The aim of this study was first to investigate the relationship at group level between approximal caries, restorations in posterior regions and consumption of sweets during the past 3 yr in 15-yr-old schoolchildren selected for different caries experience. An interview method was used with the sweets on display. Secondly, the aim was to reinterview the subjects as 18-yr-olds using the same method in order to study the correlation between caries incidence and consumption of sweets during the past 3 yr. Seventy-five individuals from three groups, 30 with 0, 25 with 4-6, and 20 with 12-15 approximal DFS, were selected for the first interview. Sixty-nine of them were reinterviewed 3 yr later. Caries incidence during the 3-yr period was expressed as a percentage of the number of caries-free approximal surfaces of premolars and molars at the age of 15. The results showed no statistically significant differences in mean number of intakes of sweets per day (2.2, 2.2, and 2.8) between the original DFS groups. The correlation coefficient between caries incidence and number of intakes of sweets in the second part of the study was 0.25 (P less than 0.05). After correction for the modifying effect of oral hygiene, the coefficient was 0.51 (P less than 0.01) in the less favorable fraction and 0.11 (P greater than 0.05) in the favorable fraction.