论文部分内容阅读
采用1∶1配对法,对湖北监利县等地104对早产儿和足月儿的母亲进行了回顾性调查。单因及条件Logistic回归分析发现早产的危险因素有双胎(呈完全分离型)、阴道出血(相对危险度为3.37)、早产儿(3.91)、生活事件(3.20)、不育史(3.00),疲劳程度(1.40),在武汉市收集非配对资料245例,通过单因和Cox线性Logistio回归分析,得到同样的结果。以上述6个危险因素做判别指标,依据极大似然判别原理,建立了判别指数表,根据该指数表对两组资料分别做了判别分类,经检验判别结果均有高度显著性。
A 1: 1 matching method was used to survey 104 mothers of preterm and term infants in Jianli County of Hubei Province. Univariate and conditional Logistic regression analysis showed that the risk factors for preterm birth were twins (completely separated), vaginal bleeding (relative risk 3.37), premature children (3.91), life events (3.20), infertility (3.00) Fatigue level (1.40). 245 unpaired data were collected in Wuhan. The same results were obtained by one-way and Cox linear Logistic regression analysis. According to the principle of maximum likelihood, a discriminant index table was set up based on the above six risk factors. According to the index table, the two groups of data were respectively classified and classified. The results of the test were highly significant.