Quality Adjusted Life Years (QALYs) have become a household word among health economists. Their use as a means of comparing the value of health programmes and medical interventions has stirred up controversy in the medical profession and the academic community. In this paper, I argue that QALY analysis does not adequately take into account the differentiated nature of the health state values it measures. Specifically, it does not distinguish between needs and preferences with respect to its valuation of health states. I defend the view that needs and preferences are clearly distinguishable, and that the concept of needs cannot be dispensed with, as many health economists suggest. It is argued that the scale along which health states are measured in QALY analysis is not a continuous interval scale, but one which concerns two distinctly different value dimensions. Measuring the values of health state intervals may reveal the weighting attached to the different value dimensions.