Importance: Colorectal cancer is a major health burden. Screening is recommended in many countries.
Objective: To estimate the effectiveness of flexible sigmoidoscopy screening on colorectal cancer incidence and mortality in a population-based trial.
Design, Setting, and Participants: Randomized clinical trial of 100 210 individuals aged 50 to 64 years, identified from the population of Oslo city and Telemark County, Norway. Screening was performed in 1999-2000 (55-64–year age group) and in 2001 (50-54–year age group), with follow-up ending December 31, 2011. Of those selected, 1415 were excluded due to prior colorectal cancer, emigration, or death, and 3 could not be traced in the population registry.
Interventions: Participants randomized to the screening group were invited to undergo screening. Within the screening group, participants were randomized 1:1 to receive once-only flexible sigmoidoscopy or combination of once-only flexible sigmoidoscopy and fecal occult blood testing (FOBT). Participants with positive screening test results (cancer, adenoma, polyp ≥10 mm, or positive FOBT) were offered colonoscopy. The control group received no intervention.
Main Outcomes and Measures: Colorectal cancer incidence and mortality.
This was a population based screening study of one time flexible sigmoidoscopy (with or without fecal occult blood testing) versus no screening. If adenomas were found, colonoscopy was offered. Everyone lived in Norway, and was between 50 and 64 years of age. They found that the screening worked. About 3.1 per 1000 people died of colon cancer over 10 years in the screening group versus 4.3 per 1000 people in the control group. The Hazard ratio for screening was 0.7. Adding fecal occult blood testing made no real difference.
And this was with only half of the invited people actually using the screening. So the use of one-time flexible sigmoidoscopy seems to work. This is evidently the fourth large randomized controlled clinical trial showing that to be the case.
Here’s where the mismatch kicks in. There’s no similar evidence for the use of colonoscopy:
Another ironic aspect of the decline of flexible sigmoidoscopy in the United States is that randomized trial–level evidence of reduced cancer-specific mortality exists for flexible sigmoidoscopy and FOBT but not for colonoscopy or the other available structural screening test, computed tomography colonograpy.
Flexible sigmoidoscopy is easier to do. It’s cheaper. So why do we do so much colonoscopy?
By far, colonoscopy has become the most commonly recommended and performed endoscopic colorectal cancer screening tool in the United States, whereas screening by sigmoidoscopy has all but vanished. Fecal occult blood testing persists unevenly—for example, in managed care organizations that actively promote it to their enrollees and in uninsured or underinsured populations without affordable access to colonoscopy. The transition to screening colonoscopy among average-risk persons in the United States accelerated in 2001, when Medicare began paying for it and many private insurers followed. Gastroenterologists began to recommend colonoscopy as the test of choice and had economic incentives to do so. As one gastroenterologist observed recently, “Colonoscopy has been really good to our specialty. It is the goose that has laid the golden egg.” In many other developed nations, FOBT remains the dominant mode of screening.
Is there ever a case when we don’t ignore evidence in order to do more invasive and more expensive testing than everyone else? Some days it feels like the answer to that is “nope”.