Abstract

This paper considers the problems that arise when death, which is a philosophically difficult concept, is incorporated into healthcare metrics, such as the quality-adjusted life year (QALY). These problems relate closely to the debate over euthanasia and assisted suicide because negative QALY scores can be taken to mean that patients would be ‘better off dead’. There is confusion in the literature about the meaning of 0 QALY, which is supposed to act as an ‘anchor’ for the surveyed preferences on which QALYs are based. In the context of the debate over euthanasia, the QALY assumes an ability to make meaningful comparisons between life-states and death. Not only is this assumption questionable, but the ethical debate is much more broad than the question of whether death is preferable to a state of living. QALYs are derived from preferences about health states, so do not necessarily reflect preferences about events (eg, dying) or actions (eg, killing). This paper presents a new kind of problem for the QALY. As it stands, the QALY provides confused and unreliable information when it reports zero or negative values, and faces further problems when it appears to recommend death. This should preclude its use in the debate over euthanasia and assisted suicide. These problems only apply where the QALY involves or seems to involve a comparison between life-states and death, and are not relevant to the more general discussion of the use of QALYs as a tool for comparing the benefits derived from treatment options.