Confusing invitation with completion is a serious epidemiological error
When a major European trial appeared to cast doubt on colonoscopy's value, an Argentine gastroenterologist stepped forward to defend not the procedure itself, but the integrity of how we read the evidence. Luis Caro's intervention reminds us that in medicine, as in life, the difference between what we offer and what people accept is not a footnote—it is the whole story. A tool that works only when used cannot be judged by the silence of those who never arrived.
- The NordICC trial's headline figures—a modest 19% reduction in colorectal cancer incidence—spread quickly, feeding public doubt about a procedure that screens millions each year.
- Gastroenterologist Luis Caro identified the fault line: the study measured whether people accepted an invitation, not whether the colonoscopy itself worked—a conflation he calls a serious epidemiological error.
- The corrected figure tells a starkly different story: among those who actually completed the procedure, colorectal cancer incidence fell by approximately 45%, more than double the headline number.
- Colonoscopy's dual power—detecting existing cancers and preventing future ones by removing precancerous polyps in the same visit—sets it apart from every other screening method available.
- The real crisis the data exposes is not the tool's efficacy but adherence: with only 42% of invited participants completing the procedure, the urgent challenge is building prevention programs that actually reach people.
A landmark European trial tracking more than 84,000 people over thirteen years recently published findings that some interpreted as undermining the case for colonoscopy. Luis Caro, gastroenterologist and president of Argentina's Gedyt Foundation, argues that those readings confuse two entirely different questions—and that the confusion carries real public health consequences.
The NordICC study reported a 19 percent relative reduction in colorectal cancer incidence among people invited to undergo colonoscopy. But only 42 percent of those invited actually completed the procedure. When researchers adjusted for actual participation, the reduction climbed to roughly 45 percent. Caro's point is methodological and precise: measuring whether an invitation works is not the same as measuring whether the procedure works. One tests human behavior; the other tests medicine.
The mortality data reinforces the nuance. Colorectal cancer deaths were low in both groups—reflecting improved treatments over two decades—and only 16 percent of cancers in the screening group were caught through early detection. Most tumors in both groups were found after symptoms appeared. For Caro, this underscores colonoscopy's most distinctive feature: it does not merely detect cancer, it prevents it, by identifying and removing precancerous polyps before they become malignant. No other screening method accomplishes both in a single procedure.
The debate Caro wants to have is not about whether colonoscopy works. It is about why so many people decline to use it. Colorectal cancer is rising globally, including among younger populations, and the gap between what prevention programs offer and what patients accept remains the central problem. Simplifying a complex trial into a headline that discourages screening, he warns, is not just journalistically careless—it is medically irresponsible.
A major study on colorectal cancer screening has sparked a sharp disagreement about what the data actually shows. The NordICC trial, which followed more than 84,000 people across Europe for thirteen years, was published recently in The Lancet with findings that some have interpreted as questioning the value of colonoscopy. But Luis Caro, a gastroenterologist and president of the Gedyt Foundation in Argentina, says those simplified readings miss a crucial distinction—and risk misleading the public about a procedure that remains one of medicine's most important cancer prevention tools.
The study itself reported that inviting people to undergo colonoscopy was associated with a 19 percent relative reduction in colorectal cancer incidence. That sounds modest. But here is the number that changes everything: only 42 percent of the people invited actually completed the procedure. When researchers calculated what the reduction would have been if everyone invited had actually shown up and had the colonoscopy done, the figure jumped to approximately 45 percent. This is not a small difference. It is the difference between a procedure that works and a procedure that appears not to work—when the real problem is that most people never had it in the first place.
Caro's objection is precise and methodological. "Confusing an invitation to get a colonoscopy with actually getting one is a serious epidemiological error," he said in an interview. The study measured a population-level screening strategy, not the efficacy of the procedure itself in people who underwent it. When you measure whether telling people to do something works, you are not measuring whether the thing itself works. You are measuring whether people listen. These are not the same question.
The mortality data tells a similar story. In the screening group, colorectal cancer death occurred in 0.41 percent of participants. In the unscreened group, it was 0.47 percent—a difference of 0.06 percentage points. Both rates were low, reflecting advances in cancer treatment over the past two decades. But the study also found that only 16 percent of cancers detected in the screening group were caught by colonoscopy or other early detection methods. Most were diagnosed after symptoms appeared in both groups. This suggests that the real value of colonoscopy lies not in catching every cancer early, but in preventing cancer altogether by removing precancerous polyps before they become tumors.
Caro emphasized this distinction. "Detecting more cancers during a screening program is not a failure of the method. It is precisely one of its goals: finding hidden lesions and early-stage cancer before they advance." Colonoscopy does two things simultaneously. It detects cancer, yes. But it also prevents it by identifying and removing adenomas and other precancerous growths in the same procedure. No other screening method does both.
The real debate, Caro argues, is not about whether colonoscopy works. It is about how to get more people to actually use it. Colorectal cancer continues to rise globally, including among younger people. The challenge is not the tool. It is adherence—convincing people to show up, to complete the procedure, to participate in prevention programs that could save their lives. When a study shows that a screening strategy has limited impact because most people decline to participate, the lesson is not that the screening is useless. The lesson is that we need better ways to encourage participation and organize prevention programs that actually reach people.
Caro warned against reducing a complex scientific question to a simple headline. "It may be journalistically attractive to simplify this into a catchy title, but it is medically irresponsible," he said. The colonoscope itself—a long, flexible tube with a camera at its tip—remains a remarkable instrument. It allows doctors to see directly inside the colon, to spot inflammation, bleeding, polyps, and abnormalities that other methods might miss. And crucially, it allows them to remove those polyps immediately, in the same visit, preventing the progression to malignancy. In a world where colorectal cancer is one of the leading causes of cancer death, and where prevention is always preferable to treatment, that capability matters enormously.
Citações Notáveis
Confusing an invitation to get a colonoscopy with actually getting one is a serious epidemiological error.— Luis Caro, gastroenterologist and president of the Gedyt Foundation
Colonoscopy is the only method that simultaneously detects and prevents cancer by removing precancerous adenomas and lesions.— Luis Caro
A Conversa do Hearth Outra perspectiva sobre a história
So the NordICC study is being read as bad news for colonoscopy. But you're saying the study itself doesn't actually say that?
Exactly. The study measured what happens when you invite a population to screening. It found that 42 percent accepted. That's a finding about human behavior and program design, not about colonoscopy itself. When you look only at the people who actually had the procedure, the cancer reduction jumps from 19 percent to 45 percent.
Why does that distinction matter so much? Isn't the real-world outcome what counts—the fact that most people don't show up?
Both matter, but they're different questions. Yes, real-world adherence is a problem we need to solve. But if we tell people colonoscopy doesn't work based on a study that shows most people didn't do it, we're solving the wrong problem. We're discouraging the tool instead of improving how we deliver it.
What about the mortality numbers? The study showed very small differences between screened and unscreened groups.
Both groups had low mortality—0.41 percent versus 0.47 percent. That's because cancer treatment has improved dramatically. But the study also showed that only 16 percent of detected cancers came from screening. Most appeared as symptoms. That doesn't mean screening failed. It means the real value is in prevention—removing polyps before they ever become cancer.
So you're saying colonoscopy is more about prevention than detection?
It's both, but prevention is the bigger story. You can detect cancer, sure. But the unique thing colonoscopy does is find precancerous growths and remove them in the same procedure. That's not detection. That's prevention. No other screening method does that.
What worries you most about how this study is being reported?
That people will hear "colonoscopy doesn't work" and stop getting screened. Colorectal cancer is rising, especially in younger people. We don't need to discourage the tool. We need to figure out how to get more people to actually use it.