Welcome to today's ReSound Symposium, Meeting the Needs of Aging Adults. We put this symposium together because we think it is very important to have good educational offerings in order to keep up with what is currently going on in hearing aid technology. We will be presenting five seminars this week. Today I will give a broad overview on where are we with modern hearing aids with the elderly. Then throughout the week you will hear about cognition from Kathy Pichora-Fuller, and Lorienne Jenstad will talk about modern signal processing. On the clinical side, Hillary Snapp will address wireless fittings in your practice. I think you will also really enjoy hearing from Matt Perry, who will talk to us about marketing and using the Internet to market to the older demographic. I encourage you to view the other courses in the symposium, which can be accessed via the ReSound course library on AudiologyOnline.
Today I'll discuss the question, "Are modern technologies meeting the amplification needs of aging adults?" This is an interesting question. Where are we today? Are we doing as much as we need to be doing to meet this huge growing demographic? This whole symposium is entitled Meeting the Needs of Aging Adults, but who are aging adults? There are a couple of definitions. Aging or elderly adults are chronologically aged at 65 years or older. More specifically, age 65 to 74 is referred to as the early elderly. Over 75 years is referred to as the late elderly. Either way, these are people 65 years or older, and they make up the vast majority of the population that acquires amplification every year. Obviously, we have other demographics, such as pediatrics and people who have hearing loss throughout their lifetime. But the vast majority of people we fit with hearing aids are elderly.
If you look in the United States, there are about 35 million Americans, and more than 18 million of these are elderly. This group is growing, of course, with the Baby Boomers coming of age. What is the average age of the first-time hearing aid user in the United States? It is 71 years old. As much as we have tried to get that age to come down, it continues to hold steady. Why is that? First we have to understand what is going on with hearing loss and aging.
Many elderly patients suffer from hearing loss associated with presbycusis. This is hearing loss due to aging, but it is never pure. It is also complicated by genetic factors. It can be complicated by noise-induced hearing loss. As we age, most of us are going to lose our hearing. We first lose it in the high frequencies and then it spreads into mid frequencies. Ultimately, it can spread even into the lower frequencies. So what are the characteristics of presbycusis? With presbycusis, individuals lose the ability to hear soft sounds and may experience loudness recruitment where loud sounds can become uncomfortable. They have a reduced dynamic range, wherein we have to squeeze the entire hearing range for a hearing aid fitting. Survey after survey, hearing in background noise is the greatest problem for these people (Kochkin, 2010).
Over the years, we found ways to quantify this problem. We call it signal to noise ratio loss. You can measure this using the Hearing in Noise Test (HINT; Nilsson, Soli, and Sullivan, 1994) or the QuickSIN (Etymotic, 2001). The QuickSIN is a test that was developed at Etymotic Research, and it is a quick clinical way how to find out how much trouble someone is having hearing in noise, measured in dB. It is a test that will help you counsel the patient on what kind of directional options you can give on the hearing aid. For the most part, this group has an average signal to noise ratio loss of 5dB. I am going to talk about that more as we move on in the presentation here.
This group also has acceptable noise levels that are 5 to 10 dB lower for older listeners. What does that mean? The acceptable noise level is a test that has become very popular over the last few years. It is a good test that examines how much noise listeners are willing to put up with. Listeners with presbycusis cannot stand to be in a lot of noise, so their listening preferences require a greater signal to noise ratio. They have other issues associated with aging, manual dexterity being a huge one for us fitting hearing aids. They have very low tactile sensitivity. They cannot feel the ends of their fingers. I had someone come up to me the other day because they were not too happy with one of the buttons on a ReSound hearing aid and said to me, "I am going to put tape on your figures to show you how hard it is for these elderly people." Of course I know how hard it is for these people. It is a tradeoff we deal with in this industry when people want the miniaturization for cosmetic reasons, yet the cosmetics come with some difficulty for many of our patients. You have to use remote controls for these people.
There is reduced flexibility of joints. Often there is impaired vision. We are dealing with illnesses that sometimes reduce cognition in this population. It is more difficult to learn some of the operations of new digital hearing aids. We have pretty complicated hearing aid systems that can greatly benefit this population, but not if they cannot figure out how to use them. There are a lot of things that we have to think about when we are dealing with aging.
If we consider hearing loss configurations due to aging, we know that we have a tendency to lose hearing in the higher frequencies as we get older. If you look at average hearing loss as a function of age and gender, you will see that men start to lose hearing at a younger age, and on average have more significant hearing loss than women of the same age.
There are more people that have complications with genetics and noise-induced hearing loss, but the vast majority of people get into their 70s before they lose quite a bit of hearing. I have had multiple conversations with people about how we can get 65 to be the average age of the first-time hearing aid user, and I am not so sure that you can. I think many people do not present with problems enough to feel like they need a hearing aid until they are 70 or 71. I think that is why 71 continues to be our average age of the first time hearing aid user. Let's look beyond the hearing loss to some of the other issues.
This is a great example (Figure 1) of a woman's who's hands have tactile dexterity and joint issues. You can see how we have to take into account how this person might manipulate a hearing aid with small buttons and controls and why that is important. So as we ask the question, "Are modern hearing aids meeting the needs of the aging," much of the answer lies in making sure you are you doing everything you can when you look at the hearing loss and all the other complicating issues to give the patient something that is very useable.
Figure 1. Example of eldery hands with joint and dexterity difficulties.
Let's take a look at the hearing loss in Figure 2. I have been in the industry for 25 years, and I can tell you that this hearing loss used to be an extremely difficult loss to fit. Today if someone walked in with this hearing loss, I would hope that all of you would say, "Oh good. That is an easy hearing loss to fit." That ought to answer our question of modern technology meeting the amplification needs of aging adults right away. We have made enormous strides in fitting this population.
Figure 2. Audiogram demonstrating a sloping hearing loss due to presbycusis.
But the sloping configuration of this presbycusis hearing loss can still pose challenges. This person has essentially normal hearing in the lows, yet has a signal-to-noise-ratio loss. One of the largest problems in fitting hearing aids to this group with similar hearing losses has been the occlusion and feedback and overall cosmetics. Because we are now able to fit this hearing loss quite easily, audiologically speaking, we have to deal with instrument issues relating to occlusion and feedback because of the open-fit technologies that are available to us now.
There are other factors that have taken this hearing loss from difficult-to-fit to entirely possible. One advantage is wide dynamic range compression (WDRC). That advance was made quite a long time ago. We first had K amp processing which widened compression in the high frequencies and absolutely revolutionized our industry. Around the same time, ReSound Corporation put out WDRC throughout the entire frequency range. What did this do? It allowed us to give gain for soft sounds and leave loud sounds transparent or give gain for loud sounds only if the hearing loss was so much that they actually needed gains for loud sounds.
WDRC has made a huge impact in sound quality. Before WDRC we had digital signal processing. When digital hearing aids first came out, there were many questions as to how much better digital hearing aids were going to be over the analog hearing aids that were on the market at the time. At the time, they were not a whole lot better and probably made the sound quality worse. However, advancements in digital signal processing completely revolutionized hearing aid fittings.
The greatest impact has been felt in this aging demographic. Because of digital signal processing, we have feedback cancellation. Only over the last few years were we able to cancel feedback, which then allowed us to create a completely open fitting and help these people able to hear naturally. Other things that I will talk about as we go on in the presentation are noise reduction algorithms and helping with the acceptance of noise, directionality and improving the signal to noise ratio. We are making great strides in what we can do in a hearing aid, but it is what we can do outside the hearing aid that ultimately will meet the needs of this population. That also brings us to wireless connectivity, which is the next revolution in hearing aid development today. We can do more than we have ever done in multiple listening environments, which is so important to satisfying the needs of this population. Let's talk more in detail about each of these areas.
Wide Dynamic Range Compression
Figure 3 is a picture that shows the area of normal hearing, and then what has to happen to move sound into the compressed auditory area of someone who has loudness recruitment and sensorineural hearing loss. This is extremely typical of the patients that we fit. We want soft sounds to still sound and moderate sounds to be comfortable. However, we do not want loud sounds to ever exceed a patient's uncomfortable loudness level, and that becomes a different task to fit all this sound into a reduced dynamic range where sounds are still comfortable without getting too loud.
Figure 3. Schematic of the difference between a normal dynamic range and the compression that is required to fit the spectrum of soft, medium and loud speech into a reduced dynamic range typical of a person with hearing loss.
Before WDRC, it was almost impossible to accomplish this in linear hearing aids with one amount of gain. Say you had 30dB of gain set in a linear hearing aid. This meant that 30dB gain was applied for a 50 dB input as well as an 85dB input. When the output was too loud, we used peak clipping or very heavy compression limiting. The sound quality was terrible, however, and people did not like it.
Today, WDRC amplifies only the sounds that will need amplification which gives patients comfort for even loud sounds and much better sound quality. Components in hearing aids have improved over the years to the point where there is very little distortion. Even the noise floor of the actual computer chip in modern hearing aids continues to get lower and lower. In the last chip ReSound put out, we were able to get down to a 13 dB internal noise floor. You do not even hear anything when you put a stethoscope on and listen to a hearing aid anymore because the internal noise of the hearing aid has gotten so low. That is a great thing for most patients. Something to keep in mind, however, is that this is not such a great thing for patients with tinnitus. We were able to put a hearing aid on many patients with tinnitus and actually solve their problems with tinnitus because of the louder steady-state internal noise that used to exist in hearing aids. We do not have that any more. So on one hand this is an extremely good benefit for this population, but you have to think a little bit differently with tinnitus patients.
Advancements in Fittings
Moving on, I mentioned that digital signal processing has made fitting aging adults easier. I think the biggest revolution here has been in feedback controls, where we do not have the whistling hearing aids that we once had. Historically, there was always a tradeoff between occlusion and feedback for this aging kind of presbycusis hearing loss. I am sure many of you remember that the only tool we had at our disposal to alleviate the occlusion effect was venting. We could use select-a-vents or IROS vents which were much larger and we could use tube fittings. The problem with all of those was that the more open the vent became the more feedback you had. If you opened the vent enough so patients did not complain about the sound of their own voice you could get about 15dB of high-frequency gain. Think about this. The first-time user comes in at age 71 with thresholds around 40 or 50 dB in the 2 to 4 kHz range. If you look at even a half-gain rule, 15 dB of gain at 2, 3 and 4 kHz is not getting you anywhere near the audibility for high-frequency sounds required in order to make speech intelligible. This tradeoff kept us from ever being successful when fitting these patients.
If you look at insertion gain before feedback in an occluded response, you can get almost 30dB of high-frequency gain at 4 kHz out of a completely occluded fitting. But the minute you start opening up the vent, even 2 millimeters, you decrease the gain to around 20 dB. An IROS vent brings the gain down even more to 18 dB. If you did a slim tube fitting with an open dome, you may have only 15 dB of insertion gain at 4 kHz, and in many cases, probably less. Previously, the only way to get the gain you needed was to occlude, and occluding created the problem where patients did not like their hearing aids. Gus Mueller in The Hearing Journal (2003) published a study where 12 to 16 dB of occlusion was what most people experienced. The more severe the hearing loss was in the low frequencies, the less bothersome the occlusion effect was.
Today we are talking about presbycusis hearing losses. Most of these people do not have severe hearing loss in the low frequencies. Giving them a hearing aid that occluded 12 to 16 dB never made them happy. These occlusion effects were huge. Think about this as well. What happened in 2003 when this article was published? The very first open hearing aid was introduced to the market, the ReSound Air. In the same year, Wayne Staab (2003) reported that hearing aids that seal in the cartilaginous canal, such as custom products, normally result in occlusion effect measurements in the range of 20 to 30 dB at 250 Hz. We spent a lot of time telling our patients that they would get used to the sound of their voice but they never did. Data out of Sweden in 2002 noted in a survey that 34% of respondents said that the sound of their own voice sounded hollow or like it was echoing. These problems were experienced by these patients six months after they were fit. This demonstrates that a large group of hearing aid users do not get used to the sound of their own voice, causing even more problems.
What is happening with these open devices is 0 dB of occlusion. There is nothing to measure. Not only did the occlusion go away, but the way these devices were designed with the thin tube, the cosmetics became less of an issue for patients. We initially made them for occlusion reduction, but it came with the added benefit of smaller hearing aids. In a normal unoccluded ear canal, when low frequency sound is generated it is allowed to go right back out of the ear. By using an open mushroom dome, we allow that sound to continue to go in and out.
Consider a completely-in-the-ear (CIC) hearing instrument. They are oftentimes seated in the cartilaginous portion of the ear canal. They were not fit in such a way that could eliminate the occlusion effect. In the United States, we have moved to fitting the vast majority of our hearing aids as BTEs. We were able to accomplish all the open fittings because we had feedback cancellation in the hearing aids.
Feedback Cancellation
Most current systems use what is called phase-inversion feedback cancellation. When you have static feedback in a hearing aid, you can put a phase inverted filter into the hearing aid so that every time the feedback enters the hearing aid, it passes through the inverted filter and is canceled out. These feedback systems continue to get better and better. You cannot underestimate how important even the first generation of cancellation feedback systems were. They were actually very good, with up to 7dB of head room, which was quite a bit. Today, it is amazing the amount of head room that you can get out of these systems. Head room is the additional gain in the hearing aid that can be given because of the feedback cancellation mechanism.
Beyond head room, we have also been able to cancel out dynamic feedback. That is feedback due to putting your hand to your ear, putting a hat on, giving a close hug, et cetera. Talking on the phone, of course, would be very dynamic situation. Feedback cancellation systems have gotten so good today they work even well in dynamic cancellation. We have come leaps and bounds in our ability to fit an open hearing aid and get the kind of gain that a patient needs in order to understand speech in a comfortable way without occlusion or feedback. We can achieve close to 30 dB gain, much like in the occulded ear, with these modern feedback systems. This is the whole premise behind modern technologies meeting the amplification needs of aging adults.
We have talked about dynamic range compression, reduction in occlusion and feedback cancellation. Those three things together have made fitting the presbycusis hearing loss so much easier and better in the last few years. You probably know where I am going with this.
Noise Reduction
Let's move to noise reduction. I have said a couple of times that acceptable noise levels for aging adults who are wearing hearing aids are 5 to 10 dB lower than what they are for the general population. There is not only a need to improve the signal to noise ratio, like we do with directional microphones, but there is a need to lower the background noise all together for this population. They do not want to have a lot of background noise in their hearing aids. I have spent many months of my life wearing hearing aids, and I can tell you, even as a normal hearing person without acceptable noise level problems, the added amplification in the hearing aid is very difficult without noise reduction systems.
There are about two types of noise reduction algorithms on the market today. The first is modulation counters. Speech is an extremely modulated signal where you can actually count the modulations. In contrast, noise does not have these rapid, frequent amplitude changes so there is less modulation. If you want to tell the difference between speech and noise, you can count the number of modulations. Figure 4 illustrates the differences between speech in quiet and noise. You can easily see the number of modulations in the speech. Modulation counting systems will count the modulations in each frequency band of the hearing aid and. If it does not count several modulations, it specifies the input as noise and will take the noise down in those channels.
Figure 4. Demonstration of modulations between speech in a quiet background versus steady speech noise. Typical speech is more dynamic than speech noise.
Many hearing aids today come with something a bit more sophisticated called spectral subtraction noise reduction. The problem with modulation counting in our hearing aids was that when they reduced the gain, they tended to reduce the gain throughout the entire frequency range, and patients were often left with the experience that people were mumbling. A lot of the high-frequency cues that they needed to understand speech were taken away when the noise reduction systems kicked in. A spectral subtraction system attempts to look between the words in running speech. In running speech, you are always going to have a lull. The task for spectral subtraction is to look at only the lulls. It makes a running average estimate of what that noise in the background looks like and then tries to subtract it out, although it is impossible to subtract all of it out. In reality, however, you would never want to subtract it all out because that would sound very odd to the patient.
When you are talking clinically about noise reduction, you have to be very careful when you are fitting these presbycusis hearing losses with noise reduction systems. My advice in doing so is to make sure that you are fitting something that has spectral subtraction. That is very important now because it has been shown that it is a much clearer way to subtract out the noise without compromising audibility and the ability to understand the high-frequency consonants. What is important to be successful is knowing how much attenuation is actually applied. Noise reduction uses an algorithm that is important to verify on a patient's ear because it is possible that the hearing aid manufacturer labels the reduction as mild, moderate, or strong, but you cannot know for sure what that means until you verify it. You have to understand at a very basic level that there is not any way patients can understand speech and noise unless the speech is audible. You have to get out that noise without adversely affecting the audibility of speech.
For people with hearing loss due to presbycusis, you can often give a bit more noise reduction, so you can often go up to the moderate and strong settings without compromising audibility because these patients have so much good normal low-frequency hearing. As they get a little more low-frequency hearing loss with those strong noise reduction settings, you begin to compromise their audibility because you are taking out some of the gain they need, especially in the lows. There are some hearing aids on the market today that allow different amounts of noise reduction for different environments. So if it is completely a noise only environment, such as a city street with traffic, you could set the hearing aid to put a very strong amount of noise reduction there, yet provide much less in cases where there is a lot of speech in the environment and the patients need to understand. We continue to make advances in the types of noise reduction and how we can apply that noise reduction to patients.
Signal-to-noise ratio
Beyond just the comfort issues with hearing in noise, directionality is also extremely important for this patient population. There are many things that we can do using digital signal processing and directionality. We can apply fixed directionality solely, which is the cardioid or hyper cardioid pattern. The microphones and directional pattern are fixed and do not move at all. You can apply adaptive kinds of directionality that changes depending on the environment or situations. You can do asymmetric directionality where you fit one ear with as omnidirectional and one ear as directional. There is now something called split-band directionality. I will expand on these strategies as well as some of the newest things we can do to improve signal-to-noise-ratio loss.
Signal-to-noise-ratio loss suggests the amount of gain in dB that a person requires to understand at least 50% of the spoken passage in the presence of background noise. The stimulus is presented, on average, at 65 or 70 dB or sometimes louds so we can ensure audibility, even for hearing-impaired individuals. For example, a person with normal hearing may have a 2 dB signal-to-noise-ratio loss. They only need about a 2 dB to benefit or to have completely no signal to noise ratio loss in noise at all. The lower the number, the less difficulty they have. Ideally, negative numbers are best. An ear with a 40 dB hearing loss, however, will likely have a signal-to-noise-ratio loss on the order of 5 dB or greater. That is part of the sensorineural hearing loss. Now consider that you present the same material to the same patient with hearing loss through a hearing aid that employs linear peak clipping. We will actually create more signal-to noise-ratio loss with hearing aids that use peak clipping, and that is why so many hearing aids ended up in the drawer. People took those hearing aids into noisy environments, and we made what was already a problem for these people a much worse problem.
Now let's say that we take the same hearing loss again and apply a modern hearing aid. What happens as you get to that 60 to 70dB range, we remain at that 5 dB signal-to-noise-ratio loss, even as the background noise gets louder and louder.
If you are not that familiar with these kinds of tests or with signal-to-noise-ratio loss, we will review how you can measure it. Figure 5 shows how normal hearing compares to hearing loss in terms of signal-to-noise-ratio loss. On the QuickSIN test (Etymotic, 2001), normals score about 2 dB for the 50%-correct mark. The dotted white line happens to be one hearing-impaired patient that scored about 8 dB on the test. In order to find out what the loss is, you would take 8 dB minus the normal 2 dB and you get a 6dB loss. This is a characteristic of people who have presbycusis hearing loss, but it can range all over. Measuring signal-to-noise ratio loss is extremely important and is something that you will want to do for these aging adults.
Figure 5. Comparison and calculation of signal-to-noise-ratio loss for normal and hearing-impaired listeners.
There are varying degrees of signal-to-noise ratio loss (Figure 6). If they have loss in the 3 to 7 dB range, a mild loss, you will put directional microphones on them and they will hear almost as well as normals. But if you get to the severe group that has 12 to 15 dB of loss, you are going need to do a whole lot more than directional microphones alone.
Figure 6. Signal-to-noise-ratio loss categories and the expected outcome by adding directional technology.
Directional Microphones
Adaptive directionality will attenuate two noises at one time. These adaptive systems can do different kinds of directionality with two or even more noise sources at the same time. This is a big advancement. Another advancement is split-band directionality, which is where you have a blending point between systems. You can do omni directional processing in the low frequencies and directional processing in the high frequencies. This allows a patient to not have a tinny sound quality. It gives overall better sound quality to elderly patients wearing directional hearing aids.
Asymmetric fittings allow you to put a person in two different directional strategies. If you have a patient listening in a group setting in the middle of the room in directional mode, they would only be able to listen within a small cone of the room. By putting one hearing aid on omni directional and one in directional, they can also hear everything going on around them as well as having the speaker focused in directionally. This can be extremely beneficial to giving someone back that more "normal" ability to hear.
We also have the ability to change the beamwidth of the directional system. Picture a classroom teacher with lots of kids in front of her. She would want the beam or tunnel cone to be very wide open to be able to hear all her students. But when she starts to talk to one student, she is going to want a much narrower beam. There are hearing aids on the market today that will automatically do this, or there are things that you can set up to manually change the beamwidths.
Wireless Connectivity
Much advancement has been made in technology, as we have already discussed. But it is the wireless connectivity that is going to allow patients to hear better in multiple listening environments. Today on the market there are two different kinds of wireless systems. The first is near-field magnetic induction (NFMI). Bluetooth is an example of this, where a signal is transmitted to an intermediary device and that signal is sent to the hearing aids. With a near-field system, you actually have to wear the intermediary device because distance is a key factor. It is a short transmission distance to do that. You do get quite a delay in the audio using a Bluetooth signal, up to even 130 milliseconds delay. But if you want to have something quick, you have look at something that is more on the order of 18 milliseconds, and that can be accomplished using real radio frequency (RF) technology.
There are a couple of RF systems on the market today where a radio frequency of 2.4 gigahertz or 900 megahertz is being used to go directly from an accessory right into the hearing aids. What this means for patients is that there is no intermediary device and a much longer range of connection of up to seven meters. RF transmission does not have echo or lip synchronization issues when watching TV. Now we can add accessories for patients that will solve some of these problems that elderly patients have.
In particular, we have companion mic systems. Figure 7 shows speech-understanding-in-noise scores for three different distances. The left point is just at one-and-a-half meters, the middle point is 3 meters and the right point is 6 meters. The bottom curve is a basic hearing aid with directional microphones. You can see as you go further away in distance, the performance worsens in noise. The two top curves represent two ReSound hearing aids with companion microphone. What you will notice is that you receive a 20dB SNR improvement, just like an FM system. I think the future with this technology is extremely bright.
Figure 7. Speech reception thresholds using three different directional microphone strategies in a digital hearing aid. Click Here for larger view (PDF)
My last point is about the top 10 factors related to overall customer satisfaction (Kochkin, 2010). Overall benefit was the number one factor related to overall satisfaction. Following next was clarity of sound, value, natural sound, reliability of the hearing aid, richness or fidelity of sound, use in noisy situations, the ability to hear in small groups, comfort with loud sounds, and the sound of their own voice. MarkeTrak (Kochkin, 2010) pointed out that improvements in these areas provide improvements in overall satisfaction.
If you think about what I talked about throughout the last hour, we have improved on hearing aids sound natural because we have band split directionality today. We have improved reliability. I have not talked about this today, but most hearing aids now have nano coating or water protection; we have also improved in that area. We have improved in the richness and fidelity of sound with low distortion and band split directionality. We have directionality for noise reduction and the connectivity with these mini microphones only improving this more as time goes on. In satisfaction trends, we are up six points in overall satisfaction from 2004 where it was 74% to 80% today (Kochkin, 2010). If you look at many BTE users with open fittings, their satisfaction rates are 85% today (Kochkin, 2010).
Let's return to that initial question, "Are we meeting the needs of aging individuals who have hearing loss?" I think we are doing a pretty good job meeting needs for individuals with presbycusis kinds of hearing losses. Can we do better? Absolutely. Of course there are needs still to be met. Of course there are still problems. Not every one of your patients is happy. But we have made amazing strides.
References
Etymotic Research. (2001). Quick Speech-in-Noise Test. [Audio CD]. Elk Grove Village, IL: Author.
Kochkin, S. (2010). MarkeTrak VIII: Consumer satisfaction with hearing aids is slowly increasing. The Hearing Journal, 63(1), 19-20,22,24,26,28,30-32.
Mueller, H.G. (2003). There's less talking in barrels, but the occlusion effect is still with us. The Hearing Journal, 56(1), 10-18.
Nilsson, M., Soli, S.D., & Sullivan, J.A. (1994). Development of the Hearing in Noise Test for the measurement of speech reception thresholds in quiet and in noise. Journal of the Acoustical Society of America, 95(2), 1085-1099.
Staab, W. (2003). Solving the occlusion effect. Hearing Review. Retrieved from: https://www.hearingreview.com/issues/articles/2003-08_05.asp
