Quality-adjusted life-years have been used in economic analyses as a measure of health outcomes, one that reflects both lives saved and patients' valuations of quality of life in alternative health states. The concept of "cost per quality-adjusted life year" as a guideline for resource allocation is founded on six ethical assumptions: quality of life can be accurately measured and used, utilitarianism is acceptable, equity and efficiency are compatible, projections of community preferences can substitute for individual preferences, the old have less "capacity to benefit" than the young, and physicians will not use quality-adjusted life-years as clinical maxims. Quality-adjusted life-years signal two shifts in the locus of control and the nature of the clinical encounter: first, formal expressions of community preferences and societal usefulness would counterbalance patient autonomy, and second, formal tools of resource allocation and applied decision analysis would counterbalance the use of clinical judgment. These shifts reflect and reinforce a new financial ethos in medical decision making. Presently using quality-adjusted life-years for health policy decisions is problematic and speculative; using quality-adjusted life-years at the bedside is dangerous.