论文部分内容阅读
目的 探索和建立一种适用于临床与科学研究的SARS确诊病例的计算机分型方法。 方法 ①分析病案首页变量与患者病情的相关性 ,采用逐步判别分析 ,确定与病情相关的变量 ,构建函数判别方程 ,建立病例分型计算机模型 ;②对比分析计算机分型与临床分型的患者基本情况、临床表现、实验室结果、预后、住院天数、医疗费用等情况。 结果 ①临床分型 :本组 6 80例中 ,普通型 6 4 2例 (94 4 1% ) ,重症型 38例 (5 5 9% ) ;计算机分型 :A型 4 36例 (6 4 12 % ) ,C型 2 37例 (34 85 % ) ,D型 7例 (1 0 3% ) ,无B型病例。②临床分型与计算机分型比较 ,其各组患者之间的一般情况 (年龄、治愈率、死亡率和平均住院日 )、实验室部分指标 (总蛋白、白蛋白和尿素氮 )与医疗费用均有统计学差异。③两种分型方法对照分析 :在计算机分型为A型 (一般普通型 )的患者中 ,99 77%为临床分型的普通型 ;在临床分型为重症型的患者中 ,97 36 %为计算机分型的CD型。 结论 计算机分型与临床分型方法之间有较好同源性 ,其绝大多数指标的组间差异基本相似 ;两种分型方法在普通型与危重型的结果有统计学上的差异 ,计算机分型中危重型患者的例数比临床分型高 6 4 2倍 ,与临床分型比较 ,计算机分型的各组间有差异大、且危重型病例?
Objective To explore and establish a computer-based method for the diagnosis of SARS in clinical and scientific research. Methods ①To analyze the correlation between the first page of the medical record and the patient’s condition, and to determine the variable related to the disease by stepwise discriminant analysis, to construct the discriminant function equation and to establish the computer model of the parting. ②Comparison of the basic type Conditions, clinical manifestations, laboratory results, prognosis, days of hospitalization, medical expenses and so on. Results ① Clinical classification: Among the 80 cases in this group, 642 (94.41%) were common type and 38 (55.9%) were severe type. The computerized type A was 436 377 cases (34 85%) of type C and 7 cases (10%) of type D without type B cases. ②Comparison of clinical classification and computer classification, the general situation (age, cure rate, mortality and average length of stay) among the patients in each group, some indicators of laboratory (total protein, albumin and urea nitrogen) and medical expenses There are statistical differences. ③ Comparison of the two types of typing analysis: Among the patients with computer type A (general type), 99.77% of them were common type of clinical classification; among the patients with clinical classification of severe type, 97.36% CD type for computer type. Conclusion There is a good homology between computer typing and clinical typing methods, and most of the differences between the two groups are similar. The results of two typing methods are statistically different between common type and critical type, The number of critically ill patients in computer typing was 644 times higher than that of clinical types. Compared with clinical classification, there were significant differences between the groups in computer typing and in critically ill cases.