OBJECTIVES: Several studies have used the Fluorosis Risk Index (FRI) to assess risk factors for dental fluorosis. This study reports fluorosis prevalence estimates in a birth cohort using the FRI in different ways. METHODS: Subjects (n=443) were participants in the Iowa Fluoride Study, a birth cohort study of fluoride exposures and intake, fluorosis, and caries. Early-erupting permanent teeth were assessed for fluorosis using the FRI at approximately age 9. Fluorosis prevalence rates were calculated separately for specific teeth and different combinations of teeth, with varied FRI zones included and using different thresholds. Mean numbers of teeth with fluorosis were calculated. RESULTS: Tooth-specific fluorosis prevalence varied, with maxillary central incisor fluorosis most prevalent and mandibular incisors least affected. Considering three zones of each tooth (cervical zones excluded), 40.6% overall had at least one tooth with mild or more involved fluorosis, 30.2% were questionable fluorosis, and 29.1% had no fluorosis. When only FRI zone I areas were considered, the corresponding percentages were 33.2%, 29.3%, and 37.5%, respectively. When different combinations of teeth were used to define fluorosis cases, the prevalences estimated using three zones were generally 1 to 9 percentage points higher than those estimated from FRI zone 1 only. Most fluorosis was mild, with only 7 individuals (1.6%) having FRI severe fluorosis. CONCLUSIONS: The FRI has advantages for use in analytical epidemiologic studies of dental fluorosis. However, the population prevalence estimates vary, depending on the index and case definition used. Therefore, it is recommended that consideration be given to concurrent use of another index (i.e., TSIF, TF, Dean’s) if prevalence estimates are an important study outcome.