OBJECTIVES 1) To assess the associations of EQ-5D index score, EQ-VAS, and seven estimated utilities with self-reported stroke in a U.S. national representative sample, and 2) To compare estimated utilities with EQ-5D and EQ-VAS in both stroke and non-stroke samples. METHODS Data were extracted from the 2001 Household Component of the Medical Expenditure Panel Survey(MEPS). Seven estimated utilities were derived from the SF-12v1®, including HUI3/VAS item models(IM) and categorical models(CM) from the Sengupta-Nichol, Brazier SF-6D, Lundberg VAS, and Sullivan EQ-5D algorithms. An analysis of covariance was used to determine differences in mean utility scores between individuals with and without stroke. Covariate(social-demographic and comorbidity) adjusted effect sizes(ES) were calculated for utilities between individuals with and without stroke, as well as between estimated utilities and EQ-5D/EQ-VAS. RESULTS A total of 19,475 individuals completed SF-12v1®, EQ-5D, and EQ-VAS. Mean age was 42.2 years, 40% were female, 14% had self-reported stroke. Individuals with stroke had significantly lower covariate adjusted mean EQ-5D(0.16 points difference), EQ-VAS(0.14 points difference), and estimated utilities(0.11 or 0.12 points differences) when compared to those without stroke(all p<0.0001). Stroke individuals showed large effect in utility scores when compared to non-stroke individuals. HUI3-IM(ES=1.20) and HUI3-CM(ES=1.22) showed ES similar to EQ-5D(ES=1.23), and both VAS-IM and VAS-CM(ES=1.09) had ES similar to EQ-VAS(ES=1.08). When comparing estimated utilities with EQ-5D, Sullivan EQ-5D(ES=0.26) and SF-6D(ES=-0.29) showed small effect, while other estimated utilities showed no effect in stroke individuals; all estimated utilities showed small(ES=0.23 for Sullivan EQ-5D) to large effect(ES=-1.00 for SF-6D) in non-stroke individuals. CONCLUSIONS EQ-5D, EQ-VAS and estimated utilities in stroke individuals displayed clinical meaningful differences from those without stroke. Derived HUI3/VAS showed the similarity with the EQ-5D/EQ-VAS in stroke individuals. Our findings indicated that EQ-5D, EQ-VAS, and estimated HUI3 and VAS were sensitive to health state and could be used for cost-effectiveness analysis in stroke individuals.