Welfare, QALYs, and costs - a comment

What costs and what effects should be included in cost-effectiveness analysis from a welfare theoretic perspective? There is a broad agreement to include health-care costs and to measure the effects as Quality-Adjusted Life-Years (QALYs). However, a hot topic has been how to handle survivor consumption and leisure foregone. It is well established that the costs for these aspects should be included only if their associated benefits are also included. The key question is, then, if these benefits are included in QALYs. In a recent paper by Nyman (Health Econ., in press) it was argued that these benefits are not generally included in the empirical assessment of QALYs and that these costs therefore should be excluded. Even if this recommendation is correct, the reasons for it can be questioned. It is here instead argued that this decision should be based on theoretical - and not empirical - considerations. Thus, these costs should be excluded because QALYs are unlikely to be consistent with a utility function also including consumption and leisure. If so, then it becomes more important to instruct individuals not to include these benefits in their QALY assessments than to consider to what extent they may or may not be implicitly included. It is also demonstrated that these results have bearings on non-medical costs for the case when survival is not affected.