[Establishment of a nomogram for predicting the high frequency hearing loss of workers exposed to noise]

Objective: To explore the related influencing factors of high frequency hearing loss (HFHL) in workers exposed to noise and establish a prediction nomogram for HFHL. Methods: A total of 822 workers exposed to noise from 46 enterprises were included. A questionnaire survey and a pure-tone hearing test were conducted for the workers. The data of noise level of the workers exposed was also collected. After single factor analysis of related influencing factors, the multivariate Logistic regression analysis was performed to identify the final independent influencing factors of HFHL. Finally, a nomogram model was established by R software to achieve individual prediction of HFHL. Results: Among the 822 workers exposed to noise, 166 (20.2%) workers had HFHL. In multivariate Logistic regression analysis, increasing age, men, increasing wearing earphone time, less wearing earplugs, and high noise level were the independent risk factors for HFHL. The C-index of the nomogram model for predicting HFHL was 0.834 (95%CI: 0.748~0.903) . The area under the predictive power curve of nomogram model was 0.834 (95%CI: 0.799~0.869, P<0.001) . Conclusion: Age, sex, wearing earphone time, wearing earplugs, and noise level are independent influence factors for HFHL. The nomogram model is successfully established as a accurate and visible tool for individually predicting the HFHL risk in workers exposed to noise.