OBJECTIVES: Conventionally, generic quality of life health states, defined within multi-attribute utility instruments, have been valued using a standard gamble or time trade-off. Preference elicitation tasks for both are complicated, limiting the number of health states each respondent can value, and therefore that can be valued overall. Additionally, the two techniques have been shown to introduce bias into health state valuation due to issues of (for example) discounting of future events or risk preference. The objective of this research is to test an alternative method for eliciting preferences for generic health states, and to produce an algorithm for use in Australian cost-utility analyses. METHODS: We designed a DCE to obtain values for SF-6D health states, and implemented it in an Australia-representative online panel (n=1,017). A range of specifications were estimated using a random-effects probit model. Utility weights were then derived such that full health and death were valued at 1 and 0 respectively. RESULTS: The results reflect the broadly monotonic design of the SF-6D, and combination of levels to remove illogical orderings did not lead to a poorer model fit. The most important dimensions in the SF-6D were identified as physical functioning, pain, mental health and vitality. Australian values differed from those reported elsewhere. CONCLUSIONS: This research provides an Australian algorithm for cost-utility analyses using the SF-6D. DCEs are a valuable alternative approach to the time trade-off or standard gamble. The explicit consideration of design principles is a major strength of the study reported here. More work, however, is required to identify how DCEs can best be employed to estimate utility weights for health states.