1.Application of random forest algorithm and logistic regression in predicting the risk of macrosomia
Xuan LIU ; Ruiyi LIU ; Yimin QU ; Yuping WANG ; Yu JIANG
Journal of Public Health and Preventive Medicine 2022;33(3):17-21
		                        		
		                        			
		                        			Objective  To establish macrosomia risk prediction models based on a cohort study, and to analyze and compare the results.  Methods  The research subjects were the pregnant women of the Chinese Pregnant Women Cohort Study. The general demographic information and clinical data of pregnant women were collected through the questionnaire and physical examination, and the related outcomes of newborns were obtained by follow-up. The dataset was divided into training set and test set by a 3:1 ratio. Multivariate logistic regression analysis (LR) and random forest algorithm (RF) were used to construct macrosomia risk prediction models in the training set, and the models were verified in the test set. The prediction efficiency of the models was evaluated by Kappa and the area under the receiver operating characteristic curve (ROC).  Results  Among 5544 pregnant women, 397 women delivered macrosomia, and the incidence of macrosomia was 7.16%. Among the pregnant women who delivered macrosomia, 10.08% (40/397) were over 35 years old, 27.46% (109/397) were overweight or obese, and 60.96% (242/397) were excessive gestational weight gain (GWG). LR was used to establish a macrosomia risk prediction model to predict the test set, with the accuracy of 0.716, the sensitivity of 0.719, the specificity of 0.715, the Kappa value of 0.428, the Yoden index of 0.393, and the AUC of 0.796 (95% CI: 0.777-0.815). RF was used to construct a risk prediction model to predict the test set, with the accuracy of 0.819, the sensitivity of 0.782, the specificity of 0.846, the Kappa value of 0.629, the Yoden index of 0.439, and the AUC of 0.897 (95% CI: 0.883-0.910).  Conclusion  The prediction effect of the two models is satisfactory. The random forest algorithm has a higher predictive effect on the risk of macrosomia in this cohort, but the multivariate logistic regression analysis can directly explain the influencing factors of the macrosomia. It is suggested to integrate the advantages of the two models in the future, so that they can play a more important role in macrosomia risk prediction.
		                        		
		                        		
		                        		
		                        	
2.Brief introduction of population medicine and value-based healthcare
Xuan LIU ; Yongle ZHAN ; Hexin YUE ; Yimin QU ; Yu JIANG ; Muir GRAY
Chinese Journal of Epidemiology 2021;42(5):923-927
		                        		
		                        			
		                        			Three healthcare revolutions and four medical paradigm shifts have had a profound impact on the development of healthcare system, which has greatly improved human health, however, the COVID-19 pandemic has exposed hidden dangers and problems in the construction of the healthcare system. In this paper, we made a brief introduction of population medicine and value-based healthcare for the purpose of suggesting new ideas and directions for the future development of healthcare system.
		                        		
		                        		
		                        		
		                        	
3. The spatial distribution and epidemic trend of silicosis in Guangdong province
Xudong LI ; Hongying QU ; Xianzhong WEN ; Hongwei TU ; Yan YUAN ; Hongwei YU ; Zhiting LIU ; Shanyu ZHOU ; Yazhou QI ; Yimin LIU
Chinese Journal of Industrial Hygiene and Occupational Diseases 2018;36(9):662-666
		                        		
		                        			 Objective:
		                        			To investigate the characteristics, temporal trend of silicosis, and provide basis for risk assessment and precise prevention and control of occupational diseases.
		                        		
		                        			Methods:
		                        			Using descriptive statistics to analyze the reported cases of silicosis by SPSS 20.0 software. Reported silicosis cases, the constituent ratio, the incidence age and the working age at onset were analyzed by a linear trend test. Analyzing the variation trends of regional, industry, economic type and enterprise scale distributions by the chi-square trend test. Moreover, using Moran's I method for spatial autocorrelation analysis and trend-surface analysis.
		                        		
		                        			Results:
		                        			(1) During 2006 to 2015, Guangdong province had reported 1, 428 cases of silicosis, mainly gathered in Foshan, Zhongshan, Guangzhou, Shenzhen, which included 1391 male cases accounting for 97.41%. And the average incidence age was 45 (39, 51) . The average working age of onset was 9 (5.5, 15) . In economic type distribution, the private economy took the main part, accounting for 59.1%. In enterprise scale distribution, it was dominated by small and medium enterprises (SMEs) , accounting for 32.4% and 37.3% respectively. In industry distribution, most cases were gathered in materials and mining industry, accounting for 32.1% and 22.9% respectively. (2) The number of silicosis cases, the incidence age and the working age of onset showed a rising trend (
		                        		
		                        	
4. Combined effects of noise exposure and SOD gene polymorphism on susceptibility of occupational noise-induced hearing loss
Xudong LI ; Hongying QU ; Yimin LIU ; Qianling ZHENG ; Xianzhong WEN ; Shijie HU ; Lin CHEN
China Occupational Medicine 2018;45(03):290-296
		                        		
		                        			
		                        			 OBJECTIVE: To explore the association of interaction of noise exposure and superoxide dismutase( SOD) gene polymorphism on the susceptibility of occupational noise-induced hearing loss( ONHL) in Chinese Han population.METHODS: A simple random sampling method was used to select 2 400 Han workers as study subjects.These workers were exposed to 75.0-120.0 dB( A) of normalized continuous A-weighted sound with pressure level equivalent to a 40 hworking-week( L_(EX,W)) in Guangzhou City.A model method was set up to define 201 sensitive workers( sensitive group)and 202 resistant workers( resistant group) by combining results with hearing pure tone threshold test and the cumulative noise exposure( CNE).The genomic DNA was extracted from peripheral blood cells.The single nucleotide polymorphisms analysis of SOD was carried out by using the TaqMan probe with chemical fluorescence allelic identification test.Binary logistic regression method was used to analyze the interaction of noise exposure and SOD.RESULTS: After adjusting confounding factors such as gender,heavy metal exposure,dust exposure and high temperature exposure,the risk for ONHL in individuals interactively carry rs2040724 AG or GG and rs4880( CC + CT) was significantly higher than that in individuals interactively carry rs2040724 AA and rs4880 TT( P < 0.05).The risk for ONHL in individuals interactively carry rs10432782 GT or GG and rs4880( CC + CT) was significantly higher than that in individuals interactively carry rs10432782 TT and rs4880 TT( P < 0.05).The risk for ONHL in individuals exposed to L_(EX,W)≥85 dB( A) and interactively carried with rs2040724 GG and rs4880( CC + CT) was significantly higher than that in individuals exposed to L_(EX,W)< 85 dB( A) and interactively carry rs2040724 AA and rs4880 TT( P < 0.01).The risk for ONHL in individuals exposed to L_(EX,W)≥85 dB( A) and interactively carry rs10432782 GT or GG and rs4880( CC + CT) was significantly higher than that in individuals exposed to L_(EX,W)< 85 dB( A) and interactively carry r10432782 TT and rs4880 TT( P < 0.05).CONCLUSION: In Chinese Han population,noise exposure intensity and SOD1( rs2040724,rs10432782),SOD2( rs4880)gene polymorphism had interacting effects on ONHL susceptibility. 
		                        		
		                        		
		                        		
		                        	
5.Relationship between pesticide exposure and adverse pregnancy outcomes among reproductive couples in rural areas of China
Yimin QU ; Shi CHEN ; Juanjuan LI ; Rongrong JIN ; Hui PAN ; Yu JIANG
Chinese Journal of Epidemiology 2017;38(6):732-736
		                        		
		                        			
		                        			Objective To analyze the association between pesticide exposure and adverse pregnancy outcomes in women from the rural areas of China.Methods Data of "National Free Preconception Health Examination Project (NFPHEP)" from January 2010 to December 2012 was used for analysis.A total of 248 501 families that were planning to deliver a baby in the next 6 months were enrolled.Data on paternal exposure to pesticides before or during pregnancy was collected through questionnaires,with related outcomes on pregnancy recorded by doctors.Results Among all the 248 501 participants,1 806 (0.74%) women and 2 653 (1.09%) men reported to have been exposed to pesticide before pregnancy,with 505 (0.21%) reported of having been exposed to pesticide during the period of pregnancy.Maternal exposure to pesticide was found a risk factor related to stillbirth (OR=3.37,95 % CI:2.05-5.55),peculiar smell pregnancy (OR =3.17,95%CI:1.18-8.55) and low birth weight (OR=7.56,95% CI:5.36-10.66).Paternal exposure to pesticide was also found related to miscarriage (OR=1.37,95% CI:1.03-1.80),low birth weight (OR=3.65,95% CI:1.51-8.84),or giant infant (OR?=0.64,95% CI:0.44-0.93).Maternal exposure to pesticide during pregnancy appeared a risk factor on miscarriage (OR=4.65,95% CI:3.47-6.24).Other adverse outcomes on pregnancy would include premature birth and high birth weight.Conclusion Parental pesticide exposure appeared a risk factor on stillbirth,peculiar smell pregnancy,low birth weight and miscarriage.
		                        		
		                        		
		                        		
		                        	
6.Current status of folic acid supplementation in women at child-bearing age in suburban area of Beijing
Rongrong JIN ; Yimin QU ; Juanjuan LI ; Shuai MA ; Liangkun MA ; Yu JIANG
Chinese Journal of Epidemiology 2017;38(10):1347-1351
		                        		
		                        			
		                        			Objective To access the current status of folic acid supplementation in women at childbearing age in suburban area of Beijing,2016.Methods In this cross-sectional study,a total of 722 women aged 17-41 years from six hospitals in suburban areas of Beijing were selected through cluster sampling.A questionnaire survey was conducted.Results The rates of folic acid intake and regular folic acid intake among women under survey were 63.9% (456/714) and 83.3% (380/456).In women who remembered the duration of folic acid supplementation,the overall rate of preconceptional folic acid supplementation was 35.5% (159/450),in whom 27.3% (123/450) began to take folic acid 0-3 months before pregnancy;8.0% (36/450) began to take folic acid 3 months before pregnancy.Multivariate logistic regression analysis indicated that household registering type,preparation for pregnancy or pregnancy,parity,smoking and alcohol use were the influencing factors for folic acid supplementation.Conclusions The rate of folic acid supplementation remained low in childbearing aged women in suburban area of Beijing.It is necessary to further strengthen the health education about the importance of folic acid supplementation to improve the intake rate and the regular intake rate of folic acid in pregnant women and to promote maternal infant health and reduce the related complications.
		                        		
		                        		
		                        		
		                        	
7.Application of autoregressive integrated moving average model in predicting the reported notifiable communicable diseases in China
Zhongzhou SHEN ; Shuai MA ; Yimin QU ; Yu JIANG
Chinese Journal of Epidemiology 2017;38(12):1708-1712
		                        		
		                        			
		                        			Objective To develop the models for predicting the reported legally notifiable diseases in China.Autoregressive integrated moving average (ARIMA) model was applied to forecast the trend of diseases.Methods Cases used for building the model were from of the records of Notifiable Infectious Diseases in China from May 2009 to July 2016 with R software and the model's predictive ability was tested by the data from August 2016 to January 2017.Results A strong seasonal nature was seen in the reported cases of notifiable communicable diseases,with the lowest point in February and highest peak in June.ARIMA (4,1,0) (1,1,1)12 model was established by the team to forecast the notifiable communicable diseases.Data showed that the biggest and lowest relative errors appeared as 9.78% and 2.21%,respectively,with the mean of the relative error as 5.39%.Conclusion Based on the results of this study,the ARIMA (4,1,0) (1,1,1)12 model seemed to have had the sound prediction of notifiable communicable diseases in China.
		                        		
		                        		
		                        		
		                        	
8.Analysis on the hospitalization costs differences and its effects on health outcome among the stroke patients
Yingnan CAO ; Tianjia GUAN ; Jinlei LI ; Yimin QU ; Ban WANG ; Yuanli LIU
Chinese Journal of Health Policy 2016;9(5):54-60
		                        		
		                        			
		                        			Objective:The paper is designed to reveal differences in stroke patients'hospitalization costs crea-ted by different socioeconomic factors , health insurance and occupational background , and analyze those differences on health outcome .Methods:18879 cases of hospitalized patients with stroke in five tertiary hospitals from four prov-inces from 2011 to 2014 were included in the database .Descriptive statistics was used to describe patients'hospitali-zation costs and health outcome .Multiple linear regression model and logistic regression were used to evaluate the im-pact of insurance and occupation .Results:After controlling for patients'social characteristics , health risks , premise during hospitalization process and clinical characteristics of patients with free healthcare costs , the socialized medi-cine was charged 19.7% higher than the private ones ( P <0.001 ), the retired people and civil servants were charged 4.2%(P<0.001) and 2.9%(P=0.049) more than farmers.The socialized medicine was also associated with health outcome.The risk of death in patients with free medical care is (OR=4.901) compared with private pa-tients (95%CI 1.652~14.537), and the retired people had increased risk of death compared with farmers (OR=2 .145 , 95%CI 1 .287 ~3 .573 ) .Conclusions: Due to the impact of social background factors , some groups are found to have a higher level of expenses than their counterparts during hospitalization , but the more hospitalization costs are not spent the better the health outcome of stroke was achieved .
		                        		
		                        		
		                        		
		                        	
9.Nutritional intake analysis of hospital diets in elderly inpatients in Peking Union Medical College Hospi-tal
Yanping LIU ; Shanshan LI ; Yuanyuan BAO ; Xiaotong YANG ; Wenjing WANG ; Rui LI ; Kang YU ; Yimin QU
Chinese Journal of Clinical Nutrition 2016;24(2):101-105
		                        		
		                        			
		                        			Objective To investigate the nutritional management level of Department of Clinical Nutrition in elderly inpatients through analyzing the actual hospital dietary intakes in elderly inpatients in Peking Union Medical College Hospital from 2013-2015.Methods Using continuous sampling, the nutrients contents and the eating rates of hospital meals ordered by the elderly inpatients on the 1st, 15th, 29th of every month from May 2013 to October 2015 were selected.The intake levels of energy, protein, calcium, vitamin A, vitamin B1 , and vitamin C were calculated and compared with the Chinese dietary reference intakes (DRIs).The nutritional in-take differences between diabetes diet and the basic hospital diet were also compared.Results A total of 90 days of hospital dietary data from 8 402 elderly inpatients were included in the study.The average eating rate was (50.1 ±4.2)%, which was lower than the general eating rate of the hospital [ (59.0 ±4.0)%, P<0.001]. Only protein intake from hospital diet reached the requirement target [male (103.1 ±47.3)%, femal (98.3 ± 33.8)%].Individual compliance rates were 63.2%and 59.8%, respectively, while the other kinds of nutrients were lower.The compliance rates of energy [ (73.3 ±26.3)%vs.(62.1 ±38.2)%, P<0.001] and ma-jor nutrient intake [ protein: ( 119.1 ±41.2 )% vs.( 93.3 ±65.1 )%, P <0.001; calcium: ( 55.5 ± 26.7)% vs.(34.3 ±34.2)%, P <0.001; vitamin A: (75.2 ±48.3)% vs.(57.4 ±97.1)%, P<0.001;vitamin B1:(76.4 ±38.2)%vs.(52.1 ±46.6)%, P<0.001;vitamin C:(92.2 ±51.4)%vs. (49.3 ±55.0)%, P<0.001) in diabetes diet group were significantly higher than those in the basic hospi-tal diet group.Conclusions The nutritional intakes from hospital diets may not meet the nutritional require-ments based on DRIs in elderly inpatients.Medical diet designed by physicians and dietitians should be ex-panded to improve the nutritional management level for these patients.
		                        		
		                        		
		                        		
		                        	
10.Relationship between maternal exposure factors and small for gestational age infant
Rongrong JIN ; Juanjuan LI ; Yimin QU ; Liangkun MA ; Yu JIANG
Chinese Journal of Epidemiology 2016;37(9):1306-1310
		                        		
		                        			
		                        			Through literature study,this paper summarizes the current status ofreseareh on the relationship between maternal exposure factors and small for gestational age infant both in China and abroad.And at the same time,the relationship between pregnancy outcome of small for gestational age and maternal exposure factors was analyzed.This might contribute to reduce the incidence of small for gestational age infant and provide important guidance and references for perinatal care.
		                        		
		                        		
		                        		
		                        	
            

Result Analysis
Print
Save
E-mail