Assessing Uncertainty in Microsimulation Model Projections

• Activity Number: 458
• Type: Invited
• Date/Time: Wednesday, August 7, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Health Policy Statistics Section
• Author(s): Michael Wolfson*+
• Companies: University of Ottawa
• Keywords: uncertainty analysis ; microsimulation model ; cancer ; technology assessment ; health policy ; projections

Abstract

Microsimulation models are used increasingly to support health policy decisions. Typically, these models are used to compare the impacts of new policy options to status quo projections. The results are subject to a range of uncertainties. Some are inherently unknowable, since they pertain to the future. But others derive from the data used as inputs, the way the data are analyzed (e.g. functional forms used for estimation) to provide algorithms (e.g. transition equations) used by the model, and the Monte Carlo processes used by the model to generate its projections. Moreover, in a number of institutional settings, such as formal technology assessments, there are specific guidelines for quantifying uncertainty where new drugs or devices are proposed for adoption in routine health care. But these guidelines, which were developed for much simpler models, are generally inappropriate for detailed microsimulation model-based analyses In this paper, we propose new guidelines, based on experiences with the Canadian Partnership Against Cancer's Cancer Risk Management Model.