Deloitte maps health data with future projections
Deloitte has launched a new online service allowing users to create localised maps of current disease prevalence and incidence, along with forward projections to 2050, to allow for better health services planning.
Deloitte's lead partner for health economics and social policy, Lynne Pezzullo, said HealthEdata covers a range of therapeutic areas and forward projections, along with very specific levels of location. It includes a geographic overlay that allows users to map services such as public and private hospitals and general practices over the numbers of people with particular conditions in each location.
“The need for the service was evident to us because clients kept coming to us and asking for data about prevalence and incidence of particular conditions that they could rely on and which had a local focus,” Ms Pezzullo said.
“For example, in my catchment area, how many people have arthritis, how many osteoporotic fractures are there likely to be this year but also in five years' time so we can do some services planning? Over many years we have assimilated all of this epidemiological data from numerous studies and now we've decided to put all of this information together in an online service.”
She said HealthEdata includes an historical time series with projections from 2010 to 2050 on the incidence and prevalence of disease by both numbers and by cases per thousand. It covers over 50 health conditions, including the national health priority areas of arthritis and musculo-skeletal conditions, asthma, cancer control, diabetes, injury prevention and control, cardiovascular health, mental health and obesity.
In terms of populations, it covers local government areas, Medicare Locals, federal and state electoral divisions and jurisdictional health districts. “The services information is also available by region so it can be overlaid, so it shows the numbers of hospitals, the numbers of general practices, optometry practices, pharmacies and dentists by those different regions. And we've also got health expenditure data for households, so you can have a look at what the expenditure is relative to the prevalence of disease.”
It also includes the Australian Bureau of Statistics' (ABS) socio-economic indexes for areas (SEIFA), which is hugely important in determining disease prevalence and future growth. “There are four different SEIFA indices of disadvantage, so you could look at what is happening in Blacktown in relation to its SEIFA, which is quite low, and it has a high and increasing prevalence of chronic disease,” she said.
Using dementia as an example, Ms Pezzullo said users can request data on that particular condition and would receive by default a table and chart. Users can also request a map of, for instance, areas in NSW with the highest prevalence, mapped by Medicare Local. Each disease state is shown in quintiles of prevalence.
The service also produces bivariate maps, which allows users to see both prevalence and growth. “Where is dementia high now and growing most? Those things are very important for planning for how many services to provide in a particular area,” she said.
The data sources are vast, including ABS statistics for the demographic data, along with a Deloitte Access Economics demographic model that maps the population of local regions by age and gender using fertility, mortality and migration patterns at a micro level.
Prevalence and incidence data is sourced from the Australian Institute for Health and Welfare along with published epidemiological studies and meta-analyses of these. “We have links on the website to those studies so people can read which peer-reviewed journals have been used and how the meta-analyses have been done.”
The services overlay – pinpointing GPs, optometrists, pharmacies and dentists – is sourced from the Yellow Pages, while hospital data comes from the Department of Health and Ageing. The expenditure data comes from the Household Expenditure Survey and the SEIFA data comes from the ABS.
Ms Pezzullo said Deloitte expects a range of groups would find the service useful, including state and federal governments and their agencies, Medicare Locals, local hospital networks, healthcare peak bodies and professional organisations, healthcare service providers such as hospitals and general practices, health insurers and their peak bodies and regulators, and organisations representing particular conditions.
The service was piloted with the help of Alzheimer's Australia, which had a need for up to date and customisable data on disease prevalence. It is a subscription service, with a year's subscription costing $2200 for 50 downloads. A one-off download costs $400, but if users want to overlay more data, more downloads are required.
She used the example of prevalence per thousand of COPD in NSW by Medicare Local for 2010-2020. “That would be one download. If you want a SEIFA overlay of disadvantage that would be another download. And if you want a map – either growth rate or bivariate – that would give you three downloads in total.”
Ms Pezzullo said there are some limitations with the data as it is impossible to control for all risk factors. HealthEdata controls for age and gender, as they are the primary determinants of prevalence, and the quirk of nature that means rates of multiple sclerosis vary due to distance from the equator is included. She said controls for variation due to ethnicity may be included in future.
Posted in Australian eHealth