Background and study aims National societies recommend colorectal cancer (CRC) screening 10 years after a normal ("negative") colonoscopy in low-risk individuals. We studied the impact of a 10-year repeat colonoscopy on the risk of early incident CRC.
Patients and methods We used health administrative data from Ontario, Canada, to conduct a population-based retrospective cohort study in 50 - 74-year-old individuals at low-to-moderate risk of CRC who had a negative colonoscopy between 1996 and 2001. We approximated exposure to repeat colonoscopy using an 8 - 12-year window. We excluded individuals who underwent lower endoscopy or colectomy, developed CRC, or were lost to follow-up between the baseline and repeat colonoscopies. We matched exposed individuals 1:1 to individuals who did not undergo lower endoscopy within 12 years for age, sex, and calendar year of baseline colonoscopy, and followed matched pairs for incident CRC. The primary analysis was multivariable hazards regression, adjusting for competing risks.
Results A total of 13 350 matched pairs were observed for a median of 4.5 years (interquartile range 3.2 - 5.9 years). The cumulative probability of CRC following the matching date was 0.70 % (95 % confidence interval [CI] 0.42 % - 1.11 %) in individuals who underwent repeat colonoscopy and 0.77 % (95 %CI 0.48 % - 1.2 %) in individuals who did not undergo repeat colonoscopy. The adjusted hazard ratio for CRC was 0.91 (95 %CI 0.68 - 1.22).
Conclusions We did not find an association between a second colonoscopy performed 10 years after a negative colonoscopy and early incident CRC. Our findings support the need for further studies on the utility of 10-year re-screening with colonoscopy in this setting.