This study used a single binary-gamble question per health state per respondent to obtain societal preferences for the health states intermittent claudication and major amputation and compare those with Health Utilities Indices obtained from patients, to test the feasibility of this method, and to investigate whether the utility depends on the presentation of a vignette as generic vs disease-specific. A random sample of the general U.S. population (n = 1,003) was randomly divided into ten subgroups. In telephone interviews, subjects answered one binary-gamble question in a standard-gamble format for each of two health states. The risks of death varied across subgroups but not between health states. Mean utility was estimated by the area above the proportional distribution of responses indicating acceptance of the gamble. The method is based on the binary-choice method used in contingent-valuation studies of willingness to pay. The health states were alternatively described by generic and disease-specific vignettes in two subsamples. The results suggest that the binary-gamble question can be used to obtain societal preferences for health states, and that disease-specific descriptions yield lower utilities compared with generic descriptions of health states.