HOW ROBUST ARE PROPOSED MITIGATION STRATEGIES TO MISSPECIFICATION OF AGE-RELATED CONTACT RATES? - AN EXAMPLE FROM THE CONTROL OF INFLUENZA (A/H1N1) SPREAD
Romarie Morales1, Jay Taylor2.
1Arizona State University, Tempe, AZ, 2School of Mathematical and Statistical Sciences, Arizona State University, Tempe, AZ.
The research objective is to find the best mitigation strategies for fast spreading influenza (A/H1N1) virus by identifying the right levels of aggregation (age-windows) to incorporate into an informational, mathematical model of the disease. A simple extension of the susceptible-infected-recovered (SIR) age-structure model of influenza transmission is used to illustrate the use of the dynamic optimization methodology. We used the interior point method approach which allows a more effective inclusion of real life constraints (i.e., limited vaccines) to assess the potential discrepancies in the age structure component. Our results show light on the growing concern that contact matrices, which determine how disease is spread within and between age groups, are formulated without ample data collection or specificity. If we apply a contact matrix that does not originate from the country of study, we expect to see an intervention policy that differs greatly from the policies implemented when a more accurate matrix is used. This work will motivate the collection of data regarding current individual contacts per day in different countries.