Public hospital organisation in Australia: a game theoretic & mechanism design-based approach

Background: Public hospitals are expected to provide high quality services and also reduce or maintain the costs. Therefore, every aspect of healthcare service delivery has to be thoroughly examined. One of the ways to conduct a research inquiry into the organisation of healthcare service delivery is to analyse the interactions between healthcare providers. Game theory and the theory of mechanism design have been used to analyse the interactions between different parties that may have a principal-agent relationship. The delivery of healthcare services by public hospitals has policy implications because their activities are controlled by politicians and bureaucrats. Purpose: This study seeks to investigate the strategic interaction-driven organisation or organising of healthcare services delivery by public hospitals by integrating: (a) institutional design, system thinking, and a structure-based approach; and (b) game theory, the theory of mechanism design and the principal-agent paradigm (delegation, policy implementation and bureaucratic capacity). Methodology: This study uses average length of stay (ALOS) data for game theoretic policy implementation modelling. MATLAB was used to operationalise two game theoretic models: (1) a simplified social or prisoner’s dilemma game; and (2) a prisoner’s dilemma game with evolving cooperation in a social network. Findings: The results of the two models presented in this study indicate that the more that hospitals play (interact or engage with each other), the more the defection rate decreases. Also, competition increases due to their interactions because the players’ performance is relative to the maximum points scored at the conclusion of a set of rounds. Therefore, early cooperation increases a player’s scores and also the overall performance, yielding a higher benefit. In a public hospital system, hospitals do not need to compete with each other because all patients are covered by public insurance cover. However, it is in their best interest (and the interests of the general public) that they engage with each other and reduce the policy implementation error (PiE). These results also indicate that the higher the size of a group (N) is, the lower the cooperation rate would be. Theoretical Implications: This research study brings together theories (game theory and the theory of mechanism design) and the three approaches to organisation or organising using game theoretical modelling. This study implicitly advances the goal of innovation in techniques and applications of empirical game theory (with the use of hospital data for simulation) by extending it to the healthcare research domain from computer sciences. It also extends the models of delegation, policy implementation error and bureaucratic capacity to healthcare research. Practical implications: The models used in this research study simulate the results based on average length of stay (ALOS) in a hospital. A reduction in not only reduces spending, it also frees up beds for patients on the waiting list. It could become a foundation of a new regulatory framework. Originality: There have been demands for theory integration both in organisational and healthcare research domains. This study integrates three approaches to organisation, game theory, the theory of mechanism design and the principal-agent paradigm to operationalise two game theoretic models of policy implementation. To the best of the researcher’s knowledge, it is the first game theoretic study on the organisation of healthcare service delivery by public hospitals.