应用Cox比例危险度模型对影响老年肺癌患者长期生存因素的探讨  

A study on factors influencing survival of senile patients with lung cancer by Cox's proportional hazard model

作　　者：吴胜其[1] 侯媛淑[1] 

机构地区：[1]湖南省肿瘤医院

出　　处：《中华老年医学杂志》1998年第4期198-200,共3页Chinese Journal of Geriatrics

摘　　要：目的应用Ｃｏｘ比例危险度模型，对影响老年人肺癌治疗后生存的可能因素进行多因素分析，并确立治疗后生存预测模型。方法对３１３例老年人原发肺癌初治患者进行了随访，观察１７个指标与老年人肺癌长期生存的关系。采用单因素和多因素分析、拟合曲线分析及残差图检验。结果多因素分析结果显示５个明显影响老年人肺癌预后的因素，其作用强度依次为身体状况、近期疗效、临床分期、年龄和治疗方法。据此计算了每个患者的预后指数（ＰＶ），根据其大小将患者分成５个危险组，结果显示随着ＰＶ的升高生存率逐步下降。结论影响老年人肺癌治疗后生存率的主要因素是身体状况、近期疗效、临床分期、年龄和治疗方法；且随着ＰＶ的升高。Objective To analyse multivariately the possible factors influencing survival of the senile patients with lung cancer, and to establish models for predicting survival by Cox's proportional hazard model. Methods A prospective followup study of 313 senile lung cancer patients was undertaken from 1 January 1987 to 31 December 1991. Seventeen factors which might influence long term survival were studied by univariate analysis (KruskalWalli's test for KaplanMeier survival curve) and by multivariate analysis (Cox's proportional hazard model). The goodness fit of estimated predicting models of survival were checked by fitting curve and residual plot. Results Multivariate analysis using Cox's model revealed 5 significantly prognostic factors: body status, shortterm response to treatment , clinical stage, age and method of treatment. According to their regression coefficients, the predicting value(PV) of each patient was calculated, then all patients were divided into five risk groups according to PV, their predicting models of survival rates were established in groups, the survival rate of lung cancer patients decrease with the rise of PV. Conclusions These results suggest that the body status, shortterm response to treatment, clinical stage, age and method of treatment are the major factors affecting prognosis. The survival probability of lung cancer patients decreases with the rise of PV.

关 键 词：肺肿瘤 诊断 存活率 比例危险度模型 

分 类 号：R734.204[医药卫生—肿瘤]