The utility approach to the measurement of health-related quality of life is discussed and its theoretical foundations and compatibility with economic evaluation are explained. The advantages of the approach include its generalizability, comprehensiveness, ability to integrate mortality and morbidity effects, ability to represent multiple viewpoints, and its incorporation of time and risk preferences in the scores. Disadvantages include interviewer administration, limitations on obtaining scores from all types of subjects, some lack of precision, and attenuated clinical sensibility. Use of the utility approach as an outcome measure in two recent clinical trials is also described. In a trial of an oral gold compound for arthritis, the utility measures were responsive and added information to that from standard clinical and other quality-of-life measures. The utility approach and clinical decision making are being used in an on-going trial of two procedures for prenatal diagnosis in which major potential differences include quality-of-life effects. The comprehensiveness of the utility approach overcomes difficulties in interpreting results from more narrowly focused psychosocial measures of outcome. In sum, the utility model provides a reliable, valid, and responsive approach to measuring quality of life as an outcome in clinical trials.