Complexity in Mathematical Models of Public Health Policies: A Guide for Consumers of Models

PLOS: 10/29/13

Mathematical models are now routinely used to inform public health policies. In addition to being useful for theoretical simulations of disease pathogenesis, models can be used to estimate the impact of approaches to control epidemic diseases like pandemic influenza or HIV, as well the health impact and cost-effectiveness of interventions ranging from knee surgeries to new pharmaceuticals [1],[2].

General medical and public health readers face a dilemma when presented with increasingly complex models used for public health policy questions: how do we know whether to trust the results of a model-based analysis, and potentially alter health policies on the basis of those results? Models are valuable for planning interventions that cannot be tested through randomized controlled trials (ethically or practically), simulating the implications of alternative theories about disease pathogenesis or control strategies, and estimating population-wide costs and consequences of public health programs. Since every public health policy decision implicitly involves assumptions, simply avoiding models because they have assumptions is not a logical approach to health policy. Read more