As a longtime Valley News reader, I am aware of the recent controversy about the use of puns in newspaper article headlines. While avoiding taking any stance on that issue, I do fear that the Valley News got it wrong when it recently (Oct. 11) headlined an article covering a New England Journal of Medicine publication on colonoscopy: “Study: Little Benefit to Colon Screenings.”
I am sensitive to the perception created by that article, as my name was used within the piece. I take no issue with that. I did co-author an editorial that accompanied the recent publication, and the quote attributed to me was accurately taken from the editorial. However, that short excerpt, in isolation, does not adequately reflect my larger takeaways from that publication, so I will summarize those here.
With regards to the scientific publication itself that was the focus of the brief report in the Valley News, the study reported results of a trial in which 84,585 individuals in Poland, Norway and Sweden were randomized to an invitation to either colonoscopy or “usual care” for colorectal cancer screening. The study was carried out in countries where screening for colorectal cancer is less common, and the investigators targeted areas within those countries where screening was less robust.
Overall, after 10 years of follow-up, those randomized to colonoscopy (relative to “usual care”) were 18% less likely to get colorectal cancer and about 10% less likely to die from it. The latter finding about mortality did not reach “statistical significance” leaving open the possibility that the observed reduction in mortality may be more the result of chance than the intervention itself.
The observed benefits of colonoscopy in this study are less than those from prior work that have estimated reductions in cancer incidence and mortality in the range of 60% or even higher. There are several potential explanations for this. Most importantly, of the 28,220 individuals randomized to colonoscopy, only 42% got the test. Screening with colonoscopy only works when it is performed and in more than half the cases that just didn’t happen. In fact, when the authors limited their analysis to those who got the test (i.e. the so called “per protocol” analysis) the benefits of colonoscopy are much more significant. Specifically, both cancer incidence and mortality were significantly reduced, with cancer mortality reduced by about 50%. Given that colorectal cancer is the second-most-common cause of cancer death in the United States (approximately 50,000 deaths per year), I would consider these reductions important and not of “little benefit.”
A second important consideration when interpreting this study is understanding the quality of those colonoscopy exams. There is outstanding evidence that high quality colonoscopy is needed if it is going to find cancer early and/or prevent subsequent cancer through the finding and removal of polyps. We know from prior publications that roughly one-third of the endoscopists performing exams in this study did not meet accepted benchmarks for detecting polyps.
A third consideration here is simply the length of the study. Finding and removing polyps is one important way that colonoscopy works to prevent subsequent cancer and cancer death. But, since polyps grow slowly, it takes many years to see that benefit. Here, only 10 years of follow-up has accrued. Further follow-up time (which is planned by the investigators) will likely result in larger benefits in those randomized to the colonoscopy.
So, what do I recommend to individuals considering screening for colorectal cancer considering the above trial results? I strongly recommend that they be screened for colorectal cancer. We know from many large, randomized, controlled trials that screening reduces the chance that you will get and die from colorectal cancer. Because of the strength of the evidence, the United States Preventive Services Task Force recommends colorectal cancer screening with a “Grade A” recommendation.
Should you specifically undergo colonoscopy for screening? Currently, a panel of tests are recommended for colorectal cancer screening, including simpler at-home stool tests (like the fecal immunochemical test or FIT) and more invasive structural tests that directly look at the colon, such as sigmoidoscopy and colonoscopy.
The US Preventive Services Task Force recommends these tests equally. So how does one choose between them? Well perhaps someday there will be direct evidence to support choosing one test over the other. In fact, I co-lead a large trial in the Department of Veterans Affairs that directly compares the two most used tests (FIT and colonoscopy). But in the interim, individuals should simply choose to undergo the test they are most likely to complete.
I recognize the difficulties in being a health care consumer these days trying to determine what is best to do for our health. There is so much information in print, electronic and social media that making educated decisions can be challenging to say the least. But in this case, I think it is critically important to get beyond the headline. When utilized, colorectal cancer screening works. Talk to your provider, choose a test, and, most importantly, follow through with it.
Douglas J. Robertson is the chief of the Section of Gastroenterology at the VA Medical Center in White River Junction and a professor of medicine at Geisel School of Medicine at Dartmouth; The Dartmouth Institute. The views expressed are those of Dr Robertson and do not necessarily represent the views of the US Department of Veterans Affairs.