摘要
目的分析精神分裂症(SCZ)患者出现自杀的危险因素,建立预测模型指导临床早期识别高危患者,并验证模型预测价值。方法选取143例SCZ患者,按照是否出现自杀行为将其分为自杀组和非自杀组。比较两组临床资料和实验室资料,采用多因素Logistic回归分析患者自杀风险影响因素,并构建风险列线图模型。结果本研究SCZ患者自杀发生率为31.47%;两组受教育年限、吸烟与饮酒史、心境障碍家族史、家庭居住地、家庭关系、病前性格、匹兹堡睡眠质量指数(PSQI)评分、领悟社会支持量表(PSSS)评分、汉密尔顿焦虑量表(HAMA)评分、汉密尔顿抑郁量表(HAMD)评分、促甲状腺激素(TSH)、游离三碘甲状腺原氨酸(FT3)、游离四碘甲状腺原氨酸(FT4)及中性粒细胞/淋巴细胞比值(NLR)比较,差异均有统计学意义(P<0.05)。回归分析结果显示,PSQI评分、HAMD评分、TSH及FT4为SCZ患者出现自杀独立危险因素(P<0.05),受教育年限、PSSS评分及NLR为其保护因素(P<0.05);SCZ患者出现自杀列线图模型的预测值与实测值基本一致,C-index指数为0.936(95%CI:0.894~0.971),列线图模型的曲线下面积(AUC)为0.936,高风险阈值在1%~76%范围内时,运用本列线图来预测SCZ患者出现自杀风险具有较高的净获益值;经验证得出该列线图模型敏感度为91.43%,特异度为91.18%,准确率为91.26%。结论受教育年限短、PSQI评分高、PSSS评分低、HAMD评分高、TSH与FT4水平过高、NLR过低均为SCZ患者出现自杀独立危险因素;基于独立危险因素构建的列线图模型有助于预测SCZ患者自杀发生风险。
Objective To analyze the risk factors of suicide in patients with schizophrenia(SCZ),and to establish a predictive model to early identify high-risk patients,and verify the predictive value of the model.Methods 143 patients with SCZ were selected and divided into suicide group and non-suicide group according to whether they had suicidal behaviors.The clinical data and laboratory data of the two groups were compared,and the influencing factors of suicide risk were analyzed by multi-factor logistic regression analysis,and the risk nomogram model was constructed.Results The suicide rate of SCZ patients was 31.47%.There existed significant differences in educational levels,smoking and drinking histories,family histories of mood disorders,family residence places,family relationships,preillness personalities,scores of Pittsburgh Sleep Quality Index(PSQI),Perceived Social Support Scale(PSSS),Hamilton Anxiety Scale(HAMA),Hamilton Depression Scale(HAMD),levels of thyroid stimulating hormone(TSH),free triiodothyronine(FT3),free tetraiodothyronine(FT4)and neutrophil/lymphocyte ratio(NLR)between two groups(P<0.05).The results of regression analysis showed that PSQI score,HAMD score,TSH and FT4 were independent risk factors for suicide in SCZ patients(P<0.05),while educational level,PSSS score and NLR were protective factors(P<0.05).The predicted value of the suicide nomogram model in patients with SCZ was basically consistent with the measured value,and the C-index was 0.936(95%CI:0.894~0.971).When the area under the curve(AUC)of the nomogram model was 0.936 and the high risk threshold was in the range of 1%~76%,the use of this nomogram to predict the risk of suicide in SCZ patients had a high net benefit value.The results showed that the sensitivity,specificity and accuracy of the model were 91.43%,91.18%and 91.26%respectively.Conclusion Low educational level,high PSQI score,low PSSS score,high HAMD score,high TSH and FT4 levels,and low NLR are independent risk factors for suicide in SCZ patients.A nomogram model based on independent risk factors can help to predict the risk of suicide in patients with SCZ.
作者
王晓明
高丽梅
时丽娟
WANG Xiaoming;GAO Limei;SHI Lijuan(The Eighth Hospital of Shijiazhuang,Shijiazhuang 050000,China)
出处
《精神医学杂志》
2023年第2期123-127,共5页
Journal of Psychiatry
基金
河北省中医药管理局中医药类科研计划项目(编号:2022197)。
