Objectives: To longitudinally examine gender and racial disparities in driving cessation among older adults. Methods: Data came from the Advanced Cognitive Training for Independent and Vital Elderly (ACTIVE) Study (N = 1,789). Logistic generalized estimating equations (GEE) were used to identify predictors of driving cessation; stratified analysis and interaction terms were used to determine whether factors differed by gender and race. Results: Two hundred and five (11.5%) participants stopped driving over the study period. Education was associated with increased risk of cessation for men (adjusted odds ratio [AOR] =1.40, 95% confidence interval [CI] =1.10 to 1.78), but decreased risk for women (AOR = 0.90, 95% CI = 0.82-0.98). Being married was associated with lower risk of cessation for men (AOR = 0.18, 95% CI = 0.06-0.56) but was unrelated to cessation for women (AOR = 1.00, 95% CI = 0.56-1.80). Results were consistent with the hypothesis that racial disparities in cessation widen with increasing age. Discussion: Factors predictive of driving cessation vary by gender. Racial disparities in cessation are wider at older ages. Transportation policies and programs should account for social determinants and aim to address social disparities in driving mobility among older adults.