Factors influencing community prehypertension and establishment of an incidence risk prediction model
Objective To explore the risk factors of community prehypertension,and to develop and validate a visualization evaluation tool which can assist grassroots general practitioners in predicting the risk of prehypertension.Methods From September 2021 to September 2022,residents of a street in Urumqi City who participated in the national health examination were selected as the research subjects.A large-scale questionnaire survey,health checkups and laboratory tests were conducted to collect the surveyed subjects'general and biochemical data.According to the subjects'blood pressure status,3,324 patients with hypertension were excluded,the remaining 9,879 subjects were divided into the normal blood pressure group and the prehypertension group,and their baseline data were analyzed by single factor analysis.After strict data filtering and preprocessing,the above-mentioned 9,879 subjects were randomly divided into the training group(n=6,586)and the verification group(n=3,293)in a ratio of 2:1,and then the subjects in the training group were subgrouped into the normal blood pressure group and the prehypertension group according to their blood pressure levels,with prehypertension as the outcome variable.Multivariate logistic regression analysis was performed to explore and establish a nomogram prediction model,and the model was verified by the data of the verification group.Based on the training group and the verification group,the area under the curve(AUC)for the receiver operator characteristic(ROC)analysis,calibration curve-based analysis and decision curve analysis(DCA)were employed to evaluate the identification ability,accuracy and applicability of the nomogram prediction model.Results A total of 13,203 valid questionnaires were retrieved in this study,including 5,599 subjects withi normal blood pressure,4,280 subjects with prehypertension and 3,324 subjects with hypertension.The detection rates of prehypertension and hypertension were 32.42％(4,280/13,203)and 25.18％(3,324/13,203)respectively.The results of multivariate logistic regression analysis displayed that fasting blood glucose(OR=1.29,95％C/:1.07-1.54),total cholesterol(OR=2.68,95％CI:2.06-3.48),low density lipoprotein(OR=2.75,95％CI:2.15-3.52),hyperuricemia(OR=1.56,95％CI:1.29-1.88),central obesity(OR=1.66,95％CI:1.21-2.29),salt intake(OR=1.30,95％CI:1.27-1.33),smoking(OR=1.88,95％CI:1.56-2.26),high density lipoprotein(OR=0.33,95％CI:0.21-0.54),thousand steps equivalent per day(OR=0.59,95％CI:0.57-0.61)and body mass index(BMI)were independent factors influencing the occurrence of prehypertension.The risk of developing prehypertension in subjects with BMI 24.0 was 1.52 times that of subjects with normal BMI(95％CI:1.25-1.84)and 8.46 times that of subjects with low BMI(95％CI:6.67-10.72).The nomogram prediction model for prehypertension was established based on the above-mentioned influencing factors selected by multivariate logistic regression analysis.The AUC of the nomogram prediction model for prehypertension in the training group was 0.895(95％CI:0.888-0.903),and that in the validation group was 0.892(95％CI:0.881-0.904).The nomogram prediction model was found with a high goodness of fit by the Hosmer-Lemeshow test(P＞0.05).DCA showed that when the subjects'threshold probability was 0-0.9,using the nomogram prediction model resulted in higher net benefit of predicting the risk of prehypertension.Conclusion We successfully established and verified a nomogram prediction model(with the predictive variables like fasting blood glucose,total cholesterol,low density lipoprotein,high density lipoprotein,hyperuricemia,central obesity,BMI,salt intake,smoking and thousand steps equivalent per day)with a high accuracy in this study,which is conducive to improving grassroots general practitioners'abilities to identify and screen high-risk patients with prehypertension.
NETL
NSTL
万方数据
