The social tariff of EQ-5D is not adequate to assess quality of life in patients with low back pain

Abstract

We compared the quality of life perceived by patients with non-specific low back pain with that predicted by the social tariff of the Spanish version of EQ-5D questionnaire. For each health state of the EQ-5D, an adjusted tariff for patients with back pain was obtained using a linear regression model in which the linear effect of the three levels of response for each of the five domains of the EQ-5D was assumed. These coefficients were compared with those obtained for the general Spanish population. In another model, equal in structure to the standard "Dolan N3" model, the linear effect of the five domains was not assumed. In 633 patients, 93 health states were recorded. Significant differences in the coefficients of self-care (p = 0.003) and the maximum level of severity in any dimension (p < 0.0001) were observed. The social tariff of the healthy population is different from the tariff of low back pain patients, with general population values being lower than those of patients, particularly in the 211 health states in which any dimension is at level 3. Weights of the different EQ-5D dimensions showed a non-linear effect on the patients' quality of life. Methods used to develop the social tariff for the Spanish version of EQ-5D were inadequate. In addition, this study shows that values given by the general population are different from those of low back pain patients, further confirming that the social tariff of EQ-5D should not be used with actual patients.