社区老年冠心病患者死亡危险因素评估.docxVIP

社区老年冠心病患者死亡危险因素评估 【摘要】目的：探究社区老年冠心病患者死亡危险因素，为相关医务人员提供 参考。方法：选择社区350名老年冠心病患者作为研究对象，系统收集疾病的 相关信息。首先进行单因素Logistic回归分析,筛选有意义的变量作为潜在指 标，将入选的变量作为协变量进行逐步向前多因素Logistic回归分析，制作模 型预测ROC曲线并进行检验，同时对模型进行Hosmer?Lemeshow ( H )拟 合优度检验。结果:单因素Logistic回归分析共有15项指标纳入，相关系数 B在0.02到0.48之间,Exp ( B )即OR值在1.02 ~ 1.62范围内。逐步向前 多因素Logistic回归分析共纳入8项指标，B值范围为0.96 ~ 2.25 , Exp ( B ) 范围为2.61 ~ 9.49 , Logistic回归建模以后的预测符合度图形ROC曲线下面 积为 0.88 , Hosmer-Lemeshow ( H )检验为 P二0.39。结论:Logistic 回归 显示以下8项指标为老年冠心病死亡的危险因素：配偶是否健在、周围血管病 变、脑血管病变、IgG、冠心病史、糖尿病史、肾功能不全、缺乏锻炼，本次 回归模型的拟合效果较好。 【关键词】老年冠心病；死亡；危险因素；Logistic回归分析;社区 Assessment of risk factors of death in elderly patients with coronary heart disease in the comm unity Abstract [Objective]: To explore the risk factors of death in elderly patients with coronary heart disease in community, and provide ref ere nee for re leva nt medical staff. [Methods]: We selected community of 350 elderly patients with coronary heart disease as the object of study, systematically collecting the disease-related information. First, single factor Logistic regression analysis was performed to select meaningful variables as potential indicators, and the selected variables as covariates were analyzed by stepwise forward multiple Logistic regression analysis. The ROC curve was predicted and tested, and the goodness of fit of the model was tested by Hosmer- Lemeshow (H). [Results]: A total of 15 indexes were in eluded in the single factor Logistic regression analysis. The correlation coefficient B was between 0.02 and 0.48, and Exp (B) was within the range of 1.02 ~ 1.62. 8 in dexes were in eluded in the stepwise multiple factor Logistic regression analysis. The range of B value was 0.96 to 2.25, and the range of Exp (B) was 2.61 ~ 9.49. After the Logistic regression modeling, the area under the ROC curve is 0.88, and the Hosmer-Lemeshow (H) test is P=0.39. [Conclusion]: Logistic regression showed that the following 8 indicators were risk factors for the death of elderly coronary heart disease