Data & Modelling
Policy modelling and simulation will be undertaken to examine which interventions are likely to create the most impactful solutions and provide the best value for money. This hub will be led by CIs Petrie and Aitken who have extensive experience in modelling the social determinants of health and estimating the cost effectiveness of alternative policy options. CIs Petrie, Aitken and Disney are experts in disability data and have in-depth understanding of the opportunities and challenges of different data sources including longitudinal surveys, administrative and linked data. Table 1 lists the data sources that will be used in AHEAD.
| Data Source | Sample Size | Variables available | Period |
|---|---|---|---|
|
PLIDA
Person Level Integrated Data Asset
|
> 20 Million |
Census 2011, 2016, 2021, Welfare data, NDIS enrolment, healthcare use, Death records, Tax and income records including the Single Touch Payroll. Also includes linked ABS surveys (e.g. National Health Survey, Survey of Disability, Ageing and Carers)
|
2010 – current (most data sets refreshed every 3 months)
|
|
DES Data
Disability Employment Service data
|
Aggregated by groups | Referrals, commencements, active caseload by age group, active caseload by primary disability and active caseload by income support type and other breakdowns for Australia | 2011 - 2012 |
|
NDDA
National Disability Data Asset
|
~6 million
All those with disability
|
The above data plus additional linked information on state-based disability and health services | Expected to go live 2024 |
|
HILDA
Household, Income and Labour
|
20,000 | Longitudinal panel survey of Australian households. Includes data on health and wellbeing, work, housing and neighbourhoods, and healthcare use | Annual since 2001 |
|
AHEAD panel
questionnaires
|
Expected: 1000-2000 | We will develop and maintain a panel of people with lived experience of disability to answer stated preference questionnaires. This will allow the prediction of behavioural responses to policy design. We have experience doing this from our NDIS Disability Wellbeing Index project. | 2025-2027 |
The AHEAD team have extensive experience in using administrative data, including all data sources for this grant, to develop epidemiology and health economic simulation models to predict the downstream consequences of changes in policy. CI Petrie co-led the regression modelling for the Department of Health COVID Mental Health modelling project to understand the likely consequences of COVID on the social determinants of mental health and the likely downstream impacts. CI Petrie has also been involved in using administrative and registry data to develop simulation models for Myeloma (the MRFF funded EPIMAP project) and Type 1 diabetes. CI Aitken has extensive experience in modelling the implications for health outcomes and healthcare costs of interventions that improve employment outcomes for people with disabilities. CIs Disney and Petrie have experience in modelling smoking related behaviours and the implications for morbidity, mortality and healthcare costs.
The preferences work from the Knowledge to Action Research Hub will also feed in estimating the likely behavioural responses that people with disabilities have to different policies. These will be inputs into the modelling.
CRE-DH created a Wellbeing Monitoring Framework that AHEAD will build on to monitor and evaluate progress against indicators relating to health, social determinants of health, and service systems. This uses data from PLIDA as well as other cross-sectional and longitudinal surveys to monitor progress on inequities over time. We will use this to monitor the progress of policy solutions in narrowing health inequities offering a form of accountability.