节点文献

颅脑损伤预后预测模型建立的方法探讨

Model construction method of predicting outcome after traumatic brain injury

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 时俊新张静蒋春舫王增珍

【Author】 SHI Jun-xin1,ZHANG Jing1,JIANG Chun-fang2,WANG Zeng-zhen1.1.School of Public Health,Tongji Medical College,Huazhong University of Science and Technology,Wuhan 430030,China;2.Department of Emergency,Wuhan Xiehe Hospital,Tongji Medical College,Huazhong University of Science and Technology,Wuhan 430022,China

【机构】 华中科技大学同济医学院公共卫生学院华中科技大学同济医学院伤害控制研究中心华中科技大学同济医学院附属协和医院急诊外科

【摘要】 目的回顾道路交通事故致颅脑损伤(traumatic brain injury,TBI)的病例,筛选预后影响因素,并对建立预后预测模型的可能性进行探讨。方法资料来源于1 294例因道路交通事故致TBI住院病历,用Logistic回归对可能的预测因子进行筛选,并对估计的Logistic回归模型进行评价。结果①在病情相同的条件下,年龄越大预后不良的可能性越大。年龄大于55岁的患者,预后不良的优势比是25岁以下患者的3.7倍。性别、健康状况、交通事故类型以及交通事故发生至入院抢救的间隔时间以及医疗费用负担方式等,对临床转归未发现有明显影响。②本研究得到的Logistic回归模型,回代判别ROC曲线下面积为0.83,以及拟合优度检验结果均显示模型拟合良好。结论病情严重程度和年龄是最为重要的两个预测因子,本研究得到的Logistic回归模型具有一定的临床应用价值。

【Abstract】 Objective To find factors of great potential to the outcome of traumatic brain injury(TBI) from road traffic accidents,and explore the possibility of constructing a Logistic model to predict the outcome of TBI.Methods Data were collected from medical records of 1 294 hospitalized patients of TBI resulted from road traffic accidents.Multiple Logistic regression model was used to screen independent variables statistically significant to prognosis.The fitted model were evaluated as respect to global null hypothesis,and goodness-of-fit,with likelihood ratio test and Hosmer and Lemeshow Goodness-of-Fit Test,and AUC respectively.Finally,by computing a reduced-bias estimate of the predicted probability,the predicting power of the fitted model was checked.Results After controlling for injury severity,age is the most influential predictor of bad outcomes.The probability of bad outcome(death or not improved) for elder patients(age ≥55) was 3.7 folds of those youths patients(age<25).If the injury were diagnosed as severe brain damage at admission,the OR of bad outcome would be 15 times higher than those at "general condition".We did not find significant effect of gender,general health level before injury,the type of the accidents,the time interval between accident and admission,and who will pay the hospital bill.Hosmer and Lemeshow Goodness-of-Fit Test,showed that the model fit the data well.The area under ROC was 0.83,so the power of prediction of the model was good.Conclusions Age and injury severity were the strongest predictors.The model with age and injury severity as independent variables showed moderate predicting power.It has the potential to be clinically useful.

  • 【文献出处】 中华疾病控制杂志 ,Chinese Journal of Disease Control & Prevention , 编辑部邮箱 ,2010年10期
  • 【分类号】R651.1
  • 【被引频次】2
  • 【下载频次】149
节点文献中: 

本文链接的文献网络图示:

本文的引文网络