Purpose: Recent prospective studies have found that cognition is a more salient predictor of driving cessation than physical performance or demographic factors among community-dwelling older adults. However, these studies have been limited to 5 years of follow-up. The current study used data from the Maryland Older Drivers Project to examine predictors of driving cessation in older adults over a 10-year period. Design and Methods: Participants (N = 1,248) completed baseline and 5-year assessments of physical and cognitive abilities. Driving status was ascertained at baseline and annually thereafter. Results: Cox proportional hazard models were used to examine the risk of driving cessation as a function of demographic, physical, and cognitive predictors. The final model indicated three significant predictors of driving cessation, older age at baseline (hazard ratio [HR] = 1.12, p less than 0.001), days driven per week (HR = 0.83, p = 0.05), and slower speed of processing as measured by the Useful Field of View Test (HR = 1.76, p less than 0.01). Implications: These results underscore the importance of cognitive speed of processing to the maintenance of driving. Brief cognitive assessment can be conducted in the field to potentially identify older adults at increased risk for driving cessation. Further research is needed to determine the costs and potential benefits of such screening.