OBJECTIVES: Although the EQ-5D has been used with increasing frequency to measure health-related quality of life, to date, the measure's validity has not been examined in Chinese American immigrants. We evaluated the construct validity of the EQ-5D by testing its performance with respect to sociodemographic and clinical characteristics of the population and comparing responses on the EQ dimensions with the SF-36 subscales. METHODS: The study surveyed low-income Chinese patients attending a community health center in New York City's Chinatown. Participants received self-administered versions of the EQ-5D and SF-36 and additional questions regarding demographic and clinical information. RESULTS: 856 patients were approached and data were collected from 523 patients (61%). Analysis of the EQ-5D responses by sociodemographic and clinical variables found significant differences among categories of age, gender, marital status, number of medical problems, self-rated health, and specific medical problems. Correlations between similar dimensions and subscale scores were stronger between the two measures than dissimilar ones. Patients without impairments on a given EQ dimension tended to have higher SF-36 subscale scores than patients reporting 'any' impairment (i.e. some/moderate or severe). CONCLUSIONS: The results provide initial support for the construct validity of the EQ-5D in our sample. Further studies should compare the performance of the EQ-5D with other preference-based measures in Chinese persons and examine the valuations that both low-income and Chinese persons place on core aspects of health.