formers” and thus avoid screening. This has enormous implications for societal costs, because substantial savings could be realized by reducing the number of colonoscopies in the general population. It also has implications for the gastroenterology workforce, because fewer procedures will be required. However, there are several other factors that will put considerable pressure on the gastroenterology workforce. At this time, only 30%–48% of the population at risk has actually been screened. The current number of trained endoscopists is inadequate to provide the number of screening colonoscopies required for 100% of the at-risk population. This problem was exacerbated by efforts in the mid-90s to decrease the gastroenterology workforce as a response to pressure from funding agencies such as health maintenance organizations. There was a 30% decrease in the number of gastroenterology trainees graduating from programs between 1994 and 1997, from a high of 400 trainees per year to 256 per year.148 The number of screening colonoscopies required yearly for the 2.78 million Americans turning age 50 years in 1995 was estimated to be 230,000 if done every 3 years and 80,000 if done every 5 years.149 This number is higher now and will continue to increase over the next 10–20 years. Use of virtual versus optical colonoscopy in the elderly Virtual colonoscopy has not been studied extensively in the elderly but might be a more appropriate screening test for an elderly patient than an optical colonoscopy.150, 151 The benefits include a procedure without sedation, although the insufflation of air into the colon during the study is often uncomfortable. The risk of perforation is certainly reduced, although there are reports of colonic perforation during virtual colonoscopy.152, 153 There is still a requirement for a colonoscopy preparation. A wide range in the sensitivity and specificity of virtual colonoscopy has been reported, with sensitivity of computed tomographic colography ranging from 34% to 73% depending on the interpreters and the size of the lesions detected.154 It is not clear whether this technique will be as sensitive as colonoscopy (near 99%) when used in the geriatric population. Chemoprevention of colorectal cancer in the elderly Most studies of chemoprevention have focused on use of aspirin or NSAIDs due to the observation that patients taking aspirin or mesalamine for treatment of arthritis had a significant reduction in the prevalence of colonic polyps found with screening colonoscopy. The patients enrolled in these studies tended to be younger (age 45–70 years) and did not have a history of GI bleed or other relative contraindications to NSAID use. Currently, only use of NSAIDs has been shown to significantly decrease the relative risk (RR) of colon cancer (RR, 0.66) in patients aged 50–74 years.155 The greatest RR reduction of aspirin use156 was observed with advanced neoplasms (RR, 0.59 for adenomas >1 cm or villous adenomas) and with 81 mg of aspirin daily (RR, 0.81). Interestingly, standard-dose aspirin (325 mg) had no effect in this study (RR, 0.96) but was found to decrease the RR to 0.65 in a study of aspirin chemoprophylaxis in patients with previous colon cancer157 with similar duration of follow-up (1 year). Other treatments, such as vitamin supplementation or decreased intake of red meat/increased intake of fish or poultry, seem to be much less effective in reducing RR (RR, 0.88–1.00) when surveyed in large