The Mortality Effects of Cost Containment under Universal Health Insurance: The Japanese Experience
J. Mark Ramseyer
Harvard Law School
September 1, 2008
Harvard Law and Economics Discussion Paper No. 619
For over four decades, Japan has offered universal health insurance. Despite the demand subsidy entailed, it has kept costs low by regulatorily capping the amounts it pays doctors, particularly for the most modern and sophisticated procedures. Facing subsidized demand but stringently capped prices on the most complex procedures, Japanese physicians have had little incentive to invest in specialized expertise. Instead, they have invested in small private clinics and hospitals.
The resulting proliferation of primitive clinics and hospitals has cut both the number of complex modern medical procedures performed, and the number of hospitals with any substantial experience in those procedures. With a quarter of the heart disease in the US, Japan performs less than 3 percent as many coronary bypass operations and less than 6 percent as many angioplasties. Of the 855 cities in Japan, 71 percent lack any hospital with substantial experience in the sophisticated modern treatment of cerebrovascular disease, and 83 percent lack any in heart disease.
In this article, I estimate the mortality cost of this regulatorily-driven lack of expertise. Toward that end, I combine mortality data from 855 cities with information on local hospital expertise and local demographic composition. In the typical city, I find that the addition of one hospital with substantial experience in stroke patients or in modern stroke treatment would cut annual stroke mortality by 10 to 15 deaths. The addition of one hospital with substantial experience in cardiac bypass operations or angioplasties would cut annual heart attack mortality in the city by 27 to 56 deaths.
Number of Pages in PDF File: 33
JEL Classification: I11, I18, K32, L51working papers series
Date posted: December 17, 2008
© 2014 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo5 in 0.672 seconds