A comparison of the EQ-5D and SF-6D across seven patient groups.

As the number of preference-based instruments grows, it becomes increasingly important to compare different preference-based measures of health in order to inform an important debate on the choice of instrument. This paper presents a comparison of two of them, the EQ-5D and the SF-6D (recently developed from the SF-36) across seven patient/population groups (chronic obstructive airways disease, osteoarthritis, irritable bowel syndrome, lower back pain, leg ulcers, post menopausal women and elderly). The mean SF-6D index value was found to exceed the EQ-5D by 0.045 and the intraclass correlation coefficient between them was 0.51. Whilst this convergence lends some support for the validity of these measures, the modest difference at the aggregate level masks more significant differences in agreement across the patient groups and over severity of illness, with the SF-6D having a smaller range and lower variance in values. There is evidence for floor effects in the SF-6D and ceiling effects in the EQ-5D. These discrepancies arise from differences in their health state classifications and the methods used to value them. Further research is required to fully understand the respective roles of the descriptive systems and the valuation methods and to examine the implications for estimates of the impact of health care interventions.