OBJECTIVETo establish the criteria of the disproportionate loss of Mandarin monosyllable discriminative abilities to pure tone hearing thresholds.

METHODSTotal of 165 patients with varying degrees of sensorineural hearing loss were recruited for routine audiological evaluations. The speech discrimination scores were obtained by Mandarin phonemic-balanced monosyllable lists via self-made speech audiometric software. The Performance-Intensity (P-I) function for individual ear was obtained by the same list which was administrated in ascending intensities, with 25 monosyllables presenting randomly. The lowest intensity was determined by the lowest pure tone threshold among all audiometric frequencies minus 5 dB. The intensities were increased in 5 dB step until the score was 100% or the intensity was reached to the patient's uncomfortable level. The PB(max) was obtained from the P-I plot. Three parameters about pure tone average hearing thresholds, including PTA(1) (average of 0.5, 1 and 2 kHz), PTA(2) (average of 1, 2 and 4 kHz) and PTA(3) (average of 0.5, 1, 2 and 4 kHz), as well as three parameters about audiogram slope, including Slope(0.5) (4 kHz minus 0.5 kHz), Slope(1) (4 kHz minus 1 kHz) and Slope(2) (4 kHz minus 2 kHz), were calculated respectively. The correlations between PB(max) and above parameters were analyzed by SPSS10.0 statistical software.

RESULTSThe audiogram slopes were not shown any correlation with PB(max), while the pure tone average thresholds, especially PTA(3) (r = -0.595, P = 0.000) were confirmed to correlate with PB(max). In the scatter plot based on PB(max) and PTA(3), a linear boundary was constructed encompassing approximately 99% of observed data collected from the sensorineural hearing-impaired.

CONCLUSIONAny PB(max) score falling below the boundary should be considered with high possibility and disproportionately poor comparison with pure tone hearing thresholds.