Background The frequency of early repeat and follow-up tests (RAFTs) occurring after colonoscopy has not been previously examined in the literature. RAFTs incur cost, discomfort, and inconvenience to patients who have undergone colonoscopic examination; therefore, it is important to identify factors associated with their use. Methods We identified elderly Medicare recipients who had colonoscopy performed in 1999 from the 5% Medicare administrative files (N = 69,282). We determined the number of early RAFTs (repeat colonoscopy, barium enema, flexible sigmoidoscopy) occurring within the year of initial colonoscopy. Results Of the study sample, 8.3% required at least one RAFT during the year. Using multivariable analysis, we found that RAFTs varied significantly with age, race, sex, income, comorbidity, provider type, and place of service. RAFTs were 22% higher in African Americans compared to whites. Gastroenterologists used 20–35% fewer RAFTs than the other provider types performing colonoscopy. Conclusions The frequency of early RAFTs after colonoscopy occurs in 8.3% of the Medicare population. Important differences exist in the frequency of RAFTs by race and provider type.