Cohort Study On Risk Prediction Model Of Metabolic Syndrome Based On Routine Physical Examination Biomarker By Exploratory Factor Analysis

Objective: The purpose of this study was to analyze the clinical features of cohort of medical personnel,and to clarify the epidemiological characteristics of the metabolic syndrome(MS),and to build a multivariate regression formula for better predicting the risk of MS by routine physical examination biomarkers.Methods: By collecting the date of the physical examination of the staff in a three top hospitals from June 2011 to June 2016 in Dalian City,to understand the cross-sectional survey data in the health of the epidemiological characteristics and the prevalence of each component of MS.In the vertical queue,the MS-related factors were extracted by using the exploratory factor analysis(EFA)and COX proportional regression models,and the MS prediction model was established.According to the results of the study,the relevant suggestions are put forward to provide effective intervention measures and scientific basis for the healthy management of the healthy physical examination population.It is also for non-MS crowd to develop scientific prevention and control strategies.Results:1.The difference of Blood pressure,body mass index(BMI),fasting blood glucose(FBG),glycerin three greases(TG),high density lipoprotein(HDL-C),white blood cell count(WBC),the absolute value of neutrophil(N),the absolute value of lymphocyte(L),alanine aminotransferase(ALT),aspartate aminotransferase(AST),gamma glutamyl transferase(gamma-GGT),uric acid(UA),serum creatinine(SCr),direct bilirubin(DBIL)and the prevalence of nonalcoholic fatty liver disease(NAFLD)between the two groups was statistically significant(P=0.000).There was no significant difference in total bilirubin(TBIL)and red blood cell count(RBC)between the two groups,as well as in Hemoglobin content(HB)and the incidence of thyroid nodule in males(P>0.05),however the difference of thembetween the two female groups was statistically significant(P=0.000).2.The incidence rate of MS were 1.92%,2.07%,2.50% and 3.07% in the follow-up cohort from 2012 to 2015,as we know,the overall trend was increasing year after year.The cumulative incidence rate of MS was 16.08/1000 person years(178/11073 years).The detection rate of the four component of MS from high to low was hyperglycemia,overweight / obesity,dyslipidemia,high blood pressure.3.According to the potential predictive factors extracted by EFA,10 factors were extracted from males and 8 from females.8 Predictors are the same for both male and female: obesity,fat factor(including BMI,TG,HDL-C,and NAFLD),liver enzyme factor(including ALT,AST,?-GGT),inflammatory factors(including WBC,N and L),blood lipid factors(include TC and LDL-C),bilirubin(including the TBIL and DBIL),blood pressure factor(include SBP and DBP),red blood cell factor(including RBC and Hb),renal function factor(including the UA and SCR).In addition,there are 2 potential predictive factors for males,which are glucose metabolism factor(including FBG)and thyroid factor(including thyroid nodules).4.We Used the COX regression analysis to construct the risk function of MS followers.With time going by,the cohort of people experienced a growing risk of MS.At fourth years of follow-up,the risk of MS in female was about 3.5 times higher than in the first place,and about 3 times than that in male.5.Logistic regression prediction model: Regression formula of MS risk assessment for female: Logit(P)= —4.866 + 0.286 × F1 + 0.536 × F2+1.557 ×F3 + 1.127 × F4-0.373× F5 + 1.566 × F6 + 0.604× F8;The regression formula of MS risk assessment for male :Logit(P)= —1.928 + 1.729 × F1 + 0.523× F2 + 1.667 × F7 + 1.022× F9?6.In the female cohort,the area under the ROC curve(AUC)of the model was0.973(95%CI:0.964-0.981),with statistical significance(P < 0.01).The sensitivity and specificity of the formula for assessing the risk of MS were 96.99%,90.44%.Combined with its sensitivity and specificity,0.874 is the critical point of the model.The area under the ROC curve(AUC)of the model was 0.922(95%CI:0.896-0.943),which was statistically significant(P < 0.01).The sensitivity and specificity of the formula for assessing the risk of MS were 92.25%,79.51%.Combined with its sensitivity and specificity,0.7177 is the critical point of the model.7.The ROC curve was drawn by using the combination of the fat factor and theliver enzyme factor and 4 kinds of MS components and potential predictors.The above three combinations of AUC in males were 55.5%,91.2%,and 92.2%,respectively,women were 86.2%,90.4%,97.3%.There is strongest predictive power in the potential predictors.