As longitudinal epidemiologic studies of dental caries address increasingly complex research questions, approaches to analysis of data from those studies have become more sophisticated. This review examines methods available for analyzing and reporting data from such studies. Traditional analytic methods utilize the DMFS increment as the outcome measure in longitudinal studies of caries. However, two other outcome measures may be needed to address some research issues: cumulative incidence, which quantifies caries risk; and incidence density, which quantifies caries rate. Four major analytic decisions have to be addressed when computing DMFS increment: examiner misclassification ("reversals"), teeth lost due to caries, findings from more than two examinations, and multiple events such as caries initiation and progression. We present a uniform approach for enumerating caries events that permits the same analytic decisions made in calculating DMFS increment to be applied to cumulative incidence and incidence density calculations. In view of the variety of analytic decisions that must be made when enumerating events in longitudinal studies of caries, authors should specify how all potential changes in caries status were handled. Furthermore, if a study uses more than one outcome measure, the same decisions for enumerating events should be used when computing those measures.