AbstractThe DMF Index and its components have been utilized by various researchers to measure aspects of the caries process ranging from prevalence of carious lesions to the amount of dental care needed. It is generally recognized that DMF Teeth (T) measures life‐time caries experience and is not a good predictor of the amount of treatment time needed in population groups. The purpose of this study is to evaluate the ability of certain modifications of the DMFT components to predict the amount of treatment time needed. In a sample of 109 adult cases, normal and diseased conditions were coded along with the usual DMFT measures and the number of missing teeth needing replacement (NR). Corresponding treatment procedures related to caries (periodontal treatment was not coded) were recorded and then converted to treatment times using the Relative Productivity Unit (RPU). Stepwise multiple regression and discriminant analysis procedures were used. The number of D teeth was linearly related to treatment time, but the number of NR and M teeth was not. Regression procedures indicated that the square root values of D and NR were the best predictors of treatment time (RPU) and produced a multiple r of 0.69. When patients were classified into categories of treatment time (low, medium, high), the DMF Treatment Time Index was able to correctly classify 70 % of the cases into their appropriate treatment time categor