Feasibility and Validity of the VAS and TTO for Eliciting General Population Values for Temporary Health States: A Comparative Study

Abstract

General population preferences for standardized health states are usually obtained for chronic health states. The primary objective of this study was to analyze the feasibility and validity of using time trade-off (TTO) and a visual analog scale (VAS) to elicit preference values for temporary health states (THS=1 year duration, followed by normal health). Subjects were a random sample (n=300) of the general population. 43 health states generated by the EuroQo-5D were valued. The VAS proved slightly more feasible than the TTO. At aggregate level, correlations between VAS and TTO values were high (Spearman r=0.98), and VAS ratings had slightly greater internal consistency and agreement with rank order preferences than the TTO. TTO values were higher than VAS values, and compression of TTO values suggested substantial reluctance to trade. The effect of age on values was reversed between methods, with older respondents scoring higher on the VAS and lower on the TTO, than other age groups. In conclusion, although the VAS proved to be slightly more feasible than the TTO, with slightly greater empirical validity, further research using a wider range of methods to test validity is required before a definitive conclusion on the relative empirical validity of the two valuation techniques can be drawn.