A critical perspective on the tools to support clinical decision making

Historically, physicians have been taught that economics should not come between them and the care of their patients. Regardless of expense, physicians have been expected to do what is, in theirjudgment, best for the patient. Increasingly, this naive policy has been called into question. Managed care has served as a catalyst for rethinking the relationship between the economics of clinical practice and the outcomes of patient care. Capitation has motivated physicians to better understand why they practice the way they do.’-4 Budget constraints are real when risk is assumed. Thus, in its rudimentary form, managed care forces physicians to find clinical excuses for economic decision^.^.^ In effect, physicians must now think of how to allocate the functional equivalent of a global budget. In a fee-for-service system, where physicians assumed little risk, it was easy to avoid the economic reality of clinical practice. In retrospect, it is almost amusing to think that physicians were exempt from economic acc~untability.~.~ As evidenced by practice profiling, physicians are now judged by an additional, very important criterion-economic efficiency.”I3 Given the uncertainty of the “new economics,” physicians have become enamored of a variety of tools for decision upp port.'^-^' In this regard, AuBuchon has provided a helpful introduction to two tools4ecision analysis and cost-effectiveness analysis. Unfortunately, there are some significant deficiencies associated with clinical decision support. To understand the value of these tools, we must have a clear understanding of their limitations. A panacea is often an illusion. The following paragraphs delineate the ways in which these tools are not sufficient. Economic determination is not decision analysis. There are five methods available for economic determination. These are a) cost-minimization analysis, b) costoutcome or cost-consequence analysis, c) cost-utility analysis, d) cost-effectiveness analysis, and e) cost-benefit analysis.s,22-2s Decision analysis may or may not be used in conjunction with each of these methods. Thus,