【摘要】：目的:篩選蚌埠地區(qū)婦女患乳腺癌的危險因素,初步建立符合蚌埠地區(qū)的乳腺癌風險評估模型,探討乳腺癌低、中、高風險的分界點;并用該模型評估蚌埠地區(qū)具備特定危險因素的女性患乳腺癌的機率。方法:本研究屬病例對照研究,于2015年3月到11月在蚌埠醫(yī)學院一附院(47)、二附院(9)和蚌埠市第三人民醫(yī)院(15)3家三甲醫(yī)院腫瘤外科和普外科住院并經病理確診的71例本地原發(fā)性乳腺癌患者。對照組選取同期在同一家醫(yī)院體檢中心體檢健康的本地女性,年齡(±2歲)與病例相近,共選取91例。采用二元Logistic回歸篩選出本地乳腺癌的主要危險因素,在此基礎上建立乳腺癌風險評估模型。Fisher判別分析評價模型;觀察ROC曲線下面積來判斷模型診斷效能,并利用ROC曲線尋找乳腺癌低、中、高風險的截斷值。結果:1.與乳腺癌相關的單因素有:(1)一般資料:文化程度、職業(yè)、家庭平均月收入、醫(yī)保方式和體重指數(shù)。?生殖因素:生育次數(shù)和流產次數(shù)。?飲食因素:豆類及豆制品、蛋奶及其制品、油炸燒烤類、薯類和飲用水源。(4)睡眠情況:睡眠時間、睡眠滿意度和戴胸罩睡覺。(5)行為生活習慣:體育鍛煉和運動量。(6)環(huán)境因素:居住地和居住環(huán)境周圍污染源情況。(7)心理因素:生活總體滿意度。(7)認知和篩檢行為:認知總分分組和乳腺癌篩查情況。2.多因素二元Logistic回歸的主要危險因素有:家庭經濟狀況、食用豆類及其豆制品、負性情緒的排解和乳腺癌篩查。3.用風險評估模型預測低、中、高危人群,預測概率值P≤0.49判為低危險性人群,預測概率值P≥0.51判為高危險性人群,0.49預測概率值P0.51判為中危險性人群。結論:該模型可評估蚌埠地區(qū)具備特定危險因素的女性患乳腺癌風險,為建立篩查標準提供一定依據(jù)。
[Abstract]:Objective: to screen the risk factors of breast cancer among women in Bengbu area, and to establish a risk assessment model for breast cancer in Bengbu area, and to explore the dividing point of low, middle and high risk of breast cancer. The model was used to assess the risk of breast cancer among women with specific risk factors in Bengbu. Methods: a case-control study was conducted in the first affiliated Hospital of Bengbu Medical College (47) from March to November, 2015. The second affiliated Hospital (9) and the third people's Hospital of Bengbu (15) were hospitalized in tumor surgery and general surgery and confirmed by pathology in 71 cases of local primary breast cancer. In the control group, 91 local women (鹵2 years old) were selected for physical examination in the same hospital. The main risk factors of local breast cancer were screened by binary Logistic regression, and the risk assessment model of breast cancer was established on the basis of which the Fisher discriminant analysis model was established. The area under the ROC curve was observed to determine the diagnostic effectiveness of the model and the ROC curve was used to find the truncation values of breast cancer at low, middle and high risk. Results: 1. The single factors associated with breast cancer are: (1) General data: education, occupation, average monthly household income, health care style and body mass index. Reproductive factors: number of births and times of abortion.? Dietary factors: beans and soy products, egg milk and its products, fried barbecues, potatoes and drinking water. (4) Sleep: sleep time, Sleep satisfaction and bra sleeping. (5) behavior habits: physical exercise and exercise volume. (6) Environmental factors: pollution sources around living place and living environment. (7) Psychological factors: overall life satisfaction (7) Cognitive and screening behavior: cognitive subgroup and breast cancer screening. The main risk factors for multivariate Logistic regression were as follows: family economic status, consumption of beans and their products, negative emotion excretion and breast cancer screening. The risk assessment model was used to predict the low, middle and high risk population. The predicted probability value P 鈮，

本文編號：2383500