Optimal financing structure mechanism design for healthcare insurance

Zhen Zhu, Purdue University

Abstract

According to a recent survey conducted by the Commonwealth Fund cite, The total spending in the U.S. healthcare system is the highest in the world, accounting for 16% of the U.S. GDP. The per capita healthcare spending in the U.S. is 2-3 times as much as other industrialized countries, but the outcome has yet to reflect this. By no means, we should put the blame solely on the total healthcare spending. It is always difficult to set aside between being economically optimal and being morally correct. Therefore, instead of controlling the total healthcare spending, it is more meaningful to study the extra expenditure, associated with each financing structure, which has no effect on improving the wellness of the patients. In this paper, we name such cost the structure cost. One can view the structure cost as the waste incurred by healthcare financing structure. We consider two types of structure costs: over-utilization cost incurred by healthcare providers and overhead cost incurred by private insurers. Our goal is to investigate the possible reduction of structure cost via the implementation of different healthcare financing mechanisms with the participation of the government. In this thesis, we investigate three mechanisms, all of which can be modeled as two-player Stackelberg games. In the first mechanism, the government provides a universal health care plan and the private insurer subsequently designs a supplementary plan. With the second mechanism, the government reimburses individual out-of-pocket healthcare spending via channels such as tax return. With the third mechanism, the government subsidizes individuals who purchase the private insurance plan. In each game, the government is the leader and the private insurance company is the follower. The government's objective is to minimize the structure cost subject to possible budgetary constraint. The private insurance company's objective is to maximize its gross profit. A particular risk group will decide the exogenous factors such as risk preference coefficient, insurance demand and its elasticity. We evaluate the structure cost as a percentage ratio over the basic medical cost given the input parameters. As a case study, we use the data from the RAND HIE experiment in outpatient acute, outpatient mental and dental care. We estimate the parameters in our model from available RAND data and the health economic and service literature. We analyze each Stackelberg game and compare the three mechanisms numerically. Our models are formulated as bi-level non-convex programming problems. We adapt the Mitsos' algorithm to solve them globally. With our framework, we expect to provide some insightful suggestions into the system design.

Degree

Ph.D.

Advisors

Kong, Purdue University.

Subject Area

Health care management|Operations research

COinS