【摘要】：目的構(gòu)建50歲以上健康管理隊(duì)列的白內(nèi)障發(fā)病風(fēng)險(xiǎn)預(yù)測(cè)模型。方法依托山東多中心健康管理縱向觀察數(shù)據(jù)庫(kù),采用Cox比例風(fēng)險(xiǎn)回歸構(gòu)建白內(nèi)障發(fā)病風(fēng)險(xiǎn)預(yù)測(cè)模型,通過ROC曲線下面積(AUC)評(píng)價(jià)模型的預(yù)測(cè)效果,并利用十折交叉驗(yàn)證來檢驗(yàn)?zāi)Ｐ偷姆�(wěn)定性。結(jié)果隨訪期間共新發(fā)白內(nèi)障病例1 010例,發(fā)病密度為24.76‰。預(yù)測(cè)模型最終納入年齡、性別、吸煙、高黏稠血癥、鼓膜疾患、屈光不正、糖尿病、總膽固醇和收縮壓9個(gè)變量。白內(nèi)障發(fā)病風(fēng)險(xiǎn)預(yù)測(cè)模型的AUC為0.712(95%CI:0.693~0.732)。十折交叉驗(yàn)證的平均AUC為0.714。結(jié)論研究構(gòu)建的白內(nèi)障發(fā)病風(fēng)險(xiǎn)預(yù)測(cè)模型有較好的預(yù)測(cè)效果,為白內(nèi)障高危人群的早期篩查提供了依據(jù)。
[Abstract]:Objective to establish a predictive model of cataract risk in health management cohort over 50 years old. Methods based on the longitudinal observation database of multi-center health management in Shandong province, the prediction model of cataract risk was constructed by Cox proportional risk regression. The area (AUC) model under the ROC curve was used to evaluate the predictive effect of the model. The stability of the model is tested by 10% cross-validation. Results during the follow-up period, 1 010 new cataract cases, the incidence density was 24.76 鈥，

本文編號(hào)：2296316