Skip to main content Skip to section navigation
AdobeStock/KatyFlaty

Insights

Cancer October 14, 2026

Don’t even think of skipping that colonoscopy, Stanford Medicine expert says

By Katia Savchuk

A recent trial suggested colonoscopies may not reduce colon cancer deaths, but experts call the findings questionable: ‘It sends the wrong message … and could take us backward.’

Researchers studying the long-term effects of colonoscopies published a shocking finding in May: Based on a randomized controlled trial known as NordICC, which involved tens of thousands of Europeans with an average risk of developing colorectal cancer, an invitation to undergo a colonoscopy reduced the risk of developing the disease but not the risk of dying from it.

The authors, who published their findings in The Lancet, concluded that societies may need to rethink their approach to preventive screening for this cancer: “Benefits must be weighed against other expenditures that promote public health,” they wrote. 

Uri Ladabaum, MD, a gastroenterology professor at Stanford Medicine and director of its Gastrointestinal Cancer Prevention Program, disagrees. This trial “just doesn’t fit with everything else that we know,” he said. “Screening decreases colorectal cancer mortality.”

That’s why Ladabaum and four other experts recently published a letter in The Lancet spelling out reasons for taking NordICC’s results with a grain of salt. We spoke with him to learn more about the study’s findings and what they mean for patients.

What is NordICC, and what question is it trying to answer?

This is a very important randomized controlled trial. It includes nearly 85,000 people between the ages of 55 and 64 randomly selected from the general population in Norway, Poland and Sweden. Between 2009 and 2014, some participants were invited to undergo a screening colonoscopy, while others were not (when the study started, these countries did not have national screening programs in place). The intent was to report on what happened in 15 years.

Uri Ladabaum
Uri Ladabaum

The researchers are asking: Does an invitation to do a screening colonoscopy reduce the risk of getting colorectal cancer at the population level, and most importantly, the risk of dying from it? We have a lot of evidence suggesting that colonoscopy reduces these risks, but this has never been tested in the most rigorous way — a randomized study.

What evidence do we have that colon cancer screening works?

We have trials from the 1990s in which people agreed to be randomized to undergo stool-based testing for colorectal cancer versus doing nothing. Patients with hidden blood in the stool would then get a colonoscopy. Those studies found that screening decreases the risk of dying from colorectal cancer.

In the early 2000s came trials in which patients were randomized to get sigmoidoscopy — a short scope exam of part of the large bowel — versus doing nothing. Those provide very strong evidence that scoping people decreases the risk of developing cancer in the first place and of dying from colorectal cancer. The logical conclusion is that if sigmoidoscopy works, then colonoscopy — which goes farther into the colon — should work even better. There’s all this observational data out there that supports this.

Screening works because of two mechanisms: Taking out polyps prevents some people from developing cancer in the first place, and finding cancer at earlier stages makes it easier to cure. As this evidence accumulated, many countries decided to launch screening programs. The American Cancer Society first issued screening guidelines for colorectal cancer in 1980, and colonoscopy has become the most widely used test in this country.

What have the results of NordICC been so far, and why are they significant?

In 2022, we had the 10-year results, and those were a bit of a head-scratcher. The risk of getting cancer was 18% lower in the group invited to get a colonoscopy, with 42% actually taking up the invitation, compared with the one that wasn’t. But there didn’t seem to be a reduction in deaths from colorectal cancer.  

Then in May 2026, the 13-year results were published. With the 42% participation rate, there was a 19% reduction in the risk of developing colorectal cancer, and that is statistically significant. But there is no statistically significant reduction in the risk of dying from this cancer.

This has been a bit of a shocker. Colonoscopy did not perform as expected based on the existing literature. We would expect the reduction in risk of dying from colorectal cancer to be greater than the incidence reduction, as it has been in all previous studies.

What made you decide to publish a letter urging caution in how the results are interpreted?

What compelled us to say something was the reaction within the field. People are saying, “Oh, OK, so colonoscopy doesn’t decrease colorectal cancer mortality. Maybe it’s now only about preventing cancer and sparing people the downsides of treatment. Is screening still worth it?” Even experts are accepting this at face value. I think that’s the wrong conclusion because of specific issues in the NordICC trial — and it’s the wrong message. 

Colorectal cancer is the second-leading cause of cancer deaths worldwide. The risk of a diagnosis is increasing dramatically in low- and middle-income countries and in young people. It’s highly preventable by screening and highly curable if found early. Telling the public that colonoscopy doesn’t decrease the risk of dying from colorectal cancer is not responsible. I don’t accept that NordICC has the right evidence to lead to that conclusion, and I’m worried it could take us backward.

What problems do you see with the NordICC study?

The NordICC results are completely at odds with everything else we know. They found that, through the initial eight years of the study, mortality from colorectal cancer was numerically higher in the group invited to get screened than the control group, and that is the opposite of what the sigmoidoscopy trials found. It’s an outlier result, and there may be weird reasons why they’re finding that.

Second, they’re using cancer registries to ascertain how many people died of cancer, and those may not always be reliable. In some other studies, there’s a team of people whose job it is to decide whether somebody’s cause of death was colorectal cancer.

Finally, results vary among the countries in the trial, with Norway’s results being more or less as expected but those in Poland and Sweden being highly unexpected. There may be subtle statistical issues with lumping them together.

I also find the investigators’ framing irresponsible for a host of reasons. It ignores the rich context of all the other studies.

The way that NordICC investigators think through their findings is inadequate. They say people are dying of colorectal cancer less these days, that there have been such advances in the treatment of colorectal cancer, maybe screening can’t promise to reduce mortality. So, if getting colonoscopies and other tests is about preventing pain and suffering, not about preventing death from the disease, maybe we need to think about whether this is still a health priority. This isn’t convincing. Even within NordICC, the proportion of people diagnosed with colorectal cancer who die from it is essentially the same as in past studies.

What should patients in the U.S. take away from all this?

Putting all the evidence together, for people at average risk of colorectal cancer, screening decreases the risk of getting it and decreases the risk of dying from it even more. Follow current guidelines, which recommend regular screening starting at age 45. You can do a colonoscopy and repeat it every 10 years if results are normal, or you can follow a well-done stool-based program annually or every three years, depending on the type of stool test, then get a colonoscopy if you get an abnormal result.

Current blood tests do not yet perform as well and are not ready to replace colonoscopy or stool tests. There are enough questions about NordICC that it should not be a reason to dispute everything we know about colorectal cancer screening.

About Stanford Medicine

Stanford Medicine is an integrated academic health system comprising the Stanford School of Medicine and adult and pediatric health care delivery systems. Together, they harness the full potential of biomedicine through collaborative research, education and clinical care for patients. For more information, please visit med.stanford.edu.

Katia Savchuk

Katia Savchuk is a freelance writer.