Abstract: Objective - - （ ） To investigate the influencing factors of high frequency noise induced hearing loss NIHL in young ， Methods maleworkersexposedtonoise andtoestablishariskpredictionmodelforNIHL. Theyoungmaleswhowerenewly recruitedandwithnormalhearinginnoiseoperationsinanautomobilemanufacturingenterprisefrom 2016to2018wasselected - asresearchsubjectusingaretrospectivecohortstudy.Afollow upsurveywasconducted.MultivariateCoxproportionalhazards， regressionmodelwasusedtoanalyzetheinfluencingfactorsofNIHL andthepartialregressioncoefficientwasusedtoconstruct （PI） PI ， - （PI< ） - theprognosticindex model.Accordingtothe value thesubjectsweredividedintolow riskgroup 0.00 andhigh （PI ） - Results ， risk group ≥0.00 .Kaplan Meiermethod wasused toanalyzetherisk ofNIHL. By 2020 theincidenceofNIHL （ ）， - ， was 33.4% 167/500 which was pure high frequency NIHL and no speech frequency NIHL was detected. Cox proportional - hazard regression model analysis showed that the risk of high frequency NIHL was 2.03 times higher than those with （P< ）， occupational noise exposure level exceeding the standard 0.01 1.62 times higher in those with drinking than in those （P< ）， （P< without drinking 0.05 and 1.69 times higher in those with noisy living environment than in those without drinking ） PI 0.05 . The model constructed according to the partial regression coefficients of the above three influencing factors is PI x x x - （ ， =−0.44+0.71 occupational intensity + 0.49 drinking + 0.52 The probability of high frequency NIHL 3.9% ， ， ， ） - ， ， ， ， 12.3% 19.4% 30.4% 51.1% in the high risk group at 12 24 36 48 and 52 months after taking office was higher than - （ ， ， ， ， ）， that in the low risk group 2.3% 8.3% 10.7% 15.3% and 27.5% especially in the third year after taking office.