肺癌危险因素的逐步回归分析

A STEPWISE REGRESSION ANALYSIS OF THE RISK LUNG CANCER

我国是肺癌高发国家之一。目前有关肺癌的病因虽有一些学者研究,但仍有许多争论。为此探讨肺癌的危险因素是肺癌病因学研究中迫切需要解决的问题。本文对锦州市肺癌流行因素作了调查:原发性肺癌新病例、其它癌症及非癌病例各105例。在单因素分析的基础上选出有意义的9个指标即:慢性气管炎史、精神创伤史、吸烟史、毒物接触史、生活用煤、饮酒、做饭、家族恶性肿瘤史及文化程度等。在此基础上对上述9个因素在 apple—Ⅱ型微机上进行逐步回归分析,其结果按标准回归系数大小被选入回归方程的因素有慢性气管炎、精神创伤史、吸烟史、毒物接触史和做饭史。复相关系数(R)为0.4430,标准估计误差(SY)为0.05。作者对肺癌进行有的放矢的预防工作提供出一定的科学依据。

The incidence of lung cancer is very high in our country.At present,in spite of the studies by some scholars,there is much controversy about the cause of lung cancer.It is urgent to study the cause of its high incidence. In this paper an investigation of the cause of lung cancer spread in Jinzhou Distract was reported 315 cases were studied.There were 105 cases with primary lung cancer,other kinds of cancer and non-cancer cases respectively. On the basis of single-factor analysis,we found 9 meaningful factors,that is,chronic bronchitis,serious psychic trauma, smoking,exposure to poison, coal fuel,drinking,cooking,family malignant tumour,culture level,etc.We made stepwise regression analysis by apple-Ⅱ electronic computer. The result of analysis is as follows:The factors selected into the regres- sion according to the normal regression coefficient from big to small,include those dne to chronic bronchitis,serious psychic trauma,smoking,exposure to poison,cooking,etc.The coefficient of multiple correlation is R=0.4430; the standard error of estimatation is SY=0.05.The result of this paper gives a scientific evidence for purposeful prevention of lung cancer in units concer- ned.