- Purchase/rental options available:
Perspectives in Biology and Medicine 46.1 (2003) 24-37
[Access article in PDF]
Indirect Restraints on the Provision of Health Care Quality
Richard A. Epstein
THE OVERARCHING THEME OF THE CONFERENCE in November 2001 was directed toward the evaluation and delivery of health care services. We ask about the quality of health care: can we afford it? And can we identify it? Most discussions of the quality of care emphasize its immediate determinants, such as the kinds of protocols or tests that should be adopted in response to particular conditions. Thus, the Institute of Medicine's publication Crossing the Quality Chasm: A New Health System for the 21st Century (2001) has defined quality as "the degree to which health services for individuals and populations increase the likelihood of desired health outcomes and are consistent with the current professional knowledge." This definition entails an emphasis on matters of short-term delivery of health care, relative to competent technique, shared decision-making, and cultural sensitivity. Often the key questions ask about the underuse of certain needed procedures (e.g., vaccinations) or the overuse of others (e.g., antibiotics).
There is little that any lawyer can say sensibly about the design of professional medical standards. But the influences on health care are indirect as well as direct, [End Page 24] and this paper will focus on the indirect influences. As the title of the conference suggests, issues that relate to the quality of health care go beyond the immediate matters of health care protocols to larger questions of system design and organization, and the impediments that these might place in the path of quality health care. Furthermore, the question marks in the conference title reveal a widespread skepticism about our collective ability to solve these structural problems.
The standard drill when addressing such issues takes a familiar two-step form, and has been applied to such markets as agriculture, education, and housing, to name only a few. We start our analysis by positing a universe with hothouse conditions that are ideal for the emergence of rational markets, and then we relentlessly show how these conditions are not, and cannot be, satisfied. Our myopic remedy usually calls for the infusion of additional cash into the sector that we champion—here health care—with scant regard for the dislocations that these funding decisions will have on those activities exposed to the new taxes, or on those rival activities that do not receive an infusion of funds equal to that committed to the preferred activity. It is all too common to ignore the "excess burden" associated with all taxes—the reduced output that comes when certain activities are foregone because the consumer pays more than the producer receives (Gwartney and Stroup 1987, pp. 110-11). The tax increment will presumptively reduce the resource base. It is equally misguided to forget that selective subsidies will distort the relative prices of goods and services that should compete with each other on level ground. Finally, it is wrong to assume that we can form any social consensus on which activities should be benefited and which should be taxed. The political infighting could easily exhaust any subsidy that might be provided.
However, whatever its political dangers, I think that the first part of this two-step approach works methodologically. Begin with the idealized universe, and then ask whether it offers a benchmark for evaluating the imperfect institutions with which we inevitably work. What to do then requires more consideration, because the second half of the problem is equally important. Given the constraints that we face, what compromises do we make with the ideal? Which unhappy outcomes are we prepared to accept in one arena in order to improve outcomes elsewhere? To see how the process often goes and should go, let us start with the idealized model and then examine the extent to which it survives more real-world assumptions.
Idealized and Realistic Models
Idealized models are always risky because they are artificial. Yet these models are always necessary because they illuminate relationships that would otherwise be too complex to...