VIVARIUM – next-generation proportional multi-state life-table simulation model

Project Details

We know much about many exposure-outcome associations in health, but we seldom systematically answer questions like: “when applied to the population, how much health gain will intervention X achieve compared to intervention Y?  Over what time period?  At what cost and what cost-effectiveness?”

The Assessing Cost-Effectiveness (ACE) projects in Australia and the Burden of Disease Epidemiology, Equity and Cost-Effectiveness Programme (BODE3) in New Zealand have led the world in simulating the health and cost impacts of 100s of public health interventions (see our online interactive league table for a collation of these evaluations).  These evaluations have mostly been undertaken with multistate life-table models, in Excel.

The Institute of Health Metrics and Evaluation (IHME,) University of Washington, USA – home of the Global Burden of Disease Study (GBD) – is developing VIVARIUM, a Python-based intervention simulation framework.  We are collaborating with IHME, and the University of Otago NZ, to develop proportional multi-state life-table models (a type of Markov macro-simulation) within VIVARIUM.

The advantages and potential of VIVARIUM include:

  • Easier to build, adapt and reuse components
  • Easier interface with GBD data to populate models
  • Faster and less error prone than spreadsheet-based models
  • Easier to scale up to conducting evaluations in multiple countries.

We are initially focusing on tobacco and e-cigarette interventions in Australia, New Zealand, Pacific Island Countries and Territories, and Asian countries. The next priority is public transport and urban design, working with collaborators at the MSPGH, Melbourne School of Design and internationally.

VIVARIUM is ideally suited to prospective PhD and Masters students, interested in:

  • Computational and model building aspects (e.g. implementing new modules that disaggregate a population by socioeconomic strata to assess inequality impacts; efficient approaches to model validation and calibration)
  • Epidemiological aspects (e.g. converting GBD epidemiological parameters into coherent future parameter sets)
  • Costing aspects (e.g. can data on disease costs from one or a few countries be merged with another country’s national health accounts, demographic and epidemiological data to generate disease costs?)
  • Substantive public health aspects (e.g. application of models for tobacco control, urban design, dietary, physical activity and other interventions); pioneering cross-national comparisons of obesity and tobacco interventions; sub-population interventions (e.g. health inequality impacts)



Prof Abie Flaxman, Health Metrics Sciences, Institute of Health Metrics and Evaluation, University of Washington

James Collins, Software Engineer

Dr Anja Mizdrak, Public health intervention simulation modelling

Prof Nick Wilson, Professor, Epidemiology

Research Group

Faculty Research Themes


School Research Themes

Prevention and management of non-communicable diseases (including cancer), and promotion of mental health, Data science, health metrics and disease modeling

Key Contact

For further information about this research, please contact the research group leader.

Department / Centre

Centre for Epidemiology and Biostatistics

MDHS Research library
Explore by researcher, school, project or topic.