Determinants of Shorter Surveillance Colonoscopy Intervals: A Nationwide Real-World Study in Japan

Colonoscopy can reduce colorectal cancer mortality. Although guidelines recommend surveillance intervals, clinical practice often varies according to physician discretion and patient factors. Inappropriate shortening of intervals results in unnecessary procedures and increased healthcare costs. This study aimed to examine the real-world practice of colonoscopy intervals and identify the factors that influence shorter surveillance intervals to optimize surveillance strategies.

We conducted a retrospective cohort study using a claims database of patients aged ≥ 40 years who underwent initial colonoscopy between April 2012 and December 2019. Based on the findings from the index colonoscopy, patients were categorized into four groups: diagnostic, endoscopic resection of polyps < 2 cm (ER < 2 cm), ER ≥ 2 cm, and endoscopic submucosal dissection (ESD). Colonoscopy intervals were estimated as time ratios (TRs) using an accelerated failure time model.

Among 147 390 patients (diagnostic: 97 621; ER < 2 cm: 47 303; ER ≥ 2 cm: 2190; and ESD: 276), the surveillance intervals were significantly shorter in the ER < 2 cm (TR = 0.56), ER ≥ 2 cm (TR = 0.33), and ESD (TR = 0.17) groups than in the diagnostic group (all p < 0.001). Older age, male sex, higher body mass index, nonsmoking status, regular alcohol consumption, and smaller facility size were independently associated with shorter intervals (p < 0.05).

This study revealed substantial variation in surveillance colonoscopy intervals in Japan, influenced by procedural, patient, and institutional factors. Shorter intervals observed in smaller facilities and specific subgroups may indicate inefficient use of medical resources and unequal access to care.