From the Desk of Gus Mueller
There are probably a handful of you readers out there who remember the old days, when we would bring patients with unusual basic audiometric findings back to the clinic for “special testing.” This usually was a one-hour appointment, and this testing included Bekesy audiometry (looking for a Type III or Type IV pattern), the SISI, tone decay, and the ABLB if air conduction results were asymmetrical. We sort of knew that these tests had very poor sensitivity and specificity, but it was all we had at the time.
Our differential diagnostic lives changed significantly, however, when immittance audiometry became available in the 1970s. Assessing and comparing ipsilateral and contralateral acoustic reflexes added a wealth of information regarding the patient’s auditory status. And then, to make things even better, OAE testing was added a decade later. Gone were the tests of old, as we finally had the tools to conduct an objective and effective audiologic evaluation.
But I’ve heard rumors, that over the past 20 years or so, somewhere along the way this new successful test battery lost its appeal, is no longer considered routine, and some, if not many audiologists have gone back to relying on the old tests of the 1950s. Could this be true?
To talk about what really is happening in the world of diagnostic audiology, and what should be happening to make this specialty both effective and efficient, we’re bringing in an expert to give us the straight scoop. Not just any expert, but someone who honed his skills while training with the father of diagnostic audiology, James Jerger.
James W. “Jay” Hall III, PhD, is a name you all know. He resides much of the time in his home state of Maine, but has stopped by to visit 20Q on several previous occasions. Dr. Hall holds academic appointments from Salus University, the University of Hawaii, and the University of Pretoria South Africa. He is internationally recognized for his research, publications and workshops.
Most of you probably own at least one of Jay’s 10 books, and probably have attended several of his seminars at state and national meetings. He is one of the founders of the American Academy of Audiology, served on the Academy’s Board of Directors, has held several leadership positions within the Academy, and has received the Academy’s Distinguished Achievement Award, Presidential Service Award and the Jerger Career Award for Research.
In this 20Q, Jay takes a critical look at our routine audiologic procedures, what we do and why we do it, and how we can be both more effective and efficient. I think that’s what we’re all looking for.
Gus Mueller, PhD
Browse the complete collection of 20Q with Gus Mueller CEU articles at www.audiologyonline.com/20Q
20Q: Healthy Hearing Over the Lifespan Part 1- Effective and Efficient Diagnostic Assessment
After reading this article, professionals will be able to:
- List clinical limitations for pure-tone audiometry.
- List guidelines for when bone-conduction audiometry is not needed.
- List three clinical questions before including the SRT in their test battery.
1. Diagnostic assessment? I’ve used the same test battery since I finished my AuD training. Am I doing something wrong?
If you use the same diagnostic test battery in assessing each patient who comes to your clinic, yes, there may be room for improvement. Tell me … what tests do you typically perform for an adult patient complaining of hearing problems?
2. I always perform air- and bone-conduction pure-tone audiometry and then some speech audiometry tests. You know, the tests described in CPT code 92557 for “comprehensive audiometry threshold evaluation and speech recognition.”
Well, you’re not alone. Our colleagues Ian Windmill and Barry Freeman (Windmill & Freeman, 2019) analyzed Medicare data for over 1,300,000 older adults who were referred to an audiologist for a hearing assessment. The analysis showed that 90% of audiologists rely almost exclusively on the same rather dated test battery. That is, air- and bone conduction pure-tone thresholds, speech recognition thresholds, and word recognition in quiet. Unfortunately, the traditional air-bone-speech test approach doesn’t always provide us with the information we need to accurately diagnose and manage. Pure-tone audiometry, for example, has some serious drawbacks as a diagnostic test of auditory function.
3. It’s reassuring to know that I’m not bucking that trend, but wait a minute. Pure-tone audiometry has been around for about 100 years. What kind of drawbacks are you talking about?
Despite its longstanding and almost exalted stature as a relied-upon procedure for routine hearing assessment, pure-tone audiometry with air conduction stimuli is characterized by at least six serious clinical limitations:
- Pure tone audiometry is a measure of one of the simplest of auditory processing …. detection of sound in quiet … for sinusoids, the most basic of sounds. The audiogram doesn’t reflect real-world hearing demands. That is, in day-to-day activities, including communication settings, no one needs to listen to pure tones in a quiet environment.
- Pure tone audiometry yields data for a remarkably tiny and inadequate sample of test frequencies. The typical audiogram depicts thresholds for six octave frequencies or, perhaps eight frequencies if inter-octave stimuli at 3000 and 6000 Hz are included. The normal human ear is capable of detecting frequencies over the range of 20 to 20 KHz, or 19,980 frequencies. The audiogram reflects hearing thresholds for a minuscule proportion of the frequency range for normal human hearing sensitivity, specifically 8/19,980 or 0.0004%!
4. Mathematics has never been my strong suit, but I can deal with small numbers. You mentioned six limitations. I’m curious to hear about the other four?
Sure. The audiogram has little relation to self-perceived hearing handicap or everyday hearing abilities (e.g., Weinstein & Ventry, 1983; Hall, 2014). In other words, persons with a normal audiogram may have very serious hearing and communication problems. Obvious examples include patients with central auditory dysfunction, including auditory processing disorders, and some with auditory neuropathy spectrum disorder (ANSD). Conversely, the majority of people with hearing loss as indicated with an abnormal audiogram do not seek audiologic assessment or management, such as amplification. Two people with the same audiogram may report very different experiences with communication.
Also, an assortment of listener variables may compromise the reliability and validity of pure-tone audiometry, and really, most any behavioral audiologic procedure. The variables can include motivation, cognitive functioning (including attention and memory), fatigue, and language factors that interfere with instructions for the task.
5. Okay, I’m still with you, keep going...
The audiogram is a graph of hearing sensitivity, mostly dependent on cochlear function. Hearing, on the other hand, requires rapid processing of often complex and rapidly changing acoustical information throughout the auditory system, from the middle ear to the cerebral cortex.
6. And what’s the final limitation?
Pure-tone audiometry provides essentially no information on listening abilities that are essential for effective human communication. Listening is an active process requiring effort, attention, and other cognitive functions. There’s not enough space in this article for a more detailed critique of air conduction pure tone audiometry and the multiple shortcomings of the audiogram as a primary measure of hearing status.
When you have a chance for a little professional reading, I suggest that you take a look at an article by Frank Musiek and colleagues for an excellent critical perspective on the pure tone audiogram (Musiek et al, 2017). I also address this same topic in a review article that came out in April of this year (Hall, 2021).
7. I must admit, that was a sobering discussion about air-conduction audiometry. The “comprehensive audiometry” CPT code also calls for bone-conduction testing. Do you have any additional concerns about this test?
It’s true that audiologists typically measure both air- and bone-conduction hearing thresholds in the initial assessment of most patients. Indeed, data reveal that almost every patient undergoes this testing (e.g., Windmill & Freeman, 2019). I really question the clinical rationale for this practice. For the majority of adult patients encountered in an audiology clinic, bone-conduction audiometry does not add value to the diagnosis or management of hearing loss.
8. Wait a minute! Ever since graduate school, I’ve automatically reached for the bone oscillator as soon as I finish air conduction testing. Are you saying that we don’t need to conduct this test with every patient?
That’s exactly what I’m saying. First of all, middle ear dysfunction and associated conductive hearing loss is highly unusual in an adult population—remember, back at Question #1 I qualified that our discussion here was about adult patients. In a study of more than 1500 older adult patients initially seen in an audiology clinic, David Zapala and colleagues at the Mayo Clinic in Jacksonville Florida (2010) reported that only 4.2% patients required referral to otolaryngology, and only a handful had possible middle ear disease. Most of the patients in the small referral group required otologic workup for possible retrocochear pathology or assorted sensorineural etiologies like Meniere’s disease or sudden onset hearing loss.
9. Yes, but we can’t just assume that all adult patients have normal middle ear function. Isn’t comparison of air- versus bone conduction thresholds a good way to rule out middle ear problems?
Many audiologists do rely on comparison of hearing thresholds for air- versus bone conduction … looking for the air-bone gap … for identification middle ear dysfunction and quantification of conductive hearing loss. However, the air-bone gap is not a reliable or valid index of middle ear status.
10. How about you just give me some simple guidelines for when I don’t need to go ahead with bone conduction audiometry.
Absolutely. Bone conduction audiometry is not indicated or clinically justified for patients meeting two or more of the following evidence-based criteria:
- No patient history of middle ear disease, including no medical evaluation or management
- Normal otoscopic findings
- No mention of middle ear abnormality in the physician examination report
- Normal tympanometry
- Normal ipsilateral or contralateral acoustic reflexes observed with the probe in each ear
- Normal otoacoustic emissions (OAEs) for low test frequencies
Of course, your test battery should include bone-conduction audiometry for patients at risk for or with a history of middle ear disease. And, you should keep the bone oscillator handy for those patients with abnormal findings on direct measures of middle ear function, such as tympanometry or wide band reflectance/absorbance.
11. That makes sense, but what’s the downside of simply always doing bone conduction testing.
There are at least four practical disadvantages or drawbacks to routinely performing bone conduction audiometry in patients lacking risk factors or clinical findings associated with middle ear dysfunction. First, the investment of precious test time yields no diagnostic return. Bilateral bone conduction audiometry with masking of the contralateral ear requires more than 10 minutes of test time (Basar & Canbaz, 2015). This is time that would be better spent on tests that contribute to validation of the patient’s complaints, to the accurate diagnosis of auditory dysfunction, and to effective management. Second, in many health care systems either the patient or a third-party health insurance carrier will be obligated to cover the cost associated with bone conduction audiometry.
12. Well, we’re back to another list of drawbacks. What’s the next one?
You’ll probably like #3. If you always conduct air- and bone conduction audiometry, even when it’s not clinically justified, you might incorrectly suspect conductive hearing loss due to false or spurious air-bone gaps, or simply because mastoids vary considerably in density. Robert Margolis extensively examined the statistical chance of recording air-bone gaps or bone-air gaps at different pure tone frequencies in persons with clinically documented normal middle ear dysfunction (Margolis, 2010 a, b). Apparent air-bone gaps or bone-air gaps of 10 dB, 15, and even 20 HL are entirely predictable from a statistical perspective, even in patients with normal middle ear status. Moreover, he calculated that a finding of no air-bone gap … 0 dB difference between air- and bone conduction thresholds for all four primary test frequencies (500, 1000, 2000, and 4000 Hz) occurs in less than 20% of patients with entirely normal middle ear dysfunction. And if I remember correctly, a perfect match of air and bone for both ears will happen once in every 250,000 patients (Margolis, 2008), about once in your career!
13. Yes. I never know what to say in my reports when a patient has an air-bone gap at only 4000 Hz, or when bone is better than air. You’re probably saving the most important problem for last.
You’re right … the final concern is perhaps the most important. There is a chance that routinely performing bone conduction audiometry in a futile attempt to document non-existent conductive hearing loss in a patient with strong evidence of normal middle ear function may serve to undermine patient and physician confidence in the competence and even the professional integrity of the audiologist. It would be entirely reasonable for a patient and/or the patient’s physician to seriously question why an audiologist went to considerable efforts to perform a test to document middle ear dysfunction that was not suspected based on patient history, physician examination, or other audiological test findings.
14. I get your points about pure tone audiometry … but what about the speech recognition threshold and supra-threshold speech recognition. They’re also required for me to bill patients with the comprehensive audiometry CPT code. You’re not going to now tell me that I shouldn’t perform speech audiometry with my patients …are you?
No, but you really don’t need to spend time on finding speech recognition threshold for all of your patients. It’s just not a good use of your precious test time. At least five minutes of test time is consumed with an explanation of the task to the patient plus the actual time required to estimate spondee threshold for each ear using rather detailed guidelines for measuring SRT (ASHA, 1988). There’s no question that the SRT provides useful information in the hearing assessment of selected patient populations, especially young or difficult-to-test children, older patients with possible cognitive decline, and patients of any age where there is a suspicion of false or exaggerated hearing loss. However, the SRT does not contribute to the diagnosis of hearing loss or to decisions about management for the majority of older pediatric and adult patients undergoing clinical audiological assessment. The SRT and PTA are almost always in close agreement for patients with normal hearing thresholds. There is really no physiological or psychoacoustic explanation for why the SRT would be significantly poorer than the PTA in a cognitively-intact adult with hearing thresholds < 20 dB HL.
I suggest that ask a few simple clinical questions before including the SRT in your test battery: 1) Will information from the SRT tell me more than what I already know about this patient’s hearing from other tests, such as pure tone audiometry or OAEs? 2) Will the SRT contribute to my diagnosis for this patient? 3) Will I alter the management plan for this patient based on the SRT? If the answer is “no” for each of these questions regarding a specific patient, then we would be well advised to “just say no” for measuring the SRT.
15. You make a good point. But what about word recognition?
The most common and serious clinical limitations or drawbacks associated with measurement of word recognition in quiet resemble those for air conduction audiometry. And, there is remarkable inconsistency among audiologists in the procedures used to measure word recognition (Mueller & Hornsby, 2020). Recognition of single syllable words in an atypically quiet setting is not consistent with real-world listening demands. Excellent word recognition scores do not rule out deficits in central auditory processing and, specifically, daily struggles in perceiving and understanding complex speech in noisy settings. Audiologists commonly encounter patients who emphatically state: “I can hear you easily in this quiet room, but I really have problems understanding people speak when there is background noise.”
As a result, if you rely exclusively on word recognition in quiet you’ll underestimate the real-world problems that some patients experience every day. Many patients whose chief complaint is difficulty with speech perception in noise understandably might question why an audiologist would spend time evaluating their word recognition in quiet. For such patients, word recognition scores in quiet will have little relation to the communication disorder that brought them to the clinic. In addition, word recognition scores in quiet generally lack sensitivity to neural and central auditory dysfunction
16. True enough. Repeating single words in a quiet sound treated room isn’t a very typical listening experience. So, what’s the alternative?
Speech perception in noise tests are a more logical, sensitive, and effective measure of communication abilities than tests of word recognition in quiet. The relatively limited information about speech perception available from tests of word recognition in quiet is readily available in speech in noise tests (Wilson 2011), but speech in noise tests provide additional clinically valuable information about real-world communicative skills and deficits (see Hall, 2014 for review).
Results are superior in determining amplification needs and options, and also for detection of neural auditory dysfunction. The latter diagnostic benefit is substantial. As Vaisbuch et al (2019) note, speech in noise tests “…can replace word-recognition in quiet in most instances in the convention audiologic test battery”… “allowing for better diagnosis and management of individuals with hearing loss. (Vaisbuch et al, 2019, p. S1). I agree!
17. According to my mathematical calculations, bypassing bone conduction and SRT testing in many of my patients will save about at least 10 minutes of test time. You probably have some ideas for how I can better use that time.
I certainly do have an idea, but it’s not mine and it’s definitely not new. The time saved should be invested in objective auditory tests, particularly aural immittance measures like tympanometry and acoustic reflexes plus otoacoustic emissions. I’m sure you remember the old crosscheck principle (Jerger & Hayes, 1976; Hall, 2016). Expanding the crosscheck principle to patients of all ages, and taking the liberty to replace the word “children” with patients, I’ll rephrase one of the time-tested statements in the classic crosscheck paper: “… simply observing the auditory behavior of [patients] does not always yield an accurate description of hearing loss. In our own experience, we have seen too many [patients] who have been misdiagnosed and mismanaged on the basis of behavioral test results alone” (Jerger & Hayes, 1976, p. 614).
18. I’m all for objective auditory tests, especially in children. How about some clinical reasons for including them in my test battery for adult patients?
The optimal test battery for most patients, including adults, combines behavioral and objective measures of auditory function in a patient-specific test battery based on the patient’s history and chief complaint. The rationale for including one or more of these procedures in a test battery is quite straightforward and logical. If a patient complains about difficulty hearing in noisy settings, then a test of speech perception in noise is included in the battery. Otoacoustic emissions are recorded with patients who are at risk for cochlear dysfunction, such as those with a history of noise or music exposure or with diseases known to increase the chance of cochlear abnormalities (e.g., diabetes). Tympanometry quickly rules out or confirms middle ear dysfunction. Acoustic reflex measurement with tonal and noise stimuli provides valuable objective information on auditory status in patients at risk for cognitive impairment, false hearing loss, loudness recruitment, and neural dysfunction.
19. This has been an interesting discussion. I’m almost out of questions, but I’m always worried about the bottom line. How will this new approach to diagnostic audiologic assessment affect my clinical revenue?
I have really good news for you. An efficient and clinically-justified diagnostic test battery based on patient history and chief complaint doesn’t add appreciable time to the assessment but it will increase your clinical revenue. Rather than bundling diagnostic audiology services with the comprehensive audiometry CPT code, you’re better off billing for each of the individual procedures that you perform. For example, for many patients you bill for pure-tone audiometry (air only), tympanometry, acoustic reflex testing (threshold), and distortion product OAES (comprehensive diagnostic evaluation). As long as you estimate speech threshold, you can use code 92556 (“speech audiometry threshold, with speech recognition” for speech perception in noise. All supra-threshold speech audiometry tests involve speech recognition. The fee schedules for the added procedures, such as aural immittance measures (tympanometry and acoustic reflexes) and otoacoustic emissions, invariably enhances your revenue for appropriate patients.
20. Makes sense. I’m interested in learning more about the relation of patient history to assessment and management. Can we have another discussion focusing on that general topic?
Great idea. Let’s get together next month and we’ll discuss promoting healthy hearing. We’ll focus on comorbid conditions associated with hearing loss and related disorders, like tinnitus. We maybe also could talk about the role of lifestyle factors in hearing health, such as diet and smoking. Never fear—You’ll have the opportunity to ask plenty of questions about these topics. See you then!
American Speech-Language-Hearing Association. (1988). Determining threshold level for speech [Guidelines]. Available from www.asha.org/policy.
Basar, F., & Canbaz, S. (2015). What is the audiological evaluation time for those aged 0 – 5 years and older. The Journal of International Advanced Otology, 42, 42-47
Fowler E.P., & Wegel, R.L. (1922). Audiometric methods and the application. Transactions of the 28th Annual Meeting of the American Laryngological, Rhinological and Otological Society (pp. 98–132).
Hall, J.W. III. (2014). Introduction to audiology today. Boston: Pearson Educational.
Hall, J.W. III. (2016). Crosscheck principle in pediatric audiology today: A 40-year perspective. Journal of Audiology and Otology, 20, 59-67.
Hall, J.W. III. (2021). Promoting healthy hearing over the lifespan. Aud Vestib Res, 30, 74-94.
Jerger,J., & Hayes, D. (1976). The cross-check principle in pediatric audiology. Archives of Otolaryngology, 102, 614-620.
Margolis, R.H. (2008). The vanishing air-bone gap – audiology’s dirty little secret. AudiologyOnline, Article 901. Retrieved from www.audiologyonline.com
Margolis, R.H. (2010 a). A few secrets about bone-conduction testing. The Hearing Journal, 63, 10-17.
Margolis, R.H., Glasberg, B.R., Creeke, S., & Moore, B.C. (2010 b). AMTAS: Automated method for testing auditory sensitivity: validation studies. International Journal of Audiology, 49 (3), 185-194.
Mueller, H.G., & Hornsby, B.Y.W. (2020). 20Q: Word recognition testing --- Let’s just agree to do it right! AudiologyOnline, Article 26478. Retrieved from www.audiologyonline.com
Musiek, F.E., Shinn, J., Chermak, G.D., & Bamiou, D.-E. (2017). Perspectives on the pure-tone audiogram. Journal of the American Academy of Audiology, 28, 655-671.
Vaisbuch, Y., Ali, N., Qian, S.Z., Gianakas, S.P., & Fitzgerald, M.B. (2019). Speech in noise understanding in patients with vestibular schwannoma. J Neurol Surg B Skull Base, 80, S1-S2.
Vogel, D.A., McCarthy, P.A., Bratt, G.W., & Brewer, C. (2007). The clinical audiogram: Its history and current use. Communication Disorders Review, 1(2), 81-94.
Wilson, R.H. (2011). Clinical experience with the Words-in-Noise Test on 3430 veterans: Comparison with pure-tone thresholds and word recognition in quiet. Journal of the American Academy of Audiology, 22, 405-423.
Windmill, I., & Freeman, B. (2019). Medicare, Hearing Care, and Audiology: Data-Driven Perspectives. Audiology Today, 31, 16-26.
Zapala, D.A., Stamper, G.C., Shelfer, J.S., Walker, D.A., Karatayli-Ozgursoy, S., Ozgursoy, O.B., & Hawkins, D.B. (2010). Safety of audiology direct access for Medicare patients complaining of hearing impairment. Journal of the American Academy of Audiology, 21, 365–379.
Hall, J.W. III. (2021). 20Q: Healthy hearing over the lifespan - effective and efficient diagnostic assessment. AudiologyOnline, Article 27964. Available at www.audiologyonline.com