OBJECTIVES: It has been recognised that EQ-5D, the most widely used generic measure used in the UK, is not appropriate in many areas of mental health.  We report the development of a classification amenable to valuation for use in mental health.    METHODS: The themes were identified through a review of the qualitative literature on the quality of life of people (QoL) with mental health difficulties complemented with in-depth interviews with service users. Framework analysis was used to identify the relevant themes. Items were selected under each theme from interview transcripts, revised versions of those from existing measures and items were newly coined by the team. Items shortlisted using a list of set criteria were presented to service users to assess their content validity. Service users from the primary, secondary and tertiary sectors were asked to complete a pool of items at baseline and follow-up.   Factor analysis was used to determine the dimensional structure of the ReQoL. Item Response theory was used to select items through item information functions.   RESULTS: Thirteen relevant papers from the 200 set retrieved and 19 in-depth interviews yielded seven themes - 6 mental health and one physical health. Out of 1600 potential items 100 items were tested for content validity with 76 service users -59 adults and 17 aged 16-18. Over 3600 service users completed a pool of 40 items. Factor analyses and Classical psychometric analyses have informed the selection of items for a six-dimension classification for ReQoL each item containing five levels.  CONCLUSIONS: : ReQoL provides an important basis for a new QALY measure that better reflects the way different mental health problems impact on the lives of people who use services. The next stage is to obtain preference weights so that the measure can be used in the economic evaluation of interventions in mental health.