Clinical Interpretation of Word-Recognition Scores for Listeners with Sensorineural Hearing Loss: Confidence Intervals, Limits, and Levels.

OBJECTIVES: Confidence levels were established to determine if a word-recognition score is within the expected range for a hearing loss group (based on 3-frequency pure-tone average) or significantly below or above the expected range. DESIGN: Two large clinical databases were mined to produce data-sets composed of word-recognition scores obtained with Q/MASS NU-6 materials and VA NU-6 materials for patients with average hearing losses ranging from 0 to 70 dB HL. Percentiles representing scores that are below (2.5%, 5%, and 10%) and above (90%, 95%, and 97.5%) the expected range (defined as the 80% confidence interval) were established. To estimate the distribution of scores and percentiles for the Auditec NU-6 materials (for which a large database is not available) Q/MASS scores were transformed to Auditec scores based on published psychometric functions. RESULTS: The resulting confidence levels and expected ranges of word-recognition scores should be useful for interpreting the relationship between a score and the distribution of scores for the patient's hearing loss severity. Confidence levels are described as low, moderate, and high corresponding to the statistical level of assurance that a score is lower or higher than the expected score. CONCLUSIONS: The confidence levels and expected ranges may be helpful for interpreting word-recognition scores obtained with three widely used sets of NU-6 test materials.

Non-linguistic auditory speech processing.

OBJECTIVES: A method for testing auditory processing of non-linguistic speech-like stimuli was developed and evaluated. DESIGN: Monosyllabic words were temporally reversed and distorted. Stimuli were matched for spectrum and level. Listeners discriminated between distorted and undistorted stimuli. STUDY SAMPLE: Three groups were tested. The Normal group was comprised of 12 normal-hearing participants. The Senior group was comprised of 12 seniors. The Hearing Loss group was comprised of 12 participants with thresholds of at least 35 dB HL at one or more frequencies. RESULTS: The Senior group scored lower than the Normal group, and the Hearing Loss group scored lower than the Senior group. Scores for forward compressed speech were slightly higher than backward compressed speech but the difference was not statistically significant. Retest scores were slightly higher than scores on the first test, but the difference was not statistically significant. CONCLUSIONS: Large differences in discrimination of distorted speech were observed among the three groups. Age and hearing loss separately affected performance. The depressed performance of the Senior group may be a result of "hidden hearing loss" that is attributed to cochlear synaptopathy. The backward-distorted speech task may be a useful non-linguistic test of speech processing that is language independent.

Ambient Noise Monitoring during Pure-Tone Audiometry.

BACKGROUND: There is an increasing need to administer hearing tests outside of sound-attenuating rooms. Maximum permissible ambient noise levels (MPANLs) from published in standards (Occupational Health and Safety Administration [OSHA] 1983; American National Standards Institute [ANSI] S3.1-1999 (R2018)) can be modified to account for the additional attenuation provided by circumaural earphones (relative to supra-aural earphones) that are used for pure-tone audiometry. Ambient noise can influence the results of pure-tone audiometry by elevating thresholds by direct masking and by producing distractions that affect the accuracy of the test. The effects of these distractions have not been studied in relation to pure-tone audiometry in adult listeners. PURPOSE: In Part I MPANLs provided by ANSI and OSHA standards are extended to account for the greater attenuation provided by circumaural earphones. Rules ("alerts") were developed taking into account the listeners' thresholds. In Part II effects of distracting noise on pure-tone thresholds are reported. METHODS AND RESULTS: In Part I MPANLs two standards were modified for circumaural earphones by adding the additional attenuation provided by three circumaural earphones (relative to supra-aural earphones). A set of rules ("alerts") is provided for identifying masking effects from ambient noise in a variety of conditions (earphone type, threshold elevation, uncovered ear). In Part II the distracting effects of an industrial noise sample on thresholds obtained from five listeners with normal hearing are described. Pure-tone thresholds were measured in quiet and in distracting noise presented at various levels. The effects of the distracting noise on the following variables were measured: time per trial, number of trials required to measure threshold, threshold shift, and perceived distractibility of the noise. Time per trial was unaffected by distracting noise. Number of trials required for threshold, threshold shift, and perceived distractibility increased with distracting noise level. CONCLUSION: Part I: The modified MPANLs provide more relevant determinations of the potential effects of ambient noise on pure-tone thresholds than the values in the standards. Part II: Distracting noise affects pure-tone threshold measurements in a manner that is different from direct masking. The potential contaminating effect of distracting noise can be measured and reported.

Hearing Health Care Utilization Following Automated Hearing Screening.

BACKGROUND: The study examined follow-up rates for pursuing hearing health care (HHC) 6 to 8 months after participants self-administered one of three hearing screening methods: an automated method for testing of auditory sensitivity (AMTAS), a four-frequency pure-tone screener (FFS), or a digits-in-noise test (DIN), with and without the presentation of a 2-minute educational video about hearing. PURPOSE: The study aims to determine if the type of self-administered hearing screening method (with or without an educational video) affects HHC follow-up rates. RESEARCH DESIGN: The study is a randomized controlled trial of three automated hearing screening methods, plus control group, with and without an educational video. The control group completed questionnaires and provided follow-up data but did not undergo a hearing screening test. STUDY SAMPLE: The study sample includes 1,665 participants (mean age 50.8 years; 935 males) at two VA Medical Centers and at university and community centers in Portland, OR; Bay Pines, FL; Minneapolis, MN; Mauston, WI; and Columbus, OH. DATA COLLECTION AND ANALYSIS: HHC follow-up data at 6 to 8 months were obtained by contacting participants by phone or mail. Screening methods and participant characteristics were compared in relation to the probability of participants pursuing HHC during the follow-up period. RESULTS: The 2-minute educational video did not have a significant effect on HHC follow-up rates. When all participants who provided follow-up data are considered (n = 1012), the FFS was the only test that resulted in a significantly greater percentage of HHC follow-up (24.6%) compared with the control group (16.8%); p = 0.03. However, for participants who failed a hearing screening (n = 467), follow-up results for all screening methods were significantly greater than for controls. The FFS resulted in a greater probability for HHC follow-up overall than the other two screening methods. Moreover, veterans had higher follow-up rates for all screening methods than non-veterans. CONCLUSION: The FFS resulted in a greater HHC follow-up rate compared with the other screening methods. This self-administered test may be more motivational for HHC follow-up because participants who fail the screening are aware of sounds they could not hear which does not occur with adaptive assessments like AMTAS or the DIN test. It is likely that access to and reduced personal cost of audiological services for veterans contributed to higher HHC follow-up rates in this group compared with non-veteran participants.

Automated Forced-Choice Tests of Speech Recognition.

PURPOSE: This project was undertaken to develop automated tests of speech recognition, including speech-recognition threshold (SRT) and word-recognition test, using forced-choice responses and computerized scoring of responses. Specific aims were (1) to develop an automated method for measuring SRT for spondaic words that produces scores that are in close agreement with average pure-tone thresholds and (2) to develop an automated test of word recognition that distinguishes listeners with normal hearing from those with sensorineural hearing loss and which informs the hearing aid evaluation process. METHOD: An automated SRT protocol was designed to converge on the lowest level at which the listener responds correctly to two out of two spondees presented monaurally. A word-recognition test was conducted with monosyllabic words (female speaker) presented monaurally at a fixed level. For each word, there were three rhyming foils, displayed on a touchscreen with the test word. The listeners touched the word they thought they heard. Participants were young listeners with normal hearing and listeners with sensorineural hearing loss. Words were also presented with nonrhyming foils and in an open-set paradigm. The open-set responses were scored by a graduate student research assistant. RESULTS: The SRT results agreed closely with the pure-tone average (PTA) obtained by automated audiometry. The agreement was similar to results obtained with the conventional SRT scoring method. Word-recognition scores were highest for the closed-set, nonrhyming lists and lowest for open-set responses. For the hearing loss participants, the scores varied widely. There was a moderate correlation between word-recognition scores and pure-tone thresholds which increased as more high frequencies were brought into the PTA. Based on the findings of this study, a clinical protocol was designed that determines if a listener's performance was in the normal range and if the listener benefited from increasing the level of the stimuli. CONCLUSION: SRTs obtained using the automated procedure are comparable to the results obtained by the conventional clinical method that is in common use. The automated closed-set word-recognition test results show clear differentiation between scores for the normal and hearing loss groups. These procedures provide clinical test results that are not dependent on the availability of an audiologist to perform the tests.

Home Hearing Test: Within-Subjects Threshold Variability.

BACKGROUND: The Home Hearing Test (HHT) is an automated pure-tone threshold test that obtains an air conduction audiogram at five test frequencies. It was developed to provide increased access to hearing testing and support home telehealth programs. PURPOSE: Test and retest thresholds for 1000-Hz stimuli were analyzed to determine intrasubject variability from two independent data sets. RESEARCH DESIGN: Prospective, repeated measures. STUDY SAMPLE: In the Veterans Affairs (VA) study, results from 26 subjects 44 to 88 years of age (mean = 65) recruited from the Nashville VA audiology clinic were analyzed. Subjects were required to have a Windows PC in the home and were self-reported to be comfortable with using computers. Two subjects had normal hearing, and 24 had hearing losses of various severities and configurations. The National Center for Rehabilitative Auditory Research (NCRAR) sample included 100 subjects (68 males; 32 females) with a complaint of hearing difficulty recruited from the local community and Veteran population. Subjects ranged in age from 32 to 87 years (mean = 63.7 years). They were tested in a quiet room at the NCRAR. DATA COLLECTION AND ANALYSIS: Subjects in the VA study were provided kits for installing HHT on their home computers. HHT was installed on a computer at NCRAR to test subjects in the NCRAR study. HHT obtains a five-frequency air conduction audiogram with a retest of 1000 Hz in both ears. Only the 1000-Hz test-retest results are analyzed in this report. Six statistical measures of test-retest variability are reported. RESULTS: Test and retest thresholds were highly correlated in both studies (r ≥ 0.96). Test-retest differences were within ±5 dB ≥92% of the time in the two studies. Standard deviations of absolute test-retest difference were ≤3.5 dB in the two studies. CONCLUSIONS: Intrasubject variability is comparable to that obtained with manual testing by audiologists in sound-treated test rooms.

Validation of the Home Hearing Test™.

BACKGROUND: The Home Hearing Test™ (HHT) is an automated pure-tone threshold test that obtains an air-conduction audiogram at five test frequencies. It was developed to provide increased access to hearing testing and support home telehealth programs. PURPOSE: The study was conducted as part of an audiology telehealth trial based at the U.S. Department of Veterans Affairs Tennessee Valley Healthcare System, Nashville, TN. Air-conduction audiograms obtained by the HHT were compared to results obtained in the clinic. RESEARCH DESIGN: Prospective, repeated measures. STUDY SAMPLE: Twenty-eight participants, aged 44-88 yr (mean = 65) were recruited from the Nashville U.S. Department of Veterans Affairs audiology clinic. Participants were required to have a Windows personal computer in the home and were self-reported to be comfortable with using computers. Two participants had normal hearing and 26 had hearing loss of various severities and configurations. DATA COLLECTION AND ANALYSIS: Audiograms were obtained in the audiology clinic by experienced audiologists following standard clinical protocols. Participants were provided with a kit for installing HHT on their home computers. The HHT air-conduction audiogram is obtained with Automated Method for Testing Auditory Sensitivity (AMTAS(®)), described in previous publications. Threshold pairs (clinic versus HHT) were analyzed by determining distributions of threshold differences and absolute differences. These were compared to distributions of differences between manual threshold pairs obtained by two audiologists and AMTAS(®) versus manual threshold pairs obtained under laboratory conditions. RESULTS: Threshold differences (clinic versus HHT) were slightly larger than differences between thresholds obtained by two audiologists and AMTAS(®) versus manual threshold differences obtained under laboratory conditions. The differences were not statistically significant. CONCLUSIONS: HHT air-conduction audiograms agree well with audiograms obtained in the clinic. HHT is well suited to home telehealth applications and personal use.

Distribution Characteristics of Air-Bone Gaps: Evidence of Bias in Manual Audiometry.

OBJECTIVES: Five databases were mined to examine distributions of air-bone gaps obtained by automated and manual audiometry. Differences in distribution characteristics were examined for evidence of influences unrelated to the audibility of test signals. DESIGN: The databases provided air- and bone-conduction thresholds that permitted examination of air-bone gap distributions that were free of ceiling and floor effects. Cases with conductive hearing loss were eliminated based on air-bone gaps, tympanometry, and otoscopy, when available. The analysis is based on 2,378,921 threshold determinations from 721,831 subjects from five databases. RESULTS: Automated audiometry produced air-bone gaps that were normally distributed suggesting that air- and bone-conduction thresholds are normally distributed. Manual audiometry produced air-bone gaps that were not normally distributed and show evidence of biasing effects of assumptions of expected results. In one database, the form of the distributions showed evidence of inclusion of conductive hearing losses. CONCLUSIONS: Thresholds obtained by manual audiometry show tester bias effects from assumptions of the patient's hearing loss characteristics. Tester bias artificially reduces the variance of bone-conduction thresholds and the resulting air-bone gaps. Because the automated method is free of bias from assumptions of expected results, these distributions are hypothesized to reflect the true variability of air- and bone-conduction thresholds and the resulting air-bone gaps.

Distribution characteristics of normal pure-tone thresholds.

OBJECTIVE: This study examined the statistical properties of normal air-conduction thresholds obtained with automated and manual audiometry to test the hypothesis that thresholds are normally distributed and to examine the distributions for evidence of bias in manual testing. DESIGN: Four databases were mined for normal thresholds. One contained audiograms obtained with an automated method. The other three were obtained with manual audiometry. Frequency distributions were examined for four test frequencies (250, 500, 1000, and 2000 Hz). STUDY SAMPLE: The analysis is based on 317 569 threshold determinations of 80 547 subjects from four clinical databases. RESULTS: Frequency distributions of thresholds obtained with automated audiometry are normal in form. Corrected for age, the mean thresholds are within 1.5 dB of reference equivalent threshold sound pressure levels. Frequency distributions of thresholds obtained by manual audiometry are shifted toward higher thresholds. Two of the three datasets obtained by manual audiometry are positively skewed. CONCLUSIONS: The positive shift and skew of the manual audiometry data may result from tester bias. The striking scarcity of thresholds below 0 dB HL suggests that audiologists place less importance on identifying low thresholds than they do for higher-level thresholds. We refer to this as the Good enough bias and suggest that it may be responsible for differences in distributions of thresholds obtained by automated and manual audiometry.

Distribution of hearing loss characteristics in a clinical population.

OBJECTIVES: This study was designed to characterize a large database of audiometric records from an academic health center Audiology clinic for the purpose of determining the distributions of hearing loss configuration, severity, and site of lesion. DESIGN: : Using AMCLASS, a validated tool for classifying audiograms described in a previous report, the audiometric configuration, severity, and site of lesion was determined for patients grouped according to the completeness of the audiometric record. Complete air conduction testing at six octave frequencies in both ears was required for inclusion. Patients were grouped according to the quantity of bone conduction thresholds in the record: both ears tested, one ear tested, or neither ear tested. All other records were discarded leaving 23,798 records of 16,818 patients for analysis. All analyses were conducted for all remaining records and with repeat tests excluded. RESULTS: The effect of removing repeat audiograms had remarkably little effect on the distributions. Sloping hearing losses dominated the distributions of configuration. One-third of all records indicated normal hearing in at least one ear and one fourth had normal hearing in both ears. Mild and moderate hearing losses were equally prevalent, each contributing 40% to 45% of the cases with hearing loss. Sensorineural was the most prevalent site of lesion, representing about 45% of cases and just over half the cases with hearing loss. CONCLUSIONS: Prevalence rates of hearing loss configurations, severities, and sites of lesion are provided against which analyses of other databases can be compared. The results may be useful for counseling patients regarding the relationship of their hearing loss to that of a large population. The high number normal-hearing people in the database suggests that estimates of the need for hearing testing based on prevalence of hearing loss may underestimate the number of people who seek or are referred for hearing evaluation.

Asymmetric hearing loss: definition, validation, and prevalence.

HYPOTHESIS: An algorithm for identifying asymmetric hearing loss (AHL) can be constructed that performs as well or better than expert judges. BACKGROUND: AMCLASS is a method for classifying audiograms based on configuration, severity, site of lesion, and interaural asymmetry. The development and clinician validation for all but asymmetry were reported separately. In this report, an algorithm for identifying AHL is described. Using the clinician-validated algorithm, the prevalence of AHL in a database from an academic health center audiology clinic was analyzed. METHODS: : Five expert clinicians classified 199 audiograms as symmetric or asymmetric. Interjudge agreement was analyzed for each pair of judges and between each judge and the consensus of the panel. An algorithm was constructed based on the set of rules that maximized agreement between AMCLASS and judges. Using the clinician-validated algorithm, the prevalence of AHL was analyzed for groups based on quantity of bone conduction testing, hearing loss configuration, severity, and site of lesion. RESULTS: There was substantial disagreement among judges that was similar to interjudge comparisons for other medical tests. Average agreement between AMCLASS and the judges was higher than agreement between the best judge and the consensus of the judges. Approximately 50% of all patients and 55% of patients with sensorineural hearing loss were classified as AHL by the clinician-validated algorithm. CONCLUSION: The algorithm met the goal of equaling or exceeding the performance of expert judges. The prevalence of AHL was higher than expected and suggests that the algorithm is not useful for screening for acoustic neuroma or other conditions. Perhaps, a criterion based on the magnitude of the asymmetry would better serve that purpose. The symmetry category provided by AMCLASS provides a determination of clinically significant AHL that agrees with the consensus of expert judges.

Toward a standard description of hearing loss.

Hearing losses are frequently described by categories that characterize the configuration, severity, and site of lesion from a pure-tone audiogram. Although many category descriptors are in common use, there are no standard definitions of those terms, nor have the category definitions been validated against current clinical practice. The development and validation of AMCLASStrade mark is described. To validate the classification method, five expert judges selected configuration, severity, and site of lesion categories for 231 audiograms that varied widely in audiometric configuration. Interjudge comparisons indicated that expert judges frequently disagree on how they describe an audiogram. Category definitions were adjusted to maximize agreement between AMCLASStrade mark and the consensus of the judges. The final set of category definitions produced categories that agreed with the consensus more often than the average agreement between pairs of judges.

Qualind: A method for assessing the accuracy of automated tests.

As audiology strives for cost containment, standardization, accuracy of tests, and accountability, greater use of automated tests is likely. Highly skilled audiologists employ quality control factors that contribute to test accuracy, but they are not formally included in test protocols, resulting in a wide range of accuracy, owing to the various skill and experience levels of clinicians. A method that incorporates validated quality indicators may increase accuracy and enhance access to accurate hearing tests. This report describes a quality assessment method that can be applied to any test that (1) requires behavioral or physiologic responses, (2) is associated with factors that correlate with accuracy, and (3) has an available independent measure of the dimension being assessed, including tests of sensory sensitivity, cognitive function, aptitude, academic achievement, and personality. In this report the method is applied to AMTAS, an automated method for diagnostic pure-tone audiometry.