Colonoscopy controversy highlights poor screening uptake

4 minute read

The poor efficacy of screening found in a recent study was met with surprise and criticism.

Surprisingly low results in the first RCT to measure effectiveness of colonoscopy as a population screening tool have elicited strong reactions and raised questions about conclusions drawn when only 42% of those invited to screen actually did so.

The NordicICC randomised control study, Effect of Colonoscopy Screening on Risks of Colorectal Cancer and Related Death, was published in the NEJM earlier this month. It involved almost 85,000 people in Norway, Poland and Sweden, some of whom received an invitation to have a colonoscopy.

The researchers found that colonoscopy reduced the risk of colorectal cancer by only 18% over 10 years and did not significantly reduce the risk of death from CRC. In order to prevent one case, 455 people would need to be invited to screen. Additional per-protocol analysis – a statistical method for estimating the effect had all the invitees accepted – brought the figures to 31% and 50% respectively.

But previous observational studies had estimated incidence to be reduced by 40-69% and mortality by 29-88% – much higher numbers than those reported in this research. The authors said their results suggested colonoscopy screening might not be much better than sigmoidoscopy at reducing the risk of colorectal cancer.

Commentary in the NEJM posed several reasons why effectiveness may have been underestimated – less acceptance of the procedure than in other countries like the US, not enough time elapsed to see long-term benefit, poor adenoma detection rates for individuals performing the colonoscopies, and a tendency in one of the countries, Poland, for high-risk people to accept the invitation to screen.

“It’s a great study if you’re testing how effective a rollout of a nation-wide colonoscopy screening program is going to be, but it’s not a great measure of benefit for an individual or for a clinician trying to decide whether or not to recommend their own patient for a colonoscopy,” said Professor Mark Jenkins, director of the Centre for Epidemiology & Biostatistics at the University of Melbourne.

“I think the per protocol analysis, which is the most favourable light you can put on this study, still indicates that it is not cost effective compared to our current program.”

As many have pointed out, the study’s main problem was that only 42% of the group randomised to have a colonoscopy actually had one, “so it’s a test of invitation to have a colonoscopy rather than a test of the colonoscopy itself,” the cancer epidemiologist told The Medical Republic.

That number is very similar to Australia’s National Bowel Screening Program participation rate, reported at 43.5% in 2021 by the Australian Institute of Health and Welfare. The screening rate for Aboriginal and Torres Strait Islander People is only 27% and for people from culturally and linguistically diverse background it’s as low as 25%.

The faecal immunochemical test (FIT) is sent to every Australian aged 50-74, every two years. It’s not known why more people don’t take the test, said Professor Jenkins, who is leading a program to decrease bowel cancer in Australia by increasing participation rates in the screening program.

“I’d be very surprised if you got a 42% uptake in Australia for a national colonoscopy program, given that you only get around that for the FIT home test kit, which is less involved,” he said.

Offering a one-off colonoscopy was considered as an option when the Australian program was first being investigated, said Professor Jenkins, but the FIT tests were more cost-effective at $25 compared to around $1000 for a colonoscopy. The tests detect up to 85% of cancers and only people who get a positive result, around 7% of those who test, are referred for colonoscopy.

However, the program only works if people actually take part. “If you’re not screening, you’re not getting any benefit,” Professor Jenkins pointed out. “That’s the problem.”

NEJM 2022, online 9 October

End of content

No more pages to load

Log In Register ×