Audiometric configuration as a reflection of low plasma glucose and diabetes.

Forty-five subjects with a rising, progressively improving (RPI) audiometric pattern were considered for a 5-hour oral-glucose tolerance test. Results revealed that 2 nondiabetic subjects (4%) had a plasma glucose nadir of 49 mg/dl or lower and 7 nondiabetic subjects had a nadir between 50 and 57 mg/dl. A total of 21% of 42 subjects with data indicated abnormally low readings. An additional 7 subjects were found to be diabetic; 5 subjects had impaired glucose tolerance, and 9 other subjects had nondiagnostic glucose intolerance, totalling 47% of 45 subjects in the study with abnormally elevated plasma glucose levels. Therefore a total of approximately 68% of all subjects in our study indicated metabolic dysfunction based specifically on a RPI pattern.

Effects of white noise on the most comfortable level for speech with normal listeners.

This study assessed the effects of noise level on the most comfortable level (MCL) for discursive speech. Normal young adults (N:25) were instructed to search thoroughly around their MCL by Bekesy tracking. Mean MCL in quiet was 82.5 db SPL (S.D. = 11.1 db) which was statistically indistinguishable from group MCLs in noises of 55 and 70 db SPL. However, MCL rose to 90.9 and 100.3 db in noises of 85 and 100 db SPL, respectively. These mean MCLs were somewhat higher than observed previously and may be due to differences in instructions and procedures in threshold-tracking.

Loudness discomfort level for speech: comparison of two instructional sets for saturation sound pressure level selection.

This study was undertaken to compare the speech loudness discomfort levels (LDL's) with two instructional sets which have been proposed for saturation sound pressure level selection of hearing aids. The phraseology recommended by McCandless and by Berger was presented to normal-hearing and hearing-impaired listeners. The normal-hearing subjects obtained mean LDL's of 94.6 and 111.9 dB SPL for these respective instructions, which was statistically significant. The hearing-impaired listeners also showed LDL's with Berger's instructions (114.7 dB SPL) to be significantly higher than with McCandless' instructional set (109.3 dB SPL). Consequently, this investigation suggests that these two instructional sets may lead to substantially different saturation sound pressure levels. Further studies are needed to determine the most appropriate phraseology for LDL measurement, including the assessment of speech intelligibility at various saturation sound pressure levels. Another instructional set was constructed which (1) includes an explanation to patients of the purpose and importance of the test, (2) requests listeners to indicate the upper level they are "willing" to listen as opposed to the level they are "able" to listen, (3) instructs patients to search thoroughly around their LDL before making a final judgment, and (4) contains a statement that the LDL should be made with the understanding that the speech could be listened to for a period of time. Whatever instructions are used, clinicians are advised to interpret their LDL's very cautiously until validational studies are available.

Comparison of speech detection and spondee thresholds and half- versus full-list intelligibility scores with MLV and taped presentations of NU-6.

Several speech audiometric measurements were made on 212 ears with mild sensorineural hearing loss. An 8-dB difference between speech detection and spondee thresholds was observed, which is the same relationship that has been found in normal ears. No significant differences in speech discrimination scores (SDS's) were observed when NU-6 was administered via monitored live voice (MLV) and the Auditec recordings. Although our data support the use of MLV testing, verification with a standardized recording should be considered when unusually poor SDS's are obtained. Half-list and full-list SDS's were analyzed for both taped and MLV presentation modes. This analysis showed that both the MLV and taped stimuli exhibited very similar variability and that about 96% of the half-list scores were within 6% of the full-list scores. The clinician should be cautious, however, because 4% of the ears had half-list/full-list discrepancies ranging from 8 to 14% and differences as large as 28% have been reported by Raffin and Thornton (1977). Furthermore, variability between half-list and full-list SDS's varies as a function of intelligibility impairment, being least for scores approaching the extremes of 0 and 100% and greatest for scores in the 30 to 70% range. Finally, our data suggest that half-list testing can be an effective screening procedure to determine it full-list testing is advisable.