidepressants by sociodemographic variables (2011–2018) | Cohort study (administrative dataset) | Antidepressant use (SES) | 880,900 medical records | No | Among male participants with newly recorded coronary heart disease and depression, antidepressant prescribing was more frequent in major cities or inner regional areas (~81%) than in outer/remote Australia (66.6%; 95%CI: 52.8–80.4%). No effect of SES. | Reliance on an administrative dataset. | Differences in prescribing were identified across geographic locations and need to be considered. | |
| Justo et al. (2017) 40 | Queensland | To investigate paediatric cardiac surgical outcomes in the Australian Aboriginal and Torres Strait Islander peoples (2006–2014) | Cohort study (administrative dataset) | Cardiac surgical outcomes (SES) | 123 Aboriginal and Torres Strait Islander peoples; 1405 non-Aboriginal and Torres Strait Islander peoples | Yes | 52.7% (62) of Aboriginal and Torres Strait Islander peoples were in the lowest third of the socio-economic index compared with 28.2% (456) of non-Aboriginal and Torres Strait Islander peoples (P ≤ 0.001). No difference was noted between the groups in 30-day mortality. | Referral bias cannot be excluded. Statistical power may be limited because the Aboriginal and Torres Strait Islander peoples cohort represented a small proportion of the total population. | The Aboriginal and Torres Strait Islander population had a higher 6-year mortality. This apparent relationship is explained by increased patient complexity, which may reflect negative social and environmental factors. | |
| Kang et al. (2021) 41 | Queensland | To examine differences in disease burden between Aboriginal and Torres Strait Islander peoples and to evaluate the care of individuals living in rural and remote – rather than urban – locations (1997–2017. | Ecological cohort study (administrative dataset) | Rheumatic heart disease incidence, hospitalisations, and surgery (SES; rurality) | 622 Aboriginal and Torres Strait Islander peoples; 64 non-Aboriginal and Torres Strait Islander peoples | Yes | An inverse correlation between an area’s SEIFA score and its rheumatic heart disease prevalence (rho = −0.77, P = 0.005) was identified. | The use of the SEIFA score may explain why there was greater correlation between the SEIFA score and the prevalence of rheumatic heart disease in different communities than the clinical endpoints in individual patients. | The burden of rheumatic heart disease remains high and is disproportionately experienced by socio-economically disadvantaged Aboriginal and Torres Strait Islander peoples. | |
| Kawai et al. (2022) 42 | Victoria | To explore regional trends in transient ischaemic attack hotspots using spatial regression followed by spatiotemporal analysis (2001–2011) | Ecological study (data linkage) | Transient ischaemic attack incidence (SES; rurality) | Not reported | No | Choropleth maps showed higher standardised transient ischaemic attack ratios in North- west rural region. | Individual-level data were not used, rather an aggregate count for each local government area, which can make findings susceptible to ecological inference fallacy. | A statistically significant spatial component to transient ischaemic attack rate over regional areas was identified but no temporal changes or yearly trends were supported. | |
| Korda et al. (2016) 43 | New South Wales | To quantify socio-economic variation in rates of primary and secondary CVD events in mid-age and older Australians (2006–2009) | Retrospective cohort study (data linkage) | Major CVD event (socio-economic position; rurality) | 266,684 participants | No | For primary and secondary events, incidence increased with decreasing education. For area-level disadvantage, CVD gradients were weak and non-significant in persons over 64 years of age. | Administrative data may not have captured all CVD events and there was reliance on some self-reported data. | Individual-level data are important for quantifying socio-economic variation in CVD incidence. | |
| Mariajoseph et al. (2022) 44 | Australia | To examine whether low SES may affect aneurysmal subarachnoid haemorrhage incidence and outcomes (2008–2018) | Ecological cross-sectional study (data linkage) | Incidence; clinical recovery (rurality; SES) | 7,209 cases | No | 3591 low-SES patients (49.8%) were identified. Average crude incidence of aneurysmal subarachnoid haemorrhage was persistently higher among the SES disadvantaged (6.6 cases per 100,000 person-years, 95%CI: 6.3–6.8), compared to the SES advantaged group (4.1 cases per 100,000 person-years, 95%CI: 4.0–4.2) (P < 0.0001). | Use of administrative data and use of an area-level measure of SES. | Aneurysmal subarachnoid haemorrhage occurs more frequently in low-SES communities. | |
| Mather et al. (2014) 45 | New South Wales | To examine variation in the magnitude of socio-economic inequalities in health and age-related variations in inequalities, according to the SES measure used (2006–2008) | Ecological cross-sectional study (data linkage) | Annual household income; highest education obtained (SES) | 205,709 participants | No | The relative index of inequality was largest for income and smallest for SEIFA; they were generally largest in the youngest age group and smallest in the oldest group. | Access to data on other markers of individual SES (i.e. wealth) was not possible. Sensitivity on individual vs area measures of SES will be dependent on precision of measurement and level of aggregation (e.g. at which level SEIFA scores were assigned). | Socio-economic inequality varies substantially according to the type of SES measure used and age (individual vs area). Researchers and policy makers should be aware of the extent to which SEIFA-based estimates underestimate the magnitude of health inequality compared with individual-level measures, especially in younger age groups. | |
| Mnatzaganian et al. (2018) 46 | Victoria | To inspect socio-economic gradients in admission to a CCU or an ICU for adult patients presenting with non-traumatic chest pain in three acute-care public hospitals in Victoria, Australia (2009–2013). | Ecological panel study (administrative data) | Admissions (SES) | 53,177 patients | No | A dose–response effect of socio-economic disadvantage and admission to CCU or ICU was identified, with risk of admission increasing as SES declined. Patients from the lowest SES locations were 27% more likely to be admitted to these units compared with those coming from the least disadvantaged locations, P < 0.001. | Use of administrative dataset with limited available data and no use of individual SES measures. | A dose–response effect was identified for socio-economic gradients in admissions to CCU and ICU, supporting increased cardiovascular morbidity as socio-economic disadvantage increases. | |
| Mnatzaganian et al. (2021) 47 | Australia | To investigate whether disparities in the management of CHD exist based on socio-economic indicators and remoteness of patient’s residence (2016–2018). | Cross-sectional study (administrative dataset) | Secondary prevention prescriptions; risk factors; treatment targets (SES; rurality) | 137,408 patients | No (Aboriginal and Torres Strait Islander peoples status noted) | Compared with patients from the highest SES fifth, those from the lowest SES fifth were 8% more likely to be prescribed more medications for secondary prevention (incidence rate ratio (95%CI): 1.08 (1.04–1.12)) but 4% less likely to achieve treatment targets (incidence rate ratio: 0.96 (95%CI: 0.95–0.98)). | Results may not be representative of population and information may be missing. | Despite being more likely to be prescribed medications for secondary prevention, those who are most socio-economically disadvantaged are less likely to achieve treatment targets. | |
| Morton et al. (2022) 48 | Victoria | To evaluate treatment disparities for myocardial infarction, as well as 1-year re-admission and mortality rates following myocardial infarction, by diabetes status, sex and socio-economic disadvantage (2012–2017). | Cohort study (data linkage) | Treatment disparities; 1-year re-admission; mortality rates (SES; diabetes) | 43,272 people | No | Male participants and people residing in more disadvantaged areas were at increased risk of re-admission and mortality following myocardial infarction. | Use of administrative datasets and potential for missing information and use of an area-level measure as a proxy for individual SES. | Inequalities attributed to socio-economic disadvantage are likely to continue without adaptations of policy. | |
| Morton et al. (2022) 49 | Victoria | To quantify 1-year re-admission and mortality rates following ischaemic stroke, and variation by diabetes status, sex, and socio-economic disadvantage (2012–2017). | Cohort study (data linkage) | One-year re-admission rates; mortality rates (SES; diabetes status) | 25,421 people | No | No relationship between socio-economic disadvantage and risk of cardiovascular or ischaemic stroke re-admission were identified, while 1-year mortality risk did increase with increasing socio-economic disadvantage (HR for most vs least disadvantaged quintile: 1.15 [95%CI: 1.03–1.27]; P trend = 0.006), and all-cause re-admission risk decreased (sub-HR: 0.94 [95%CI: 0.90–0.99]; P trend = 0.001). | Use of administrative datasets and potential for missing information. | A high risk of re-admissions following ischaemic stroke was identified. | |
| Nembhard et al. (2016) 50 | Western Australia | To describe survival into adulthood for Aboriginal and Torres Strait Islander children and non-Aboriginal and Torres Strait Islander children with selected congenital heart defects and determine whether Aboriginal and Torres Strait Islander children experience increased risk of mortality (1980–2010). | Cohort study (data linkage) | Mortality; 5- and 25-year survival (SES; rurality) | 4339 infants | Yes | Aboriginal and Torres Strait Islander children had lower survival rates than non-Aboriginal and Torres Strait Islander children for all congenital heart defects. | Possible underestimation of mortality because did not use the Australian National Death Index. | Long-term survival was lower for Aboriginal and Torres Strait Islander children with congenital heart defects. Increased risk may be due to SES and environmental factors. | |
| Nghiem et al. (2022) 51 | Queensland | To present the baseline characteristics of index cardiovascular hospitalisations between first time and recurrent admissions (2010–2015). | Cohort study (data linkage) | Admission types; Charlson comorbidity index; hospital characteristics (SES) | 132,343 hospitalisations | No (Aboriginal and Torres Strait Islander peoples status noted) | SEIFA quintiles were evenly distributed for recurrent admissions, whereas higher quintiles were overrepresented for first time admissions. | Only follows those with a cardiovascular hospitalisation during 2010. | Demonstrates that linked health data is a useful tool to examine factors mediating with CVD progression. | |
| Nichols et al. (2021) 52 | Tasmania | To understand early (<24 h post ictus) and late (up to 12 months) survival post aneurysmal subarachnoid haemorrhage with a focus on rurality and SES (2010–2014). | Ecological cohort study (data linkage) | Aneurysmal subarachnoid haemorrhage- related death (SES; rurality) | 237 cases | No | 12-month mortality was 52.3% with 54.0% of these deaths occurring within 24 h post-ictus. In univariable analysis of 12-month survival, outcome was not influenced by SES, but rural geographical location was associated with a non-significant increase in death. | Use of an administrative dataset with limited variables and missing data. | Survival to 12 months was not related to geographical location or SES. | |
| Nichols et al. (2018) 53 | Tasmania | To define a new baseline of the incidence and temporal trends of aneurysmal subarachnoid haemorrhage within an Australian population (2010–2014). | Ecological cohort study (data linkage) | Incidence (SES; rurality) | 237 persons | No | A significant association between area-level socio-economic disadvantage and incidence was identified, with the rate of aneurysmal subarachnoid haemorrhage in disadvantaged geographical areas being 1.40-fold higher than that in advantaged areas (95%CI: 1.11–1.82; P = 0.012). | Use of an area-level measure of SES may have underestimated findings and subject to ecological fallacy. | A high incidence of aneurysmal subarachnoid haemorrhage was identified with socio-economic variations. Addressing this is imperative for improving disease prevention and management. | |
| Nichols et al. (2020) 54 | Tasmania | To examine the effect of geographical location, SES, and inter-hospital transfer on time to treatment following an aneurysmal subarachnoid haemorrhage (2010–2014). | Ecological cohort study (data linkage) | Time to treatment (rurality; SES; inter-hospital transfer time) | 205 cases | No | The median (IQR) time to intervention was 13.78 (6.48–20.63) h. Socio-economic disadvantage was associated with a 1.52-fold increase in the time to hospital (P < 0.05) and a 1.76-fold increase in time to neurosurgical admission (P < 0.05). | Findings may not be generalisable to other healthcare systems. | Time to treatment was negatively influenced by socio-economic disadvantage, geographical location, and inter-hospital transfers, which requires attention. | |
| Pemberton et al. (2019) 55 | Queensland | To describe temporal trends in incidence of pre-hospital outcomes from adult out-of-hospital cardiac arrest attended by Queensland Ambulance Service paramedics (2012–2014). | Cohort study (data linkage) | Resuscitation status (rurality; SES) | 30,541 cases | No | Crude incidence significantly increased over time for No-Resus and Sustained-return of spontaneous circulation, and significantly decreased for No-return of spontaneous circulation. These trends were reflected in major cities, inner and outer regional areas. The incidence of out-of-hospital cardiac arrest increased in areas categorised as lower relative advantage. | Confounders not accounted for, and use of residential postcode to estimate SES and remoteness. | Factors to be addressed include being of middle age, more rural location, and lower SES. | |
| Pemberton et al. (2019) 56 | Queensland | To describe incidence in pre-hospital outcomes of adult out-of-hospital cardiac arrest of presumed cardiac aetiology, attended by QAS paramedics (2012–2014). | Cohort study (data linkage) | Resuscitation status (rurality; SES) | 30,560 cases | No | Incidence was significantly greater in male than in female participants and incrementally increased with age. An inverse association between incidence and SES was identified (SEIFA 1 and 2: 81.34 per 100,000 [95%CI:79.28–83.40]; SEIFA 9 and 10: 61.57 per 100,000 [95%CI: 59.67–63.46]). | Confounders not accounted for, and use of residential postcode to estimate SES and remoteness. | Prevention and management strategies for out-of-hospital cardiac arrests are required for lower socio-economic groups. | |
| Pemberton et al. (2021) 57 | Queensland | To describe annual incidence and temporal trends (2002–2014) in incidence of long-term outcomes of adult out-of-hospital cardiac arrest of presumed cardiac aetiology attended by QAS paramedics (2012–2014). | Cohort study (data linkage) | Survival (SES; rurality) | 4393 cases | No | Incidence of total admitted events, survival 30–364 days, and survival 365+ days, increased over time; no trends were observed for survival <30 days. | Confounders not accounted for, and use of residential postcode to estimate SES and remoteness. | Prevention and management strategies for out-of-hospital cardiac arrests are required for lower socio-economic groups. | |
| Rachele et al. (2016) 58 | Queensland | To examine associations between neighbourhood socio-economic disadvantage and self-reported type 2 diabetes and heart disease (2007). | Cross-sectional study (survey and data linkage) | Self-reported type 2 diabetes, heart disease and comorbidity (SES) | 10,620 participants | No | Compared with the most advantaged neighbourhoods, residents of the most-disadvantaged neighbourhoods were more likely to report type 2 diabetes, heart disease, and comorbidity. This weakened after adjustment for individual-level socio-economic position but remained statistically significant for type 2 diabetes and comorbidity. | The study had a 31.5% survey non-response rate, higher among persons residing in lower socio-economic areas, which may underestimate findings. | There is a need to establish why persons residing in disadvantaged areas are more likely to have heart disease and type 2 diabetes independent of their individual socio-economic position. | |
| Ramkumar et al. (2019) 59 | Tasmania | To investigate whether IRSAD, IEO and IER were associated with incident atrial fibrillation, independent of risk factors and cardiac function (not reported). | Cohort study (administrative dataset) | Proportion of new-onset atrial fibrillation (SES) | 379 participants (atrial fibrillation n = 50; sinus rhythm n = 329) | No | Persons with atrial fibrillation (n = 50, 13%) were more likely to be male (64% vs 42%, P = 0.003). Areas with lower SES (IAD (assumed to be SEIFA 2011 IRSAD) and IEO) had a higher risk of incident atrial fibrillation. | Potential for population selection bias owing to recruitment through media outlets and a small sample. | Area SES was associated with the risk of incident atrial fibrillation, independent of clinical risk, indicating that additional resources may be required for people residing in these areas. | |
| Randall et al. (2016) 60 | Western Australia | To investigate acute myocardial infarction incidence in Australia in more detail, including both hospitalisations and out of hospital deaths in the Western Australian population (1993–2012). | Ecological study (data linkage) | Incidence of myocardial infarction (rurality; SES) | 97,638 cases | No | Myocardial infarction incidence decreased in Western Australia from 1993 to 2012 by 1.2% per year (95%CI: −1.7 to −0.8). There was a large effect of SES, with those from the lowest quintile having a 68% higher acute myocardial infarction incidence than those from the highest socio-economic quintile. | Diagnostic process for acute myocardial infarction changed, which may mediate with results. | Focus on sub-populations is required for the primary care prevention of acute myocardial infarction. | |
| Roberts et al. (2015) 61 | Queensland, Northern Territory and Western Australia | To compare regional differences in the prevalence of rheumatic heart disease detected by echocardiographic screening in high-risk Aboriginal and Torres Strait Islander children (2008–2010). | Cross-sectional study (screening) | Rheumatic heart disease prevalence (SES) | 3946 participants | Yes | Prevalence of rheumatic heart disease differed between regions. Evaluation of socio-economic data suggests that the Top End group was the most disadvantaged in our study population. | Selection bias may contribute to differences in prevalence. | Prevalence of rheumatic heart disease in and Torres Strait Islander children residing in remote settings is significant. Regional variations in prevalence were observed and need to be considered. | |
| Robins et al. (2017) 62 | Victoria | To investigate the differences in hypertensive disease hospitalisations across rural and urban Victoria, and to determine predicting variables (2010–2015). | Ecological study (data linkage) | Hypertensive disease hospitalisations (SES) | 11,205 admissions | No (Aboriginal and Torres Strait Islander peoples status noted) | Hospitalisation rates were consistently higher in rural areas than in urban areas, and rural residents on average stayed in hospital for longer. Significant predictor variables for hypertensive disease hospitalisation included various indicators of socio-economic disadvantage, GPs per 1000 population and GP attendance per 1000 population. | Limited variables available for analysis in administrative datasets. | Hypertensive disease hospitalisation is increasing in Victoria, with rates in rural areas exceeding that of urban areas. Further research is required. | |
| Roseleur et al. (2021) 63 | Australia | To investigate the prevalence of diagnosed hypertension in Australian general practice and explore whether hypertension control is influenced by sociodemographic characteristics, duration since diagnosis, or prescription of antihypertensive medications (2017). | Cross-sectional study (administrative dataset) | Hypertension control (SES) | 1,198,199 patients | No | Blood pressure control was lower in female participants (54.1%) than in male participants (55.7%) and in the oldest age group (52.0%), but there were no differences by SES. | Potential for missing information in administrative datasets. | Prevalence of hypertension varied by sociodemographic, but there were no differences in assessment or control of hypertension by SES. | |
| Saghapour et al. (2021) 64 | Queensland | To examine the total effect of neighbourhood disadvantage on CVD to address the limitations of previous research by examining the indirect effects of neighbourhood disadvantage on CVD (2007–2016). | Cohort study (data linkage) | Self-reported CVD (neighbourhood disadvantage) | 11,035 participants | No | The incidence of CVD was found to be significantly higher in the most disadvantaged neighbourhoods (OR 1.50; HR 1.29) compared with the least disadvantaged. Physical activity was a significant mediator of this. | Use of self-reported CVD. | Further research is required around the social and built environment, physical activity, and CVD. | |
| Shi et al. (2014) 65 | Victoria | To examine the clinical profile, early outcomes and late survival of patients presenting for coronary surgery, to identify whether rurality and SES were predictors of early and late outcome (2001–2009). | Ecological study (data linkage) | Coronary surgery outcomes (rurality; SES) | 14,150 patients | No | Patients from socio-economically-disadvantaged areas had a greater burden of cardiovascular risk factors including diabetes, obesity and current smoking. Thirty-day mortality (disadvantaged 1.6% vs advantaged 1.6%, P > 0.99) was similar between groups as was late survival (7 years: 83 ± 0.9% vs 84 ± 1.0%, P = 0.79). Propensity analysis did not show SES or rurality to be associated with late outcomes. | SES assigned by postcodes and limitation of retrospective study design. | Targeted strategies to promote early recognition and referral of patients with coronary disease are required for those residing in areas characterised by socio-economic disadvantage. | |
| Smurthwaite and Bagheri (2017) 66 | South Australia | To determine the geographic variation of obesity, CVD and type 2 diabetes, using general practice clinical data (2010–2014). | Ecological cross-sectional study (data linkage) | Prevalence of obesity, CVD and type 2 diabetes (SES) | 20,594 patients | No | Spatial distribution of obesity, CVD, and type 2 diabetes varied across geographical areas. An inverse relationship was observed between area-level prevalence of CVD, obesity, and type 2 diabetes with SES. | Potential for selection bias in use of general practice records and potential for lack of generalisability. | Further research is required around community profiles and disease distribution. | |
| Straney et al. (2016) 67 | Victoria | To identify population-based demographic and health factors associated with (1) high incidence and (2) low bystander CPR, and to examine the contribution of these factors to the variation seen across Victoria (2011–2013). | Ecological cross-sectional study (data linkage) | Rates of out-of-hospital cardiac arrest; bystander cardiac pulmonary resuscitation (LGA; SES) | 15,830 cardiac arrests | No | Incidence rates varied across the state between 41.9 and 104.0 cases/100,000 population. The proportion of the population over 65, SES, smoking prevalence and education level were significant predictors of incidence in the multivariable model, explaining 93.9% of the variation in incidence between LGAs. | Study population may not be representative of people at risk, therefore incidence rate may be overestimated. | Characteristics identified will be useful in targeting regions for interventions. | |
| Tideman et al. (2013) 68 | South Australia and Victoria | To understand causes of geographical CVD mortality disparities (2004–2006). | Ecological cross-sectional study (data linkage) | Age group-specific measures of absolute CVD risk; mortality rates (SES; rurality) | 1563 Greater Green Triangle participants; 3036 North West Adelaide Study participants | No (Aboriginal and Torres Strait Islander peoples status noted) | Few significant differences in CVD risk between the study regions were identified, with absolute CVD risk ranging from approximately 5% to 30% in the 35–39 and 70–74 age groups, respectively. Lower measures of SES were associated with worse cardiovascular outcomes regardless of geographic location. | Study populations not necessarily representative. | Further research is required around the determinants of CVD and targeted strategies. | |
| Xu et al. (2021) 69 | Queensland | To assess the effects of extreme temperatures on hospitalisations and post-discharge deaths for stroke in individuals with and without pre-existing hyperlipidaemia, and examine whether individual- and community-level characteristics modified the temperature–stroke relationship (2005–2013). | Ecological cohort study (data linkage) | Hospitalisations; post-discharge deaths for stroke (SES; greenspace) | 11,469 cases | No | People living in suburbs with the lowest socio-economic advantage level or the lowest economic resources level were most vulnerable to the effects of heat and cold on hospitalisations for stroke. | Area-level SES used as a proxy for individual SES. | Need for targeted interventions for individuals with hyperlipidaemia to reduce heat-related stroke burden, especially for those residing in socio-economic areas characterised by disadvantage. | |
| Yiallourou et al. (2022) 70 | Victoria | To compared age-adjusted all-cause and CVD mortality for women registered for fertility treatment who received it, compared with those who did not (1975–2014). | Cohort study (data linkage) | Standardised mortality rates for all-cause and CVD mortality stratified by area-level socio-economic disadvantage | 44,149 women | No | All-cause and CVD mortality was lower for the study participant when compared to the general female population. The standardised mortality rate was lowest for both groups in the fifth IRSD quantile (least disadvantaged) | CVD risk was not assessed and SEIFA used as a proxy. | Fertility treatment does not increase long-term all-cause of CVD mortality risk. |
CCU, coronary care unit; CHD, coronary heart disease; CPR, cardiopulmonary resuscitation; GP, general practitioner; HR, Hazard Ratio; ICU, intensive care unit; IEO, Index of Education and Occupation; IER, Index of Economic Resources; IRSAD, Index of Relative Socio‐Economic Advantage and Disadvantage; IRSD, Index of Relative Socio‐Economic Disadvantage; LGA, local government area; PCI, percutaneous coronary intervention; QAS, Queensland Ambulance Service; RR, relative risk; SA1/2, statistical area level 1/2; SLA, statistical local area.
Study design varied and included ecological studies using data linkage methods (n = 22: n = 10 study design not specified, n = 5 cohort study, n = 5 cross-sectional study, n = 1 time series and spatial analysis, n = 1 panel study), cohort studies using data linkage methods (n = 19) or analysis of trial datasets (n = 1), cross-sectional studies using screening (n = 1) or data linkage and/or surveys (n = 4), an economic modelling study using data linkage (n = 1), and a case study (n = 1).
The included studies cited 78 data sources. These included existing state or territory databases (n = 32; e.g. Victorian Admitted Episodes Dataset, Tasmanian Death Registry), national databases (n = 19; e.g. Australian Pharmaceutical Benefits Scheme database), research and/or quality improvement databases (n = 22; e.g. Medicine Insight database), health service databases (n = 4; e.g. hospital specific), and other administrative databases not otherwise named (n = 1).
Application of Socio-Economic Indexes for Areas indexes
All of the studies cited the use of SEIFA in the methods section (Table 3). One study did not cite which SEIFA version (year of release) or SEIFA index was applied,69 13 studies did not report which specific SEIFA index was used but reported the SEIFA version applied, and two studies did not report which version was used. Studies that reported the version and index applied most frequently, used the Index of Relative Socio-economic Advantage and Disadvantage (IRSAD) 2011.
| Citation | SEIFA indexes (raw scores or deciles) | Spatial unit | How SEIFA was applied in methods and results | |
|---|---|---|---|---|
| Adair and Lopez (2021) 22 | IRSAD 2006 for deaths registered in 2006–2010 | SLA for deaths registered in 2006–2010 | Indirect: Australian Cause of Death Unit Record File place of usual residence was linked to IRSAD. Inequalities were assessed by area socio-economic quintile (Q1 lowest SES to Q5 highest SES). | |
| IRSAD 2011 for deaths registered in 2011–2017 | SA2 for deaths registered in 2011–2017 | |||
| Astley et al. (2020) 23 | IRSAD 2016 | Not reported | Direct: Used linkage key identifiers held separately from any personal demographic information to determine IRSAD score. Presented for each cohort (not referred, referred/declined, and attended) as an IRSAD median score with interquartile range. | |
| Atkins et al. (2013) 24 | SEIFA 2006 (index used not reported) | If available, collection district. If not available, SLA or LGA (non-ABS structure) | Indirect: A SEIFA score was assigned by the data linkage program at index admission using residential address at hospital admission. For analysis, deciles were grouped into quintiles at the Collection District level (available for 89% of the sample, or SLA level or LGA (11%). | |
| Baker et al. (2017) 25 | IRSAD 2001 | SLA | Indirect: Data were obtained from the 2001 versions specific to each of the 21 SAs of resident included in the study. Raw scores for each index were split into quartiles for analysis. | |
| IRSD 2001 | ||||
| IEO 2001 | ||||
| IER 2001 | ||||
| Biswas et al. (2019) 26 | IRSD 2011 | Postcodes | Indirect: Patients’ residential postcode was used to assign an IRSD decile, which were merged into quintiles for analysis with Q1 as patients with the lowest SES, and Q5 as patients with the highest SES. | |
| Busija et al. (2013) 27 | IRSAD 2006 | Postcodes | Indirect: Assigned using participant postcode, and classified into tertiles for analysis (low SES, middle SES, and high SES). Not reported whether SEIFA raw scores or deciles were used. | |
| IRSD 2006 | ||||
| IEO 2006 | ||||
| IER 2006 | ||||
| Carter et al. (2019) 28 | SEIFA 2011 (index used not reported) | Not reported | Indirect: Area of usual residence (undefined) converted to a SEIFA quintile. | |
| Cheng et al. (2020) 29 | IRSAD 2011 | Postal areas | Indirect: Deciles for postal areas were obtained from IRSAD 2011 and grouped into quintiles for analysis. | |
| Chew et al.(2016) 30 | SEIFA 2011 (index used not reported) | Not reported | Indirect: Not reported. Mean SEIFA raw score and standard deviation presented in analysis by geographical area (metropolitan, regional, and rural). Not stated how geographical areas were classified. | |
| Close et al. (2014) 31 | IRSD 2001 | LGA | Direct: Outcomes by LGA were correlated with SEIFA Index of Disadvantage scores. | |
| Dawson et al. (2022) 32 | IRSD 2016 | Postcodes | Indirect: Scores were divided into quintiles and allocated for postcodes. | |
| Gutman et al. (2019) 33 | IRSAD 2011 | Not reported | Indirect: Patient residential address was used to determine IRSAD quintiles. | |
| Hanigan et al. (2017) 34 | IRSD 2011 | SA1 | Indirect: The SA1 and SA2 of residence was used to assign an area-level measure of socio-economic disadvantage using the IRSD. Original data were standardised to Z-scores because SA1 values are known to have a national mean of 1000 and standard deviation of 100. Given that the principal components used to construct the index are arbitrary with respect to their sign (positive or negative), the index was rescaled to improve intuitive interpretation. More-disadvantaged areas were assigned positive scores, and less-disadvantage areas, negative scores. | |
| Hastings et al. (2022) 35 | SEIFA 2011 (index used not reported) | Not reported | Indirect: SEIFA deciles were grouped in quintiles, defining five socio-economic levels (SE 1–SE 5). The model population was profiled on the latest available demographic data for the Australian population in 2020 divided into five quintiles, each representing a socio-economic quintile, with SE 1 being the lowest socio-economic quintile (most disadvantaged) and SE 5 being the highest quintile (least disadvantaged). | |
| Huynh et al. (2018) 36 | IRSAD 2011 | Not reported | Indirect: Patient’s residential address was used to determine SES using the four indexes of SEIFA. Grouped into tertiles for analysis by index used. | |
| IRSD 2011 | ||||
| IER 2011 | ||||
| IEO 2011 | ||||
| Hyun et al. (2018) 37 | IRSD 2011 | Postal areas | Indirect: Patient’s postcode of usual residence was matched with the IRSD postcode. IRSD deciles were further stratified into four groups, where group 1 includes 20% of the lowest SES areas and group 4 includes 40% of the highest SES areas. | |
| Jacobs et al. (2018) 38 | IRSAD 2011 | SA2 | Indirect: IRSAD scores for each SA2 were obtained. Sex-specific proportions of SA2s were tabulated by remoteness and IRSAD category (quintiles). To improve clarity of maps, data were aggregated to an SA3 level (an SA3 is composed of multiple SA2s, with populations ranging from 30,000 to 130,000), and death rates were then calculated and age-standardised in the same manner as at the SA2 level. | |
| Jahan et al. (2022) 39 | IRSAD 2016 | Not reported | Indirect: Used IRSAD quintiles and assigned to patients. Not reported explicitly how this occurred, but the authors refer to IRSAD quintiles based on postcodes in the methods section. | |
| Justo et al. (2017) 40 | SEIFA 2011 (index used not reported) | Postcodes | Indirect: The SES of the usual residence of children was assessed at the postcode level and were presented as tertiles in the results section. | |
| Kang et al. (2021) 41 | SEIFA 2011 (index used not reported) | Not reported | Direct: Used SEIFA (index not reported) scores. Area level not reported. | |
| Kawai et al. (2022) 42 | IRSAD (year not reported) | LGA | Direct: Decile assigned according to the IRSAD in each LGA. | |
| Korda et al. (2016) 43 | IRSD 2006 | Postcodes | Indirect: IRSD was categorised into population-based quintiles and assigned to individuals using their postcode of residence. | |
| Mariajoseph et al. (2022) 44 | SEIFA 2016 (index used not reported | Not reported | Indirect: Quintiles collapsed into two categories: socio-economic advantage was defined to encompass middle and high socio-economic groups (quintile 3, 4, 5), while socio-economic disadvantage was limited to quintiles 1 and 2. | |
| Mather et al. (2014) 45 | IRSD 2006 | Postcodes | Indirect: IRSD was based on postcode of residence and categorised into population-based quintiles using cut-off scores from the 2006 Australian Census. | |
| Mnatzaganian et al. (2018) 46 | SEIFA 2011 (index used not reported) | Postcodes | Indirect: Individual’s residential postcode was used to assign a SEIFA score. The SEIFA was further used to calculate a Relative Index of Inequality (RII), which is a regression-derived index summarising the magnitude of socio-economic disadvantage while taking into account the sample size and the relative disadvantage experienced by each individual. The estimated RII was further introduced as quintiles categorised according to the score’s distribution in the sample. | |
| Mnatzaganian et al. (2021) 47 | IRSD 2016 | Postcodes | Indirect: Patients’ most recent residential addresses were used because these were recorded in the last patient–GP encounter during the 2-year study period. The SEIFA-IRSD deciles were categorised into five groups. | |
| Morton et al. (2022) 48 | IRSD 2016 | Postcodes | Indirect: IRSD was assigned based on the individual’s last known postcode. IRSD was classified into quintiles, where a higher IRSD indicates a lower proportion of disadvantaged people in an area. | |
| Morton et al. (2022) 49 | IRSD 2016 | Postcodes | Indirect: IRSD was assigned based on the individual’s last known postcode. IRSD was classified into quintiles, where a higher IRSD indicates a lower proportion of disadvantaged people in an area. | |
| Nembhard et al. (2016) 50 | IRSD 1996, 2011 and 2006 | Collection district | Indirect: Obtained maternal community-level social class information by linking to the SEIFA collection district-level data using maternal household postcode at the time of delivery. | |
| Nghiem et al. (2022) 51 | SEIFA 2011 – index used not reported | Postcodes | Indirect: Used quintiles to represent socio-economic advantage of a region using postcode data. | |
| Nichols et al. (2021) 52 | IRSAD 2011 | SA2 | Indirect: Assigned from geocoded residential street addresses using SA2 data. Deciles were dichotomised with scores of less than or equal to 3 representing disadvantage. | |
| Nichols et al. (2018) 53 | SEIFA 2011 (index used not reported) | SA2 | Indirect: Assigned using participant’s geocoded residential street address at the time of the haemorrhage. Data were available at SA2. To assess the association between socio-economic disadvantage and incidence, the SEIFA deciles were analysed both on a linear basis across the deciles and dichotomously (split at the 3rd decile). | |
| Nichols et al. (2020) 54 | SEIFA 2011 (index used not reported) | SA2 | Indirect: Applied using participant’s geocoded residential street address at the time of the haemorrhage. SEIFA deciles measuring socio-economic advantage/disadvantage were dichotomised (score ≤ 3 = disadvantaged). | |
| Pemberton et al. (2019) 55 | IRSAD 2011 | SA2 | Indirect: Determined using residential postcode. Analysis presented as quintiles. | |
| Pemberton et al. (2019) 56 | IRSAD 2011 | SA2 | Indirect: Determined using residential postcode. Analysis presented as quintiles. | |
| Pemberton et al. (2021) 57 | IRSAD 2011 | SA2 | Indirect: Determined using residential postcode. Analysis presented as quintiles. | |
| Rachele et al. (2016) 58 | IRSD 2006 | Collection district | Indirect: Neighbourhood socio-economic disadvantage was derived using a weighted linear regression, using scores from the ABS IRSD from each of the previous six censuses from 1986 to 2011. Derived socio-economic scores from each of the HABITAT neighbourhoods were then quantised as percentiles, relative to all of Brisbane. The 200 HABITAT neighbourhoods were then grouped into quintiles with Q1 denoting the 20% least disadvantaged areas relative to the whole of Brisbane and Q5 the most disadvantaged 20%. | |
| Ramkumar et al. (2019) 59 | IAD 2011 (assumed to be SEIFA 2011 IRSAD) | Postcodes | Indirect: Indexes applied using participants’ postcode. Reported using deciles. | |
| IER 2011 | ||||
| IEO 2011 | ||||
| Randall et al. (2016) 60 | SEIFA 2001 2006 and 2011 (index used not reported) | Not reported | Indirect: Index was classified into quintiles. | |
| Roberts et al. (2015) 61 | IRSD 2011 | Not reported | Indirect: Information about school attendance and the Aboriginal and Torres Strait Islander peoples status of enrolled students, as well as Index of Community Socio-Educational Advantage (ICSEA 10) scores were obtained for each participating school from the Australian Government’s MySchool website. SEIFA scores were obtained for each participating community from the ABS 2011 census data and presented as mean and median scores. | |
| IRSAD 2011 | ||||
| Robins et al. (2017) 62 | IRSD (year not reported) | LGA | Direct: Data sourced according to LGAs. Not clear how indexes were applied in results. | |
| IER (year not reported) | ||||
| IEO (year not reported) | ||||
| Roseleur et al. (2021) 63 | IRSAD 2011 | Not reported | Indirect: Not clear how SEIFA was applied. Presented in IRSAD quintiles in results. | |
| Saghapour et al. (2021) 64 | IRSD 2011 | Collection districts | Indirect: Each of the 200 neighbourhoods was assigned a socio-economic score using the IRSD. The derived IRSD scores for the HABITAT neighbourhoods were then grouped into quintiles, with Q5 denoting the 20% least disadvantaged areas relative to the whole of Brisbane and Q1 denoting the 20% most disadvantaged areas. | |
| Shi et al. (2014) 65 | IRSAD 2011 | Postcodes | Indirect: Determined by linking patient residential postcode to the IRSAD from the ABS. To enable a sizeable group for statistical comparison, the deciles of socio-economic disadvantage were collapsed into quintiles. In the socio-economic analysis, outcomes were compared between the quintiles of most and least socio-economic disadvantage. | |
| Smurthwaite and Bagheri (2017) 66 | SEIFA 2011 (index used not reported) | SA1 | Indirect: SES was classified into tertiles based on ABS SEIFA data, including low socio-economic, moderate socio-economic, and high socio-economic regions. | |
| Straney et al. (2016) 67 | IRSAD 2011 | LGA | Direct: Scores were assigned for each LGA. | |
| Tideman et al. (2013) 68 | IRSD 2006 | SLA | Direct: The distribution of IRSD scores between the two groups was compared and the relationship between IRSD and CVD mortality rates were explored. | |
| Xu et al. (2021) 69 | SEIFA (year and index not reported) | Not reported | Not reported. | |
| Yiallourou et al. (2022) 70 | IRSD 2016 | Postcodes | Indirect: Standardised mortality ratios were stratified by IRSD quintiles. |
HABITAT, How Areas in Brisbane Influence healTh And acTivity project; LGA, local government area; SA1/2, statistical area level 1/2; SLA, statistical local area.
Of the included studies, seven studies used a direct approach that involved assigning SEIFA scores or deciles to study populations for analysis. Most studies (n = 41) applied an indirect approach, which involved collapsing SEIFA scores or deciles into further categories, including quintiles, quartiles and tertiles. One study did not report how SEIFA was applied.69
Studies varied in the spatial unit applied for analysis. Of the studies, 22 applied a non-ABS spatial unit for analysis (n = 16 applied SEIFA using postal areas (POAs) or postcodes (terms used interchangeably, noting that SEIFA is available only for POA); n = 4 used local government area (LGA); n = 2 used statistical local area), 14 used an ABS spatial unit for analysis (n = 2 statistical area level 1 (SA1); n = 7 SA2; n = 1 multiple statistical area levels; n = 4 collection district), and 13 studies did not report what the spatial unit of analysis was. One study analysed data using multiple spatial units (SA1 and SA2), and identified that analysis at a finer spatial resolution supported a stronger association between SES as an exposure and the outcome of interest (i.e. rates of CVD).34 Another study examined the association between temperature and myocardial infarction hospitalisations in a metropolitan centre using postal area levels (non-ABS structure) rather than the finest spatial resolution available for SEIFA (i.e. SA1).29
Quality assessment and study limitations
The quality of studies varied. A common limitation of cross-sectional and cohort studies was that confounders were not identified or adjusted for (Tables 4, 5 and 6).
| Citation | Were the criteria for inclusion in the sample clearly defined? | Were the study subjects and the setting described in detail? | Was the exposure measured in a valid and reliable way? | Were objective, standard criteria used for measurement of the condition? | Were confounding factors identified? | Were strategies to deal with the confounding factors stated? | Were the outcomes measured in a valid and reliable way? | Was appropriate statistical analysis used? | |
|---|---|---|---|---|---|---|---|---|---|
| Adair and Lopez (2021) 22 | Y | Y | Y | Y | N | N | Y | Y | |
| Busija et al. (2013) 27 | Y | Y | Y | Y | Y | Y | Y | Y | |
| Chew et al.(2016) 30 | Y | Y | Y | Y | Y | Y | Y | Y | |
| Hanigan et al. (2017) 34 | Y | Y | Y | Y | Y | Y | Y | Y | |
| Jacobs et al. (2018) 38 | Y | Y | Y | Y | N | N | Y | Y | |
| Kawai et al. (2022) 42 | Y | Y | Y | Y | Y | Y | Y | Y | |
| Mariajoseph et al. (2022) 44 | Y | Y | Y | Y | Y | Y | Y | Y | |
| Mather et al. (2014) 45 | Y | Y | Y | Y | N | N | Y | Y | |
| Mnatzaganian et al. (2021) 47 | Y | Y | Y | Y | Y | Y | Y | Y | |
| Rachele et al. (2016) 58 | Y | Y | Y | Y | Y | Y | Y | Y | |
| Randall et al. (2016) 60 | Y | Y | Y | Y | N | N | Y | Y | |
| Roberts et al. (2015) 61 | Y | Y | Y | Y | N | N | Y | Y | |
| Robins et al. (2017) 62 | Y | Y | Y | Y | Y | Y | Y | Y | |
| Roseleur et al. (2021) 63 | Y | Y | Y | Y | N | N | Y | Y | |
| Smurthwaite and Bagheri (2017) 66 | Y | Y | Y | Y | N | N | Y | Y | |
| Straney et al. (2016) 67 | Y | Y | Y | Y | N | N | Y | Y | |
| Tideman et al. (2013) 68 | Y | Y | Y | Y | N | N | Y | Y |
N, no; NA, not applicable; Y, yes.
| Citation | Were the two groups similar and recruited from the same population? | Were the exposures measured similarly to assign people to both exposed and unexposed groups? | Was the exposure measured in a valid and reliable way? | Were confounding factors identified? | Were strategies to deal with confounding factors stated? | Were the groups/participants free of the outcome at the start of the study (or at the moment of exposure)? | Were the outcomes measured in a valid and reliable way? | Was the follow up time reported and sufficient to be long enough for outcomes to occur? | Was follow up complete, and if not, were the reasons for loss to follow up described and explored? | Were strategies to address incomplete follow up used? | Was appropriate statistical analysis used? | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Astley et al. (2020) 23 | Y | Y | Y | N | N | Y | Y | Y | Y | Y | Y | |
| Atkins et al. (2013) 24 | Y | Y | Y | Y | Y | Y | Y | Y | NA | NA | Y | |
| Biswas et al. (2019) 26 | Y | Y | Y | Y | Y | Y | Y | Y | NA | NA | Y | |
| Carter et al. (2019) 28 | NA | NA | Y | N | N | N | Y | Y | Y | NA | Y | |
| Cheng et al. (2020) 29 | NA | NA | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Close et al. (2014) 31 | NA | NA | Y | N | N | Y | Y | Y | Y | Y | Y | |
| Dawson et al. (2022) 32 | NA | NA | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Gutman et al. (2019) 33 | NA | NA | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Hastings et al. (2022) 35 | NA | NA | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Huynh et al. (2018) 36 | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Hyun et al. (2018) 37 | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Jahan et al. (2022) 39 | Y | Y | Y | Y | Y | N | Y | Y | Y | Y | Y | |
| Justo et al. (2017) 40 | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Kang et al. (2021) 41 | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Korda et al. (2016) 43 | Y | Y | Y | N | N | Y | Y | Y | Y | Y | Y | |
| Mnatzaganian et al. (2018) 46 | NA | NA | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Morton et al. (2022) 48 | NA | NA | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Morton et al. (2022) 49 | NA | NA | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Nembhard et al. (2016) 50 | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Nghiem et al. (2022) 51 | Y | Y | Y | N | N | Y | Y | Y | Y | Y | Y | |
| Nichols et al. (2021) 52 | NA | NA | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Nichols et al. (2018) 53 | NA | NA | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Nichols et al. (2020) 54 | NA | NA | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Pemberton et al. (2019) 55 | NA | NA | Y | N | N | Y | Y | Y | Y | Y | Y | |
| Pemberton et al. (2019) 56 | NA | NA | Y | Y | N | Y | Y | Y | Y | Y | Y | |
| Pemberton et al. (2021) 57 | NA | NA | Y | N | N | Y | Y | Y | Y | Y | Y | |
| Ramkumar et al. (2019) 59 | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | Y | |
| Saghapour et al. (2021) 64 | NA | NA | Y | N | N | Y | Y | Y | Y | Y | Y | |
| Shi et al. (2014) 65 | Y | Y | Y | N | N | Y | Y | Y | Y | Y | Y | |
| Xu et al. (2021) 69 | Y | Y | N | Y | Y | Y | Y | Y | Y | Y | Y | |
| Yiallourou et al. (2022) 70 | Y | Y | Y | N | N | Y | Y | Y | Y | Y | Y |
N, no; NA, not applicable; Y, yes.
| Citation | Were there clear criteria for inclusion in the case series? | Was the condition measured in a standard, reliable way for all participants included in the case series? | Were valid methods used for identification of the condition for all participants included in the case series? | Did the case series have consecutive inclusion of participants? | Was there clear reporting of the demographics of the participants in the study? | Was there clear reporting of clinical information of the participants? | Were the outcomes or follow up results of cases clearly reported? | Was there clear reporting of the presenting site(s)/clinic(s) demographic information? | Was statistical analysis appropriate? | |
|---|---|---|---|---|---|---|---|---|---|---|
| Baker et al. (2017) 25 | Y | Y | Y | N | N | N | Y | N | Y |
N, no; Y, yes.
Other limitations included the constraints of analysing administrative datasets (e.g. missing data, set data points),23,34,37 which included the inability to use address-level data due to privacy issues, or lack of availability to geocode to a smaller spatial unit42 or inability to confirm whether the address was current,24 and caveats of using an area-based approach as a proxy for individual SES (e.g. ecological fallacy, unable to account for individual confounders, underestimates inequality),26,28,37,38,42,45,48 particularly for studies including Aboriginal and Torres Strait Islander peoples.50 A longitudinal study cited the limitation of using SEIFA as a proxy measure for SES over time, which required multiple versions of SEIFA to be used.22
Discussion
The application of the ABS SEIFA indexes in Australian peer-reviewed CVD research to measure SES as an exposure has been widespread since 2013. Variations in the application of the SEIFA indexes were identified. It was not always clear which of the four SEIFA indexes (IRSAD, IRSD, Index of Education and Occupation, and Index of Economic Resources) were applied, how they were applied, and at what spatial unit. It is likely that the terms 'postcode' and 'POA' have been used interchangeably in included studies. These findings are important because how SEIFA is applied affects study outcomes (e.g. due to modifiable areal unit problem; MAUP), which has implications for the generalisability of CVD research in Australia.
Findings expand on the established complexity of measuring SES as an exposure, and quantifying the relationship between SES and health outcomes.2,5,71 Internationally, there has been much discussion around measuring SES, with the strengths and limitations of individual measures (e.g. educational attainment) discussed at length71–73 and compared with area-level measures of SES.74 In Australia, population health research identified that socio-economic inequality varied according to how SES was measured, with a SEIFA-based approach found to underestimate inequality when compared with the use of individual-level measures.45
What is clear is that there is no single measure of SES that is appropriate for all study designs, disease groups, settings and countries. The application of area-level measures is useful when individual-level data are not available and for the understanding of potential areas in need of population-level interventions. In CVD research it has been identified that a single measure of SES is unlikely to be sufficient to predict CVD risk or outcomes due to regional differences in SES, changes in individual measures of SES across the lifespan, and the complex interplay of factors known to contribute to CVD risks.5,38 The application of the SEIFA indexes as a composite area-level measure of SES in CVD research remains important.
There are some key issues identified in studies that should be considered by researchers. First, although SEIFA indexes have been applied as a proxy for individual SES in CVD research, there are limitations to this because the indexes do not account for individual measures of SES,75 confounders, or diversity in a geographical region.76 A second consideration is that SEIFA indexes are constructed after each 5-yearly national census, making indexes cross-sectional only. Due to this, SEIFA indexes cannot be applied to populations outside of Australia. Given that each version of SEIFA is different (e.g. between SEIFA 2006 and SEIFA 2011, the smallest spatial unit changed from collection districts to SA1), longitudinal application is problematic because versions are not comparable. Studies applying SEIFA are also subject to spatial bias, specifically the ecological fallacy of assuming SES homogeneity in a region when applying an area-level approach, which is more problematic for sparsely populated regions (e.g. rural areas)77 and the MAUP (i.e. effects of applying different spatial units to analysis).78
Given the tendency for studies to assign a SEIFA decile or score based on postcode or POA (an approximation for postcodes),79 and categorise into tertiles, quartiles or quintiles for analysis, it is important to consider what constitutes appropriate application. Supported by the ABS aligning with the release of SEIFA 2021,13 it is appropriate to use SEIFA as an area-level measure to describe the characteristics of a study cohort in a defined geographical region. If used as a proxy for individual SES, it is important to interpret findings correctly and identify the limitations of this approach.75 If individual-level measures of SES are available as variables, it is recommended to analyse these and compare findings to that of area-level SES; a recommendation supported internationally.80
It is also necessary to apply the version of SEIFA that corresponds to the study period and consider multiple versions for longitudinal studies. It is also recommended to opt for the smallest spatial unit available34 to mitigate bias attributed to the ecological fallacy (e.g. avoid postcodes or large areas, particularly for rural studies) and geocode the participant addresses into the smallest spatial unit (e.g. SA1, noting that the purpose of SA2 units is to represent a community that interacts together socially and economically).81–83 This may not be possible due to the use of administrative datasets (a widely cited limitation).84,85 There are other appropriate applications of SEIFA, such as application for the prospective sampling of study populations for area-level SES representativeness.86,87
A clear and transparent explanation and justification should be provided in the methods section (or as a Supplementary File) as to which SEIFA version and index was applied, how it was applied (direct or indirect), at what spatial unit it was applied, and whether the spatial unit was an ABS or non-ABS unit. Presenting data prior to collapsing into categories for analysis88 and providing information as to which data sources were analysed, is also recommended. Future efforts should examine the appropriateness of SEIFA as an area-level composite measure for specific populations, the implications of application from a policy perspective, and opportunities for other approaches to spatial modelling that account for a finer spatial resolution.89
Limitations
The scope of this review was limited to peer-reviewed literature. An analysis of grey literature, including evaluation reports, government documents, theses and policy documents would be of value to examine consistency in application beyond the peer-reviewed literature. Although the focus of this review was CVD, it is probable that the findings will be of relevance to researchers examining other highly prevalent non-communicable diseases, which is an area requiring further investigation. There is a possibility that not all Australian CVD research was retrieved due to publication bias.
Conclusion
The use of SEIFA in Australian CVD peer-reviewed research has been widespread with variations in the application to measure SES as an exposure. There is a need to improve the reporting of how SEIFA is applied in the methods sections of research for greater transparency. This is important to ensure that the application is consistent and the research findings are generalisable, so that they can accurately inform population-level interventions and investment to address the burden of CVD.
Data availability
No new data were created for this study, instead information was obtained from included studies. Data sharing is not applicable in this article.
Declaration of funding
No funding was received for this review. HB, SMW, SC, SW, LA, NTC and VLV are funded by the Australian Government’s Rural Health Multidisciplinary Training program. LA is funded by the National Heart Foundation Post-doctoral fellowship (reference: 102530).
Acknowledgements
We thank Deakin University librarian, Chrissy Freestone, for assistance in reviewing the search strategies used for this review.
Author contributions
HB and VLV conceptualised the study. HB drafted the study design and search strategies, screened studies, extracted data, and drafted the findings and the review. NTC contributed to the study design and provided a critical review of the manuscript. All authors provided input into the study design. SMW, SW, SC and LA took part in screening the studies, extracting the data and doing the quality assessments. All authors have read and approved the final manuscript.
References
1 Dugravot A, Fayosse A, Dumurgier J, Bouillon K, Rayana TB, Schnitzler A, et al. Social inequalities in multimorbidity, frailty, disability, and transitions to mortality: a 24-year follow-up of the Whitehall II cohort study. Lancet Public Health 2020; 5(1): e42-50.
| Crossref | Google Scholar | PubMed |
2 Lago S, Cantarero D, Rivera B, Pascual M, Blázquez-Fernández C, Casal B, et al. Socioeconomic status, health inequalities and non-communicable diseases: a systematic review. J Public Health 2018; 26(1): 1-14.
| Crossref | Google Scholar | PubMed |
3 World Health Organization. Cardiovascular Diseases. Geneva: World Health Organization; 2021. Available at https://www.who.int/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds) [cited 4 July 2023].
4 World Heart Federation. World Heart Report 2023: Confronting the World’s Number One Killer. Geneva: World Heart Federation; 2023. Available at https://world-heart-federation.org/wp-content/uploads/World-Heart-Report-2023.pdf [cited 10 September 2023].
5 Schultz WM, Kelli HM, Lisko JC, Varghese T, Shen J, Sandesara P, et al. Socioeconomic Status and Cardiovascular Outcomes. Circulation 2018; 137(20): 2166-78.
| Crossref | Google Scholar | PubMed |
6 Backholer K, Peters SAE, Bots SH, Peeters A, Huxley RR, Woodward M. Sex differences in the relationship between socioeconomic status and cardiovascular disease: a systematic review and meta-analysis. J Epidemiol Community Health 2017; 71(6): 550-7.
| Crossref | Google Scholar | PubMed |
7 Bagheri N, Gilmour B, McRae I, Konings P, Dawda P, Del Fante P, et al. Community cardiovascular disease risk from cross-sectional general practice clinical data: a spatial analysis. Prev Chronic Dis 2015; 12: 140379.
| Crossref | Google Scholar |
8 Australian Institute of Health and Welfare. Indicators of socioeconomic inequalities in cardiovascular disease, diabetes and chronic kidney disease. Canberra: Australian Institute of Health and Welfare; 2019. Available at https://www.aihw.gov.au/reports/social-determinants/indicators-socioeconomic-inequalities/summary [cited 4 July 2023].
9 Ritte RE, Lawton P, Hughes JT, Barzi F, Brown A, Mills P, et al. Chronic kidney disease and socio-economic status: a cross sectional study. Ethn Health 2020; 25(1): 93-109.
| Crossref | Google Scholar | PubMed |
10 Australian Institute of Health and Welfare. Health across socioeconomic groups. Canberra: Australian Institute of Health and Welfare; 2022. Available at https://www.aihw.gov.au/reports/australias-health/health-across-socioeconomic-groups [cited 10 September 2023].
11 Beks H, Wood SM, Clark RA, Vincent VL. Spatial methods for measuring access to health care. Eur J Cardiovasc Nurs 2023; 22: 832-40.
| Crossref | Google Scholar | PubMed |
12 Australian Bureau of Statistics. Socio-Economic Indexes for Areas (SEIFA): Technical Paper. Available at https://www.abs.gov.au/statistics/detailed-methodology-information/concepts-sources-methods/socio-economic-indexes-areas-seifa-technical-paper/2021 [cited 15 September 2023].
13 Australian Bureau of Statistics. Socio-Economic Indexes for Areas (SEIFA), Australia. 2023. Available at https://www.abs.gov.au/statistics/people/people-and-communities/socio-economic-indexes-areas-seifa-australia/latest-release [cited 4 July 2023].
14 Australian Bureau of Statistics. Australian Statistical Geography Standard (ASGS) Edition 3. Canberra: Australian Bureau of Statistics; 2021. Available at https://www.abs.gov.au/statistics/standards/australian-statistical-geography-standard-asgs-edition-3/jul2021-jun2026 [cited 7 August 2023].
15 McCracken K. Into a SEIFA SES cul-de-sac? Aust N Z J Public Health 2001; 25(4): 305-6.
| Crossref | Google Scholar | PubMed |
16 Peters MDJ, Godfrey C, McInerney P, Munn Z, Tricco AC, Khalil H. Chapter 11: Scoping Reviews. JBI Manual for Evidence Synthesis; 2020. Available at https://jbi-global-wiki.refined.site/space/MANUAL/4687342/Chapter+11%3A+Scoping+reviews [cited 11 July 2023].
17 Kyoon-Achan G, Lavoie J, Avery Kinew K, Phillips-Beck W, Ibrahim N, Sinclair S, et al. Innovating for Transformation in First Nations Health Using Community-Based Participatory Research. Qual Health Res 2018; 28(7): 1036-49.
| Crossref | Google Scholar | PubMed |
18 Peters MDJ, Marnie C, Tricco AC, Pollock D, Munn Z, Alexander L, McInerney P, Godfrey CM, Khalil H. Updated methodological guidance for the conduct of scoping reviews. JBI Evid Synth 2020; 18(10): 2119-26.
| Crossref | Google Scholar | PubMed |
19 World Health Organization. ICD-11 for Mortality and Morbidity Statistics. Geneva: World Health Organization; 2022. Available at https://www.who.int/standards/classifications/classification-of-diseases [cited 2 February 2023].
20 Wood SM, Alston L, Beks H, McNamara K, Coffee NT, Clark RA, et al. The application of spatial measures to analyse health service accessibility in Australia: a systematic review and recommendations for future practice. BMC Health Serv Res 2023; 23(1): 330.
| Crossref | Google Scholar | PubMed |
21 Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021; 372: n71.
| Crossref | Google Scholar | PubMed |
22 Adair T, Lopez AD. An egalitarian society? Widening inequalities in premature mortality from non-communicable diseases in Australia, 2006–16. Int J Epidemiol 2021; 50(3): 783-96.
| Crossref | Google Scholar | PubMed |
23 Astley CM, Chew DP, Keech W, Nicholls S, Beltrame J, Horsfall M, et al. The Impact of Cardiac Rehabilitation and Secondary Prevention Programs on 12-Month Clinical Outcomes: A Linked Data Analysis. Heart Lung Circ 2020; 29(3): 475-82.
| Crossref | Google Scholar | PubMed |
24 Atkins ER, Geelhoed EA, Nedkoff L, Briffa TG. Disparities in equity and access for hospitalised atherothrombotic disease. Aust Health Rev 2013; 37(4): 488-94.
| Crossref | Google Scholar | PubMed |
25 Baker J, White N, Mengersen K, Rolfe M, Morgan GG. Joint modelling of potentially avoidable hospitalisation for five diseases accounting for spatiotemporal effects: A case study in New South Wales, Australia. PLoS One 2017; 12(8): e0183653.
| Crossref | Google Scholar | PubMed |
26 Biswas S, Andrianopoulos N, Duffy SJ, Lefkovits J, Brennan A, Walton A, et al. Impact of Socioeconomic Status on Clinical Outcomes in Patients With ST-Segment-Elevation Myocardial Infarction. Circ Cardiovasc Qual Outcomes 2019; 12(1): e004979.
| Crossref | Google Scholar | PubMed |
27 Busija L, Tao LW, Liew D, Weir L, Yan B, Silver G, et al. Do patients who take part in stroke research differ from non-participants? Implications for generalizability of results. Cerebrovasc Dis 2013; 35(5): 483-91.
| Crossref | Google Scholar | PubMed |
28 Carter HE, Schofield D, Shrestha R. Productivity costs of cardiovascular disease mortality across disease types and socioeconomic groups. Open Heart 2019; 6(1): e000939.
| Crossref | Google Scholar | PubMed |
29 Cheng J, Bambrick H, Tong S, Su H, Xu Z, Hu W. Winter temperature and myocardial infarction in Brisbane, Australia: Spatial and temporal analyses. Sci Total Environ 2020; 715: 136860.
| Crossref | Google Scholar | PubMed |
30 Chew DP, MacIsaac AI, Lefkovits J, Harper RW, Slawomirski L, Braddock D, et al. Variation in coronary angiography rates in Australia: correlations with socio-demographic, health service and disease burden indices. Med J Aust 2016; 205(3): 114-20.
| Crossref | Google Scholar | PubMed |
31 Close GR, Newton PJ, Fung SC, Denniss AR, Halcomb EJ, Kovoor P, et al. Socioeconomic status and heart failure in Sydney. Heart Lung Circ 2014; 23(4): 320-4.
| Crossref | Google Scholar | PubMed |
32 Dawson LP, Andrew E, Nehme Z, Bloom J, Biswas S, Cox S, et al. Association of Socioeconomic Status With Outcomes and Care Quality in Patients Presenting With Undifferentiated Chest Pain in the Setting of Universal Health Care Coverage. J Am Heart Assoc 2022; 11(7): e024923.
| Crossref | Google Scholar | PubMed |
33 Gutman SJ, Costello BT, Papapostolou S, Iles L, Ja J, Hare JL, et al. Impact of sex, socio-economic status, and remoteness on therapy and survival in heart failure. ESC Heart Fail 2019; 6(5): 944-52.
| Crossref | Google Scholar | PubMed |
34 Hanigan IC, Cochrane T, Davey R. Impact of scale of aggregation on associations of cardiovascular hospitalization and socio-economic disadvantage. PLoS ONE 2017; 12(11): e0188161.
| Crossref | Google Scholar | PubMed |
35 Hastings K, Marquina C, Morton J, Abushanab D, Berkovic D, Talic S, et al. Projected New-Onset Cardiovascular Disease by Socioeconomic Group in Australia. Pharmacoeconomics 2022; 40(4): 449-60.
| Crossref | Google Scholar | PubMed |
36 Huynh Q, Venn AJ, Marwick TH. Socioeconomic Disadvantage and Days at Home After Hospital Discharge of Patients with Heart Failure. Am J Cardiol 2018; 122(4): 616-24.
| Crossref | Google Scholar | PubMed |
37 Hyun K, Redfern J, Woodward M, D’Souza M, Shetty P, Chew D, et al. Socioeconomic Equity in the Receipt of In-Hospital Care and Outcomes in Australian Acute Coronary Syndrome Patients: The CONCORDANCE Registry. Heart Lung Circ 2018; 27(12): 1398-405.
| Crossref | Google Scholar | PubMed |
38 Jacobs J, Peterson KL, Allender S, Alston LV, Nichols M. Regional variation in cardiovascular mortality in Australia 2009-2012: the impact of remoteness and socioeconomic status. Aust N Z J Public Health 2018; 42(5): 467-73.
| Crossref | Google Scholar | PubMed |
39 Jahan H, Bernardo C, Gonzalez-Chica D, Benson J, Stocks N. General practice management of depression among patients with coronary heart disease in Australia. BMC Prim Care 2022; 23(1): 329.
| Crossref | Google Scholar | PubMed |
40 Justo ER, Reeves BM, Ware RS, Johnson JC, Karl TR, Alphonso ND, et al. Comparison of outcomes in Australian indigenous and non-indigenous children and adolescents undergoing cardiac surgery. Cardiol Young 2017; 27(9): 1694-700.
| Crossref | Google Scholar | PubMed |
41 Kang K, Chau KWT, Howell E, Anderson M, Smith S, Davis TJ, et al. The temporospatial epidemiology of rheumatic heart disease in Far North Queensland, tropical Australia 1997-2017; impact of socioeconomic status on disease burden, severity and access to care. PLoS Negl Trop Dis 2021; 15(1): e0008990.
| Crossref | Google Scholar | PubMed |
42 Kawai A, Hui S, Beare R, Srikanth VK, Sundararajan V, Ma H, et al. Spatiotemporal analysis of regional TIA trends. Front Neurol 2022; 13: 983512.
| Crossref | Google Scholar | PubMed |
43 Korda RJ, Soga K, Joshy G, Calabria B, Attia J, Wong D, et al. Socioeconomic variation in incidence of primary and secondary major cardiovascular disease events: an Australian population-based prospective cohort study. Int J Equity Health 2016; 15(1): 189.
| Crossref | Google Scholar | PubMed |
44 Mariajoseph FP, Huang H, Lai LT. Influence of socioeconomic status on the incidence of aneurysmal subarachnoid haemorrhage and clinical recovery. J Clin Neurosci 2022; 95: 70-4.
| Crossref | Google Scholar | PubMed |
45 Mather T, Banks E, Joshy G, Bauman A, Phongsavan P, Korda RJ. Variation in health inequalities according to measures of socioeconomic status and age. Aust N Z J Public Health 2014; 38(5): 436-40.
| Crossref | Google Scholar | PubMed |
46 Mnatzaganian G, Hiller JE, Fletcher J, Putland M, Knott C, Braitberg G, et al. Socioeconomic gradients in admission to coronary or intensive care units among Australians presenting with non-traumatic chest pain in emergency departments. BMC Emerg Med 2018; 18(1): 32.
| Crossref | Google Scholar | PubMed |
47 Mnatzaganian G, Lee CMY, Robinson S, Sitas F, Chow CK, Woodward M, et al. Socioeconomic disparities in the management of coronary heart disease in 438 general practices in Australia. Eur J Prev Cardiol 2021; 28(4): 400-7.
| Crossref | Google Scholar | PubMed |
48 Morton JI, Ilomäki J, Wood SJ, Bell JS, Huynh Q, Magliano DJ, et al. Treatment gaps, 1-year readmission and mortality following myocardial infarction by diabetes status, sex and socioeconomic disadvantage. J Epidemiol Community Health 2022; 76(7): 637-45.
| Crossref | Google Scholar | PubMed |
49 Morton JI, Ilomäki J, Wood SJ, Bell JS, Shaw JE, Magliano DJ. One-year readmission and mortality following ischaemic stroke by diabetes status, sex, and socioeconomic disadvantage: An analysis of 27,802 strokes from 2012 to 2017. J Neurol Sci 2022; 434: 120149.
| Crossref | Google Scholar | PubMed |
50 Nembhard WN, Bourke J, Leonard H, Eckersley L, Li J, Bower C. Twenty-five-year survival for Aboriginal and Caucasian children with congenital heart defects in Western Australia, 1980 to 2010. Birth Defects Res A Clin Mol Teratol 2016; 106(12): 1016-31.
| Crossref | Google Scholar | PubMed |
51 Nghiem S, Afoakwah C, Scuffham P, Byrnes J. A baseline profile of the Queensland Cardiac Record Linkage Cohort (QCard) study. BMC Cardiovasc Disord 2022; 22(1): 35.
| Crossref | Google Scholar | PubMed |
52 Nichols L, Gall S, Stankovich J, Stirling C. Associations between socioeconomic status and place of residence with survival after aneurysmal subarachnoid haemorrhage. Intern Med J 2021; 51(12): 2095-103.
| Crossref | Google Scholar | PubMed |
53 Nichols L, Stirling C, Otahal P, Stankovich J, Gall S. Socioeconomic Disadvantage Is Associated with a Higher Incidence of Aneurysmal Subarachnoid Hemorrhage. J Stroke Cerebrovasc Dis 2018; 27(3): 660-8.
| Crossref | Google Scholar | PubMed |
54 Nichols L, Stirling C, Stankovich J, Gall S. Time to treatment following an aneurysmal subarachnoid hemorrhage, rural place of residence and inter-hospital transfers. Australas Emerg Care 2020; 23(4): 225-32.
| Crossref | Google Scholar | PubMed |
55 Pemberton K, Bosley E, Franklin RC, Watt K. Pre-hospital outcomes of adult out-of-hospital cardiac arrest of presumed cardiac aetiology in Queensland, Australia (2002-2014): Trends over time. Emerg Med Australas 2019; 31(5): 813-20.
| Crossref | Google Scholar | PubMed |
56 Pemberton K, Bosley E, Franklin RC, Watt K. Epidemiology of pre-hospital outcomes of out-of-hospital cardiac arrest in Queensland, Australia. Emerg Med Australas 2019; 31(5): 821-9.
| Crossref | Google Scholar | PubMed |
57 Pemberton K, Franklin RC, Bosley E, Watt K. Long-term outcomes of adult out-of-hospital cardiac arrest in Queensland, Australia (2002-2014): incidence and temporal trends. Heart 2021; 107(16): 1310-9.
| Crossref | Google Scholar | PubMed |
58 Rachele JN, Giles-Corti B, Turrell G. Neighbourhood disadvantage and self-reported type 2 diabetes, heart disease and comorbidity: a cross-sectional multilevel study. Ann Epidemiol 2016; 26(2): 146-50.
| Crossref | Google Scholar | PubMed |
59 Ramkumar S, Ochi A, Yang H, Nerlekar N, D’Elia N, Potter EL, et al. Association between socioeconomic status and incident atrial fibrillation. Intern Med J 2019; 49(10): 1244-51.
| Crossref | Google Scholar | PubMed |
60 Randall SM, Zilkens R, Duke JM, Boyd JH. Western Australia population trends in the incidence of acute myocardial infarction between 1993 and 2012. Int J Cardiol 2016; 222: 678-82.
| Crossref | Google Scholar | PubMed |
61 Roberts KV, Maguire GP, Brown A, Atkinson DN, Remenyi B, Wheaton G, et al. Rheumatic heart disease in Indigenous children in northern Australia: differences in prevalence and the challenges of screening. Med J Aust 2015; 203(5): 221-7.
| Crossref | Google Scholar | PubMed |
62 Robins S, Gardiner S, Terry DR. The urban-rural divide: Hypertensive disease hospitalisations in Victoria 2010–2015. Australas Med J 2017; 10:.
| Crossref | Google Scholar |
63 Roseleur J, Gonzalez-Chica DA, Bernardo CO, Geisler BP, Karnon J, Stocks NP. Blood pressure control in Australian general practice: analysis using general practice records of 1.2 million patients from the MedicineInsight database. J Hypertens 2021; 39(6): 1134-42.
| Crossref | Google Scholar | PubMed |
64 Saghapour T, Giles-Corti B, Rachele J, Turrell G. A cross-sectional and longitudinal study of neighbourhood disadvantage and cardiovascular disease and the mediating role of physical activity. Prev Med 2021; 147: 106506.
| Crossref | Google Scholar | PubMed |
65 Shi WY, Yap CH, Newcomb AE, Hayward PA, Tran L, Reid CM, et al. Impact of socioeconomic status and rurality on early outcomes and mid-term survival after CABG: insights from a multicentre registry. Heart Lung Circ 2014; 23(8): 726-36.
| Crossref | Google Scholar | PubMed |
66 Smurthwaite K, Bagheri N. Using Geographical Convergence of Obesity, Cardiovascular Disease, and Type 2 Diabetes at the Neighborhood Level to Inform Policy and Practice. Prev Chronic Dis 2017; 14: E91.
| Crossref | Google Scholar | PubMed |
67 Straney LD, Bray JE, Beck B, Bernard S, Lijovic M, Smith K. Are sociodemographic characteristics associated with spatial variation in the incidence of OHCA and bystander CPR rates? A population-based observational study in Victoria, Australia. BMJ Open 2016; 6(11): e012434.
| Crossref | Google Scholar | PubMed |
68 Tideman P, Anne WT, Edward J, Ben P, Robyn C, Elizabeth P, et al. A comparison of Australian rural and metropolitan cardiovascular risk and mortality: the Greater Green Triangle and North West Adelaide population surveys. BMJ Open 2013; 3(8): e003203.
| Crossref | Google Scholar | PubMed |
69 Xu Z, Tong S, Pan H, Cheng J. Associations of extreme temperatures with hospitalizations and post-discharge deaths for stroke: What is the role of pre-existing hyperlipidemia? Environ Res 2021; 193: 110391.
| Crossref | Google Scholar | PubMed |
70 Yiallourou SR, Magliano D, Haregu TN, Carrington MJ, Rolnik DL, Rombauts L, et al. Long term all-cause and cardiovascular disease mortality among women who undergo fertility treatment. Med J Aust 2022; 217(10): 532-7.
| Crossref | Google Scholar | PubMed |
71 Shavers VL. Measurement of socioeconomic status in health disparities research. J Natl Med Assoc 2007; 99(9): 1013-23.
| Google Scholar | PubMed |
72 Rosengren A, Smyth A, Rangarajan S, Ramasundarahettige C, Bangdiwala SI, AlHabib KF, et al. Socioeconomic status and risk of cardiovascular disease in 20 low-income, middle-income, and high-income countries: the Prospective Urban Rural Epidemiologic (PURE) study. Lancet Glob Health 2019; 7(6): e748-60.
| Crossref | Google Scholar | PubMed |
73 Mark W, Sanne AEP, Batty GD, Hirotsugu U, Jean W, Graham GG, et al. Socioeconomic status in relation to cardiovascular disease and cause-specific mortality: a comparison of Asian and Australasian populations in a pooled analysis. BMJ Open 2015; 5(3): e006408.
| Crossref | Google Scholar | PubMed |
74 Moss JL, Johnson NJ, Yu M, Altekruse SF, Cronin KA. Comparisons of individual- and area-level socioeconomic status as proxies for individual-level measures: evidence from the Mortality Disparities in American Communities study. Popul Health Metr 2021; 19(1): 1.
| Crossref | Google Scholar | PubMed |
75 Australian Bureau of Statistics. Socio-Economic Indexes For Areas: Getting a Handle on Individual Diversity Within Areas - 1351.0.55.036. Canberra: Australian Bureau of Statistics; 2011. Available at https://www.ausstats.abs.gov.au/ausstats/subscriber.nsf/0/C523F80A0B938ACBCA25790600138037/$File/1351055036_sep%202011.pdf [cited 6 July 2023].
76 Kerr J, Mavoa S, Schroers R-D, Eagleson S, Exeter D, Watkins A, et al. Measuring area-level disadvantage in Australia: Development of a locally sensitive indicator. Aust Popul Stud 2021; 5(2): 15-28.
| Crossref | Google Scholar |
77 Shih Y-CT, Bradley C, Yabroff KR. Ecological and individualistic fallacies in health disparities research. J Natl Cancer Inst 2023; 115(5): 488-91.
| Crossref | Google Scholar | PubMed |
78 Briz-Redón Á. A Bayesian shared-effects modeling framework to quantify the modifiable areal unit problem. Spatial Stat 2022; 51: 100689.
| Crossref | Google Scholar |
79 Australian Bureau of Statistics. Postal Areas. Canberra: Australian Bureau of Statistics; 2021. Available at https://www.abs.gov.au/statistics/standards/australian-statistical-geography-standard-asgs-edition-3/jul2021-jun2026/non-abs-structures/postal-areas [cited 15 September 2023].
80 Pichora E, Polsky JY, Catley C, Perumal N, Jin J, Allin S. Comparing individual and area-based income measures: impact on analysis of inequality in smoking, obesity, and diabetes rates in Canadians 2003-2013. Can J Public Health 2018; 109(3): 410-8.
| Crossref | Google Scholar | PubMed |
81 Australian Bureau of Statistics. Using and interpreting SEIFA. Canberra: Australian Bureau of Statistics; 2023. Available at https://www.abs.gov.au/statistics/detailed-methodology-information/concepts-sources-methods/socio-economic-indexes-areas-seifa-technical-paper/2021/using-and-interpreting-seifa [cited 6 July 2023].
82 Versace VL, Skinner TC, Bourke L, Harvey P, Barnett T. National analysis of the Modified Monash Model, population distribution and a socio-economic index to inform rural health workforce planning. Aust J Rural Health 2021; 29(5): 801-10.
| Crossref | Google Scholar | PubMed |
83 Australian Bureau of Statistics. Statistical Area Level 2. Canberra Australian Bureau of Statistics; 2021. Available at https://www.abs.gov.au/statistics/standards/australian-statistical-geography-standard-asgs-edition-3/jul2021-jun2026/main-structure-and-greater-capital-city-statistical-areas/statistical-area-level-2 [cited 15 September 2023].
84 Beks H, Versace V, Mitchell F, Charles J, Chatfield T, Zwolak R. Redressing barriers to healthcare for Aboriginal and Torres Strait Islander Peoples: preliminary findings from a mobile clinic in rural Victoria. Public Health Res Pract 33: e33012301.
| Crossref | Google Scholar | PubMed |
85 Beks H, Mitchell F, Charles J, Wong Shee A, McNamara K, Versace VL. Implementation of telehealth primary health care services in a rural Aboriginal Community-Controlled Health Organisation during the COVID-19 pandemic: a mixed-methods study. Rural Remote Health 2023; 23: 7521.
| Crossref | Google Scholar | PubMed |
86 Ines K, Rob C, Bernadette M, Mohammadreza M, Sophy TFS, Peta T, et al. Pharmacy Diabetes Screening Trial: protocol for a pragmatic cluster-randomised controlled trial to compare three screening methods for undiagnosed type 2 diabetes in Australian community pharmacy. BMJ Open 2017; 7(12): e017725.
| Crossref | Google Scholar | PubMed |
87 Krass I, Carter R, Mitchell B, Mohebbi M, Shih STF, Trinder P, et al. Pharmacy diabetes screening trial (PDST): Outcomes of a national clustered RCT comparing three screening methods for undiagnosed type 2 diabetes (T2DM) in community pharmacy. Diabetes Res Clin Pract 2023; 197: 110566.
| Crossref | Google Scholar | PubMed |
88 Versace VL, Beks H, Charles J. Towards consistent geographic reporting of Australian health research. Med J Aust 2021; 215(11): 525.
| Crossref | Google Scholar | PubMed |
89 Australian Research Council. LE220100028. Deakin University; 2023. Available at https://dataportal.arc.gov.au/RGS/Web/Grants/LE220100028 [cited 6 July 2023]
