Long-term care comparative studies by agent-based simulation: a computational framework and a case study

Shuang Chang, Koji Maruhashi

Abstract: Long-term care (LTC) services provide various types of assistance to elderly persons who may have difficulty conducting daily activities due to deterioration in health. There has been a growing interest in comparing different LTC designs to generate knowledge on how to deal with the diverse needs of such services. However, the system complexity arising from micro-level interactions among involved stakeholders complicates the comparison and creates the urgent need for deploying new approaches. This work aims to formalize and construct a computational framework that enables the deployment of bottom-up simulation approaches to compare and evaluate LTC models. It can be used to investigate the extent to which various LTC designs may influence LTC system outcomes, particularly in terms of equity and efficiency. To illustrate the application of this framework, we present a comparison of the effects of four different LTC designs considering the heterogeneity of individuals and their interactions with care providers using an agent-based simulation approach. This framework is among the first to support such comparative LTC studies by bottom-up simulation approaches. It is expected to be extended and customized by LTC domain experts for comparing the policy alternatives of their interests and thus to shed light on designing better LTC systems for benefiting the elderly.