| Objective:To understand the basic information,occupation-related situation of dust operators in Changchun City and the detection of individual indexes of lung function of dust exposure personnel.This paper explores the relationship between lung function injury of dust workers in Changchun City and its related influencing factors,builds logistic regression model and decision tree model respectively,analyzes the influencing factors of lung function injury of dust workers in Changchun City,and compares the effects,advantages and disadvantages of the two models.To provide reference for the prevention and control of occupational diseases in Changchun,to find a better prediction model for the prevention and control of lung function injury,and to provide a new method for the early intervention of pneumoconiosis.Method:Using the data of occupational health examination of dust workers in Changchun,a total of 2373 subjects were included according to the inclusion and exclusion criteria determined in this study.Descriptive analysis method was used to analyze the basic information and lung function injury of dust workers.Quantitative data conforming to normal distribution were used mean ± standard deviation,while quantitative data not conforming to normal distribution were used median ± quartile distance.Qualitative data are expressed as relative numbers.The Chi-square test was used to compare the distribution differences between groups of classified variables,and the logistics regression analysis was used to analyze the influencing factors of lung function injury in dust workers.The influence of the size of small and medium-sized enterprises on lung function injury and the influence of dust exposure type on lung function injury were analyzed by stratification of the years of service with statistical significance in multivariate regression.The risk prediction model of lung function injury in dust workers was constructed by logistics regression and decision tree method,and the prediction effect of different models was evaluated and compared.Results:(1)Basic information of dust handling personnelIn this study,the physical examination reports of dust handling personnel in various districts of Changchun City were included.The researchers were 2373 dust handling personnel of various jobs,of which 2274 were male,accounting for 95.8%,and 99 were female,accounting for 4.2%.The main reason was that men were mostly engaged in this kind of work,and the subjects included were mainly male.The subjects ranged in age from 21 to 71 years old,with an average age of 49±8years old.Of all the subjects,227 had high blood pressure,accounting for 9.6percent,and 260 had abnormal electrocardiograms,accounting for 11 percent.The average length of service of dust receiving personnel is 10±9 years.The enterprises to which the researchers belong are mostly medium-sized enterprises with 2,052 employees,accounting for 86.5% of the total number of employees;small enterprises have 289 employees,accounting for 12.2%;large enterprises have the least,with only 32 employees,accounting for only 1.3%.The number of registered enterprises is mainly state-owned enterprises,with 1,932 people,accounting for81.4%.The number of private enterprises and foreign-funded enterprises is 388 people and 53 people,accounting for 16.4% and 2.2% respectively.In terms of industry classification,the subjects are mainly engaged in mining industry with1,817 people,accounting for 76.6%,followed by power and heat supply industry with 220 people accounting for 9.3%,building materials industry with 92 people accounting for 3.9%,non-ferrous metal smelting industry with 90 people accounting for 3.8%,automobile manufacturing industry with 81 people accounting for 3.4%.Food and drug processing industry 73 people accounted for 3.1%.In this study,coal dust was the main type of dust exposure,with 1,965 persons exposed to dust,accounting for 82.8%;silicon dust exposure occurred in 374 persons,accounting for15.8% of all researchers;asbestos workers were the least exposed,accounting for 34 persons,accounting for 1.4%.(2)Detection of individual indexes of lung function of dust-exposed personnelAmong 2373 physical examination reports,3 cases of pneumoconiosis were detected,accounting for 0.11%.There were 181 cases of chest radiographic abnormalities detected,accounting for 7.6%.In terms of the abnormal detection rate of single index of lung function injury of dust exposed workers,84 cases of abnormal FEV1 were detected,accounting for 3.5%,the highest abnormal rate;84cases of abnormal FVC were detected,accounting for 3.5%;and 26 cases of abnormal FEV1/FVC were detected,accounting for 1.1%.Among the pulmonary dysfunction,the number of restricted pulmonary dysfunction was the largest,83 people,accounting for 3.5%,the number of obstructive pulmonary dysfunction was25 people,accounting for 1.1% of the total number,and the number of mixed pulmonary dysfunction was 1 person.(3)Influencing factors of lung function injuryAmong the 2373 dust exposed workers who participated in physical examination,109 people were found to have abnormal lung function,with an abnormal rate of 4.6%.In chest X-ray examination,a total of 2373 people participated in physical examination,with a total of 181 people with abnormal chest X-ray,accounting for 7.6%.The results of this study showed that the age of exposure to dust,type of exposure to dust,industry,enterprise size,enterprise registration type,and blood pressure of coal dust workers in Changchun all independently affected the lung function injury of dust workers.Age,age of dust exposure,dust exposure type,dust exposure industry and enterprise size,electrocardiogram all independently affect whether the chest X-ray is abnormal.The results of multi-factor analysis showed that working age,enterprise size,industry name and dust exposure type jointly affected the lung function injury of dust workers.In addition,age and working age jointly affected the chest X-ray injury of dust workers.(4)Hierarchical analysis of occupation-related factors for dust operatorsThe risk of lung function impairment in small enterprises with less than five years of working experience was 8.7 times that of medium-sized enterprises,and the risk of lung function impairment in dust-exposed workers in small enterprises with more than five years of working experience was 3.3 times that of medium-sized enterprises.The risk of lung function impairment was 2.819 times that of coal dust workers with less than 5 years of working experience.After adjusting for multiple confounding factors(gender,age,enterprise size,registration type,industry,electrocardiogram and blood pressure),there was no significant difference in lung function impairment between coal dust workers and silicon dust workers with over 5years of working experience.(5)Prediction model of lung function injury in dust workersAs can be seen from the prediction results of decision tree model,three explanatory variables are screened out,which are industry,enterprise size and dust exposure years.Industry is the root node,and the next best predictive variables are enterprise size and industry.logistic regression model also showed that the type of dust exposure also affected the lung function of dust exposure workers,but this factor was removed from the decision tree model,which also reflected the difference between the two models.By comparing the two methods,it can be seen that the evaluation efficiency of the two models is close,and the final effect of the two models is better.However,the logistic regression model focuses on the interdependence of variables,while the decision tree plays a guiding role in the study of the interaction of variables.The decision tree can show the function form of each variable in the group in detail.If the two are combined,the prediction effect will be better,and the influencing factors of abnormal lung function can be found more efficiently and accurately.Conclusion:(1)Among the 2373 results,pneumoconiosis changes were detected in 3 cases,accounting for 0.11%;abnormal chest radiographs were detected in 181 cases,accounting for 7.6%;abnormal lung function was detected in 109 cases,accounting for 4.6%,all within the normal range reported in China.(2)Among the specific types of pulmonary dysfunction,the number of patients with restrictive pulmonary dysfunction was the largest,while the number of patients with mixed pulmonary dysfunction was the least.(3)The lung function injury of dust workers in Changchun was significantly affected by related occupational hazard factors.With the increase of working years,the risk of lung function injury of dust workers increased,and the abnormal rate of lung function and chest radiographs increased.In small businesses,workers exposed to dust have the greatest risk of lung damage.Of all industries,automobile manufacturing has the highest risk of lung injury.The abnormal rate of lung function varies with different types of dust exposure.(4)The constructed decision tree model and logistic model have similar evaluation effectiveness,and both have good final effect.The logistic regression prediction model has high sensitivity and the decision tree model has high specificity.Therefore,if the two are combined,the prediction effect will be better and the influencing factors of abnormal lung function can be found more efficiently and accurately. |
PDF Full Text Request