PHILADELPHIA — A common screening test failed to detect potentially cancerous colon growths 95 percent of the time, falsely reassuring patients and doctors, according to a new study.
Researchers found that the digital, in-office test on stool samples was not as reliable as a six-sample test given to patients to do on their own at home — although even that test detected potentially cancerous growths less than 24 percent of the time.
"What we found is that it was pretty worthless," Dr. David Lieberman, one of the study's authors, said of the in-office test. "It's a wake-up call that we shouldn't be relying on this test."
The study, published Tuesday in the Annals of Internal Medicine, was conducted at 13 Veterans Affairs medical centers and involved 2,665 patients — most of them men — who were given the at-home test and the in-office test followed by a colonoscopy.
The digital fecal occult blood test was positive in only 5 percent of patients with tumors or large, precancerous growths called polyps; the take-home test found 24 percent.
The reliability of the at-home test, however, improves if patients use it every year, said Lieberman, gastroenterogy chief at Portland VA Medical Center in Oregon.
Because polyps and tumors often bleed intermittently, the chances of detecting them increase as more samples are taken over periods of days and years, he said.
Early testing can find growths before they turn cancerous, a process that can take five to seven years, Lieberman said. The take-home test was positive for 43 percent of patients who actually had cancer, the study said.
An accompanying survey of physicians also published in the Annals of Internal Medicine found that about a third used only the office test, a fourth used only the home test, and about 40 percent used both.
The survey, funded by the National Cancer Institute and the Centers for Disease Control and Prevention, said one possible reason for using the office tests was concern that patients won't complete the home tests. The completion rate varies greatly, with the median between 40 and 50 percent.
"My guess is that many physicians may not be aware that the in-office test is such a poor test," said Marion R. Nadel, a CDC epidemiologist who worked on the survey. Nadel said doctors should be trying to find more ways to get patients to complete the home test.
But while the new research should make doctors and patients more skeptical of a negative result from an in-office digital test, it shouldn't be written off altogether, said Dr. Christine Berg, chief of the early detection research group of the cancer institute.
Doctors shouldn't miss an opportunity to give a patient an annual digital FOBT while they're in the office, she said.
"I wouldn't say it's worthless," Berg said. "I would say you shouldn't rely on it solely."
© 2013 The Associated Press. All rights reserved. This material may not be published, broadcast, rewritten or redistributed.