Risk Factors Analysis And Risk Prediction Of Cerebrovascular Disease In Community Residents In Qing Yun Pu District Of Nan Chang

Objectives:The thesis is supposed to grasp the epidemiological status quo and influencing factors of the prevalence of cerebrovascular disease in community residents in Qingyunpu District,Nanchang City,and to build a risk index model of cerebrovascular disease suitable for local community residents.In addition,it is also aimed at providing support for local community health service institutions to carry out personalized interventions for cerebrovascular diseases and providing scientific basis for local development of cerebrovascular disease prevention and control.Methods:The stratified cluster sampling method was used to randomly select 2147permanent residents of the community aged 40 and above in Qingyunpu District of Nanchang City for investigation and some physiological index detection or measurement.The sampling residents were randomly assigned to the modeling group and verification group.The survey included the basic situation,lifestyle,family history of past diseases and some diseases,and the main indicators such as height,weight,blood pressure,pulse,blood sugar,blood lipids and so on.The survey was completed by trained investigators using face-to-face inquiries.The Poisson regression model of the modeling group was used to screen for the factors and relative risks that affected the local residents’cerebrovascular disease.The Harvard Cancer Index Risk Method was used to construct the local residents’cerebrovascular disease risk prediction model.The verification group was used to predict the risk of the disease and verify the effectiveness of the model.Statistical methods such as t test,χ² test,and trend test were also used in the comparison between data groups.Results:（1）Survey of the general situation of residents:2,147 local community residents aged 40 and over were randomly selected,with 2138 effective samples and an effective rate of 96.0%;among them,men accounted for 38.45%and women accounted for 61.55%;the average age was 65.57±9.63 years old;99.92%are Han nationality;the proportion of residents with education level of elementary school and below,junior high school,junior high school/high school,college/university,master’s degree and above are 22.22%,41.21%,25.58%,16.65%,and 0.19%respectively.The health care types of the sampling residents are various:urban employee medical insurance accounts for 79.18%,urban residents medical insurance14.5%,new rural cooperative medical insurance 5.52%,all self-financed 0.61%,and others 0.19%.The smoking rate of residents is 14.97%.85.78%of the residents do not drink alcohol,13.24%drink a small amount,and 0.98%drink regularly.The residents with a moderate taste account for 85.27%,the ones with a light taste 11.09%,and the ones with a salty taste stand at 3.65%.The residents who eat vegetables in more than five days a week account for 88.49%,those who eat vegetables in 3 to 4day of a week account for 10.71%,and those who eat vegetables for less than 2 days in a week account for 0.80%.14.13%of the residents have a family history of stroke.The morbidity rate of hypertension is 55.33%,diabetes 14.27%,dyslipidemia 43.17%and heart disease 12.35%.（2）The prevalence of cerebrovascular disease:The proportion of residents with cerebrovascular disease stands at 4.49%,and the morbidity rate after standardization is 2.46%.The morbitity rate of males is 6.33%,and that of the females stands at 3.34%（?²=10.495,P<0.001）.The higher the age,the more poosible the prevalence of cerebrovascular disease is(?_Trend ²=26.403,P<0.001).The rate of cerebrovascular disease in smokers is 8.44%,higher than that in non-smokers（3.80%）（?²=13.673,P<0.001）;the frequency of physical exercise is related to the rate of cerebrovascular disease（?²=13.420,P<0.001）.The rate of cerebrovascular disease in people with a family history of stroke is 7.95%,which is significantly higher than that in those without a family history of 3.92%（?²=9.799,P<0.001）.The rate of cerebrovascular disease in hypertensive patients is 6.59%,which is significantly higher than that in non-hypertensive patients（?²=27.318,P<0.001）.The rate of cerebrovascular disease in diabetic patients is 10.49%,higher than that in non-diabetic patients（3.49%）（?²=29.879,P<0.001）.（3）Influencing factors of cerebrovascular disease:Multiple stepwise Poisson regression found that age,exercise habits,hypertension,diabetes,smoking and family history of stroke are the six factors that affect cerebrovascular disease in community residents of Qingyunpu District.The relative risk OR of 6 cerebrovascular disease factors were found to be 1.051（95%CI:1.060¹.086）,1.998（95%CI:1.098³.634）,2.324（95%CI:1.121⁴.819）,4.343（95%CI:2.267 to 8.320）,2.315（95%CI:1.167 to 4.593）,and 2.131（95%CI:1.013 to 4.484）respectively.（4）The risk assessment and prediction model of community residents’cerebrovascular disease:the risk scores of 6 factors of age,exercise habits,hypertension,diabetes,smoking,and family history of stroke are 2.5,10,10,25,10,and 10 respectively.The exposure rates are 0.48,0.82,0.28,0.11,0.28,and 0.14respectively.The average risk score of local residents for cerebrovascular disease is19.15.（5）Verification of the risk assessment and prediction model:10 variables including gender,age,occupation,smoking,and exercise in the modeling and verification groups constitute a balance（all P>0.05）.（1）Evaluation of the stability of the model:the proportion of the individual cerebrovascular disease in the modeling group with lower,low,general,high,and higher risks is 26.19%,39.29%,6.83%,23.37%,and 4.30%,respectively.The proportion of the verification group is 32.48%,42.47%,5.52%,23.11%,and 5.43%,respectively.The risk level composition of individual cerebrovascular disease predicted by the modeling group is not different from that of the verification group（P>0.05）.（2）Evaluation of the classification of risk prediction model:the risk levels of cerebrovascular disease of the modeling group and the verification group are directly proportional to the actual morbidity rate,and the actual morbidity rate of cerebrovascular disease shows an upward trend with the increase of the risk level（All P<0.001）.（3）Analysis of the effect of the cerebrovascular disease risk prediction model in the verification group:the AUC under the ROC curve of the risk prediction model is 0.715（95%CI:0.638⁰.792）.R,which is 1.082,is the best positive cut-off point predicted by the model,with a sensitivity of 63.64%and a specificity of 69.74%.Conclusion:（1）Grasp the prevalence of cerebrovascular disease in community residents over40 years old in Qingyunpu District of Nanchang City,making up the data gap of cerebrovascular disease in this area.（2）Explore the six factors of the age,exercise habits,hypertension,diabetes,smoking,and family history of stroket that affect the morbidity rate among the residents,and find out the links between these risk factors and cerebrovascular disease（relative risk level）,provide basic data for local prevention of cerebrovascular disease.（3）Construct an index model of the risk level of cerebrovascular disease in accordance with the characteristics of residents in Qingyunpu District and regional subcultural influence.In this study,the Harvard Cancer Index method was used to preliminarily construct a cerebrovascular disease risk assessment model,and the model has been tested to have good stability,reasonable grade standards,and predictable results,successfully supporting the personalized cerebrovascular disease interventions for the local area.The high morbidity,high invalidity and high mortality have made the cerebrovascular diseases become one of the important public health issues in modern society.Constructing a prediction model of the risk index of cerebrovascular diseases within a certain range is conducive to screening high-risk populations of cerebrovascular diseases.It is of great practical significance to carry out specialized intervention on this high-risk group.