Objective: Conflicting requirements are made for health outcome measures reported in the USA and in Europe. Draft guidance issued by the FDA takes the narrow prescriptive view that a PRO is a measurement of any aspect of a patient's health status that comes directly from the patient (i.e. without the interpretation of the patient's responses by a physician or anyone else). This runs counter to the requirement stipulated by NICE that health benefits should be weighted using population preference values. The absolute need for hypothetical health state valuations is questionable and this paper reports on a novel scoring system for EQ-5D based entirely on self-assessments obtained from “real” people. Methods: Data were pooled from several different UK sources, including national population surveys conducted by post and patients self-assessment in clinical studies. A total of 23,679 respondents contributed EQ-5D data that included the health state defined by respondent's reported problem level on the 5 dimensions (OSTATE) and the 0-100 VAS rating of their own health status (EQ-5DVAS). A total of 139 EQ-5D health states were identified. The mean EQ-5DVAS was computed for the 73 OSTATEs for which more than 5 observations were available. OLS regression analysis was performed on the micro-level data taking the OSTATE problem levels as independent variable and the self-rated VAS rating as the dependent variable. Results: The model fit proved to be very good (r2=0.985) when forced through the origin. Level decrements within dimension were monotonic and consistent. A value for the state dead of 12.0 obtained from similar UK national survey data enabled these “real” VAS-based scores to be converted into the 0-1 format required in economic evaluation. Conclusion: Values for EQ-5D based on own-health (“real”) ratings are preferred to hypothetical values by some decision-makers. This simple methodology contrasts markedly with the more complex requirements of utility estimation techniques.