Font Size: a A A

Study On The Prevalence And Risk Factors Of Stroke Among Middle-aged And Elderly In Community Based On Decision Tree Model

Posted on:2021-03-11Degree:MasterType:Thesis
Country:ChinaCandidate:Q NanFull Text:PDF
GTID:2404330614964415Subject:Epidemiology and Health Statistics
Abstract/Summary:PDF Full Text Request
Objective Stroke had the characteristics of high disability and mortality,which brings severe economic burden to society and families.However,the risk of stroke could be reduced by improving the controllable factors,and the identification of factors affecting stroke was particularly important.The purpose of this study aimed to explore the factors associated with stroke among middle-aged and elderly in the community based the decision tree model,and provided a basis for prevention strategies on stroke in community.Method Residents aged 45 years and above were included in the study from August 2019 to January 2020,in Qiaohua Community Health Service Station and Xing'an South Road Community Health Service Station in Hohhot.Questionnaire,scale,and dietary surveys were used to collect demographic characteristics,lifestyles,diseases,and dietary information of the subjects.Physical examination data and diagnosis of stroke were provided by the Community Health Service Stations.Measurement data used mean ± standard deviation(X±S)and rate(ratio).The comparison between groups used t test or chi-square test.The decision tree model was constructed to explore the factors related to stroke,which trimmed globally and 10-fold cross-validation were set,which the error classification rate and receiver operating characteristic curve were used to evaluate the model.All analysis used SPSS 19.0 and SPSS Modeler 18.0 software.MedCalc16.0 software was used to draw receiver operating characteristic curve.Results(1)A total of 307 residents participated in this study,including 143 men and 164 women,with an average age of 68.22±7.55 years old.60.53%of residents with stroke were low-level SES participants.(2)Residents who never smoked,no drinking,were not exposed to secondhand smoke,had good sleep quality,did not control weight,and had moderate physical activity accounted for a relatively large of proportion(70.76%?98.70%?76.15%?44.41%?50.49%?40.72%).The quality of sleep was significantly different between residents with stroke and without stroke.The average duration of sleep was 6.58 hours among residents in the communities,and the average sleep duration of residents without stroke was higher than that of stroke.(3)The intake of vegetables,fruits,cereals,poultry and meat,eggs,milk and dairy products,beans and nuts,and aquatic products were 66.123 g/d?11.378 g/d?181.191 g/d?22.470 g/d?19.166 g/d?66.753 g/d?15.290 g/d?2.074 g/d,respectively,which was significantly lower than the recommendation.However,the intake of oil and salt were higher than the recommendation of Dietary Guidelines for Chinese Residents.(4)The awareness rate of knowledge related to stroke among middle-aged and elderly residents in the communities was low,and there was no difference in the knowledge related to stroke between residents with stroke and non-stroke residents.(5)The decision tree screened out seven variables which had effect on stroke.The importance of the variables was ranked in descending order of SES,sleep quality,intake of egg,ABSI,hypertension,diabetes and nuts and legume.The sequence of importance were 0.25?0.22?0.15?0.14?0.10?0.08?0.06,respectively.(6)The decision tree model with C5.0 algorithm grew into 4 layers with a total of 18 nodes.The root node of the decision tree model was divided from intake of eggs.A total of eleven decision paths were generated.(7)the misclassification rate of the decision tree model,which explored the association between risk factors among middle-aged and elderly in community,was 30.60%,and the area under the ROC curve was 0.714(95%CI:0.663-0.764).Conclusions The probability of suffering from stroke was highest among participant with low level SES in decision tree model,which was the most important predictor of stroke,compared with middle and high level SES.The dietary was imbalanced among middle-aged and elderly residents in the communities in Hohhot,and the intake of foods,such as cereals,vegetables,and fruits,were insufficient,however,the intake of oil and salt was excessively.More knowledge about healthy diet and balanced diet should be publicized to residents in communities in health education.Insufficient intake of eggs was more likely to associate with high risk of stroke,which doctors in communities should encourage residents to consume eggs in an appropriate amount to ensure protein intake adequately.Sleep quality would affect the occurrence of stroke,and poor sleep quality was related to a high probability of stroke.It was important to attach importance to middle-aged and elderly residents and help them improve their quality of sleep.The middle-aged and elderly residents in communities should be encouraged to control weight through physical activity or other ways so that they could keep lower ABSI.The prevalence of stroke was higher among residents with diabetes or hypertension,and the decision path also showed that the probability of suffering from stroke was higher among residents with hypertension or diabetes.Residents with hypertension and diabetes should be paid more attention,which they had better take measures to improve their lifestyle.The knowledge related to stroke prevention was poor among residents,which was not beneficial to stroke prevention.Participants with stroke did not have enough knowledge on risk factors of stroke,which they still did not pay attention to stroke prevention.
Keywords/Search Tags:Stroke, ABSI, SES, Diet, Knowledge, Decision tree, Factors of stroke
PDF Full Text Request
Related items