Starkey Research & Clinical Blog

Evidence for the Value of Real-Ear Measurement

Abrams, H.B., Chisolm, T.H., McManus, M., & McArdle, R. (2012). Initial-fit approach versus verified prescription: Comparing self-perceived hearing aid benefit. Journal of the American Academy of Audiology, 23(10), 768-778.

Audiology best practice guidelines state that probe microphone verification measures should be done to ensure that hearing aid gain and output characteristics meet prescribed targets for the individual. In the American Academy of Audiology’s Guidelines for the Audiologic Management of Adult Hearing Impairment, an expert task force recommends that “prescribed gain from a validated prescriptive method should be verified using a probe microphone approach that is referenced to ear canal SPL” (Valente, et al., 2006). Similarly, the Academy’s Pediatric Amplification Protocol (AAA, 2003) states that hearing aid output characteristics should be verified with real-ear measures or with real-ear-to-coupler-difference (RECD) calculations when lengthy adjustments subsequent to real-ear measurement are not possible.

In contrast to these recommendations, the majority of hearing aid providers are not routinely conducting real-ear verification measures. In a survey of audiologists and hearing instrument specialists, Mueller and Picou (2010) found that respondents used real-ear verification only about 40% of the time and Bamford (2001) reported that only about 20% of individuals fitting pediatric patients used real-ear measures. The reasons most often cited for skipping probe microphone measures are based on financial, time, or space constraints.

When probe microphone measures are not conducted, other verification techniques may be used such as aided word recognition, but these not likely to provide reliable information (Thornton & Raffin, 1978).  Or, verification may not be attempted at all, with fitting parameters being chosen based on the manufacturer’s initial-fit specifications. Although most fitting software allows for entry of age, experience and acoustic information such as canal length and venting characteristics, their predictions are based on average data and cannot account for individual ear canal effects.

Numerous studies have shown that initial-fit algorithms often deviate significantly from prescribed targets, usually underestimating required gain, especially in the high frequencies. Hawkins & Cook (2003) found that simulated fittings from one manufacturer’s initial-fit algorithm over-estimated the coupler gain and in-situ response by as much as 20dB, especially in the low and high frequencies.  Bentler (2004) compared the 2cc coupler response from six different hearing aids programmed with initial-fit algorithms and found that the responses were different for each manufacturer and deviated from prescriptive targets by as much as 15dB, usually falling below prescribed targets. Similarly, Bretz (2006) studied three manufacturers’ pediatric first-fit algorithms and found that the average output varied by about 20dB and initial-fit gain values were below both NAL-NL1 and DSL (i/o) targets. This is of particular concern because pediatric patients may be less able than adults to provide subjective responses to hearing aid settings, rendering objective measures such as real-ear verification even more important.

These studies and others illuminate the potential difference between first-fit hearing aid settings and those verified by objective measures, but it is not well known how this affects the user’s perceived benefit.  Some early reports using linear amplification targets indicated that verification did not predict perceived benefit (Nerbonne et al., 1995; Weinstein et al., 1995), but more recent work indicates that adults fit to DSL v5.0a targets demonstrated benefit as measured by the Client Oriented Scale of Improvement (COSI, Dillon & Ginis, 1997). A recent survey by Kochkin et al. (2010) found that patients whose fittings were verified with a comprehensive protocol including real-ear verification reported increased hearing aid usage, benefit and satisfaction. Furthermore, these respondents were more likely to recommend their hearing care professional to friends and family than were the respondents who were not fitted with real-ear verification.

The purpose of the study discussed here was to determine if perceived hearing aid benefit differed based on whether the user was fitted with an initial-fit algorithm only or with modified settings based on probe-microphone verification. Twenty-two experienced hearing aid users with mild to moderately-severe hearing loss participated in the study. All were fitted with binaural hearing aids, though a variety of hearing aid styles and manufacturers were represented.  Probe microphone measurements were conducted on all subjects, but  those in the initial-fitting group did not receive adjustments based on the verification measures.

Perceived hearing aid benefit was measured using the Abbreviated Profile of Hearing Aid Benefit (APHAB, Cox & Alexander, 1995). The APHAB consists of 24 items in four subscales: ease of communication (EC), reverberation (RV), background noise (BN) and aversiveness of sounds (AV).  In addition to subscale scores, an average global score can be calculated, as well as a benefit score which represents the difference between unaided and aided responses.

Prior to being fitted with their hearing aids, participants completed the APHAB questionnaire. Because all were experienced hearing aid users, they were asked to base their answers on their experiences without amplification.  Hearing aid fittings and probe microphone verification were then conducted on all subjects, but half of the subjects received adjustments to match prescribed targets and half of the subjects maintained their first-fit settings. Efforts were made to ensure that subjects were not aware of the difference between the initial-fit and verified fitting methods. The only adjustments that subjects in the initial-fit group received were based on issues that could affect their willingness to wear the hearing aids, such as loudness discomfort or feedback.

One month following the first appointment, subjects returned to the clinic and were administered the APHAB again. They were given their initial “unaided” APHAB responses to use as a comparison. After completion of the APHAB, the subjects who had been fitted with the initial-fit algorithms were switched to verified fittings and those had been fitted to prescribed targets were switched to the manufacturer’s initial-fit settings. All subjects were re-tested with probe microphone measures and those with loudness or feedback complaints received minor adjustments.

One month after the second appointment, subjects returned to complete the APHAB and were again allowed to use their original APHAB responses as a basis for comparison. They were not allowed to view their responses to the APHAB that was administered after the first hearing aid trial. Participants were also asked to indicate which fitting method (Session 1 or Session 2) they preferred and would want permanently programmed into their hearing aids.

Analysis of the probe microphone measurements indicated, not surprisingly, that the verified fittings were more closely matched to prescriptive targets than the fittings based on the first-fit algorithms, even after minor adjustments based on comfort and user preferences.  For three of the APHAB subscales – ease of communication, reverberation and background noise – scores obtained with verified fittings were superior to those obtained with the initial-fit approach and the main effect of fitting approach was found to be statistically significant. There was no interaction between fitting approach and APHAB subscale, indicating that the better outcomes obtained with verified fittings were not related to any specific listening environment.

When asked to indicate their preferred fitting method, 7 of the 22 participants selected the initial-fit approach, whereas more than twice as many subjects, 15 out of 22, selected the verified fitting. For all but 5 subjects, the global difference score on the APHAB predicted their preferred fitting method, and the relationship between global score and final preference was statistically significant.

The findings of this study and of related reports bring up some philosophical and practical considerations for audiologists. One of our primary goals is to provide effective rehabilitation for hearing-impaired patients and this is most often accomplished by fitting and dispensing quality hearing instruments. Clinical and research data repeatedly indicates the importance of probe microphone verification. It serves the best interest of our patients to offer them the most effective fitting approach, so it follows that probe microphone verification measures should be a routine, essential part of our clinical protocol.

The reports that a minority of hearing aid fittings are being verified with real-ear measures indicates that many clinicians are not following recommended best practices. Indeed, Palmer (2009) points out that failure to follow best practice guidelines is a departure from the ethical standards of professional competence. Failure to provide the recommended objective verification for hearing aid fittings does run counter to our clinical goals and as Palmer suggests may even be damaging to our “collective reputation” as a profession.

Philosophical arguments notwithstanding, there are also practical reasons to incorporate real-ear measures into the fitting protocol. In the MarkeTrak VIII survey, Kochkin reported that hearing aid users who received probe microphone verification testing as part of a detailed fitting protocol were more satisfied with their hearing instruments and were more likely to refer their clinician to friends. In the current field of hearing aid service provision, it is important for audiologists to consider ways that they can meaningfully distinguish themselves from online, mail-order and big-box retail competitors. Hearing aid users are becoming well-informed consumers and it is clear that establishing a base of satisfied patients who feel they have received comprehensive, competent care is crucial for growing a private practice. Probe microphone verification is a brief yet effective part of ensuring successful hearing aid fittings and it benefits our patients and our profession to provide this essential service.

References

Abrams, H.B., Chisolm, T.H., McManus, M., & McArdle, R. (2012). Initial-fit approach versus verified prescription: Comparing self-perceived hearing aid benefit. Journal of the American Academy of Audiology, 23(10), 768-778.

American Academy of Audiology (2003). Pediatric Amplification Protocol. www.audiology.org, (accessed 3-3-13).

Bamford,  J., Beresford, D., Mencher, G.(2001). Provision and fitting of new technology hearing aids: implications from a survey of some “good practice services” in UK and USA. In: Seewald, R.C., Gravel, J.S., eds. A Sound Foundation Through Early Amplification: Proceedings of an International Conference. Stafa, Switzerland: Phonak AG, 213–219.

Bentler, R. (2004). Advanced hearing aid features: Do they work? Paper presented at the convention of the American Speech-Language-Hearing Association, Washington, D.C.

Bretz, K. (2006). A comparison of three hearing aid manufacturers’ recommended first fit to two generic prescriptive targets with the pediatric population. Independent Studies and Capstones, Paper 189. Program in Audiology and Communication Sciences, Washington University School of Medicine. http://digitalcommons.wustl.edu/pacs_capstones/189.

Cox, R. & Alexander, G. (1995). The abbreviated profile of hearing aid benefit. Ear and Hearing 16, 176-183.

Dillon, H. & Ginis, J. (1997). Client Oriented Scale of Improvement (COSI) and its relationship to several other measures of benefit and satisfaction provided by hearing aids. Journal of the American Academy of Audiology 8: 27-43.

Hawkins, D. & Cook, J. (2003). Hearing aid software predictive gain values: How accurate are they? Hearing Journal 56, 26-34.

Kochkin, S., Beck, D., & Christensen, L. (2010). MarkeTrak VIII: The impact of the hearing health care professional on hearing aid user success. Hearing Review 17, 12-34.

Mueller, H., & Picou, E. (2010). Survey examines popularity of real-ear probe-microphone measures. Hearing Journal 63, 27-32.

Nerbonne, M., Christman, W. & Fleschner, C. (1995). Comparing objective and subjective measures of hearing aid benefit. Poster presentation at the annual convention of the American Academy of Audiology, Dallas, TX.

Palmer, C.V. (2006). Best practice: it’s a matter of ethics. Audiology Today, Sept-Oct.,31-35.

Thornton, A. & Raffin, M. (1978) Speech-discrimination scores modeled as a binomial variable. Journal of Speech and Hearing Research 21, 507–518.

Valente, M., Abrams, H., Benson, D., Chisolm, T., Citron, D., Hampton, D., Loavenbruck, A., Ricketts, T., Solodar, H. &  Sweetow, R. (2006). Guidelines for the Audiological Management of Adult Hearing Impairment. Audiology Today, Vol 18.

Weinstein, B., Newman, C. & Montano, J. (1995). A multidimensional analysis of hearing aid benefit. Paper presented at the 1st Biennial Hearing Aid Research & Development Conference, Bethesda, MD.

Can Aided Audibility Predict Pediatric Lexical Development?

Stiles, D.J., Bentler, R.A., & McGregor, K.K. (2012). The speech intelligibility index and the pure-tone average as predictors of lexical ability in children fit with hearing aids. Journal of Speech Language and Hearing Research, 55, 764-778.

This editorial discusses the clinical implications of an independent research study and does not represent the opinions of the original authors.

Despite advances in early hearing loss identification, hearing aid technology, and fitting and verification tools, children with hearing loss consistently demonstrate limited lexical abilities compared to children with normal hearing.  These limitations have been illustrated by poorer performance on tests of vocabulary (Davis et al., 1986), word learning (Gilbertson & Kamhi, 1995; Stelmachowicz et al., 2004), phonological discrimination, and non-word repetition (Briscoe et al., 2001; Delage & Tuller, 2007; Norbury, et al., 2001).

There are a number of variables that may predict hearing-impaired children’s performance on speech and language tasks, including the age at which they were first fitted with hearing aids and the degree of hearing loss.  Moeller (2000) found that children who received earlier aural rehabilitation intervention demonstrated significantly larger receptive vocabularies than those who received later intervention.  Degree of hearing loss, which is typically defined in studies by the pure-tone average (PTA) or the average of pure-tone hearing thresholds at 500Hz, 1000Hz, and 2000Hz (Fletcher, 1929), has been significantly correlated with speech recognition (Davis et al., 1986; Gilbertson & Kamhi, 1995), receptive vocabulary (Fitzpatrick et al., 2007; Wake et al., 2005), expressive grammar, and word recognition (Delage & Tuller, 2007) in some studies comparing hearing-impaired children to those with normal hearing.

In contrast, other studies have reported that pure-tone average (PTA) did not predict language ability in hearing-impaired children.  Davis et al. (1986) tested hearing-impaired subjects between five and18 years of age and found no significant relationship between PTA and vocabulary, verbal ability, reasoning, and reading.  However, all subjects scored below average on these measures, regardless of their degree of hearing loss.  Similarly, Moeller (2000) found that age of intervention affected vocabulary and verbal reasoning, but PTA did not.  Gilbertson and Kamhi (1995) studied novel word learning in hearing-impaired children ranging in age from  seven to 10 years and found that neither PTA nor unaided speech recognition threshold was correlated to receptive vocabulary level or word learning.

At a glance, it seems likely that degree of hearing loss should affect language development and ability, because hearing loss affects audibility, and speech must be audible in order to be processed and learned.  However, the typical PTA of thresholds at 500Hz, 1000Hz, and 2000Hz does not take into account high-frequency speech information beyond 2000Hz.  Some studies using averages of high-frequency pure-tone thresholds (HFPTA) have found a significant relationship between degree of loss and speech recognition (Amos & Humes, 2007; Glista et al., 2009).

Because most hearing-impaired children now benefit from early identification and intervention, their pure-tone hearing threshold averages (PTA or HFTPA) might not be the best predictors of speech and language abilities in every-day situations.  Rather, a measure that combines degree of hearing loss as well as hearing aid characteristics might be a better predictor of speech and language ability in hearing-impaired children.  The Speech Intelligibility Index (SII; ANSI,2007), a measure of audibility that computes  the importance of different frequency regions based on the phonemic content of a given speech test, has proven to be predictive of performance on speech perception tasks for adults and children (Dubno et al., 1989; Pavlovic et al., 1986; Stelmachowicz et al., 2000).  Hearing aid gain characteristics can be incorporated into the SII algorithm to yield an aided SII, which has been reported to predict performance on word repetition (Magnusson et al., 2001) and nonsense syllable repetition ability in adults (Souza & Turner, 1999).  Because an aided SII includes the individual’s hearing loss and hearing aid characteristics into the calculations, it better represents how audibility affects an individual’s daily functioning.

The purpose of the current study was to evaluate the aided SII as a predictor of performance on measures of word recognition, phonological working memory, receptive vocabulary, and word learning.  Because development in these areas establishes a base for later achievements in language learning and reading (Tomasello, 2000; Stanovich, 1986), it is important to determine how audibility affects lexical development in hearing-impaired children.  Though the SII is usually calculated based on the particular speech test to be studied, the current investigation used aided SII values based on average speech spectra.  The authors explained that vocabulary acquisition is a cumulative process, and they intended to use the aided SII as a measure of cumulative, rather than test-specific, audibility.

Sixteen hearing-impaired children with hearing aids (CHA) and 24 children with normal hearing (CNH) between six and nine years of age participated in the study.  All of the hearing-impaired children had bilateral hearing loss and had used amplification for at least one year.  All participants used spoken language as their primary form of communication.  Real-ear measurements were used to calculate the aided SII at user settings.  Because the goal was to evaluate the children’s actual audibility as opposed to optimal audibility, their current user settings were used in the experiment whether or not they met DSL prescriptive targets (Scollie et al., 2005).

Subjects participated in tasks designed to assess four lexical domains.  Word recognition was measured by the Lexical Neighborhood Test and Multisyllabic Lexical Neighborhood Test (LNT and MLNT; Kirk & Pisoni, 2000).  These tests each contain “easy” and “hard” lists, based on how frequently the words occur in English and how many lexical neighbors they have.  Children with normal lexical development are expected to show a gradient in performance with the best scores on the easy MLNT and poorest scores on the hard LNT.  Non-word repetition was measured by a task prepared specifically for this study, using non-words selected based on adult ratings of “wordlikeness”.  In the word recognition and non-word repetition tasks, children were simply asked to repeat the words that they heard.  Responses were scored according to the number of phonemes correct for both tasks.  Additionally, the LNT and MLNT tests were scored based on number of words correct.  Receptive vocabulary was measured by the Peabody Picture Vocabulary Test (PPVT-III; Dunn & Dunn, 1997) in which the children were asked to view four images and select the one that corresponds to the presented word.  Raw scores are determined as the number of items correctly identified and norms are applied based on the subject’s age.  Novel word learning was assessed using the same stimuli from the non-word repetition task, after the children were given sentence context and visual imagery to teach them the “meaning” of the novel words.  Their ability to learn the novel words was evaluated in two ways: a production task in which they were asked to say the word when prompted by a corresponding picture and an identification task in which they were presented with an array of four items and were asked to select the item that corresponded to the word that was presented.

On the word recognition tests, the children with hearing aids (CHA) demonstrated poorer performance than the children with normal hearing (CNH) for measures of word and phoneme accuracy, though both groups demonstrated the expected gradient, with performance improving in parallel fashion from the hard LNT test through the easy MLNT test.  There was a correlation between aided SII and word recognition scores, but PTA and aided SII were equally good at predicting performance.

On the non-word repetition task, which requires auditory perception, phonological analysis, and phonological storage (Gathercole, 2006), CHA again demonstrated significantly poorer performance than CNH, and CNH performance was near ceiling levels.  PTA and aided SII scores were correlated with non-word repetition scores.  Beyond the effect of PTA, it was determined that aided SII accounted for 20% of the variance on the non-word repetition task, which was statistically significant.

The receptive vocabulary test yielded similar results; CHA performed significantly worse than CNH and both PTA and aided SII accounted for a significant proportion of the variance.

The only variable that predicted performance on the word learning tasks was age, which only yielded a significant effect on the word production task.  On the word identification task, both the CHA and CNH groups scored only slightly better than chance and there were no significant effects of group or age.

As was expected in this study, children with hearing aids (CHA) consistently showed poorer performance than children with normal hearing (CNH), with the exception of the novel word learning task.  The pattern of results suggests that aided audibility, as measured by the aided SII, was better at predicting performance than degree of hearing loss as measured by PTA.  Greater aided SII scores were consistently associated with more accurate word recognition, more accurate non-word repetition, and larger receptive vocabulary.

Although PTA or HFPTA may represent the degree of unaided hearing loss, because the aided SII score accounts for the contribution of the individual’s hearing aids, it is likely a better representation of speech audibility and auditory perception in everyday situations.  The authors point out that depending on the audiometric configuration and hearing aid characteristics, two individuals with the same PTA could have different aided SIIs, and therefore different auditory experiences.

The results of this study underscore the importance of audibility for lexical development, which in turn has significant implications for further development of language, reading, and academic skills.  Therefore, the early provision of audibility via appropriate and verifiable amplification appears to be an important step in the development of speech and language.  The SII, which is already incorporated into some real-ear systems or is available in a standalone software package, is a verification tool that should be considered a standard part of the fitting protocol for pediatric hearing aid patients.

 

References

American National Standards Institute (2007). Methods for calculation of the Speech Intelligibility index (ANSI S3.5-1997[R2007]). New York, NY: Author.

Amos, N.E. & Humes, L.E. (2007). Contribution of high frequencies to speech recognition in quiet and noise in listeners with varying degrees of high-frequency sensorineural hearing loss. Journal of Speech, Language and Hearing Research 50, 819-834.

Briscoe, J., Bishop, D.V. & Norbury, C.F. (2001). Phonological processing, language and literacy: a comparison of children with mild-to-moderate sensorineural hearing loss and those with specific language impairment. Journal of Child Psychology and Psychiatry 42, 329-340.

Davis, J.M., Elfenbein, J., Schum, R. & Bentler, R.A. (1986). Effects of mild and moderate hearing impairments on language, educational and psychosocial behavior of children. Journal of Speech and Hearing Disorders 51, 53-62.

Delage, H. & Tuller, L. (2007). Language development and mild-to-moderate hearing loss: Does language normalize with age? Journal of Speech, Language and Hearing Research 50, 1300-1313.

Dubno, J.R., Dirks, D.D. & Schaefer, A.B. (1989). Stop-consonant recognition for normal hearing listeners and listeners with high-frequency hearing loss. II: Articulation index predictions. The Journal of the Acoustical Society of America 85, 355-364.

Dunn, L.M. & Dunn, D.M. (1997). Peabody Picture Vocabulary Test – III. Circle Pines, MN: AGS.

Fitzpatrick, E., Durieux-Smith, A., Eriks-Brophy, A., Olds., J. & Gaines, R. (2007). The impact of newborn hearing screening on communications development. Journal of Medical Screening 14, 123-131.

Fletcher, H. (1929). Speech and hearing in communication. Princeton, NJ: Van Nostrand Reinhold.

Gilbertson, M. & Kamhi, A.G. (1995). Novel word learning in children with hearing impairment. Journal of Speech and Hearing Research 38, 630-642.

Glista, D., Scollie, S., Bagatto, M., Seewald, R., Parsa, V. & Johnson, A. (2009). Evaluation of nonlinear frequency compression: Clinical outcomes. International Journal of Audiology 48, 632-644.

Kirk, K.I. & Pisoni, D.B. (2000). Lexical Neighborhood Tests. St. Louis, MO:AudiTEC.

Magnusson, L., Karlsson, M. & Leijon, A. (2001). Predicted and measured speech recognition performance in noise with linear amplification. Ear and Hearing 22, 46-57.

Moeller, M.P. (2000). Early intervention and language development in children who are deaf and hard of hearing. Pediatrics 106, e43.

Norbury, C.F., Bishop, D.V. & Briscoe, J. (2001). Production of English finite verb morphology: A comparison of SLI and mild-moderate hearing impairment. Journal of Speech, Language and Hearing Research 44, 165-178.

Pavlovic, C.V., Studebaker, G.A. & Sherbecoe, R.L. (1986). An articulation index based procedure for predicting the speech recognition performance of hearing-impaired individuals. The Journal of the Acoustical Society of America 80, 50-57.

Scollie, S.D., Seewald, R., Cornelisse, L., Moodie, S., Bagatto, M., Laurnagary, D. & Pumford, J. (2005). The desired sensation level multistage input/output algorithm. Trends in Amplification 9(4), 159-197.

Souza, P.E. & Turner, C.W. (1999). Quantifying the contribution of audibility to recognition of compression-amplified speech. Ear and Hearing 20, 12-20.

Stanovich, K.E. (1986). Matthew effects in reading: Some consequences of individual differences in the acquisition of literacy. Reading Research Quarterly 21, 360-407.

Stelmachowicz, P.G., Hoover, B.M., Lewis, D.E., Kortekaas, R.W. & Pittman, A.L. (2000). The relation between stimulus context, speech audibility and perception for normal hearing and hearing-impaired children. Journal of Speech, Language and Hearing Research 43, 902-914.

Stelmachowicz, P.G., Pittman, A.L., Hoover, B.M. & Lewis, D.E. (2004 ). Novel word learning in children with normal hearing and hearing loss. Ear and Hearing 25, 47-56.

Tomasello, M. (2000). The item-based nature of children’s early syntactic development. Trends in Cognitive Sciences 4, 156-163.

Wake, M., Poulakis, Z., Hughes, E.K., Carey-Sargeant, C. & Rickards, F.W. (2005). Hearing impairment: A population study of age at diagnosis, severity and language outcomes at 7-8 years. Archives of Disease in Childhood 90, 238-244.