个体化预测精神分裂症患者合并代谢综合征的风险Nomogram预测模型的构建及验证  

Construction and validation of a nomogram model for individualized prediction of the metabolic syndrome comorbidity risk in patients with schizophrenia

作　　者：曹娟 蒋光明[1] 苏苗苗 郭颍龙 CAO Juan;JIANG Guangming;SU Miaomiao(The Third People’s Hospital of Fuyang City,Fuyang 236000,China)

机构地区：[1]阜阳市第三人民医院,安徽阜阳236000

出　　处：《精神医学杂志》2023年第5期488-492,共5页Journal of Psychiatry

摘　　要：目的构建个体化预测精神分裂症患者合并代谢综合征(MS)风险的Nomogram模型,并验证模型的预测效能。方法选取209例精神分裂症患者作为研究对象,根据是否发生MS将患者分为MS组(n=61)和非MS组(n=148),收集患者临床资料,采用单因素和多因素Logistic回归分析精神分裂症患者合并MS的危险因素,并建立Nomogram模型。结果209例精神分裂症患者中MS的发生率为29.2%。单因素分析结果显示,两组病程、吸烟史、AAs用药时长、HDL-C、空腹血糖和高尿酸血症比较,差异均有统计学意义(P<0.01)。回归分析结果显示,病程延长、吸烟史、AAs用药时长≥12个月、HDL-C<1.0 mmol/L、空腹血糖≥6.1 mmol/L和高尿酸血症是精神分裂症患者合并MS的独立危险因素(P<0.01)。Nomogram模型验证结果显示,C-index为0.852(95%CI:0.818~0.886),校准曲线趋近于理想曲线,ROC曲线的AUC为0.850(95%CI:0.820~0.880),模型阈值>0.05时,模型提供了附加的临床收益。结论病程延长、吸烟史、AAs用药时长≥12月、HDL-C<1.0 mmol/L、空腹血糖≥6.1 mmol/L和高尿酸血症会增加精神分裂症患者合并MS的风险,本研究建立的Nomogram模型对精神分裂症患者合并MS风险具有良好的预测效能。Objective To construct a nomogram model for individualized prediction of the metabolic syndrome(MS)comorbidity risk in patients with schizophrenia,and to validate the predictive efficacy of the model.Methods A total of 209 schizophrenics were selected and divided into MS group(n=61)and non-MS group(n=148)according to the occurrence of MS,and their clinical data were collected and analyzed through unifactorial and multifactorial logistic regression to explore the risk factors for MS comorbidity in schizophrenics and construct a nomogram model.Results The incidence rate of MS among 209 schizophrenics was 29.2%.The results of univariate analysis showed that there existed significant differences in disease duration,smoking history,duration of AAs medication,HDL-C,fasting blood glucose and hyperuricemia between two groups(P<0.01).The results of multifactorial analysis showed that prolonged disease duration,smoking history,AAs medication duration≥12 months,HDL-C<1.0 mmol/L,fasting blood glucose≥6.1 mmol/L and hyperuricemia were the independent risk factors for MS comorbidity in schizophrenics(P<0.01).The results of the validation of the nomogram model showed that the C-index was 0.852(95%CI:0.818~0.886),and the calibration curve converged to the ideal curve,with the AUC of the ROC curve as 0.850(95%CI:0.820~0.880);the model provided additional clinical benefit at a modeling threshold>0.05.Conclusion Prolonged disease duration,smoking history,AAs medication duration≥12 months,HDL-C<1.0 mmol/L,fasting blood glucose≥6.1 mmol/L and hyperuricemia increase the risk of MS comorbidity in schizophrenics,and the nomogram model developed in the present study has a good predictive efficacy for the MS comorbidity risk in schizophrenics.

关 键 词：精神分裂症 代谢综合征 危险因素 NOMOGRAM 

分 类 号：R749.3[医药卫生—神经病学与精神病学]