Colorectal cancer is caused by the interaction of genetic and environment factors. Domestic and foreign scholars have attempted to develop several colorectal cancer risk prediction models, in order to identity risk factors, to screen for high risk population and evaluate the risk of developing colorectal cancer, so as to provide personalized screening protocols for individuals with different risk, and eventually reduce the incidence and mortality rate of colorectal cancer. Currently, the common colorectal cancer risk prediction models were mainly developed based on case-control study and cohort study. Models developed in European and American regions and Asia (excluding China) only include common risk factors, while Chinese models also include hereditary factors on the bases of common risk factors. However, the development and verification of each model are mainly based on local population, whether it can be applied for other population need to be determined. This article reviews the development, validation and evaluation of the risk prediction models, in order to provide a basis for developing more precise risk prediction models for colorectal cancer.
结直肠癌的发生是遗传因素和环境因素共同作用的结果。国内外学者已尝试建立多种结直肠癌风险预测模型用于识别危险因素、筛选高危人群及评估发病风险，从而为不同风险人群提供个性化的筛查方案，有效降低结直肠癌的发病率和病死率。现有的典型结直肠癌风险预测模型的建立多基于病例对照研究和队列研究。欧美地区和亚洲地区（除中国外）模型仅纳入常见风险因素；中国的模型在常见风险因素的基础上，还纳入了遗传因素。然而，各模型的建立和验证多基于本地区人群，是否适用于外部人群尚待验证。本文就各种模型的建立、验证和评价进行综述，为进一步建立精确的风险预测模型提供依据。