BACKGROUND Survival curves can provide useful information for designing cancer surveillance programs. OBJECTIVE To outline how to derive the density and hazard functions from the survival curve and to use this information to make recommendations regarding cancer surveillance in patients with ulcerative colitis. CONCLUSIONS The hazard of cancer or dysplasia remains low during the first decade of ulcerative colitis but rises exponentially thereafter. After 40 years, approximately 20% of patients with ulcerative colitis acquire cancer or dysplasia per year. CONCLUSIONS A reasonable recommendation for cancer surveillance based on information from survival curves would be a colonoscopy with biopsy approximately 10 years after the onset of ulcerative colitis, and repeated every 3 years for the next decade, every 2 years for the subsequent decade, and every year thereafter. Prophylactic proctocolectomy is an option after 40 years of disease due to the extremely high cancer risk, but data supporting this option are sparse.