Paul Konings

General practitioner (family physician) workforce in Australia: comparing geographic data from surveys, a mailing list and medicare

BackgroundGood quality spatial data on Family Physicians or General Practitioners (GPs) are key to accurately measuring geographic access to primary health care. The validity of computed associations between health outcomes and measures of GP access such as GP density is contingent on geographical data quality. This is especially true in rural and remote areas, where GPs are often small in number and geographically dispersed. However, there has been limited effort in assessing the quality of nationally comprehensive, geographically explicit, GP datasets in Australia or elsewhere.Our objective is to assess the extent of association or agreement between different spatially explicit nationwide GP workforce datasets in Australia. This is important since disagreement would imply differential relationships with primary healthcare relevant outcomes with different datasets. We also seek to enumerate these associations across categories of rurality or remoteness.MethodWe compute correlations of GP headcounts and workload contributions between four different datasets at two different geographical scales, across varying levels of rurality and remoteness.ResultsThe datasets are in general agreement with each other at two different scales. Small numbers of absolute headcounts, with relatively larger fractions of locum GPs in rural areas cause unstable statistical estimates and divergences between datasets.ConclusionIn the Australian context, many of the available geographic GP workforce datasets may be used for evaluating valid associations with health outcomes. However, caution must be exercised in interpreting associations between GP headcounts or workloads and outcomes in rural and remote areas. The methods used in these analyses may be replicated in other locales with multiple GP or physician datasets.

Protecting the privacy of individual general practice patient electronic records for geospatial epidemiology research

Background: General practitioner (GP) practices in Australia are increasingly storing patient information in electronic databases. These practice databases can be accessed by clinical audit software to generate reports that inform clinical or population health decision making and public health surveillance. Many audit software applications also have the capacity to generate de‐identified patient unit record data. However, the de‐identified nature of the extracted data means that these records often lack geographic information. Without spatial references, it is impossible to build maps reflecting the spatial distribution of patients with particular conditions and needs. Links to socioeconomic, demographic, environmental or other geographically based information are also not possible. In some cases, relatively coarse geographies such as postcode are available, but these are of limited use and researchers cannot undertake precision spatial analyses such as calculating travel times.

Social Exclusion, Deprivation and Child Health: a Spatial Analysis of Ambulatory Care Sensitive Conditions in Children Aged 0-4 years in Victoria, Australia

Recent Australian policy initiatives regarding primary health care focus on planning services around community needs and delivering these at the local area. As in many other countries, there has also been a growing concern over social inequities in health outcomes. The aims of the analysis presented here were firstly to describe small area variations in hospital admissions for ambulatory care sensitive conditions (ACSC) among children aged 0-4 years between 2003 and 2009 in the state of Victoria, Australia, and secondly to explore the relationship of ACSC hospitalisations with socio-economic disadvantage using a comparative analysis of the Child Social Exclusion (CSE) index and the Composite Score of Deprivation (CSD). This is a cross sectional secondary data analysis, with data sourced from 2003 to 2009 ACSC data from the Victorian State Government Department of Health; the Australian Standard Geographical Classification of remoteness; the Australian 2006 Census of Population and Housing; and AMPCo General Practitioner data from 2010. The relationship between the indexes and child health outcomes was examined through bivariate analysis and visually through a series of maps. The results show there is significant variation in the geographical distribution of the relationship between ACSCs and socio-economic disadvantage, with both indexes capturing important social gradients in child health conditions. However, measures of access, such as geographical accessibility and workforce supply, detect additional small area variation in child health outcomes. This research has important implications for future primary health care policy and planning of services, as these findings confirm that not all areas are the same in terms of health outcomes, and there may be benefit in tailoring mechanisms for identifying areas of need depending on the outcome intended to be affected.

Community Cardiovascular Disease Risk From Cross-Sectional General Practice Clinical Data: A Spatial Analysis

Introduction Cardiovascular disease (CVD) continues to be a leading cause of illness and death among adults worldwide. The objective of this study was to calculate a CVD risk score from general practice (GP) clinical records and assess spatial variations of CVD risk in communities. Methods We used GP clinical data for 4,740 men and women aged 30 to 74 years with no history of CVD. A 10-year absolute CVD risk score was calculated based on the Framingham risk equation. The individual risk scores were aggregated within each Statistical Area Level One (SA1) to predict the level of CVD risk in that area. Finally, the pattern of CVD risk was visualized to highlight communities with high and low risk of CVD. Results The overall 10-year risk of CVD in our sample population was 14.6% (95% confidence interval [CI], 14.3%–14.9%). Of the 4,740 patients in our study, 26.7% were at high risk, 29.8% were at moderate risk, and 43.5% were at low risk for CVD over 10 years. The proportion of patients at high risk for CVD was significantly higher in the communities of low socioeconomic status. Conclusion This study illustrates methods to further explore prevalence, location, and correlates of CVD to identify communities of high levels of unmet need for cardiovascular care and to enable geographic targeting of effective interventions for enhancing early and timely detection and management of CVD in those communities.

Undiagnosed diabetes from cross-sectional GP practice data: an approach to identify communities with high likelihood of undiagnosed diabetes

Objectives To estimate undiagnosed diabetes prevalence from general practitioner (GP) practice data and identify areas with high levels of undiagnosed and diagnosed diabetes. Design Data from the North-West Adelaide Health Survey (NWAHS) were used to develop a model which predicts total diabetes at a small area. This model was then applied to cross-sectional data from general practices to predict the total level of expected diabetes. The difference between total expected and already diagnosed diabetes was defined as undiagnosed diabetes prevalence and was estimated for each small area. The patterns of diagnosed and undiagnosed diabetes were mapped to highlight the areas of high prevalence. Setting North-West Adelaide, Australia. Participants This study used two population samples—one from the de-identified GP practice data (n=9327 active patients, aged 18 years and over) and another from NWAHS (n=4056, aged 18 years and over). Main outcome measures Total diabetes prevalence, diagnosed and undiagnosed diabetes prevalence at GP practice and Statistical Area Level 1. Results Overall, it was estimated that there was one case of undiagnosed diabetes for every 3–4 diagnosed cases among the 9327 active patients analysed. The highest prevalence of diagnosed diabetes was seen in areas of lower socioeconomic status. However, the prevalence of undiagnosed diabetes was substantially higher in the least disadvantaged areas. Conclusions The method can be used to estimate population prevalence of diabetes from general practices wherever these data are available. This approach both flags the possibility that undiagnosed diabetes may be a problem of less disadvantaged social groups, and provides a tool to identify areas with high levels of unmet need for diabetes care which would enable policy makers to apply geographic targeting of effective interventions.

A brief report on Primary Care Service Area catchment geographies in New South Wales Australia

BackgroundTo develop a method to use survey data to establish catchment areas of primary care or Primary Care Service Areas. Primary Care Service Areas are small areas, the majority of patients resident in which obtain their primary care services from within the geography.MethodsThe data are from a large health survey (n =267,153, year 2006–2009) linked to General Practitioner service use data (year 2002–2010) from New South Wales, Australia. Our methods broadly follow those used previously by researchers in the United States of America and Switzerland, with significant modifications to improve robustness. This algorithm allocates post code areas to Primary Care Service Areas that receive the plurality of patient visits from the post code area.ResultsConsistent with international findings the median Localization Index or the median percentage of patients that obtain their primary care from within a Primary Care Service Area is 55% with localization increasing with rurality.ConclusionsWith the additional methodological refinements in this study, Australian Primary Care Service Areas have great potential to be of value to policymakers and researchers.

Beyond reasonable drought: hotspots reveal a link between the ‘Big Dry’ and cryptosporidiosis in Australia's Murray Darling Basin

There is little evidence on how the health impacts of drought vary spatially and temporally. With a focus on waterborne cryptosporidiosis, we identify spatio-temporal hotspots and by using interrupted time series analysis, examine the impact of Australias Big Dry (2001-2009) in these disease clusters in the Murray Darling Drainage Basin. Analyses revealed a statistically significant hotspot in the north of the Australian Capital Territory (ACT) and a hotspot in the north-eastern end of the basin in Queensland. After controlling for long-term trend and seasonality in cryptosporidiosis, interrupted time series analysis of reported cases in these hotspots indicated a statistically significant link with the Big Dry. In both areas, the end of the Big Dry was associated with a lower risk of reported cryptosporidiosis; in the ACT, the estimated relative risk (RR) was 0.16 (95% confidence interval: 0.07; 0.33), and in Queensland the RR was 0.42 (95% confidence interval: 0.19; 0.42). Although these data do not establish a causal association, this research highlights the potential for drought-related health risks.