Interview with Aaron Thornton Ph.D.
Share:
Topic: Historical Notes: Automation in Audiology & Tympany
AO/Beck: Good Morning Dr. Thornton. It is a pleasure to meet you. Before we get to issues in automation, would you please tell me when you got your Ph.D.?
Thornton: I received my doctoral degree in 1975 from the University of Iowa, where I stayed to serve on the faculty for a couple of years.
AO/Beck: What was your dissertation on?
Thornton: My dissertation was on middle latency auditory evoked potentials, and one interesting aspect was that they could be recorded during sleep, when myogenic noise is generally at a minimum and the signal-to-noise ratio is best This offered a tremendous advantage over late-component recording. We were looking at how to use the middle latency evoked response to measure hearing, and I started thinking about using evoked potentials as a means of screening babies for hearing loss. You could set up a baby for testing, record a large amount of data on tape, and then send it to some central place to process it through a computer. By separating the raw data recording from the then expensive computer processing, the cost might be affordable.
AO/Beck: That seems very clever, with one possible flaw!
Thornton: Correct. What I didn't understand at that time was that the middle latency response is virtually non-existent in babies! So that was a red herring! Serendipitously, the extreme filtering that was in vogue for recording the middle-latency response, or MLR, was phase-shifting energy from the ABR into the MLR time domain, which gave the appearance of a response in babies. None-the-less we eventually developed an algorithm for detecting the evoked potential and built a screening instrument that ran completely automatically. This was completed about four years later when I was at the University of Wisconsin in Madison
AO/Beck: That must've been incredibly exciting.
Thornton: Well yes, it was, but it was very disappointing too. It was 1976 or so, and I couldn't get anyone interested in it. We had a false positive rate that was simply miniscule, yet no one was interested in building this device. That's when I started learning about the politics of medicine and audiology.
AO/Beck: For example?
Thornton: Well, back then, the reason you didn't test babies was because you didn't have the technology. So I developed the technology, and then I found out people really weren't interested in testing babies. They were interested in talking about testing babies. It was incredibly frustrating, and it wasn't until I met Maurizzio Liverani in 1983 that the ideas started to gel. He was trying to start a company, a Silicon Valley start-up, and a mutual acquaintance, Gian Mazzoni, put us in contact. That led to the development of the Algo-I Infant Hearing Screener, but it took until 1985 to get the first commercial infant hearing screening that actually worked and has stayed in production with little change to the underlying method to the present day.
AO/Beck: I remember that unit. In fact, we went way off label, and used it in the operating room. I really liked small and portable in those days.
Thornton: You're right, that was indeed off-label!
AO/Beck: What were the false positive and false negative rates like at that time? I seem to recall they were quite good.
Thornton: Yes, they really were. We had tremendous statistical power with that unit, which has been retained in the newer Natus instruments as well The false-negative rate was under exact control and we set that to about 1 in 10,000. You cannot let the instrument pass a deaf baby. But, for the machine to be economically feasible, we absolutely had to control the false positive rate, and we really had a nice protocol devised and it was very reliable. Our philosophy was do the test right and only do it once . With the Algo, there has never been a need for the expensive two-stage screening protocols. I would like to point out that automation in audiology isn't a new idea at all. In Jerger's old 1963 textbook, Modern Developments of Audiology , Chapter 2, by Wayne Rudmose, was devoted entirely to automated audiometry. It was over 45 pages in length, which was more than 10% of the content of the book.
AO/Beck: Excellent point. What year did you start at Harvard Medical School.
Thornton: That was in 1979. I went to the Massachusetts Eye and Ear Infirmary specifically to set up a clinical laboratory for auditory evoked potentials. It was a stand-alone clinic that had to earn its own way. It was very exciting, and I stayed with Harvard at the Infirmary until I retired in 2001. My principle interest in life seems to have been invention, and audiology has been the medium. So whatever I've worked on, I've found that the most interesting thing is to try to improve it, reduce cost, and bring in technology to reduce error. To do that, you must first understand everything about the processes that you are trying to improve, and often that can lead to some interesting discoveries and insights about what we do and why we do it.
AO/Beck: Absolutely. That makes sense, and you've applied those goals to clinical audiometry too?
Thornton: Yes, I have recently worked on a few projects that involve automation and audiometry. Of course, I had a terrific advantage at the Infirmary as the director of the audiology program. We had a patient volume that I think reached a peak of about 30,000 patient visits in a year, and at one time we had 35 full time audiologists on staff. So efficiency and automation were important issues, and we had the interest and ability to explore them. With that kind of volume the computer- based, semi-automated audiometer that we developed in house saved the Infirmary about $100,000 a year in departmental operating costs, so it easily paid for the development cost, and it improved the quality of service as well. I must give credit to the support given by Dr. Nadol, the Chief of Otolaryngology, and the many audiologists and engineers in the department who worked on this project. Special thanks to Dr. Chris Halpin for his dedication to this effort over a period of many years.
AO/Beck: I think it's important to note that you wrote an article which included the efficiencies of automation - I think in the AJA?
Thornton: That's right. I wrote an article in 1993 for the American Journal of Audiology that addressed the early audiometer project and explained many of the realized benefits of computer assistance and automation for audiometry, including its performance in the hands of audiometric technicians. The computer-assisted audiometer allowed us to set-up an audiometric technician program, which I'd like to explain. We decided to try using audiometric technicians during a period when there seemed to be a shortage of qualified audiologists and a relentless pressure from the hospital to cut operating costs. We set this up so that an audiologist would review a patient's chart and determine whether or not the audiometry could be done by a technician. The technician didn't talk with the patient, didn't counsel the patient, didn't ask questions, and didn't answer questions. They just did the hearing test after the audiologist determined it was appropriate. Then, after the technician completed the test, it was reviewed by the audiologist, and the audiologist did the counseling and whatever else was needed, and returned the patient to the physician.
We hired four technicians and we actually did a study of inter-examiner reliability that compared the four technicians, four CFY's, and four experienced staff audiologists to a reference group of three senior staff audiologists.
AO/Beck: And what did you find?
Thornton: The data showed that all three groups showed comparable variance with the reference group of senior audiologists. CFY's differed the least and experienced audiologists differed the most, with technicians falling in between these two groups. So we had solid data that with the computer-assisted audiometer, the technician's performance was as good or better than that of experienced audiologists. However, our experiment with the use of technicians was short lived, but this wasn't because of any inadequacies of their audiometry. Rather, it was due to our inability to provide the steady workload that would have made them cost effective. Audiologic services at the Infirmary were predominately unscheduled, so the number of patients needing a hearing test at a particular time could vary considerably. Audiologists could work with hearing aids, implant patients, ABR, and a wide range of support activity, but the technicians couldn't. The audiologists had no down time, but the technicians were only doing basic audiometry, and because we couldn't keep them busy enough, they actually cost more than audiologists per unit of clinical work! But we learned that the concept was good and in many environments where you have a highly regulated case load or a lot of appointments, it could work very well. Alternatively, a fully-automated version of our audiometer would not have required the technician and downtime would not have been an issue.
AO/Beck: Very good. Well, let's fast forward a bit to Tympany and the Otogram. I know you have been involved in that project as a consultant, and probably the best thing to do is to ask you to tell us something about Tympany and describe the Otogram?
Tympany is a new company that developed around a plan to improve early diagnosis of hearing loss by making audiometry accessible in a wider range of settings. For the most part, audiometry is being done much the same as it was forty years ago, though the relative costs have changed markedly since the Rudmose chapter on automation. Back then labor and space were cheap and electronic equipment was expensive. Today, labor and space are expensive and equipment is cheap. Also, our society has gradually shifted the costs of services to recipients. We pump our own gas, take our own blood pressure, measure our own blood chemistry, and many other things that were formerly be done by trained personnel. Tympany believes that technology has advanced to the point that most people can effectively test their own hearing with computerized equipment. With this concept, hearing tests can be affordable in many offices that cannot provide the patient volume to support an audiologist. Also, there are vast rural areas of the country that have never had local access to hearing testing.
Tympany's progress in realizing their plan has been astounding, especially when you compare it to any previous audiologic equipment development. When I first heard of their plans to make a self-testing diagnostic audiometer, I tried to talk them out of it, not because it was a bad idea but because it would be so difficult. The complexity of audiometry is always underestimated by those who don't do it. I could easily think of a thousand ways that it might be made to fail, and I wasn't sure if there was even one way to make it work with high criteria for performance. Bob Margolis has been working for years with NIH support to prove the concept of fully-automated diagnostic pure-tone testing. And now, Tympany wanted to set the bar even higher for the full battery of tests, not to mention implementation in several languages.
That first inquiry from Tympany was in the Fall of 2001, just as I had retired from the Infirmary. Amazingly, the first Otogram, which is Tympany's name for the instrument, was ready for shipment and was first shown the Fall of 2002 at the AAO meeting, only one year from concept to completion. That attests to the quality of the team that Tympany put together. On the instrument development side we had Jeff Harrison, a senior engineer whom I worked with on the original Algo-1 and the Nicolet NIM. Also, Jeff had worked with me on the automated audiometer developed at the Infirmary. Jon Birck, who founded the Virtual Corporation and sold the first totally computer-driven audiometer, designed and built the hardware. I have never seen such a rapid development combined with outstanding quality. Of course none of this would have been accomplished without equally fine work from everyone else at Tympany, especially Ken Stott, who designed the user interface and brought it all together under to leadership of Chris Wasden, the company president.
Every effort was made to design the Otogram around time-tested audiometric methods and to draw as much as possible from published research and standards. The result is a self-testing audiometer that tests pure-tone air- and bone-conduction thresholds, SRT, and Word Recognition, employing masking when it is needed. I don't think we have time to go into all of the specific algorithms that are used, but I will say that most are derived from the same thinking that audiologists use when testing manually. The system is sensitive to a wide range of potential error conditions, including masking dilemma, malingering, and excessive ambient noise, since the Otogram has been designed to work in office settings that do not have special sound-treating or testing chambers. By using forehead bone-conduction and insert earphones, and by calibrating for occluded bone-conduction, we can achieve a satisfactory level of sound attenuation and there is no need to move transducers during the testing. Ambient noise microphones are used to calculate the masking effects, if any, of any room noise and either retest or report on the degree of noise floor interference if it is both persistent and significant. Finally, we have an integrated paging system that is used to alert the person overseeing the test to patient requests for help, problems detected by the Otogram and the completion of a test.
AO/Beck: I wonder if you can share your thoughts on the newest version of automated hearing tests? What do you say to people who tell you they're very concerned about automated hearing tests?
Thornton: The latest version of the Otogram integrates tympanometry, acoustic reflexes and OAE's into the testing session, using a single insertion of bilateral probes. And the Otogram will now guide the patient through all of the procedures and do the speech tests in five languages. More are being prepared. It has worked accurately with preschoolers and geriatrics, and the patients like the test somewhat better than traditional testing. The power of this is quite naturally scary to some audiologists, who might wonder how this can be done and harbor doubts until they can see it with their own eyes. I don't mean to be disrespectful, but audiologists historically have been more openly distrustful of each other than have other professions. It was once common to read or hear my audiology colleagues advise others to seek out or refer to a competent audiologist. Physicians, dentists, psychologists, etc. seldom, if ever, see the need to qualify their peers in a such a manner. The practice is decreasing in frequency, but here is a direct quote taken from the AAA website. How can I find a good audiologist to provide me with hearing aids? Doesn't this imply that some might not be so good? I guess that what I am driving at is that in addition to (or in place of) a concern that the Otogram might not test reliably, some audiologists worry that the Otogram will be used improperly by others, especially non-audiologists. I guess that I have greater faith that conscientious professionals of all disciplines take care not to venture far beyond their areas of expertise. Naturally there are exceptions, but those occur with or without the Otogram.
AO/Beck: In the final analysis, do you see the Otogram as benefiting or hurting the audiology profession?
Thornton: Benefiting, definitely! First, let's address professional image. Audiology has been struggling for decades to shed the cloak of being no more than a low-paid, knob-twiddling technician who might help the office manager with filing duties when the patient load was slack. Second, the Otogram helps us to cut costs and increase the proportion of our time spent on services having higher rates of reimbursement. While we have embraced automation in tympanometry, ABR, OAE, and electro-acoustic measurement of hearing aids, we have doggedly clung to a labor-intensive approach to the basic diagnostic test battery. Labor costs are now the dominate factor in everything we do, but the most labor-intensive test that we do today, audiometry, is reimbursed at the lowest rate per unit of time spent. Salaried audiologists might not be sensitive to this issue, but their employers surely are. Third, the Otogram is needed to compensate for the impending shortage of audiologists that will accompany our transition to a doctoral entry-level for the profession. If we cannot provide for the current level of hearing testing, let alone bring it to additional patients, the medical profession will surely create technician-training programs to fill the need. I am not sure that we would retain control if that were to happen.
The Otogram is easily incorporated into the practices of most audiologists. Difficult-to-staff satellite facilities are ideal for the Otogram, but it is equally beneficial for handling fractional staffing needs, unscheduled or overscheduled patient loads, covering for staff on leave, and for the too-frequent end-of-day walk-in. Also, we have the ever-increasing non-English speaking caseload. Most physicians want an audiologist's interpretation of the audiogram, so they would like to see the cost-savings of the Otogram combined with the audiologist's management of the test. I think that the Otogram can only be a threat if we make it one. Audiologists are in the best position to make this a tool for increasing their productivity and for using it as a means for freeing more of their time for more demanding professional work that cannot be automated. The technology is here and it is our choice as to whether to make it work for us or against us. I think we will adapt and may eventually wonder how we worked without it.
AO/Beck: Thanks Aaron. I appreciate your time and thoughts on these matters.
Thornton: Thank you too, Doug.
For more information, click here.
Thornton: I received my doctoral degree in 1975 from the University of Iowa, where I stayed to serve on the faculty for a couple of years.
AO/Beck: What was your dissertation on?
Thornton: My dissertation was on middle latency auditory evoked potentials, and one interesting aspect was that they could be recorded during sleep, when myogenic noise is generally at a minimum and the signal-to-noise ratio is best This offered a tremendous advantage over late-component recording. We were looking at how to use the middle latency evoked response to measure hearing, and I started thinking about using evoked potentials as a means of screening babies for hearing loss. You could set up a baby for testing, record a large amount of data on tape, and then send it to some central place to process it through a computer. By separating the raw data recording from the then expensive computer processing, the cost might be affordable.
AO/Beck: That seems very clever, with one possible flaw!
Thornton: Correct. What I didn't understand at that time was that the middle latency response is virtually non-existent in babies! So that was a red herring! Serendipitously, the extreme filtering that was in vogue for recording the middle-latency response, or MLR, was phase-shifting energy from the ABR into the MLR time domain, which gave the appearance of a response in babies. None-the-less we eventually developed an algorithm for detecting the evoked potential and built a screening instrument that ran completely automatically. This was completed about four years later when I was at the University of Wisconsin in Madison
AO/Beck: That must've been incredibly exciting.
Thornton: Well yes, it was, but it was very disappointing too. It was 1976 or so, and I couldn't get anyone interested in it. We had a false positive rate that was simply miniscule, yet no one was interested in building this device. That's when I started learning about the politics of medicine and audiology.
AO/Beck: For example?
Thornton: Well, back then, the reason you didn't test babies was because you didn't have the technology. So I developed the technology, and then I found out people really weren't interested in testing babies. They were interested in talking about testing babies. It was incredibly frustrating, and it wasn't until I met Maurizzio Liverani in 1983 that the ideas started to gel. He was trying to start a company, a Silicon Valley start-up, and a mutual acquaintance, Gian Mazzoni, put us in contact. That led to the development of the Algo-I Infant Hearing Screener, but it took until 1985 to get the first commercial infant hearing screening that actually worked and has stayed in production with little change to the underlying method to the present day.
AO/Beck: I remember that unit. In fact, we went way off label, and used it in the operating room. I really liked small and portable in those days.
Thornton: You're right, that was indeed off-label!
AO/Beck: What were the false positive and false negative rates like at that time? I seem to recall they were quite good.
Thornton: Yes, they really were. We had tremendous statistical power with that unit, which has been retained in the newer Natus instruments as well The false-negative rate was under exact control and we set that to about 1 in 10,000. You cannot let the instrument pass a deaf baby. But, for the machine to be economically feasible, we absolutely had to control the false positive rate, and we really had a nice protocol devised and it was very reliable. Our philosophy was do the test right and only do it once . With the Algo, there has never been a need for the expensive two-stage screening protocols. I would like to point out that automation in audiology isn't a new idea at all. In Jerger's old 1963 textbook, Modern Developments of Audiology , Chapter 2, by Wayne Rudmose, was devoted entirely to automated audiometry. It was over 45 pages in length, which was more than 10% of the content of the book.
AO/Beck: Excellent point. What year did you start at Harvard Medical School.
Thornton: That was in 1979. I went to the Massachusetts Eye and Ear Infirmary specifically to set up a clinical laboratory for auditory evoked potentials. It was a stand-alone clinic that had to earn its own way. It was very exciting, and I stayed with Harvard at the Infirmary until I retired in 2001. My principle interest in life seems to have been invention, and audiology has been the medium. So whatever I've worked on, I've found that the most interesting thing is to try to improve it, reduce cost, and bring in technology to reduce error. To do that, you must first understand everything about the processes that you are trying to improve, and often that can lead to some interesting discoveries and insights about what we do and why we do it.
AO/Beck: Absolutely. That makes sense, and you've applied those goals to clinical audiometry too?
Thornton: Yes, I have recently worked on a few projects that involve automation and audiometry. Of course, I had a terrific advantage at the Infirmary as the director of the audiology program. We had a patient volume that I think reached a peak of about 30,000 patient visits in a year, and at one time we had 35 full time audiologists on staff. So efficiency and automation were important issues, and we had the interest and ability to explore them. With that kind of volume the computer- based, semi-automated audiometer that we developed in house saved the Infirmary about $100,000 a year in departmental operating costs, so it easily paid for the development cost, and it improved the quality of service as well. I must give credit to the support given by Dr. Nadol, the Chief of Otolaryngology, and the many audiologists and engineers in the department who worked on this project. Special thanks to Dr. Chris Halpin for his dedication to this effort over a period of many years.
AO/Beck: I think it's important to note that you wrote an article which included the efficiencies of automation - I think in the AJA?
Thornton: That's right. I wrote an article in 1993 for the American Journal of Audiology that addressed the early audiometer project and explained many of the realized benefits of computer assistance and automation for audiometry, including its performance in the hands of audiometric technicians. The computer-assisted audiometer allowed us to set-up an audiometric technician program, which I'd like to explain. We decided to try using audiometric technicians during a period when there seemed to be a shortage of qualified audiologists and a relentless pressure from the hospital to cut operating costs. We set this up so that an audiologist would review a patient's chart and determine whether or not the audiometry could be done by a technician. The technician didn't talk with the patient, didn't counsel the patient, didn't ask questions, and didn't answer questions. They just did the hearing test after the audiologist determined it was appropriate. Then, after the technician completed the test, it was reviewed by the audiologist, and the audiologist did the counseling and whatever else was needed, and returned the patient to the physician.
We hired four technicians and we actually did a study of inter-examiner reliability that compared the four technicians, four CFY's, and four experienced staff audiologists to a reference group of three senior staff audiologists.
AO/Beck: And what did you find?
Thornton: The data showed that all three groups showed comparable variance with the reference group of senior audiologists. CFY's differed the least and experienced audiologists differed the most, with technicians falling in between these two groups. So we had solid data that with the computer-assisted audiometer, the technician's performance was as good or better than that of experienced audiologists. However, our experiment with the use of technicians was short lived, but this wasn't because of any inadequacies of their audiometry. Rather, it was due to our inability to provide the steady workload that would have made them cost effective. Audiologic services at the Infirmary were predominately unscheduled, so the number of patients needing a hearing test at a particular time could vary considerably. Audiologists could work with hearing aids, implant patients, ABR, and a wide range of support activity, but the technicians couldn't. The audiologists had no down time, but the technicians were only doing basic audiometry, and because we couldn't keep them busy enough, they actually cost more than audiologists per unit of clinical work! But we learned that the concept was good and in many environments where you have a highly regulated case load or a lot of appointments, it could work very well. Alternatively, a fully-automated version of our audiometer would not have required the technician and downtime would not have been an issue.
AO/Beck: Very good. Well, let's fast forward a bit to Tympany and the Otogram. I know you have been involved in that project as a consultant, and probably the best thing to do is to ask you to tell us something about Tympany and describe the Otogram?
Tympany is a new company that developed around a plan to improve early diagnosis of hearing loss by making audiometry accessible in a wider range of settings. For the most part, audiometry is being done much the same as it was forty years ago, though the relative costs have changed markedly since the Rudmose chapter on automation. Back then labor and space were cheap and electronic equipment was expensive. Today, labor and space are expensive and equipment is cheap. Also, our society has gradually shifted the costs of services to recipients. We pump our own gas, take our own blood pressure, measure our own blood chemistry, and many other things that were formerly be done by trained personnel. Tympany believes that technology has advanced to the point that most people can effectively test their own hearing with computerized equipment. With this concept, hearing tests can be affordable in many offices that cannot provide the patient volume to support an audiologist. Also, there are vast rural areas of the country that have never had local access to hearing testing.
Tympany's progress in realizing their plan has been astounding, especially when you compare it to any previous audiologic equipment development. When I first heard of their plans to make a self-testing diagnostic audiometer, I tried to talk them out of it, not because it was a bad idea but because it would be so difficult. The complexity of audiometry is always underestimated by those who don't do it. I could easily think of a thousand ways that it might be made to fail, and I wasn't sure if there was even one way to make it work with high criteria for performance. Bob Margolis has been working for years with NIH support to prove the concept of fully-automated diagnostic pure-tone testing. And now, Tympany wanted to set the bar even higher for the full battery of tests, not to mention implementation in several languages.
That first inquiry from Tympany was in the Fall of 2001, just as I had retired from the Infirmary. Amazingly, the first Otogram, which is Tympany's name for the instrument, was ready for shipment and was first shown the Fall of 2002 at the AAO meeting, only one year from concept to completion. That attests to the quality of the team that Tympany put together. On the instrument development side we had Jeff Harrison, a senior engineer whom I worked with on the original Algo-1 and the Nicolet NIM. Also, Jeff had worked with me on the automated audiometer developed at the Infirmary. Jon Birck, who founded the Virtual Corporation and sold the first totally computer-driven audiometer, designed and built the hardware. I have never seen such a rapid development combined with outstanding quality. Of course none of this would have been accomplished without equally fine work from everyone else at Tympany, especially Ken Stott, who designed the user interface and brought it all together under to leadership of Chris Wasden, the company president.
Every effort was made to design the Otogram around time-tested audiometric methods and to draw as much as possible from published research and standards. The result is a self-testing audiometer that tests pure-tone air- and bone-conduction thresholds, SRT, and Word Recognition, employing masking when it is needed. I don't think we have time to go into all of the specific algorithms that are used, but I will say that most are derived from the same thinking that audiologists use when testing manually. The system is sensitive to a wide range of potential error conditions, including masking dilemma, malingering, and excessive ambient noise, since the Otogram has been designed to work in office settings that do not have special sound-treating or testing chambers. By using forehead bone-conduction and insert earphones, and by calibrating for occluded bone-conduction, we can achieve a satisfactory level of sound attenuation and there is no need to move transducers during the testing. Ambient noise microphones are used to calculate the masking effects, if any, of any room noise and either retest or report on the degree of noise floor interference if it is both persistent and significant. Finally, we have an integrated paging system that is used to alert the person overseeing the test to patient requests for help, problems detected by the Otogram and the completion of a test.
AO/Beck: I wonder if you can share your thoughts on the newest version of automated hearing tests? What do you say to people who tell you they're very concerned about automated hearing tests?
Thornton: The latest version of the Otogram integrates tympanometry, acoustic reflexes and OAE's into the testing session, using a single insertion of bilateral probes. And the Otogram will now guide the patient through all of the procedures and do the speech tests in five languages. More are being prepared. It has worked accurately with preschoolers and geriatrics, and the patients like the test somewhat better than traditional testing. The power of this is quite naturally scary to some audiologists, who might wonder how this can be done and harbor doubts until they can see it with their own eyes. I don't mean to be disrespectful, but audiologists historically have been more openly distrustful of each other than have other professions. It was once common to read or hear my audiology colleagues advise others to seek out or refer to a competent audiologist. Physicians, dentists, psychologists, etc. seldom, if ever, see the need to qualify their peers in a such a manner. The practice is decreasing in frequency, but here is a direct quote taken from the AAA website. How can I find a good audiologist to provide me with hearing aids? Doesn't this imply that some might not be so good? I guess that what I am driving at is that in addition to (or in place of) a concern that the Otogram might not test reliably, some audiologists worry that the Otogram will be used improperly by others, especially non-audiologists. I guess that I have greater faith that conscientious professionals of all disciplines take care not to venture far beyond their areas of expertise. Naturally there are exceptions, but those occur with or without the Otogram.
AO/Beck: In the final analysis, do you see the Otogram as benefiting or hurting the audiology profession?
Thornton: Benefiting, definitely! First, let's address professional image. Audiology has been struggling for decades to shed the cloak of being no more than a low-paid, knob-twiddling technician who might help the office manager with filing duties when the patient load was slack. Second, the Otogram helps us to cut costs and increase the proportion of our time spent on services having higher rates of reimbursement. While we have embraced automation in tympanometry, ABR, OAE, and electro-acoustic measurement of hearing aids, we have doggedly clung to a labor-intensive approach to the basic diagnostic test battery. Labor costs are now the dominate factor in everything we do, but the most labor-intensive test that we do today, audiometry, is reimbursed at the lowest rate per unit of time spent. Salaried audiologists might not be sensitive to this issue, but their employers surely are. Third, the Otogram is needed to compensate for the impending shortage of audiologists that will accompany our transition to a doctoral entry-level for the profession. If we cannot provide for the current level of hearing testing, let alone bring it to additional patients, the medical profession will surely create technician-training programs to fill the need. I am not sure that we would retain control if that were to happen.
The Otogram is easily incorporated into the practices of most audiologists. Difficult-to-staff satellite facilities are ideal for the Otogram, but it is equally beneficial for handling fractional staffing needs, unscheduled or overscheduled patient loads, covering for staff on leave, and for the too-frequent end-of-day walk-in. Also, we have the ever-increasing non-English speaking caseload. Most physicians want an audiologist's interpretation of the audiogram, so they would like to see the cost-savings of the Otogram combined with the audiologist's management of the test. I think that the Otogram can only be a threat if we make it one. Audiologists are in the best position to make this a tool for increasing their productivity and for using it as a means for freeing more of their time for more demanding professional work that cannot be automated. The technology is here and it is our choice as to whether to make it work for us or against us. I think we will adapt and may eventually wonder how we worked without it.
AO/Beck: Thanks Aaron. I appreciate your time and thoughts on these matters.
Thornton: Thank you too, Doug.
For more information, click here.
