To estimate the potential cost-effectiveness of colorectal cancer chemoprevention with cyclooxygenase-2–specific inhibitors (COX-2 inhibitors).

Using a decision analytic Markov model, we estimated the discounted cost per life-year saved for three strategies: a COX-2 inhibitor alone; as an adjunct to colonoscopy every 10 years in persons at average risk of colorectal cancer; and as an adjunct to colonoscopy every 5 years in persons with first-degree relatives who had colorectal cancer.

In the base case, the incremental cost per life-year saved with a COX-2 inhibitor alone compared with no screening was $233,300 in persons at average risk of colorectal cancer and $56,700 in persons with 2 first-degree relatives who had the disease. Chemoprevention was both less effective and more costly than screening. The incremental cost per life-year saved with a COX-2 inhibitor as an adjunct to screening was $823,800 in persons at average risk and $404,700 in persons with 2 first-degree relatives who had colorectal cancer. Combining a COX-2 inhibitor with less frequent screening was not as cost-effective as screening at currently recommended intervals. Cost-effectiveness estimates were highly sensitive to the cost of COX-2 inhibitors and their effect on the risk of cancer.

Chemoprevention of colorectal cancer with COX-2 inhibitors is likely to incur substantially higher costs per life-year saved than are currently recommended screening strategies. COX-2 inhibitor use as an adjunct to screening may increase life expectancy, although at prohibitive costs, and is unlikely to result in less frequent screening.