This course will provide an introduction to agent-based models and their application to population health research and health policy. Agent-based models utilize stochastic computer simulations to observe the population-level outcomes and patterns of behavior produced when heterogeneous agents interact with other agents and their environment according to preset rules. These models circumvent many of the limitations of traditional analytic approaches and are increasingly used to investigate the spread of health behaviors and outcomes, as well as to compare interventions and policies to promote population health. Topics covered will include the properties of agent-based models, including illustrations from the health and social sciences; the types of questions best answered by agent-based models; the steps involved in developing, calibrating, and validating agent-based models; and the presentation and interpretation of model results. NetLogo software will be used to demonstrate and practice agent-based modeling techniques. Participants will also have the opportunity to propose and receive feedback on an agent-based model addressing their own research question of interest.

This course is eligible for an EPIC scholarship. Visit the scholarship application page for more information.

By the end of the course, participants will be able to:

• Discuss the properties of agent-based models and their strengths and limitations for population health research
• Formulate research questions that can feasibly be evaluated using agent-based models
• Describe the steps involved in developing and verifying agent-based models, including parameterization, calibration, validation, and sensitivity analyses
• Identify the optimal presentation of results from an agent-based model
• Critically review studies using agent-based models
• Plan and begin programming an agent-based model to address a particular research question

Some introductory epidemiology and biostatistics knowledge will be assumed. No prior experience with agent-based modeling or NetLogo is expected.

• Railsback SF, Grimm V. Agent-based and individual-based modeling: a practical introduction. 2002, Princeton University Press.
• Auchincloss AH, Diez Roux AV. A new tool for epidemiology: the usefulness of dynamic-agent models in understanding place effects on health. Am J Epidemiol 2008; 168(1): 1-8.
• Gorman DM, Mezic J, Mezic I, Gruenewald PJ. Agent-based modeling of drinking behavior: a preliminary model and potential applications to theory and practice. Am J Public Health 2006; 96: 2055-2060.
• Kumar S, Grefenstette JJ, Galloway D, Albert SM, Burke DS. Policies to reduce influenza in the workplace: Impact assessments using an agent-based model. Am J Public Health 2013; 103: 1406-1411.
• Yonas MA, Burke JG, Brown ST, Borrebach JD, Garland R, Burke DS, Grefenstette JJ. Dynamic simulation of crime perpetration and reporting to examine community intervention strategies. Health Education & Behavior 2013; 40(1S): 87S-97S.