| Objective:Collect the screening data of gastric cancer in the early diagnosis and treatment project of Nanchang City,describe the distribution characteristics and detection rate of gastric cancer highrisk groups,analyze the risk factors of gastric cancer and precancerous lesions,and build a risk prediction model of gastric cancer and precancerous lesions centered on Nanchang City to provide data support for the risk screening and intervention measures of gastric cancer in Nanchang City.Methods:This article includes two parts.The first part is based on the project of early diagnosis and early treatment of cancer in Nanchang City,adopts multi-stage random sampling method,collects relevant data of residents according to the inclusion and exclusion criteria,carries out questionnaire evaluation on high-risk groups of gastric cancer,describes the distribution characteristics of high-risk groups participating in gastric cancer screening,and the compliance of gastroscope screening;The results of gastroscope screening,the detection rate of gastric cancer and precancerous lesions,and the early diagnosis rate of gastric cancer were analyzed.In the second part,a 1:3 case-control study was used.114 cases of gastric cancer and precancerous lesions were collected in the first part as the case group,and 342 cases of control group were matched according to sex and age.The data of the two groups were analyzed by single factor analysis to screen the risk factors related to gastric cancer and precancerous lesions.The selected risk factors were evaluated by multivariate logistic regression analysis to determine independent risk factors,The logistic regression model is constructed with the selected independent risk factors as the predictive variables of risk prediction.The prediction efficiency is verified by drawing ROC curve,and the model stability is verified by cross-cutting.The performance and effect of the prediction model are comprehensively evaluated by analyzing the AUC value,sensitivity,specificity,Yoden index and other indicators of the model.Results:(1)A total of 26576 research subjects were recruited,and 5494 high-risk individuals for gastric cancer were evaluated,with a high-risk rate of 20.67%;Among them,1626 participated in gastroscopy screening,with a participation rate of 29.60%.The high-risk rate of males is higher than that of females(Rao scott χ2=316.060);The high-risk rate of individuals aged 40 to 49 years old,personnel from government agencies and enterprise units,educated at university or above,married,and overweight is higher than that of other groups(Rao Scott χ2=163.755,483.693,167.775,48.027,26.807,P<0.05);The compliance of women with gastroscopy screening was higher than that of men(χ2=5.131,P<0.05);The highest age is between 50 and 59 years old,and the lowest is between 70 and 74 years old(χ2=15.083,P<0.05);The compliance with screening in high school/vocational school/technical school with the highest educational level(χ2=35.318,P<0.05);There was no statistically significant difference in compliance analysis among other screening groups(2)The results of gastroscope screening showed that the detection rate of non-atrophic gastritis was 83.0%,that of atrophic gastritis was 4.25%,that of intestinal metaplasia was 1.66%,that of gastric mucosal lesions was 0.80%,that of precancerous lesions of gastric cancer was 6.10%,and that of early gastric cancer was 0.31%.(3)The results of univariate logistic regression analysis showed that farmers,education level,HP infection,family history of cancer,fresh vegetables 2-3 days and less,fresh fruits,meat,eggs and milk,beans,garlic 2-3 days a week,fried foods 4-6 days a week and less,salty taste,hot food,a lot of indoor lampblack,smoking,drinking,and physical exercise were the influencing factors of gastric cancer and precancerous lesions.(4)Multivariate logistic regression showed that HP infection,family history of cancer,low frequency of consumption of fresh fruits and beans,preference for hot food and alcohol consumption were independent risk factors for gastric cancer and precancerous lesions(all OR>1,P<0.05);Moderate taste preference and low frequency of fried food consumption were protective factors for gastric cancer and precancerous lesions(all OR<1,P<0.05).(5)Finally,a risk prediction model for gastric cancer and precancerous lesions was established.The area under the ROC curve AUC=0.971,sensitivity=0.974,specificity=0.936,and Jordan index=0.909.The accuracy of the model is good after the cross-internal validation of ten fold.Conclusion(1)Nanchang Early Diagnosis and Treatment of Gastric Cancer Project can effectively screen out early gastric cancer and precancerous lesions.In the future,health education can be strengthened for the elderly and men to improve residents’ screening compliance,so that more residents can benefit from this project.(2)The independent risk factors for gastric cancer and precancerous lesions were HP infection,family history of cancer,low frequency of consumption of fresh fruits and beans,preference for hot food,and alcohol consumption.(3)The established risk prediction model for gastric cancer and precancerous lesions has certain predictive ability,which can provide theoretical guidance for community workers to screen high-risk groups of gastric cancer. |
PDF Full Text Request