BACKGROUND & AIMS: We investigated adenoma and colonoscopy characteristics that are associated with recurrent colorectal neoplasia based on data from community-based surveillance practice. METHODS: We analyzed data of 2990 consecutive patients (55% male; mean age 61 years) newly diagnosed with adenomas from 1988 to 2002 at 10 hospitals throughout The Netherlands. Medical records were reviewed until December 1, 2008. We excluded patients with hereditary colorectal cancer (CRC) syndromes, a history of CRC, inflammatory bowel disease, or without surveillance data. We analyzed associations among adenoma number, size, grade of dysplasia, villous histology, and location with recurrence of advanced adenoma (AA) and nonadvanced adenoma (NAA). We performed a multivariable multinomial logistic regression analysis to estimate odds ratios (ORs) and 95% confidence intervals (CIs). RESULTS: During the surveillance period, 203 (7%) patients were diagnosed with AA and 954 (32%) patients with NAA. The remaining 1833 (61%) patients had no adenomas during a median follow-up of 48 months. Factors associated with AA during the surveillance period included baseline number of adenomas (ORs ranging from 1.6 for 2 adenomas; 95% CI: 1.1 - 2.4 to 3.3 for >= 5 adenomas; 95% CI: 1.7 - 6.6), adenoma size >= 10 mm (OR = 1.7; 95% CI: 1.2 - 2.3), villous histology (OR = 2.0; 95% CI: 1.2 - 3.2), proximal location (OR = 1.6; 95% CI: 1.2 - 2.3), insufficient bowel preparation (OR = 3.4; 95% CI: 1.6 - 7.4), and only distal colonoscopy reach (OR = 3.2; 95% CI: 1.2 - 8.5). Adenoma number had the greatest association with NAA. High-grade dysplasia was not associated with AA or NAA. CONCLUSIONS: Large size and number, villous histology, proximal location of adenomas, insufficient bowel preparation, and poor colonoscopy reach were associated with detection of AA during surveillance based on data from community-based practice. These characteristics should be used jointly to develop surveillance policies for adenoma patients.