Interview with Harvey Dillon Ph.D., Director of Research at National Acoustic Laboratories (NAL), Sydney, Australia.
Share:
AO/Beck: Hi Dr. Dillon, it is a rare honor to speak with you, thank you for your time.
Dillon: The pleasure is mine Dr. Beck.
AO/Beck: If you don't mind, I'd like to start by getting a little information about you, your education and your professional career.
Dillon: Sure. I started with a degree in Electrical Engineering and then received my Ph.D. in psychoacoustics at the University of New South Wales in Sydney in 1979.
AO/Beck: After you graduated, where did you go?
Dillon: I went to NAL and although it was not my original plan, I'm still there 22 years later. I had intended to stay for only about 5 years and to then become a university lecturer. Having been studying for ten years, I thought I'd get out of the university system for a while to get another perspective on the world. Then all of a sudden the time passes, and I've been here 22 years, with no intention of going anywhere else, as I greatly enjoy working at NAL.
AO/Beck: Before we go much further, I'd like to be sure to get your website address for those readers who may want to get in touch with you.
Dillon: Of course, the website is at www.nal.gov.au
AO.Beck: Thank you. Let's move on to the topic at hand, which includes your new book and your thoughts on topics related to hearing and hearing aids. I have had the pleasure of reviewing your new book, simply titled 'Hearing Aids,' Boomerang Press (Sydney) and co-published by Thieme (NewYork and Stuttgart) and. I must say this is the single most important, thorough and comprehensive text I have ever read on the topic. I would strongly urge anyone teaching a class, or trying to find the most current book on hearing aids, to consider this book as the primary, authoritative text.
Dillon: Thank you very much, that's very kind of you.
AO/Beck: What was it that inspired you to spend years writing and editing this book?
Dillon: I had been lecturing and teaching casually at the local university for about 15 years on psychoacoustics, and more lately on hearing aids. As I looked at the texts available on hearing aids, I couldn't find anything that covered all the topics I thought were important, or that covered individual topics in the way I thought they should be covered, so I started preparing some notes and decided to turn them into a book. Many of the people I had spoken with in the USA said much the same thing; they really couldn't find a text that was adequate for the hearing aid courses they gave, so they found themselves assembling piles of information from multiple sources for the students.
AO/Beck: What I'd like to do next is give you a topic, and have you talk about the subject. For instance, if I were to mention binaural hearing aids versus monaural fittings, can you tell me your thoughts on that issue?
Dillon: The primary bilateral advantage comes from having a head between the two ears - and everyone has one - so therefore everyone has head baffle and head shadow effects. One gets an increase in loudness on the side closest to the sound source, and a decrease in loudness on the side further away from the sound source. Assuming the patient has two ears, both of which are candidates for amplification, by placing a hearing aid in each ear, you're making sure that if either side of the head is receiving a better signal-to-noise ratio, that clearer signal is available for amplification. Suppose the patient is trying to have a conversation with a person standing somewhat to the left and there is noise coming from somewhat to the right. If you average sound across the speech frequencies, the signal-to-noise ratio in the left ear can be up to 17 dB greater than the signal-to-noise ratio in the right ear. This is a tremendous difference. Of course in real life, the wanted signal will sometimes be to the left, sometimes to the right, and sometimes straight ahead. Having two hearing aids enables the patient to attend to whichever ear is receiving the better signal-to-noise ratio, without having to rearrange where everyone is standing or sitting.
AO/Beck: Do you recommend that all audiometrically eligible patients be fit with binaural hearing aids, or are you concerned about binaural interference?
Dillon: I am aware of binaural interference and I cover that in the book, but briefly, there is a small proportion of people, perhaps as high as ten percent of elderly people, for whom binaural interference is a problem. So, I don't think one can make a sweeping statement such as 'fit all patients bilaterally,' but I think we can start with the assumption that we fit patients bilaterally, and then look for reasons why it may not be appropriate in a small percentage of cases. If the proportion of patients demonstrating bilateral interference does turn out to be greater than ten percent for any population, then probably it would be worthwhile specifically testing for this prior to fitting. Given current knowledge, I would recommend fitting bilaterally whenever it is audiometrically and psychologically possible, but then be prepared to sometimes take one hearing aid back if the patient reports they do better with one hearing aid than they do with two. I came to this conclusion somewhat to my surprise, because I started with the belief that bilateral hearing aids were possibly overfit, but reversed this view after careful research of all the literature I could find.
AO/Beck: Please tell me a little about speech testing. I recall Denis Byrne writing years ago that most speech testing wasn't worthwhile. In essence, I think he wrote that although speech is the most important thing we listen to, the variability in presentation and response is simply too great to be dependable. Did I get that correct?
Dillon: Denis used speech tests in his research, but thought they had limited applications for hearing aid fitting in the clinic. An overriding consideration is that if one spends time doing something in the clinic, one must have a clear reason for doing it. If we're going to do speech audiometry in the clinic as part of a hearing rehabilitation program, we must be able to make meaningful decisions about whether to aid, or how to prescribe, based on the speech test scores. However, as you're aware, for the vast majority of the time, speech results don't impact our analysis or our treatment.
AO/Beck: So basically, if we're not going to use the information in a meaningful way, there is little point in collecting it?
Dillon: Exactly.
AO/Beck: Having said that, we all recognize that if there are gross discrepancies between the speech results of the two ears, or if the results are not in accordance with the pure tone audiogram, that may indicate a retrocochlear, or other situation requiring medical intervention.
Dillon: Yes that's certainly true; my earlier comments (and those of Denis') were in relation to the use of speech tests for hearing aid prescription and evaluation. Harking back to the earlier topic, one could certainly use speech tests to detect adverse binaural interactions. It's just a question of cost-benefit as to whether the results change the clinical decision often enough to justify the time spent on all the patients where the test has no effect.
AO/Beck: The next topic I'd like you to discuss is dual, and directional microphones.
Dillon: I think of all the features available on hearing aids, directional microphones are the most important and offer the most proven benefit. Directionality is the best way (apart from FM systems) to increase intelligibility in noise, and using dual microphones is the most flexible way to achieve directionality. If one can change from directional to omni-directional at will, there are no disadvantages to having directional mics.
AO/Beck: I was at a conference a year or two ago when one of the speakers mentioned that if the patient wearing a dual mic instrument were to move their head more than 15 degrees vertically, the dual mic benefit was negligible at that point, and may even be a detriment. What are your thoughts on that?
Dillon: The dual mic system is best when the head (or pedantically speaking, the line joining the inlet ports) is horizontal. As you move your head up and down the directivity index will decrease, but I would think it would be more than 15 degrees before a significant deterioration in directivity occurs. Eventually, there will be an inclination where having the directional mic is worse than not having it, but I don't think this is an important practical constraint. A more important limitation relates to whether the sounds heard have ever passed through the microphones. I know that sounds a little silly, but many people have normal, or at least good, hearing in the low tones, and for those people, we usually don't amplify the low tones. The low tones are coming in through the vent, or around the shell and no matter how directional the hearing aid is, such patients will be hearing predominantly vent-transmitted sound, which will eliminate any directional ability of the hearing aid. So for these patients, the directional sound applies only to the high frequencies, which are being amplified. So, if you compare a patient with only a high-frequency hearing loss to a patient who needs amplification across the entire frequency range, the patient who needs amplification across the entire spectrum will almost certainly derive more benefit from directionality than would the patient with only a high-tone loss.
AO/Beck: What about real-ear measures? Do you do that on every patient?
Dillon: We routinely do it in research studies, and we certainly recommend it for clinical patients. If it is achievable, I recommend measurements at three levels, say 50, 65 and 80 dB SPL input level with a broadband stimulus with a speech-like spectrum. Often we find that we really cannot measure the 50 dB SPL input signal unless the real-ear equipment is in a sound booth, which is typically not practical. Sometimes that is the case with a 65 dB SPL input level too. In such cases, one can measure the lowest level response one can, such as 70 dB, and make sure the compression ratios are set according to some good target. (I can suggest one!) If the compression ratios are set well, and if you have the response for 70 dB measured and adjusted appropriately, then the results should match the targets for lower and higher input levels too. What one shouldn't do is accept that the simulated curve on the fitting software is necessarily the same as what is happening in the individual ear. However, generally speaking, one can accept that if the fitting software shows a compression ration of 1.6 for example, that should be very much in accordance with what actually is happening in the ear. So for determining the real-ear frequency response, real-ear measurements are needed, but to determine compression ratios, they are not.
AO/Beck: As a general rule, do you recommend both compression and expansion?
Dillon: I certainly recommend compression, both output compression limiting for controlling the maximum output, and wide dynamic range compression for increasing intelligibility at low input levels and increasing comfort at high input levels. With wide dynamic range fittings, the internal noise from the hearing aid circuit may be audible, and in that case, expansion is useful to reduce the internal circuit noise of the hearing aid. Expansion can also reduce the softer annoying sounds in the environment, which the patient may not want to hear.
AO/Beck: And just like in real life, there is nothing new under the sun. Expansion has been around for quite a while now, and it has recently become popular again.
Dillon: Yes, that's correct. We did an experiment with expansion in the early 1980s where we had 2:1 compression above a certain value and 2:1 expansion below a certain value, in a multi-channel hearing aid. Patients did indeed report less noise when expansion was present. Unfortunately, when the expansion threshold was set too high we found that we also lost some soft speech cues. If you can get the levels just right you can reduce the noise for non-speech, without significantly affecting the perception of speech sounds. The key is really to set the expansion point very carefully, as the speech and the extraneous noises are a moving target. Some form of tracking expansion is probably optimal.
AO/Beck: I wonder if you can tell me a little about frequency transposition hearing aids? What can you tell me about where we are as far as technical aspects and where we are as far as patient satisfaction with those instruments?
Dillon: That's a tough one. We ve tried to make frequency transposition work well, in fact we've put three or four years into it. We tried to develop a processing scheme that would create minimal distortion and confusion as it transposed, but in the end we saw no additional benefits compared to a well-fit linear hearing aid. I can think of perhaps a dozen papers on this topic, and only two showed tangible benefit for the patient. One was by Chris Turner and colleagues, and the other was by Harry Levitt and colleagues. I still think there is potential there, and its possible we used subjects with the wrong sort of loss. The only thing I'm sure about is that I am not sure which subjects can benefit or how the transposition characteristics should best be adjusted.
AO/Beck: Lastly, I wonder if you can tell me your thoughts on the lack of market penetration in the USA. As I'm sure you know, about 4 out of 5 patients who could benefit from hearing aids never walk through the clinic door. Can you tell me, what must we do to attract more of our target audience?
Dillon: I can only guess. First, let me say that the proportions in Australia are the same as in the USA, despite hearing aids being free to all aged pensioners. About 1 out of 5 patients with hearing loss actually own hearing aids. The reasons I can offer are the same old cliches you are familiar with. Some people perceive hearing aids as being ugly, some associate hearing aids with aging, with the wearers not functioning well in conversation and perhaps with senility, some are put off by the anticipated expense of hearing aids, and some try them and find that hearing aids don't perform up to their expectations - either because of unrealistic expectations or because the hearing aids were poorly chosen or fit. My main suggestions are that we help individual patients work through each of the reasons that are holding them back from enthusiastically giving hearing aids a go, ensure they have realistic expectations, and then do the best possible job of selecting and fitting hearing aids for them. More enthusiastic, satisfied hearing aid wearers should lead to more positive word-of-mouth advertising about the benefits of hearing aids instead of the converse. Doubtless there are public education approaches that would also help.
AO/Beck: Dr. Dillon, it has indeed been a pleasure speaking with you and I want to thank you for your time. Again, the book ('Hearing Aids' by Harvey Dillon, Published by Thieme and Boomerang) is theoretically sound, highly practical and I strongly recommend it for all students and professionals. I particularly like the fact that the information is accessible to people who initially know nothing about hearing aids, but that it progresses to a satisfying and useful depth of information.
Dillon: Thanks Doug, our talk's been fun. Let's do it again some time.
AO/Beck: Absolutely, I'll look forward to it.
For more information on Dr. Dillon's book, HEARING AIDS, visit Thieme's web
site.
Dillon: The pleasure is mine Dr. Beck.
AO/Beck: If you don't mind, I'd like to start by getting a little information about you, your education and your professional career.
Dillon: Sure. I started with a degree in Electrical Engineering and then received my Ph.D. in psychoacoustics at the University of New South Wales in Sydney in 1979.
AO/Beck: After you graduated, where did you go?
Dillon: I went to NAL and although it was not my original plan, I'm still there 22 years later. I had intended to stay for only about 5 years and to then become a university lecturer. Having been studying for ten years, I thought I'd get out of the university system for a while to get another perspective on the world. Then all of a sudden the time passes, and I've been here 22 years, with no intention of going anywhere else, as I greatly enjoy working at NAL.
AO/Beck: Before we go much further, I'd like to be sure to get your website address for those readers who may want to get in touch with you.
Dillon: Of course, the website is at www.nal.gov.au
AO.Beck: Thank you. Let's move on to the topic at hand, which includes your new book and your thoughts on topics related to hearing and hearing aids. I have had the pleasure of reviewing your new book, simply titled 'Hearing Aids,' Boomerang Press (Sydney) and co-published by Thieme (NewYork and Stuttgart) and. I must say this is the single most important, thorough and comprehensive text I have ever read on the topic. I would strongly urge anyone teaching a class, or trying to find the most current book on hearing aids, to consider this book as the primary, authoritative text.
Dillon: Thank you very much, that's very kind of you.
AO/Beck: What was it that inspired you to spend years writing and editing this book?
Dillon: I had been lecturing and teaching casually at the local university for about 15 years on psychoacoustics, and more lately on hearing aids. As I looked at the texts available on hearing aids, I couldn't find anything that covered all the topics I thought were important, or that covered individual topics in the way I thought they should be covered, so I started preparing some notes and decided to turn them into a book. Many of the people I had spoken with in the USA said much the same thing; they really couldn't find a text that was adequate for the hearing aid courses they gave, so they found themselves assembling piles of information from multiple sources for the students.
AO/Beck: What I'd like to do next is give you a topic, and have you talk about the subject. For instance, if I were to mention binaural hearing aids versus monaural fittings, can you tell me your thoughts on that issue?
Dillon: The primary bilateral advantage comes from having a head between the two ears - and everyone has one - so therefore everyone has head baffle and head shadow effects. One gets an increase in loudness on the side closest to the sound source, and a decrease in loudness on the side further away from the sound source. Assuming the patient has two ears, both of which are candidates for amplification, by placing a hearing aid in each ear, you're making sure that if either side of the head is receiving a better signal-to-noise ratio, that clearer signal is available for amplification. Suppose the patient is trying to have a conversation with a person standing somewhat to the left and there is noise coming from somewhat to the right. If you average sound across the speech frequencies, the signal-to-noise ratio in the left ear can be up to 17 dB greater than the signal-to-noise ratio in the right ear. This is a tremendous difference. Of course in real life, the wanted signal will sometimes be to the left, sometimes to the right, and sometimes straight ahead. Having two hearing aids enables the patient to attend to whichever ear is receiving the better signal-to-noise ratio, without having to rearrange where everyone is standing or sitting.
AO/Beck: Do you recommend that all audiometrically eligible patients be fit with binaural hearing aids, or are you concerned about binaural interference?
Dillon: I am aware of binaural interference and I cover that in the book, but briefly, there is a small proportion of people, perhaps as high as ten percent of elderly people, for whom binaural interference is a problem. So, I don't think one can make a sweeping statement such as 'fit all patients bilaterally,' but I think we can start with the assumption that we fit patients bilaterally, and then look for reasons why it may not be appropriate in a small percentage of cases. If the proportion of patients demonstrating bilateral interference does turn out to be greater than ten percent for any population, then probably it would be worthwhile specifically testing for this prior to fitting. Given current knowledge, I would recommend fitting bilaterally whenever it is audiometrically and psychologically possible, but then be prepared to sometimes take one hearing aid back if the patient reports they do better with one hearing aid than they do with two. I came to this conclusion somewhat to my surprise, because I started with the belief that bilateral hearing aids were possibly overfit, but reversed this view after careful research of all the literature I could find.
AO/Beck: Please tell me a little about speech testing. I recall Denis Byrne writing years ago that most speech testing wasn't worthwhile. In essence, I think he wrote that although speech is the most important thing we listen to, the variability in presentation and response is simply too great to be dependable. Did I get that correct?
Dillon: Denis used speech tests in his research, but thought they had limited applications for hearing aid fitting in the clinic. An overriding consideration is that if one spends time doing something in the clinic, one must have a clear reason for doing it. If we're going to do speech audiometry in the clinic as part of a hearing rehabilitation program, we must be able to make meaningful decisions about whether to aid, or how to prescribe, based on the speech test scores. However, as you're aware, for the vast majority of the time, speech results don't impact our analysis or our treatment.
AO/Beck: So basically, if we're not going to use the information in a meaningful way, there is little point in collecting it?
Dillon: Exactly.
AO/Beck: Having said that, we all recognize that if there are gross discrepancies between the speech results of the two ears, or if the results are not in accordance with the pure tone audiogram, that may indicate a retrocochlear, or other situation requiring medical intervention.
Dillon: Yes that's certainly true; my earlier comments (and those of Denis') were in relation to the use of speech tests for hearing aid prescription and evaluation. Harking back to the earlier topic, one could certainly use speech tests to detect adverse binaural interactions. It's just a question of cost-benefit as to whether the results change the clinical decision often enough to justify the time spent on all the patients where the test has no effect.
AO/Beck: The next topic I'd like you to discuss is dual, and directional microphones.
Dillon: I think of all the features available on hearing aids, directional microphones are the most important and offer the most proven benefit. Directionality is the best way (apart from FM systems) to increase intelligibility in noise, and using dual microphones is the most flexible way to achieve directionality. If one can change from directional to omni-directional at will, there are no disadvantages to having directional mics.
AO/Beck: I was at a conference a year or two ago when one of the speakers mentioned that if the patient wearing a dual mic instrument were to move their head more than 15 degrees vertically, the dual mic benefit was negligible at that point, and may even be a detriment. What are your thoughts on that?
Dillon: The dual mic system is best when the head (or pedantically speaking, the line joining the inlet ports) is horizontal. As you move your head up and down the directivity index will decrease, but I would think it would be more than 15 degrees before a significant deterioration in directivity occurs. Eventually, there will be an inclination where having the directional mic is worse than not having it, but I don't think this is an important practical constraint. A more important limitation relates to whether the sounds heard have ever passed through the microphones. I know that sounds a little silly, but many people have normal, or at least good, hearing in the low tones, and for those people, we usually don't amplify the low tones. The low tones are coming in through the vent, or around the shell and no matter how directional the hearing aid is, such patients will be hearing predominantly vent-transmitted sound, which will eliminate any directional ability of the hearing aid. So for these patients, the directional sound applies only to the high frequencies, which are being amplified. So, if you compare a patient with only a high-frequency hearing loss to a patient who needs amplification across the entire frequency range, the patient who needs amplification across the entire spectrum will almost certainly derive more benefit from directionality than would the patient with only a high-tone loss.
AO/Beck: What about real-ear measures? Do you do that on every patient?
Dillon: We routinely do it in research studies, and we certainly recommend it for clinical patients. If it is achievable, I recommend measurements at three levels, say 50, 65 and 80 dB SPL input level with a broadband stimulus with a speech-like spectrum. Often we find that we really cannot measure the 50 dB SPL input signal unless the real-ear equipment is in a sound booth, which is typically not practical. Sometimes that is the case with a 65 dB SPL input level too. In such cases, one can measure the lowest level response one can, such as 70 dB, and make sure the compression ratios are set according to some good target. (I can suggest one!) If the compression ratios are set well, and if you have the response for 70 dB measured and adjusted appropriately, then the results should match the targets for lower and higher input levels too. What one shouldn't do is accept that the simulated curve on the fitting software is necessarily the same as what is happening in the individual ear. However, generally speaking, one can accept that if the fitting software shows a compression ration of 1.6 for example, that should be very much in accordance with what actually is happening in the ear. So for determining the real-ear frequency response, real-ear measurements are needed, but to determine compression ratios, they are not.
AO/Beck: As a general rule, do you recommend both compression and expansion?
Dillon: I certainly recommend compression, both output compression limiting for controlling the maximum output, and wide dynamic range compression for increasing intelligibility at low input levels and increasing comfort at high input levels. With wide dynamic range fittings, the internal noise from the hearing aid circuit may be audible, and in that case, expansion is useful to reduce the internal circuit noise of the hearing aid. Expansion can also reduce the softer annoying sounds in the environment, which the patient may not want to hear.
AO/Beck: And just like in real life, there is nothing new under the sun. Expansion has been around for quite a while now, and it has recently become popular again.
Dillon: Yes, that's correct. We did an experiment with expansion in the early 1980s where we had 2:1 compression above a certain value and 2:1 expansion below a certain value, in a multi-channel hearing aid. Patients did indeed report less noise when expansion was present. Unfortunately, when the expansion threshold was set too high we found that we also lost some soft speech cues. If you can get the levels just right you can reduce the noise for non-speech, without significantly affecting the perception of speech sounds. The key is really to set the expansion point very carefully, as the speech and the extraneous noises are a moving target. Some form of tracking expansion is probably optimal.
AO/Beck: I wonder if you can tell me a little about frequency transposition hearing aids? What can you tell me about where we are as far as technical aspects and where we are as far as patient satisfaction with those instruments?
Dillon: That's a tough one. We ve tried to make frequency transposition work well, in fact we've put three or four years into it. We tried to develop a processing scheme that would create minimal distortion and confusion as it transposed, but in the end we saw no additional benefits compared to a well-fit linear hearing aid. I can think of perhaps a dozen papers on this topic, and only two showed tangible benefit for the patient. One was by Chris Turner and colleagues, and the other was by Harry Levitt and colleagues. I still think there is potential there, and its possible we used subjects with the wrong sort of loss. The only thing I'm sure about is that I am not sure which subjects can benefit or how the transposition characteristics should best be adjusted.
AO/Beck: Lastly, I wonder if you can tell me your thoughts on the lack of market penetration in the USA. As I'm sure you know, about 4 out of 5 patients who could benefit from hearing aids never walk through the clinic door. Can you tell me, what must we do to attract more of our target audience?
Dillon: I can only guess. First, let me say that the proportions in Australia are the same as in the USA, despite hearing aids being free to all aged pensioners. About 1 out of 5 patients with hearing loss actually own hearing aids. The reasons I can offer are the same old cliches you are familiar with. Some people perceive hearing aids as being ugly, some associate hearing aids with aging, with the wearers not functioning well in conversation and perhaps with senility, some are put off by the anticipated expense of hearing aids, and some try them and find that hearing aids don't perform up to their expectations - either because of unrealistic expectations or because the hearing aids were poorly chosen or fit. My main suggestions are that we help individual patients work through each of the reasons that are holding them back from enthusiastically giving hearing aids a go, ensure they have realistic expectations, and then do the best possible job of selecting and fitting hearing aids for them. More enthusiastic, satisfied hearing aid wearers should lead to more positive word-of-mouth advertising about the benefits of hearing aids instead of the converse. Doubtless there are public education approaches that would also help.
AO/Beck: Dr. Dillon, it has indeed been a pleasure speaking with you and I want to thank you for your time. Again, the book ('Hearing Aids' by Harvey Dillon, Published by Thieme and Boomerang) is theoretically sound, highly practical and I strongly recommend it for all students and professionals. I particularly like the fact that the information is accessible to people who initially know nothing about hearing aids, but that it progresses to a satisfying and useful depth of information.
Dillon: Thanks Doug, our talk's been fun. Let's do it again some time.
AO/Beck: Absolutely, I'll look forward to it.
For more information on Dr. Dillon's book, HEARING AIDS, visit Thieme's web
site.
