The frequency specificity of the ABR threshold evoked by a 1000-Hz filtered click was determined in subjects with sloping cochlear hearing losses, both high- and low-frequency in character. The results show that the ABR threshold evoked by this stimulus is low-frequency specific. The standard error in estimating the 1000-Hz pure-tone threshold (PTT) is 10.4 dB, which equals that for estimating the 3000-Hz PTT from the routinely used click-evoked ABR threshold. The ABR threshold evoked by a 1000-Hz filtered click can therefore be regarded as an accurate tool to predict the pure-tone hearing loss at 1000-Hz. In comparison with the ABR threshold evoked by a click masked with 1590-Hz high-pass noise, the ABR threshold evoked by a 1000-Hz filtered click has a larger dynamic range, yields a larger number of useful responses and is less time consuming. For clinical low-frequency-specific ABR threshold assessment, the 1000-Hz filtered click is therefore preeminently useful.