Risk Factors For Gastric Cancer Of Rural Residents In Central And Eastern Parts Of Gansu Province:A Case-control Study

Objective: To explore the risk factors for gastric cancer in rural residents of central and eastern Gansu Province by using a case-control method to collect information on gastric cancer risks so as to analyze the risk factors and their interactions of gastric cancer and provide evidence-based evidences for establishing the "three-level" prevention system in rural areas of Gansu Province.Methods :(1)The choice of subjects in the case group and the control group.Through the new rural cooperative medical insurance information platform in some high-incidence counties(Zhuanglang County,Jingning County and Huining County)in the central and eastern parts of Gansu Province,215 patients who were diagnosed with gastric cancer discharged from December 2013 to September 2017 were included in the case group.The control groups included a source population control and a non-source population control.According to the principle of gender,age(±3 years old)and 1?2 residence-matched,430 healthy residents in the patient's place of residence were selected as the source group control.In some neighboring counties(Zhenyuan County,Xifeng District and Huachi County)near areas where the source population lives,226 healthy people who met the inclusion and exclusion criteria were selected as non-source population control.(2)Case-control study.A questionnaire was designed,and six types of risk related information on gastric cancer such as general demographic characteristics,living environment factors,personal living habits and life-style,eating habits and dietary content,digestive tract related medical history and family history of cancer,social psychological factors and so on were collected through “face-to-face” investigation method.(3)Statistical analysis methods.Conditional Logistic regression of individual matching design and univariate and multivariate models of unconditioned Logistic regression of group design were applied to analyze the risk factors related to gastric cancer,The principal component logistic regression analysis was used to optimize the former models.The interaction of risk factors for gastric cancer: the possible interactions among various risk factors of gastric cancer were explored by using the nonparametric statistical model of classification regression tree;Interaction parameter point estimation for factor items that have been predicted as possible interaction parameters was conducted by using the bifurcation analysis based on multiplication and addition models.Furthermore,the conditional Logistic regression model was used to conduct the test hypothesis of the regression coefficient based on the interaction terms of the multiplication model,and the Excel table compiled by Adersson was used to verify the confidence interval based on the parameters of the interaction model interaction parameters.Results:(1)the risk factors for gastric cancer against the source control included personality type A(OR=1.710),weak emotional regulation ability(OR=4.007),Frequently consumed preserved food(OR=1.721,),dry and hard diet(OR=2.283),irregular diet(OR=2.577),overeating(OR=2.500),concentrated water supply without chlorination disinfection(OR=2.334),smoking “dry cigarette”(OR=2.778),excessive drinking(OR=2.987),drink “pot tea”(OR=2.005).Protective factors for gastric cancer: long service life of the refrigerator(10 years)(OR=0.572).(2)risk factors for gastric cancer against non-source population controls included: emotional irritability(OR=3.413),high-temperature diet(OR=2.532),dry and hard diet(OR=3.484),history of chronic atrophic gastritis(OR=7.813),high tea drinking concentration(OR=2.381),irregular diet(OR=0.310).Protective factors for gastric cancer: regular consumption of soy products(OR=0.066),regular consumption of white meat(OR=0.261),use refrigerator(OR=0.340).(3)The analysis on the interaction among risk factors for gastric cancer: the results of qualitative classification and regression tree model suggested that there was interaction between smoking “dry cigarette” and one of three factors including dietary law,hard food and emotion regulation ability.The results of quantitative analysis based on the multiplication and addition model showed that there was a negative multiplicative interaction between the emotional adjustment ability and smoking “dry cigarette”,however not any interaction was found among the above various influencing factors.Conclusion :(1)adverse dietary preferences and habits,drinking "pot tea" and strong tea,excessive drinking and smoking,chronic atrophic gastritis and adverse psychosocial factors are risk factors for gastric cancer in rural residents of central and eastern Gansu Province e.Regular intake of white meat and soy products and the use of refrigerators are protective factors for gastric cancer.(2)There may be an interaction between smoking “dry cigarette” and dietary law,hard and dry diet or emotional regulation ability.