Establishment and verification of risk prediction model of non-alcoholic fatty liver disease for helicopter flying personnel
10.3760/cma.j.cn113854-20230523-00050
- VernacularTitle:直升机飞行人员非酒精性脂肪肝发病风险预测模型的建立与验证
- Author:
Hongwei XU
1
;
Junsong LIU
;
Yunyue XIAO
;
Yanchao TANG
;
Yanmei QIAN
Author Information
1. 空军杭州特勤疗养中心疗养三区健康管理科，杭州 310002
- Publication Type:Journal Article
- Keywords:
Fatty liver;
Models, statistical;
Nomograms;
ROC curve;
DCA curve;
Helicopter flying personnel
- From:
Chinese Journal of Aerospace Medicine
2023;34(4):198-204
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To establish and verify a nomogram model for predicting the risk of non-alcoholic fatty liver disease (NAFLD) in helicopter flying personnel.Methods:The helicopter flying personnel who recuperated at Air Force Healthcare Center for Special Services Hangzhou were selected as the study subjects. They were divided into the training group (75%) and the verification group (25%) by using the cluster random sampling method for internal verification. Multivariate Logistic regression analysis was used to analyze the factors influencing NAFLD of helicopter flying personnel, and the nomogram model was constructed by R4.2.1 software and rms package. The H-L goodness-of-fit test and receiver operating characteristic (ROC) curve were used to assess the discrimination and calibration of the prediction model, and decision curve analysis (DCA) was used to evaluate the clinical practicality of the model.Results:A total of 2 195 helicopter flying personnel were included, with 1 648 in the training group and 547 in the verification group. Seven variables, including age, BMI, hyperuricemia, hypertriglyceridemia, abnormal high-density lipoprotein, daily intakes of milk and/or its products and years of smoking were used as the predictors of the nomogram model for the risk prediction of NAFLD in helicopter flying personnel. The H-L goodness-of-fit tests of training group and verification group were χ2=8.54, 11.03, P=0.382, 0.200, respectively, indicating a good predictive accuracy. The area under the ROC curve of the 2 groups was 0.840 and 0.860, respectively, indicating a certain degree of discrimination. The incidence risk probability of DCA curve in 2 groups was about 10%-30%, and the net return rate was above 0, suggesting that it had a certain clinical applicability. Increased age ( OR=1.18, P=0.010), overweight and obesity ( OR=3.67, P<0.001), hyperuricemia ( OR=2.12, P<0.001), hypertriglyceridemia ( OR=2.96, P<0.001), abnormal high-density lipoprotein cholesterol ( OR=1.83, P=0.010), and increased years of smoking ( OR=1.20, P=0.010) were the risk factors for NAFLD in helicopter flying personnel, while increased daily intakes of milk and/or its products ( OR=0.76, P=0.020) were identified as a protective factor. Conclusions:The nomogram model based on the aforementioned factors can effectively evaluate the incidence risk of NAFLD in helicopter flying personnel, and has certain clinical practicality.
