The growing demand for subjective measurements of health in clinical studies has encouraged the rapid creation of many new scales, leading to compromises in the quality of some instruments. Quality control standards are desirable to guide the development of measures and subsequent data interpretation. The limitations in existing measures may be classified under three broad topic areas: 1) shortcomings in the design of the instrument, 2) methodological limitations in the process of its development, and 3) shortcomings in the way that the instrument is described following its initial development. We outline examples of common problems in these areas and propose guidelines for the design, development and presentation of health measurement methods. The guidelines represent a preliminary step in formalizing a discipline of health measurement.