Font Size: a A A

Screening Model For High Risk Population Of Esophageal Cancer Based On Early Detection And Treatment Program For Upper Gastrointestinal Cancer

Posted on:2020-09-14Degree:MasterType:Thesis
Country:ChinaCandidate:J H GuFull Text:PDF
GTID:2404330572489022Subject:Public health
Abstract/Summary:PDF Full Text Request
BackgroundChina is one of the countries with a high incidence of esophageal cancer in the world.In rural high-incidence areas where health resources and economic levels are relatively backward,esophageal cancer has brought a heavy disease burden to local residents.The secondary prevention based on "early detection,early diagnosis,early treatment" is the key to reducing the mortality and disease burden of esophageal cancer.In 2006,under the aegis of central financial transfer payment for early diagnosis and treatment of esophageal cancer,Feicheng People's Hospital used the technique of"endoscopic iodine staining + indicative biopsy technology" to carry out early detection and treatment of esophageal cancer on local residents aged 40-69.Since the implementation of the project,the early diagnosis rate of local upper digestive tract tumors,especially esophageal cancer,has been effectively improved.However,given that the endoscope is a traumatic examination and nearly 70%of the subjects are not to be examined,the subjects are not well compliance and the cost-effectiveness of screening still needs to be improved.Therefore,in order to providing scientific basis and improving the cost-effectiveness and efficiency of screening program,it is urgent for us to develop a new method which can find more high-risk populations.Objective1.By describing the implementation of the early detection and treatment of Feicheng City,we will analyze the detection rate of early esophageal cancer and precancerous lesions,and then evaluate the effect of screening program comprehensively.2.The screening model of esophageal cancer will be constructed to predict the risk of individual disease and screen out high-risk groups.The role of various exposure factors in the occurrence and development of the disease will be analyzed by this model as well.Materials and methodsWith support of the platform of cancer screening and early detection-treatment in Shandong Province,our study described and analyzed the detection rate of early esophageal cancer and precancerous lesions in different gender and age groups,and then evaluated the effects of the screening program comprehensively.All research materials will be obtained by establishing a cohort of early detection and treatment of esophageal cancer screening population.Based on the basis of descriptive analysis and univariate analysis in intervention group,variables which have statistical and clinical significance will be included in the multivariate logistic regression to calculate individual risk.Variable information includes general conditions(height,weight,etc.),lifestyle(smoking,drinking,etc.),history of digestive tract disease and medication.The prediction results will be evaluated by the ROC curve,AUC,specificity and sensitivity,and then the stability and prediction effect of the model will be evaluated by the ten-fold cross-validation.Result1.There were 52,977 target subjects aged 40-69 in our study from 2006 to 2016.Among them,3397 subjects were diagnosed with precancerous lesions,with the detection rate was 6.41%;140 subjects were diagnosed with early esophageal cancer,with the detection rate was 0.26%.2.Combined with the clinical expertise of esophageal cancer and the outcomes of univariate analysis,we selected appropriate factors enter the multivariate analysis model.After selected by the backward elimination method,the model of all screening population retains the following variables:age(OR=0.80),gender(OR=0.68),income(OR=0.92),drinking(OR=1.27),BMI(OR=2.10),tea(OR=1.29),esophagitis(OR=1.79),family history of esophageal cancer(OR=1.36);the model of male retains the following variables:age(OR=2.13),income(OR=0.92),drinking(OR=1.49),BMI(OR=0.81),smoking(OR=1.28),tea(OR=1.28),esophagitis(OR=1.89),family history of esophageal cancer(OR=1.36);age(OR=2.32),drinking(OR=1.36),BMI(OR=0.84),smoking(OR=1.65),tea(OR=1.27),hypertension(OR=1.35),family history of esophageal cancer(OR=1.36).3.The area under the ROC curve of the whole screening population,male group and female group was 0.693(95%CI:0.685-0.701),0.765(95%CI:0.756-0.774),0.726(95%CI:0.717-0.73 5)respectively.After ten-fold cross-validation,the AUC of the whole screening population,male group and female group was 0.697(95%CI:0.691-0.703),0.771(95%CI:0.760-0.782),0.736(95%CI:0.725-0.747).Conclusion1.Early diagnosis and early treatment screening technology based on "endoscopic iodine staining+indicative biopsy" could screen out potential early esophageal cancer patients in the population and reduce the burden of disease.2.The screening model of esophageal cancer based on information of screening population had a good discriminative ability and calibration ability to assess the risk of esophageal cancer in high-risk populations and screen out high-risk groups.
Keywords/Search Tags:Esophageal cancer, Early detection and treatment, Logistic regression, Screening model
PDF Full Text Request
Related items