Researchers at Portland, Oregon’s Veterans Affairs Medical Center have concluded that testing for the presence of blood in stools samples is not an accurate means by which to screen for colon cancer. Dr. David Lieberman, who is chief of gastroenterology at the center, warns that, 'Colon cancer screening does save lives. But it has to be done properly.' He calls this testing “inadequate” and “just about worthless.” The study is being published today in the Annals of Internal Medicine.

His group studied 2,665 people who underwent a digital rectal exam followed by a fecal occult blood test, the test that is commonly administered as a screening measure. Then all of the volunteers had a colonoscopy. Of the group, 284 were found to have either precancerous growths or cancer. Only 14, or 4.9 percent, of those who had cancer also had blood show up in the stool samples.

The study also required volunteers to use home tests for the same purpose. These tests did a little better, with 68 of the participants who ultimately were shown to have cancer, or 23.9 percent, reporting blood in the stool.

Lieberman’s group deducted that tests done in the doctor’s office only found 9.5 percent of the cancers discovered on the colonoscopy as opposed to 42.9 percent from the tests taken at home.

Colon cancer is second only to lung cancer in the death rate. The American Cancer Society reports that 146,940 cases of colorectal cancer were diagnosed in the United States in 2004, and 56,730 people died of the disease. This despite increased awareness of the disease and continuing publicity regarding the need for testing.

Doctors recommend that everyone should have a screening done at the age of 50, and then some form of testing should be done annually.