The aim of the present study was to introduce the potential tooth years of life lost (PYLL) approach to the analysis of data showing the incidence of tooth extraction according to the reason for extraction and to illustrate its use. Six sets of data from prospective nationwide questionnaire surveys of patients treated by systematic random samples of dentists in five European countries were translated into PYLL. Response rates ranged from 25 to 81 per cent and the number of extracted teeth from 959 to 29,397, according to the study. PYLL in Norway 1988 was calculated using the ages of 85 and 80 as cut-off points, or the average sex-specific remaining life expectancy at the age of extraction; otherwise the age of 85 was used. Mean PYLL for all reasons varied from 40.6 to 46.3 years for Norway in 1988 depending on the cut-off point used. For patients aged 21 and older PYLL85 ranged from 35.7 years for France in 1984 to 42.3 years for Sweden in 1959-61. Employing PYLL changed the rank order of the reasons for extraction based on the number of extracted teeth in some instances. It combined the two dimensions 'incidence of 'potential years of tooth function lost' into a continuous quantitative variable which was easy to understand and simple to handle analytically.