There is increasing use of multiattribute health-state utility systems, such as the Health Utilities Index and the EuroQol (now EQ-5D), to estimate quality-adjusted life years (QALYs) for cost-utility analysis. Whereas the preferences elicited from individuals using willingness-to-pay techniques for cost-benefit analysis would be expected to reflect those individuals' income levels, it is often suggested that cost-utility analysis can avoid this income effect by not valuing health in monetary terms. Contrary to this view, the authors argue that income can influence the measurement of utilities used to estimate QALYs. In the context of multiattribute utility instruments, two income effects can take place: 1) when individuals are asked to value health states to generate the set of utilities to apply in subsequent evaluation studies; 2) when those multiattribute systems are used to categorize individuals' (usually patients') health status in the field in applied evaluation studies. The authors review the most popular utility systems regarding how these income effects are handled and assess the implications for the measurement of utilities using these systems.