哈尔滨市宫颈癌致病因素的研究——宫颈癌致病因素的Logistic回归分析

A Conditional Logistic Regression Analysis For the Risk Factors of Cervical Cancer

对宫颈癌的病因,学者们做了大量工作,提出了不少可能的致病因素,但至今尚无定论,因此对宫颈癌的病因及致病因素的研究,仍是十分迫切而重要的课题。运用Logistic回归模型及危险状态分析法进行多变量分析,不仅可控制混杂因素,分析各因素的联合作用,并能对致病因素的定量测定进行分析。有助于宫颈癌病因探讨,在国内尚未见报导。为了探讨病因及致病因素,我们进行了回顾性病例对照研究,现将结果报告如下:

Death rates for cancer of the cervix uteri are generally higher than breast in China, Cervical cancer, despite its declinnig mortalty in China is still an important CancerIn order to elucidate the relationship between cervical cancer and suspected etiological factor, a 1: 2 matching to case-control study was carried out at tumour hospital in Harbin. One hundred cases of cervical cancer and two hundred cases of controls were selected during the two-year study period.The results of matched pair analysis revealed that cervical cancer was si nificanly highly correlated with three factors that is multiple Sexual partners, chronic cervicitis socioeconomic status, and their odds ratios were 9.28(p<0.01)4.86(p<0.01),1.99(p<0.01)resectively, The analysis of combinations of these three factors suggest that each factor might be related to cervical cancer multiplicatively as well as additively