SynthEco: Synthetic Ecosystems for Expanding Cohort Data to Full Population-Level Representation and Visualization for Decision Support: A Novel Methodology to Improve Urban Population Health

Principal Investigator: Dr. Laurette Dubé

Affiliation: McGill University

Start Year: 2020

To answer questions in urban planning and population-level health interventions such as where to implement health interventions, planners and researchers need tools that provide geographically and temporally resolved representations of the characteristics of their populations. Traditionally, population-wide data collection efforts are very difficult to execute due to the level of recruiting, cost, and methodological reliability. Additionally, they may lack any representation of real-world environments. We propose the SynthEco platform to operationalize government, private and academic research for intervention planning. As an exemplar, we are developing a SynthEco of Montreal, Quebec. SynthEco is created by first using Canadian Census data to create a realistic set of virtual people living in households. We then assign variables from other datasets (e.g., CanPath) to our virtual people – Note that once our synthetic ecosystems are created, the individual CanPath data is no longer needed and individuals will not be identifiable. For example, we get information on key indicators/variables available from CanPath that can be used to evaluate geographically, the health and well-being of the population. Variable categories from CanPath will include the Health and Risk Factor Questionnaire, Baseline Physical Measures, and Mental Health Data. SynthEco will help decision-makers elucidate inequities in health services, plan targeted interventions, and engage in urban planning with public health as a factor.