As part of the validation of the Assessment of Quality of Life (AQoL) instrument comparisons were made between five multiattribute utility (MAU) instruments, each purporting to measure health-related quality of life (HRQoL). These were the AQoL, the Canadian Health Utilities Index (HUI) 3, the Finnish 15D, the EQ-5D (formerly the EuroQoL) and the SF6D (derived from the SF-36). The paper compares absolute utility scores, instrument sensitivity, and incremental differences in measured utility between different instruments predicted by different individuals. The AQoL predicted utilities are similar to those from the HUI3 and EQ-5D. By contrast the 15D and SF6D predict systematically higher utilities, and the differences between individuals are significantly smaller. There is some evidence that the AQoL has greater sensitivity to health states than other instruments. It is concluded that at present no single MAU instrument can claim to be the 'gold standard', and that researchers should select an instrument sensitive to the health states they are investigating. Caution should be exercised in treating any of the instrument scores as representing a trade-off between length of life and HRQoL.