Binary Logistic Regression Analysis Based on Macro-,Meso-,and Micro-Levels of the Factors Associated with the Pre-Existing Evidence of Coronary Heart Disease Blood Stasis Evidence
Objective To explore the relevant protective/risk factors during the development of coronary heart disease blood stasis evidence in the process of pre-existing evidence based on the macro-,meso-,and micro-health state characterization parameter system of Chinese medicine state science.Methods 253 cases of coronary heart disease to be investigated were collected from the outpatient and inpatient departments of the Department of Cardiology in the hospitals affiliated to Hunan University of Traditional Chinese Medicine,and questionnaires were formulated according to the three dimensions of macro,meso,and micro,and the collected parameters were categorized with Python software,and the patients were diagnosed as pre-coronary heart disease blood stasis evidence(150 cases)and coronary heart disease blood stasis evidence(100 cases),and statistical analyses were performed with frequency analysis,χ2 test,and Logistic regression and other methods for statistical analysis.Results ①The results of univariate analysis showed that:age,BMI,history of smoking,history of alcohol consumption,history of hypertension,history of diabetes mellitus,average monthly high temperature,air quality,season,type of occupation,social environment,coronary artery angiographic stenosis,diastolic blood pressure,systolic blood pressure,creatinine,uric acid and total cholesterol differed between patients diagnosed as pre-Coronary artery disease blood stasis evidence and those diagnosed as Coronary artery disease blood stasis evidence,and all the differences were statistically significant(P<0.05).② Binary logistic regression analysis showed that age,BMI,history of alcohol consumption,type of occupation,coronary angiographic stenosis,diastolic blood pressure,creatinine,and dark red tongue were independent risk factors.A prediction model was established:P=1/[1+exp(16.522-1.427×age-0.975×BMI-3.55×drinking history+1.982×monthly average high temperature+0.709×season-1.827×occupational type-1.1×coronary angiographic stenosis-0.072×diastolic blood pressure-0.076×creatinine+2.398×dizziness-4.108×dark red tongue+4.169×pulse asthenia)],the model prediction rate was 90.5%.Conclusion The logistic regression model of coronary heart disease with blood stasis evidence is good with clinical diagnosis,which lays the foundation for the exploration of the state between the already diseased and undiseased of coronary heart disease,and provides important basic data for the theory of subhealth.
NETL
NSTL
万方数据
