摘要
目的 :考察影响创伤结局的重要因素 ,初步筛选创伤患者结局预测模型的构成变量 ,为建立新的适合中国患者的创伤严重度评分方法进行有关评分模型的初步探讨 .方法 :利用全国 2 0 0余所医院的创伤患者病案首页数据 ,初选出包括ISS评分所得严重度在内的 9个相关的因素 ,将患者结局作为应变量拟合Logistic回归模型 .逐步回归法筛选变量 ,并估计其影响作用的大小 .结果 :解剖严重程度、有无并发症、年龄组、身份、是否接受手术、费用类型等 6个因素对患者结局的影响有统计学意义 (P <0 .0 5 ) ,被选入Logistic回归模型 .另外 ,某些因素的交互效应也一并被选入Logistic回归模型 (P <0 .0 5 ) .结论 :解剖严重度是影响创伤结局的最重要的因素 ,其次是有无并发症和年龄等因素 .
AIM: To review some important factors affecting the outcomes of trauma patients and to screen variables for the predictive model of trauma outcomes so that a new trauma scaling model more applicable to Chinese patients can be established. METHODS: The database of discharge abstracts of trauma patients from more than 200 hospitals nationwide was used and a logistic regression model was fitted with the outcomes of patients as response and other 9 factors as predictors, including the anatomic injury severity grade by ISS value. A stepwise regression method was used to select the variables and their parameters were estimated. RESULTS: Six factors, namely, anatomic injury severity grade, complication status, age group, identity, operation status and financial support status were selected into the logistic regression model (P<0.05). Several interactive effects were also selected in the model (P<0.05). CONCLUSION: The anatomic injury severity grade is the most important factor affecting the outcomes of trauma patients, followed by the complication status and age. The effects of other factors are not so obvious and should be further analyzed.
出处
《第四军医大学学报》
北大核心
2004年第18期1714-1716,共3页
Journal of the Fourth Military Medical University
基金
陕西省自然科学基金课题 (2 0 0 3G0 9)