Modeling and Systems Approaches for Public Health Policy-Making

Mathematical models for infectious disease that include representations of people's behavior are increasingly available. They can be useful in analyzing the relative impacts of proposed interventions, both pharmaceutical and non-pharmaceutical. This talk will present results of several such analyses. It will address the need for, and limitations of, a systems approach. In particular, it will explore what we should demand from models and what we can and cannot expect them to provide.

The presentation will focus on a regional scale individual-based model for infectious disease epidemiology. I will sketch its construction using census, survey, administrative, and disease surveillance data. In the process, I will explain why we have felt compelled to develop such a complicated model. I will illustrate its use in several published studies analyzing the effectiveness of community-based interventions in a variety of situations ranging from home care-giving strategies to prevalence-dependent individual decisions. I will discuss the policy implications of the results of these studies as part of the broader point, that sophisticated models can be useful for public health policy-making.