Discriminative capacity of the EQ-5D, SF-6D, and SF-12 as measures of health status in population health survey

To compare the EQ-5D, SF-6D, and SF-12 in terms of their capacity to discriminate between groups defined by relevant socio-demographic and health characteristics in a general population survey. Data were obtained from the 2006 Catalan Health Interview Survey, a representative sample (n = 4,319) of the general population of Catalonia (Spain). Effect sizes (ES) and Receiver Operating Characteristic (ROC) curves were calculated to evaluate the instruments' capacity to distinguish between groups based on socio-demographic variables, recent health problems, perceived health, psychological distress, and selected chronic conditions. All instruments showed a similar discriminative capacity between groups based on socio-demographic variables, recent medical visit (ES = 0.47-0.55), activity limitations (ES = 0.92-0.98), perceived health (ES = 0.97-1.33), and psychological well-being (ES = 1.17-1.57). Effect sizes between respondents with and without any of fourteen selected chronic conditions were large (0.76-1.04) for 4, moderate (0.55-0.74) for 8, and small (0.17-0.39) for two on the EQ-5D index. A similar pattern was observed for the SF-12 but ES were predominantly moderate (7 conditions) or small (6 conditions) on the SF-6D. The EQ-5D and SF-12 were largely comparable in estimating the health burden of chronic conditions, recent health problems, and social inequalities. The SF-6D was less sensitive than the EQ-5D index and SF-12, particularly for physical chronic conditions.