| Purpose:By analyzing the basic demographic characteristics and occupational health related data of dust-exposed workers in small,medium and micro enterprises in Jiangxi Province,the occupational health status of dust-exposed workers was understood,the dust hazard situation in small,medium and micro enterprises in Jiangxi Province was recognized,potential pneumoconiosis patients were found,and the key care groups of pneumoconiosis were explored.Through in-depth study on the dynamic changes,regularity and characteristics of pneumoconiosis,an effective time sequence model was constructed to monitor and warn the occurrence of pneumoconiosis more accurately,and to provide data support for the prevention and treatment of pneumoconiosis in Jiangxi Province.Methods:Using the judgment sampling method,5665 dust exposure workers in Jiangxi Province were selected as research objects to analyze the anterior and posterior X-ray chest radiographs(hereinafter referred to as "chest radiographs")and lung function abnormalities and their influencing factors in 2021.The SARIMA prediction model of pneumoconiosis was constructed using the data of new cases of pneumoconiosis in Jiangxi Province from 1975 to 2017,and the data from 2018 to 2019 were used as model validation data for verification.The fitting value of the SARIMA model of pneumoconiosis was taken as the input value of the combined model,and the real data was taken as the output value.The SARIMA-GRNN combined model was constructed for fitting and prediction.The mean absolute error(MAE),mean absolute percentage error(MAPE)and root mean square error(RMSE)were used to evaluate the model fitting and prediction effect.Results:(1)The abnormal rate of chest radiographs was 5.1%(287/5665),among which96 patients(1.7%)had pneumoconiosis changes,and the abnormal rate of lung function was 26.7%(1511/5665).(2)The results of a special chest X-ray examination showed that the number of small enterprises was the largest,accounting for 61.4%,and 60.9% were workers from private enterprises.The abnormal detection rate of male chest X-ray was higher than that of females,and the abnormal rate of chest X-ray increased with the increase of age and working age.The abnormal detection rate of chest X-ray was 8.1% for the type of dust exposure,and the mining industry had the highest abnormal detection rate.The abnormal rate of chest radiographs in Hong Kong,Macao,Taiwan,and foreign enterprises was the highest(16.7%).(3)Among the dust exposure types,the number of workers exposed to silica dust and other types of dust accounted for a relatively high proportion,38.2% and 45.5%,respectively.The abnormal function rate of silica pneumoconiosis was the highest,40.5%.Among different industries,the manufacturing industry had the largest number of inspectors,with a total of 3,032,accounting for 53.5%;Among different enterprise sizes,small enterprises had the largest number of patients examined,accounting for61.4%,and micro-enterprises had the highest detection rate of abnormal lung function,accounting for 37.1%.In the sex distribution,the difference between men and women is not obvious;The abnormal lung function was proportional to age and working years,and the abnormal lung function rate was the highest in patients over 60 years old,with an abnormal rate of 32.7%.(4)Gender,economic type,age,enterprise size and dust exposure type were all independent factors leading to abnormal chest radiographs of dust exposure workers(P<0.05).The risk of abnormal chest radiographs in male dust exposed workers was higher than that in female dust exposed workers(P<0.05).In the dust exposure type,the risk of abnormal chest radiograph of the workers exposed to electric welding dust was 4.963 times that of the workers exposed to silica dust.Workers exposed to dust in Hong Kong,Macao,Taiwan and foreign enterprises had a higher risk of abnormal chest radiographs than those exposed to dust in private enterprises(P<0.05).The older the age,the higher the risk of abnormal chest radiographs(P<0.05).(5)Age,working age,enterprise scale,economic type,industry type and dust exposure type were all independent factors leading to abnormal lung function of dust exposure workers(P<0.05).The longer the age of workers,the higher the risk of abnormal lung function(P<0.05);Workers in small and micro enterprises had higher risk of abnormal lung function than workers exposed to dust in medium-sized enterprises(P<0.05).The risk of abnormal lung function in manufacturing workers was higher than that in mining workers exposed to dust(P<0.05).Workers exposed to dust in Hong Kong,Macao,Taiwan and foreign enterprises had higher risk of abnormal lung function than workers exposed to dust in private enterprises(P<0.05).Among dust exposure types,silica dust exposure workers had the highest risk of abnormal lung function(P<0.05).(6)The optimal model was determined according to the AIC value of each model,and SARIMA(1,1,1)(0,1,1)4 was finally determined as the optimal model.The residual error of this model passed the white noise test,and the Box-Jenkins Q statistic was 9.31(P=0.68).The value of the smooth factor starts from 0.001,and increases by one unit each time from 0.001 to 0.2.The selected points to be estimated are predicted,and the mean square error(MSE)value of the error sequence of predicted samples and real samples is calculated.When the smooth factor is 0.003,the MSE value reaches the minimum,so the value is 0.003.The relative errors of the SARIMA model and SARIMA GRNN combined model are 6.13% and 4.74%,respectively.Conclusion:(1)Through the single factor analysis,it was found that the gender,age,working years,economic type,and dust exposure type of dust exposure workers were all influencing factors leading to abnormal chest radiographs,and the abnormal rate of chest radiographs increased with the increase of age and working years.(2)The incidence of lung injury among dust workers is related to a variety of factors,including age,working age,industry type,dust exposure type,enterprise size,and economic type.Strengthen the supervision and management of key groups and industries to reduce the occurrence of diseases.(3)From 1975 to 2019,the incidence of pneumoconiosis in Jiangxi Province showed an overall downward trend,and the seasonal trend was obvious.The overall results showed that the reporting rate in the first quarter of each year was the lowest.(4)By using the SARIMA-GRNN combined model,we can extract the linear and nonlinear components in the time series more effectively and with higher accuracy.The SARIMA-GRNN combined model has a better effect in predicting the incidence of pneumoconiosis in Jiangxi Province,and the prediction results can provide relevant data support for the prevention and control of pneumoconiosis. |
PDF Full Text Request