Is colonoscopy the gold standard for colorectal cancer screening?

Gut Check is a periodic look at health claims made by studies, newsmakers, or conventional wisdom. We ask: Should you believe this?

The Claim: Colonoscopy is far and away the best way to screen for colorectal cancer.

The Background: This is National Colorectal Cancer Awareness Month, and it’s already seen a livestreamed colonoscopy from the Mayo Clinic, the modern version of Katie Couric’s televised colonoscopy in 2000. Back then, the “Today” show said the procedure was “considered the most effective test for colon cancer.” Many screening messages, not to mention physicians, today implicitly send the same message: getting a colonoscopy every 10 years from ages 50 to 74 is the best way to prevent the cancer.

In contrast, a fecal test to screen for colon cancer has been largely sidelined. Guidelines released in 2008 by the American Cancer Society and other medical groups warned that fecal blood tests such as FIT (the fecal immunochemical test) — which detects invisible blood in a stool sample, a sign of either benign polyps or cancer — cannot prevent colorectal cancer. They can only detect cancer, the 2008 statement said, because if FIT finds something suspicious, say a precancerous polyp, only a colonoscopy can remove it.

That’s partly why the majority of people after turning 50 are told to schedule a colonoscopy, not FIT. “Physicians seem to think if they don’t recommend colonoscopy, they’re recommending a primitive test,” said Dr. Theodore Levin, a gastroenterologist at Kaiser Permanente in northern California.

First Take: The assertion that colonoscopy reduces your chances of dying from colorectal cancer comes from indirect evidence. Trials of flexible sigmoidoscopy, which explores only the lower part of the colon, have shown it does reduce the chances of dying from the disease. If sigmoidoscopy reduces mortality, the thinking goes, so must colonoscopy, which reaches more of the colon. Those data plus mathematical models “show we can prevent the same number of colon cancer deaths with colonoscopy every 10 years and FIT annually,” said Dr. Richard Wender, chair of the National Colorectal Cancer Roundtable, and chief cancer control officer of the American Cancer Society.

However, no randomized controlled trial — the most rigorous kind of study — has shown that colonoscopy reduces colorectal cancer mortality. (Three such trials — in Spain, Sweden, and the United States — are underway.) Nor has a rigorous trial compared colonoscopies performed every 10 years with annual FITs to see which is better at preventing deaths from colorectal cancer.

The closest researchers have come is a 2013 study that followed 88,902 health care professionals for 22 years. Those who had screening sigmoidoscopy had a 41 percent lower mortality from colorectal cancer, while the colonoscopy group had a 68 percent lower mortality rate, than the no-‘oscopy group. But because participants were not randomly assigned to ‘oscopy or no-‘oscopy, it could not rule out the possibility that some other difference between the two groups (healthy habits, being plugged into the health care system, etc.) explained their cancer differences.

Second Take: Expert groups have since disavowed the 2008 statement implicitly denigrating FIT. That reflects better science.

A 2013 randomized trial of annual fecal blood screening, following participants for 30 years, found that the test reduced deaths from colorectal cancer by 32 percent. That was comparable to the 26 percent cancer-mortality reduction with sigmoidoscopy found in a 2012 randomized trial. And FIT is considered even better at detecting early-stage colorectal cancers, and, thus, preventing deaths from the disease, than the older fecal test in the 2013 study.

Based on those and other studies, the US Preventive Services Task Force, a group of medical experts who advise the government, gives equally high recommendations to colonoscopy every 10 years and FIT every year for people aged 50 to 74 at average risk of colorectal cancer. The US Centers for Disease Control and Prevention also gives FIT and colonoscopy equal billing. “We don’t need to do colonoscopy as a primary screening,” Wender said. “We have equally excellent tests.”

More radically, last month a Canadian task force recommended against colonoscopy for routine screening of average-risk adults, and instead endorsed FIT and flexible sigmoidoscopy. The lack of randomized clinical trials of colonoscopy was a big reason it got an emphatic thumbs down. Another was that colonoscopy has serious risks, such as colon perforation which, depending on the gastroenterologist, can occur in 1 out of every 1,750 procedures.

It’s not clear why physicians recommend colonoscopy more than FIT. “There is a big disconnect between thought leaders and experts, on the one hand, and the general public and primary care physicians on the other,” said Wender. “It’s hard to convince them colonoscopy isn’t the best screening.”

One reason might be that corralling people into annual FIT requires more organization and follow-up by health care providers than does every-decade colonoscopy, Kaiser’s Levin said. It’s also a way to offload work to gastroenterologists, who perform colonoscopies, whereas primary care physicians do FIT tests. “And with colonoscopy you don’t have to think about it for 10 years,” Wender said. “But we’re not going to get to 80 percent” — the percent of  50- to 74-year-olds the Roundtable has set as a screening goal by 2018 —”if we keep emphasizing colonoscopy.”

Currently, only 66 percent of the target age groups get screenings for colon cancer, but Kaiser blew past 80 percent by emphasizing FIT. It sends reminders and FIT kits to its 50-to-74-year-old patients, Levin said, with follow-up calls and coordination with electronic health records. “The best screening test is the one that gets done,” he said.

The Takeaway: Colonoscopy is not better than other tests at reducing your chance of dying of colorectal cancer, but the belief that it is superior keeps many people from getting any form of colorectal screening.