Changes in concepts of health and disease have led to increased interest in health related quality of life in medicine. Quality of life measures tend to treat quality of life as a stable construct that can be measured externally. They do not consider people's differing expectations and assume that what quality of life means to people is stable over time. This paper reports on a study which aimed to find out how measures of oral health related quality of life (OHRQoL) vary between and change within individuals. Longitudinal semi-structured open-ended interviews were carried out with twenty people with socially noticeable broken, decayed or missing teeth who were or were not seeking dental treatment. The data were collected and analysed using the iterative processes of grounded systems theory based on Glaserian (Theoretical Sensitivity, The Sociology Press, Mill Valley, CA, 1978) grounded theory and Luhmann's (Social Systems, Stanford University, Stanford, 1984) social systems theory. During the data analysis it emerged that participants' were adopting positions on seven dimensions of oral health relating to the positions that people would adopt with respect norm, attribution, trust, accessibility, commodity, authenticity, and character. The core distinction that accommodated people's varying positions was that people constructed their own margins of relevance of oral health. The margins of relevance indicated a variable from a hypothetical extreme of 'super-relevant' to the other extreme of 'not relevant'. The margins of relevance could shift, meaning that assessments of quality of life would vary. Oral health related quality of life is therefore defined as the cyclical and self-renewing interaction between the relevance and impact of oral health in everyday life.