OBJECTIVE Otoacoustic emissions (OAE) and auditory brainstem responses (ABR) are tests widely used in neonatal hearing screening. This study aimed to investigate the differences and clinical value of distortion product otoacoustic emissions (DPOAE) and ABR in hearing screening of high-risk neonates admitted to a neonatal intensive care unit (NICU). METHODS DPOAE and ABR were measured with the Smart-OAE analyser and the Smart-EP brain-stem electric response audiometry apparatus, respectively, in 600 high-risk neonates (1,200 ears). The testing results of DPOAE and ABR were compared. RESULTS Of the 600 neonates (1,200 ears), the incidence of ABR abnormality (78.6%, 943/1,200) was remarkably higher than that of DPOAE abnormality (22.3%, 268/1,200). Two hundred and forty-one ears (20.8%) were negative and 252 (21%) were positive in both DPOAE and ABR tests. A total of 707 ears (58.9%) presented with a discordant result in DPOAE and ABR. The false positive and false negative rates of the DPOAE test were 6.0% (16/268) and 74.1% (691/932) respectively. CONCLUSIONS In high-risk neonates the diagnostic value of DPOAE for identification of hearing loss, when used alone, is limited. The ABR test appears to be more reliable for hearing screening in high-risk neonates. It is suggested that hearing screening for high-risk neonates should be conducted with ABR first, followed by OAE after failure on ABR.