Editorial: On the Benefits of Modeling Using QALYs for Societal Resource Allocation: The Model Is the Message

Abstract

While moderating a debate at the recent ISPOR meeting in Athens, Professor Milt Weinstein suggested that maybe we should focus more on the “LY” (life-years) and less on the “QA” (quality-adjusted). I interpreted this comment to mean that health state utilities are only one element in the larger bioclinical models that are at the core of many cost-utility analyses. Typical disease-based cost-utility models also include intermediate end points to reflect disease progression, the probabilities of clinical events, and projected survival, as well as the cost elements. The focus of this quality-adjusted life-years (QALY) consensus development workshop was, however, on health state utilities—one element in the entire model. We have argued elsewhere that similar benefit-risk models, including health state utilities, could be very helpful in making regulatory approval deliberations more systematic and transparent [1]. In my view, the use of integrative, quantitative models to assist health-care decision-makers is more important than precisely how morbidity (i.e., health-related quality of life) is reflected in those models. On the whole, these articles provide a fair and very useful representation of the current state of the debate about health state utilities and preferences in our field: namely, there remain several persistent and contentious issues—both about the theory and the practice. As noted, the ISPOR Consensus Development Workshop was motivated by the challenge that Nobel laureate Daniel Kahneman posed in his ISPOR address in 2005. He reprises that argument about human adaptability to illness in his contribution to this collection. Although these final Workshop articles did not resolve any of these issues, the Consensus Development Group should be congratulated on fashioning eight highlevel principles [2] that capture the state of the debate as well as a way forward, including the pragmatic next step of calling for the development of a reference method for estimating QALYs. In my view, this is a useful goal to pursue, but it will be helpful to keep in mind some higher level distinctions that were either implicit or unaddressed in these articles. The Workshop articles did not explore two related foundational issues in any depth: 1) the use of the QALY in normative versus positive (or behavioral) economics; and 2) welfarism versus extra-welfarism as a basis for the use of the QALY in societal resource allocation.