For how we live.
EPoS II randomised 10,799 people with high-risk adenomas to a first surveillance colonoscopy at five years or three. At an interim analysis the colorectal-cancer rates were near-identical.
Inviting people to a single colonoscopy cut 10-year colorectal cancer risk from 1.20% to 0.98%. The reduction in cancer deaths did not reach significance — and only 42% of those invited turned up.
A Polish study found endoscopists' unaided polyp-detection fell after AI arrived. A Norwegian trial that tracked doctors before, during and after AI use found no such deskilling.
A Danish stepped-wedge trial gave endoscopists automated feedback on their technique after every procedure. Adenoma detection rose from 43.4% to 48.6% — a different tool from real-time polyp AI.
SCREESCO randomised Swedish 60-year-olds to colonoscopy, stool testing, or nothing. After a median 4.8 years, overall cancer incidence barely moved; the stage distribution did.
The EAGLE trial ran colonoscopy AI off-site over a network and aimed it at the lesions that matter. It reports a threefold gain in serrated lesions — in a field whose main US guideline recommends nothing.