A data-driven agent based simulation platform for early health economics device evaluation

Health economics is a relatively new but growing field within the discipline of economics and is concerned with making the best use of scarce resources. Early health economic estimates of new medical devices, in particular, can assist producers of health technology in making appropriate product design and investment decisions. It allows companies to understand their likely market and possible reimbursement more thoroughly. Despite the many advantages of point-of-care testing the key problem facing decision makers at the moment is the poor understanding of the potential value gained from new or alternative product or service offerings. Understanding medical device features in the wider market place can be addressed used agent based modelling and simulation (ABMS). In this paper we examine the use of ABMS underpinned by a novel datadriven approach to model generation. A sepsis use case is presented where pathway and device characteristics are defined using the ‘headroom’ method and a semantic evidence capture application. Types and sub-types are automatically extracted into agent models and subsequently executed in our own data-driven agent based simulation platform (TEASIM). A highly typed data-driven approach is evaluated in a manner that clearly presents the technical aspects of TEASIM platform and its practical usage. Initial evaluation of a data-driven approach (and the TEASIM platform) is positive. The approach offers a viable guide to product development in a cost-effective manner, especially in the earlier stages when deciding between potential product configurations or features.