Background: Interval colorectal cancer (CRC) occasionally is detected in patie nts who have recently undergone colono- scopy.Systematic evaluation of CRC detected after colonoscopy could identify w ays to improve the quality and the outcome of colonoscopy.Methods: This study ex amined cancer diagnoses in the course of the dietary Polyp Prevention Trial, a r andomized study of a dietary intervention on recurrence of adenomatous polyps. A n algorithm was developed to classify each cancer into one of 4 etiologies: (1) incomplete removal (cancer at the site of previous adenoma), (2) failed biopsy d etection (cancer in an area of suspected neoplasia with negative biopsy specimen s),(3)missed cancer (large, advanced stage cancer found at a short interval afte r colonoscopy), or (4) new cancer (sma- ll, early stage cancer after a longer time interval). Results: Of 2079 patient s,13 had cancer detected over 5810 person years of observation(PYO) (2.2 cases/1 000 PYO); 7/13 or 53.8%of patients had either a potentially “avoidable“ cancer or one detectable at an earlier time interval because of incomplete removal (4/ 13) or missed cancer (3/13). Conclusions: Interval cancer occurs despite colonos copy. Im- proved quality of colonoscopy may have reduced cancer prev- alence or resulted in earlier cancer detection in over 50%of prevalent cancer s in the dietary Polyp Prevention Trial. Background: Interval colorectal cancer (CRC) occasionally is detected in patients who have recently undergone colono- scopy. Systematic evaluation of CRC detected after colonoscopy could identify w ays to improve the quality and the outcome of colonoscopy. Methods: This study ex amined cancer diagnoses in the course of the dietary Polyp Prevention Trial, ar andomized study of a dietary intervention on recurrence of adenomatous polyps. A n algorithm was developed to classify each cancer into one of 4 etiologies: (1) incomplete removal (cancer at the site of previous adenoma, (2) failed biopsy d etection (cancer in an area of ​​suspected neoplasia with negative biopsy specimen s), (3) missed cancer (large, advanced stage cancer found at a short interval afte r colonoscopy), or (4 Results: Of 2079 patient s, 13 had cancer detected over 5810 person years of observation (PYO) (2.2 cases / 1,000 PYO); 7/13 or 53.8% of patients h ad either a potentially ”avoidable " cancer or one detectable at an earlier time interval because of incomplete removal (4/13) or missed cancer (3/13). Conclusions: Interval cancer consequen despite colonos copy. Im- proved quality of colonoscopy may have reduced cancer prev- alence or resulted in earlier cancer detection in over 50% of prevalent cancer s in the dietary Polyp Prevention Trial.