Exposure characteristics of complex industrial noise and its risk assessment model for hearing loss
Objective To study the exposure characteristics of complex industrial noise (CIN) and its risk assessment model for hearing loss. Methods 10-15 typical enterprises with serious noise hazards or noise deafness in the past 5 years were selected,and all noise-exposed positions (equivalent A-weighted sound level≥80 dB) in each enterprise were included in the study,and at least 3-5 workers (all included if<3 workers) in each noise-exposed position were selected as the measurement subjects,and a total of 1200 cases of workers were included. Conducted a survey to record and analyze clinical data on noise-induced hearing loss (NIHL) among workers,performed univariate and multivariate logistic regression analysis,and constructed a model and performed receiver operating characteristic (ROC) curve analysis on the predictive performance. Results The NIHL detection rates of workers with noise intensity ≥95 dB(A),kurtosis ≥7,cumulative noise exposure (CNE)≥100 dB(A)·year and kurtosis-corrected CNE (CNE') ≥105 dB(A)·year were higher than those with noise intensity<95 dB(A),kurtosis<7,CNE<100 dB(A)·year and CNE'<105 dB(A)·year (P<0.05). There was no statistically significant difference in the comparison of the incidence of NIHL among patients with different ages,gender,education levels,marital status,body mass index (BMI) levels and with or without diabetes,hypertension,smoking,alcohol consumption,and heat exposure (P>0.05). With NIHL as the dependent variable and noise intensity[<95 dB(A)=0;≥95 dB(A)=1],kurtosis (<7=0;≥7=1),CNE[<100 dB(A)·year=0;≥100 dB (A)·year=1],CNE'[<105 dB(A)·year=0;≥105 dB(A)·year=1]as independent variables,Logistic regression analysis was performed,and found that noise intensity ≥95 dB(A),kurtosis ≥7,CNE ≥100 dB(A)·year,and CNE'≥105 dB(A)·year were all risk factors for the occurrence of NIHL in the workers (OR=2.471,2.033,3.357,3.112>1,P<0.05). According to the regression coefficients of the variables obtained from the Logistic regression as well as the constant term in order to construct the risk assessment model of NIHL for the workers,the main formulas were:Prob=ea/(1+ea)×100％,where e was an exponential function,and a=0.905×noise intensity[<95 dB(A)=0;≥95 dB(A)=1]+0.710×kurtosis (<7=0;≥7=1)+1.211×CNE[<100 dB(A)·year=0;≥100 dB(A)·year=1]+1.135×CNE'[<105 dB(A)·year=0;≥105 dB(A)·year=1]-8.933. A fitting degree test was implemented for the above model,and the application of the H-L test revealed that x2=3.671,P=0.530>0.05,indicating that there was a better fit between the risk probabilities predicted by this model and the actual probabilities. ROC test of the above prediction model and the fitting effect generated by NIHL showed that the sensitivity was 70.36％,the specificity was 74.66％,and the area under the curve (AUC) was 0.790[95％CI=(0.688,0.869);P=0.000<0.05],the critical Prob value of the model was 0.452. Conclusion The exposure characteristics of CIN are mainly high noise intensity,high kurtosis,high CNE and high CNE',and noise intensity ≥95 dB(A),kurtosis ≥7,CNE ≥100 dB(A)·year and CNE'≥105 dB(A)·year are the risk factors for the occurrence of NIHL in workers,and the risk assessment model constructed by them has good predictive effect.
Complex industrial noiseExposure characteristicsNoise-induced hearing lossRisk assessment model
万方数据
维普
