The aim of this study was to compare the caries risk profiles of children and elderly, the actual annual caries increment and the impact of some selected caries related factors. The risk profiles were created by a computerised risk assessment program, the Cariogram, which evaluates data and presents the weighted and summarized result as one figure, illustrating the 'percent chance of avoiding caries' in the future. The data used originated from two separate longitudinal studies illustrating the Cariogram's capacity to assess caries risk. One study comprised about 400 children; the other included about 150 elderly. At baseline, information on past caries experience, diet, oral hygiene and use of fluoride was obtained. Saliva analyses included mutans streptococci and lactobacilli, buffering capacity and secretion rate. The caries risk was assessed and after two and five years, respectively, caries was re-evaluated and the incidence was compared with the predictions. Fifty percent of the children, but only two percent of the elderly appeared in the lowest caries risk group. Of the elderly, 26% belonged to the highest caries risk group versus 3 % of the children. The mean DFS increment per year for the total group of children was 0.4 +/- 0.8 (SD) and 1.2 +/- 1.9 for the elderly. Individual factors contributing significantly to the higher risk profiles for the adults were higher plaque scores, higher counts of mutans streptococci and lower buffering capacity. Over all, the risk for caries, as assessed by the Cariogram, was twice as high for the elderly.