Modeling an Agent-Based Healthcare Service System: An Exploration of the Capitation Payment Scheme of Zhishan-CHG in Taiwan

Abstract

Based on information from preventive care and capitation pilot projects sponsored by Taiwan's Bureau of National Health Insurance (BNHI), this study modeled and analyzed the Zhishan community healthcare group (CHG) service system using a multi-agent simulation. The model was built and validated according to publicly available statistical and transactional data from Zhishan-CHG, and estimated the system's financial sustainability under different conditions. Five scenarios were explored in the experiment simulating the capitation payment scheme in over a period of three years. The results can be summarized as follows: (1) medical cost expenditure per person can be reduced as more patients are enrolled in the capitation system, (2) the service model with a health promotion agent outperforms the model without one in a fee-for-service (FFS) cost control system, (3) greater quality and retention rates improve capitation income, (4) the risk of the CHG service system can be mitigated by excluding inpatient medical costs from the virtual points assigned by NHI, and (5) simulation results show a positive financial outcome for Zhishan-CHG. These simulation results suggest that for primary care institutes forming CHGs in Taiwan, combining the capitation model with proper risk estimates is cost effective.