Even though current guidelines advocate colorectal cancer screenings for those with severe illnesses, they may bring little benefit and may actually pose harm, according to a recent study by Yale School of Medicine.

The study offers a new approach for assessing who is likely to benefit from a screening so that screening recommendations can be tailored more effectively to individual patients.

First author R. Scott Braithwaite, M.D., and his colleagues developed a new method of evaluating medical screening tests like colonoscopy, called the "payoff time," which is the minimum amount of time it takes for the benefits from a test to exceed its harms (i.e., its complications and side effects). The method can also be applied to patients of any age and illness.

To estimate the payoff time for using colonoscopy to screen for colorectal cancer, the team focused on two patient groups that included 50-year-old men with HIV, and 60-year-old women with congestive heart failure.

Braithwaite said the payoff time for colorectal cancer screening was as long as five years for 50-year-old men and as long as 2.9 years for 60-year-old women. Because patients with severe congestive heart failure have a life expectancy of less than 2.9 years, they were more likely to be harmed than benefited by colorectal cancer screening, say the researchers, whereas patients with HIV have a life expectancy of greater than five years, so they were likely to benefit from colorectal cancer screening.

"This issue is only becoming increasingly important as pay-for-performance and physician 'report cards' encourage clinicians to offer screening to everyone, regardless of individual benefit," said Braithwaite, assistant professor of medicine at Yale School of Medicine and at the VA Connecticut Healthcare System. "This may have the unintended consequence of harming patients with severe illnesses."

Braithwaite added that the "cocktail" of therapies that have revolutionized HIV care have increased life expectancy so much that screening guidelines now apply to patients with HIV.

Other authors on the study included John Concato, M.D., Chung Chou Chang, Mark S. Roberts, M.D. and Amy C. Justice, M.D.

The study was supported by a grant from the National Institute of Alcohol Abuse and Alcoholism, National Institutes of Health.

Citation: Archives of Internal Medicine Vol. 167, No. 21 (November 26, 2007)