The aim of this review was to systematically assess clinical evidence in the literature to determine the predictive validity of currently available multivariate caries risk-assessment strategies (including environmental, sociodemographic, behavioral, microbiological, dietary/nutritional, and/or salivary risk factors) in: 1) primary teeth; 2) coronal surfaces of permanent teeth; and 3) root surfaces of permanent teeth. We identified 1,249 articles in the search, and selected 169 for full review. Inclusion and exclusion criteria were established prior to commencement of the literature search. Papers that conformed to these criteria were included (n = 15 for primary teeth; n = 22 for permanent teeth; and n = 6 for root surfaces), and 126 papers were excluded. Included articles were grouped by study design as: longitudinal, retrospective, and cross-sectional. The predictive validity of the models reviewed depended strongly on the caries prevalence and characteristics of the population for which they were designed. In many instances, the use of a single predictor gave equally good results as the use of a combination of predictors. Previous caries experience was an important predictor for all tooth types.