RESUMEN
BACKGROUND & AIMS: Withdrawal times and adenoma detection rates are widely used quality indicators for screening colonoscopy. More rapid withdrawal times have been associated with undetected adenomas, which can increase risk for interval colorectal cancer. METHODS: We analyzed records of 76,810 screening colonoscopies performed between 2004 and 2009, by 51 gastroenterologists practicing in Minneapolis and St Paul, MN. Colonoscopy records were linked electronically to the state cancer registry (Minnesota Cancer Surveillance System) to identify incident interval cancers that were diagnosed within 5.5 years after the screening examination. RESULTS: The physicians' mean ± SD withdrawal time was 8.6 ± 1.7 minutes and adenoma detection rates were 25% ± 9%. Longer mean withdrawal times were associated with higher adenoma detection rates (3.6% per minute; 95% confidence interval: 2.4% to 4.8%; P < .0001). We identified 78 cancers during 410,687 person-years of follow-up, for an annual rate of 0.19/1000 person-years. Physicians' mean annual withdrawal times were inversely associated with cancer incidence (P < .0001). Compared with withdrawal times ≥6 minutes, the adjusted incidence rate ratio for withdrawal times of <6 minutes was 2.3 (95% confidence interval: 1.5-3.4; P < .0001). CONCLUSIONS: Shorter mean annual withdrawal times during screening colonoscopies were independently associated with lower adenoma detection rates and increased risk of interval colorectal cancer.