Evaluation of a clinical risk index for advanced colorectal neoplasia among a North American population of screening age

Background: A clinical risk index employing age, sex, family history of colorectal cancer (CRC), smoking history and body mass index (BMI) may be useful for prioritizing screening with colonoscopy. The aim of this study was to conduct an external evaluation of a previously published risk index for advanced neoplasia (AN) in a large, well-characterized cohort. Methods: Five thousand one hundred thirty-seven asymptomatic persons aged 50 to 74 (54.9 % women) with a mean age (SD) of 58.3 (6.2) years were recruited for the study from a teaching hospital and colorectal cancer screening centre between 2003 and 2011. All participants underwent a complete screening colonoscopy and removal of all polyps. AN was defined as cancer or a tubular adenoma, traditional serrated adenoma (TSA), or sessile serrated adenoma (SSA) with villous characteristics (>= 25% villous component), and/or high-grade dysplasia and/or diameter >= 10 mm. Risk scores for each participant were summed to derive an overall score (0-8). The c-statistic was used to measure discriminating ability of the risk index. Results: The prevalence of AN in the study cohort was 6.8 %. The likelihood of detecting AN increased from 3.6 to 13.1 % for those with a risk score of 1 to 6 respectively. The c-statistic for the multivariable logistic model in our cohort was 0.64 (95 % CI = 0.61-067) indicating modest overlap between risk scores. Conclusions: The risk index for AN using age, sex, family history, smoking history and BMI was found to be of limited discriminating ability upon external validation. The index requires further refinement to better predict AN in average risk persons of screening age.