Race and ethnic disparities in valuing health

Objective. Patient preferences are often cited as a possible explanation for disparities in treatment. However, no prior studies have examined whether there are racial and ethnic differences in preferences for health states in a general population sample. Methods. Data from 21,362 adult respondents to the 2002 Medical Expenditure Panel Survey were used to study variations in valuations of health states. Respondents' health states were valued based on the self-rated Visual Analogue Scale (VAS) and the Euro-Qol-5D using the US and UK societal scoring algorithms. Regression analyses determined whether valuations in health states varied by race or Hispanic origin, controlling for socioeconomic status and place of residence. Results. Race and ethnicity were not associated with differences in valuations of health states. However, there were systematic differences in characteristics that were controlled, such as health status, age, poverty status, and region of the country. Blacks and Hispanics had slightly higher VAS scores than whites. The negative influence of pain/ discomfort on the VAS score was greater for blacks and Hispanics. Conclusions. Racial and ethnic differences in treatment preferences probably do not result from differences in health state valuations. Future research should explore whether differences in preferences for other attributes of treatment account for differences in treatment decisions. Cost-utility researchers using the EuroQol-5D or VAS need not account for blacks' and Hispanics' systematically valuing health states differently than whites do. However, caution may be warranted when considering interventions designed to manage pain or discomfort, because blacks and Hispanics gave greater weight to that domain of health status in their valuations.