We can do little more than map out the structure of our argument. Our aim is not to say that these thoughtful positions are wrong but to identify exactly where we differ.

Dr. Kosseff hopes that rationing will not be necessary if waste can be eliminated, but the fundamental reality of limited resources makes this an unavoidably false hope. The essential question is, Who should make rationing decisions? Much can be said for the centralized, rule-based approach that Drs. Gurewich and Moore seem to prefer, but Dr. Temianka helps to explain why it is not feasible or desirable for insurers or government bureaucracies to micromanage most medical decisions. The adversarial approach to medicine that Dr. Moore suggests (when he says that physicians should behave more like lawyers in court) is precisely what most physicians detest about modern practice. Dr. Volpintesta argues that physicians can be persuaded to do the right thing without financial motivation; financial incentives of some kind are always present, however, and cost containment usually fails unless incentives reinforce the information from outcomes studies and clinical guidelines. Accordingly, like it or not, physicians are being asked to, and are agreeing to, make some discretionary rationing decisions at the bedside under financial incentives.

We do not mean to insist that this arrangement is right for everyone or is socially optimal. We simply recognize the reality of this arrangement and that physicians have little ethical guidance as they enter into it. Dr. Kosseff's unrealistic suggestion that physicians "leave practice" or Dr. Moore's refusal "to do managed care's dirty work" might be persuasive to someone deciding how to structure insurance or whether to enter managed care practice at all, but they are not helpful to physicians who find themselves making treatment decisions under financial constraints.

Despite the various ideas offered, at least we all agree with Dr. Temianka that honesty is essential. Our suggestion that catches the most flak is that physicians should think in terms of the best use of resources for the group of patients they are responsible for, rather than the traditional myopic focus on the absolute best for each individual. A group-based focus is not as antithetical to medical practice as Drs. Moore, Prechter, and Katzoff maintain. It is the foundation of public health. It is also implicit in how every physician allocates time and energy in a busy practice that requires balancing the competing needs of different patients. A group focus is even more explicit in the conventional resource-constrained settings in most other countries and in government facilities here. Rhetoric aside, a group-based ethic in other professions is consistent with attention to individuals. When lawyers are in court, Dr. Moore is correct that they think only of their individual client, but in setting their schedule, deciding to take cases, advising settlement, and setting their fees, lawyers manage a case load. Similarly, investment advisors manage portfolios, artists produce collections of work, and parents raise a family. These examples illustrate why there is nothing inconsistent between the Hippocratic tradition and thinking about the best results from a body of work. This is also consistent with Dr. Prechter's preferred Golden Rule, which in essence embodies the impartiality principle we develop in our article. Although Dr. Gurewich is correct that these guidelines do not require doing "everything possible," they do provide the solace that comes from honestly telling patients and families that "I did my best." Better this than an ethic that says make as much money as possible or come as close as possible to committing insurance fraud without getting caught.