The questionable value of colorectal cancer screening

Screening for colorectal cancer is widespread in high income countries and strongly recommended by most guidelines.123 Evidence from randomised trials shows that invitation to screening with sigmoidoscopy or guaiac faecal occult blood test reduces colorectal cancer deaths,4 but evidence evaluating the most widely used screening tests—colonoscopy and faecal immunochemical test—has been limited to observational studies, modelling, and logical reasoning.5

The first randomised trial of colonoscopy screening, the NordICC trial,6 was recently published, triggering an update to the BMJ Rapid Recommendation on colorectal cancer screening (doi:10.1136/bmj.l5515).1 Some 85 000 people aged 55-64 years were randomised to either an invitation to one-time colonoscopy or no invitation. After 10 years, invitation to screening had not reduced colorectal cancer mortality: 0.31% died of colorectal cancer in the control group compared with 0.28% in the group invited to screening (risk ratio 0.90, 95% confidence interval 0.64 to 1.16).6 While screening did decrease the risk of developing colorectal cancer (risk ratio 0.82, 0.70 to 0.93), the disappointing results for colorectal cancer mortality raise …