摘要
目的建立结直肠息肉风险预测模型并进行验证,检验预测效果。方法收集我院结肠镜及相关检查的患者900例,建立模型,转化为风险评分系统并进行验证。结果通过单因素特征分析,确定了10项与结直肠息肉发病相关的因素,包括:性别、年龄、家族肠癌/息肉病史、吸烟史、酗酒史、体重指数(BMI)、高血压、糖尿病、幽门螺杆菌感染以及便潜血。进行Logistic逐步回归分析,结果显示:性别为男性、年龄≥45岁、家族有肠癌/息肉病史、BMI提示超重或肥胖、有糖尿病、既往有或现正感染HP、便潜血阳性是结直肠腺瘤的危险因素。构建结直肠息肉风险预测评分系统(-5-18分),ROC曲线下面积为0.794,具有统计学意义的预测能力,截断值设为2分时,可以进行结直肠息肉的初步筛查。结论构建的结直肠息肉发病风险预测模型(评分系统)具有一定的预测能力,可考虑将其作为本地区人群结直肠癌初步筛查的风险评估工具。
Objective To develop and validate a risk prediction model for colorectal polyps,and to evaluate its predictive performance.Methods A total of 900 patients who underwent colonoscopy and related examinations in our hospital were included.A prediction model was constructed and transformed into a risk scoring system,which was subsequently validated.Results Univariate analysis identified 10 factors associated with the development of colorectal polyps,including:gender,age,family history of colorectal cancer or polyps,smoking history,alcohol abuse history,body mass index(BMI),hypertension,diabetes,Helicobacter pylori infection,and fecal occult blood.Stepwise Logistic regression analysis revealed that male gender,age≥45 years,family history of colorectal cancer or polyps,overweight or obesity(as indicated by BMI),diabetes,past or current H.pylori infection,and positive fecal occult blood were independent risk factors for colorectal adenomas.A risk prediction scoring system for colorectal polyps was developed,with scores ranging from-5 to 18.The area under the ROC curve was 0.794,indicating statistically significant predictive ability.Using a cutoff value of 2 points,the system can be applied for preliminary screening of colorectal polyps.Conclusion The constructed risk prediction model(scoring system)for colorectal polyps demonstrates moderate predictive performance and may be considered as a risk assessment tool for preliminary screening of colorectal polyps in the local population.
作者
陈君康
蒋晚晴
方芳芳
Chen Junkang;Jiang Wanqing;Fang Fangfang(Pujiang County Hospital of Traditional Chinese Medicine,Jinhua,Zhejiang 322200,China)
出处
《首都食品与医药》
2026年第4期60-63,共4页
Capital Food Medicine
基金
浙江省中医药科技计划项目(编号:2023ZR139)。
关键词
结直肠息肉
危险因素
中医体质
LOGISTIC回归分析
预测模型
colorectal polyps
risk factors
Traditional Chinese Medicine constitution
Logistic regression analysis
prediction model
