In recent years much attention has been paid to the development of measures of subjective health status yet, although statistical criteria of reliability and validity have been quite rigourously tested, there has been little consideration of the different theories of disability which underlie the design. The sociology of disability may illuminate such tacit theories. It is suggested that the development of health status questionnaires has not been one of simple rational accumulation in response to methodological advances. Through an examination of the content of health assessment questionnaires, four distinct models of disability are identified. These are shown to influence not only the focus of the content and phrasing of the questions but also, crucially, the way that they perform and how responsive they are to change. The models (the functional, subjective distress, comparative and dependence) are illustrated and discussed in terms related to research design.