OBJECTIVES: To select and test severity labels for a new, five-level version of the EQ-5D. METHODS: The EQ-5D is a generic instrument for describing and valuing health. Each dimension (Mobility, Self-Care, Usual activities, Pain/discomfort, Anxiety/depression) is currently measured on 3 levels of health (no problems, some problems, extreme problems). A EuroQol Group task force was established to find ways of improving the instrument’s sensitivity to small and medium changes in health and reducing the ceiling effect. The study used a two-stage approach: i) response scaling was performed in the UK and Spain to explore the severity represented by potential new labels that could be used as additional or replacement levels within each dimension. Labels as close as possible to the 25th, 50th and 75th centiles were selected for further testing; ii) the face and content validity of alternative 5-level versions were investigated in 8 focus groups of healthy participants and patients in each country. Hypothetical health states based on a revised labeling system were also reviewed. RESULTS: Rank ordering of severity labels was similar across dimensions and countries. Selecting labels at approximately the 25th, 50th and 75th centiles produced two alternative 5-level versions. Focus group work showed a slight preference for the wording ‘slight-moderate-severe’ problems, with anchors of ‘no problems’ and ‘unable to do’ in the EQ-5D functional dimensions. Similar wording was used in the Pain/discomfort and Anxiety/depression dimensions. Focus group comments indicated that, although not always colloquial, the magnitude of problems represented by the new labels was well understood. Comments on hypothetical health states referred more to their internal consistency than to the labels, which were again well understood. CONCLUSIONS: This study represents the first step towards developing an official 5-level version of the EQ-5D. Testing in patient settings and development of the corresponding health state valuations is required.