The present study is a part of a prospective, longitudinal investigation of caries development in children (n = 692) living in the southern suburbs of Stockholm. The aim was to evaluate, longitudinally, the caries-predictive ability of variables describing social and immigrant background, dietary habits, microbial and oral hygiene factors, and fluoride exposure in children at 1 and 2.5 years of age with respect to caries development before the age of 3.5. The predictors for caries development in children before 2.5 years of age were mutans streptococci (p < 0.01), immigrant background (p < 0.01), and consumption of candy (p < 0.01). The predictors for developing manifest caries between 2.5 and 3.5 years of age were mutans streptococci (p < 0.001), mother's education (p < 0.001), immigrant background (p < 0.01), and consumption of candy (p < 0.05) and sugar-containing beverages (p < 0.05). The caries incidence at 2.5 years of age as well as the caries increment between 2.5 and 3.5 years of age were significantly higher in children with immigrant background compared to non-immigrants. This study indicates that the possibility of identifying children at risk for caries development increases longitudinally from 1 to 3.5 years of age and there was a synergistic effect between age and each predictor.