OBJECTIVES: To test whether individuals' responses to standard gamble (SG) and visual analogue scale (VAS) questions do not depend on the time horizon of the health scenario presented. METHODS: Face-to-face interviews were conducted in a convenience sample aged 24 to 59 years at a Southern University in the US (n = 14). Preferences were estimated using SG and VAS. Individuals rated their preferences for three health states of varying severity of post chemotherapy nausea and vomiting (PCNV) assuming different time horizons. Repeated measures analysis of variance (ANOVA) were conducted (SX6X2) for each health state to determine the affect of TIME (levels: 3 days, 3 months, 1 year, 5 years, 20 years and rest of life) and METHOD (levels: SG and VAS) on preference. RESULTS: Analysis showed that the time horizon used did affect an individual's preference rating for SG and VAS. For all three models tested (OP, ST, NV), main effects TIME (F(5,65) = 10.71, F(5,65) = 13.85, F(5,65) = 17.40) and METHOD (F(1,13) = 34.56, F(1,13) = 58.01, F(1,13) = 17.80) were statistically significant. For the least desirable heath state (NV), TIME*METHOD interaction was also significant indicating the two factors were not independent of each other, F(5,65) = 7.51. These results showed individuals rated the same health state as less desirable (lower rating) for longer time horizons and the preference elicitation method used made a difference. Furthermore, not only were health states not "timeless", but preference reversal also occurred resulting in fates perceived as worse than death. CONCLUSIONS: For the majority of respondents the utility independence assumption with respect to time for SG and VAS methods did not hold. Similar to Bala et al (1999) and Franic et al (2003) the results of this study indicated preference weights as measured by SG and VAS techniques were not "timeless". Regardless of the preference measure used: both SG and VAS yield higher scores for shorter time horizons.