DSL, RECD, and Pediatric Amplification: Interview with Richard C. Seewald, PhD
Douglas L. Beck, AuD, Web content editor, speak with Dr. Seewald about desired sensation level (DSL), real ear to coupler difference (RECD), SPLograms, and more.
Academy: Good morning, Richard. Thanks for your time.
Seewald: Hi, Doug. Good to talk with you again.
Academy: Richard, where and when did you get your doctorate?
Seewald: Well, I received my PhD from the University of Connecticut in 1981. I went there to study with Mark Ross to learn more about his work in fitting pre-verbal children with hearing aids. Interestingly, by the time he and I got together he was more interested in educational audiology. So my dissertation looked at auditory and visual speech perception as it related to speech production in school-age children.
Academy: And as that’s just about the same time period in which I was getting involved with audiology, I wonder if you had a better way of fitting children with hearing aids than I recall?
Seewald: Well, it was pretty fuzzy then. As a profession, we used the “by the seat of your pants” approach.
Academy: I remember it well. I think it started with a “half-gain” rule and then we just tweaked things from there?
Seewald: Yeah, that’s about right. I mean really, all we had was a rough half-gain rule, some really limited hearing aids and little understanding of earmold and ear-canal acoustics, and we just didn’t know very much about auditory and speech processing how to measure the outcomes of our interventions with amplification. In retrospect, it was very frustrating as there were no real fitting formulas to apply with infants and children. I thought there must be a better way to do this.
Academy: And so you started working on DSL?
Seewald: Well, yes. We thought there had to be a relationship between the hearing loss and the amount of gain prescribed as a function of frequency. Mark Ross had written an early chapter in the mid-1970s on hearing aid fitting in children, and he was using what we now refer to as “SPLograms” to bring hearing measurements and electroacoustic measurements to the same reference. This was the staring place for our work.
Academy: That was very clever. And just to clarify, in that early chapter in Michael Pollacks’s book, Mark was trying to standardize input and output, as well as SPL for hearing aids and HL for hearing tests.
Seewald: Right, he was trying to have all the information in the same “language” so that the important interrelationships between all relevant auditory and electroacoustic variables could be studied. I’ve always thought that we (i.e., audiologists) made the fitting process unnecessarily difficult on ourselves by working in these different scales and domains.
Academy: I agree... but then again, I still argue that we should’ve transitioned to the metric system decades ago. I can recall signs on the New York State Thruway in kilometers as we tried to transition, before totally giving up a few decades ago. Anyway, yes, it certainly makes sense to use SPL for all hearing and hearing aid measures. In fact, one could argue that audiograms should also have amplitude increase as it rises across the audiogram... but I think these changes are unlikely in our lifetime!
Seewald: You’re probably right. Nonetheless, that’s why we use SPLograms in DSL, to try to make the amplification fitting process more logical and clear. During my final year in Connecticut, Mark and I started to really work on a new approach to pediatric fitting and to begin to think through what DSL might look like.
Academy: So that was done via paper and pencil at that time... and I guess that must have been in the mid-1980s?
Seewald: Yes, perhaps 1985 or 1986. In addition, we hoped to have a software version soon, but it was not until 1991 when we released the first DSL software and now (2009) we are up to version 5.0.
Academy: And 5.0 has some remarkable new attributes, such as compatibility with toneburst ABR results and different prescriptions for children with congenital hearing loss and adults with acquired impairments. However, I think there are two things that still make people a bit uncomfortable; the SPL scale itself and RECD. Therefore, we have already discussed the SPL issues, so let’s talk about “real ear to coupler difference.”
Seewald: Sure. Well, Doug, as you know, the 2cc coupler is the standard into which all hearing aid measures start. However, the 2cc coupler is a hard-walled cavity and although it grossly approximates the volume of the adult ear, without actually being identical to the majority of adult ears, the 2cc coupler has very little in common with pediatric ears. The pediatric ear and particularly the newborn's ear is much smaller. If we use hearing aid data based on a 2cc coupler to fit hearing aids to a newborn or a child, the chances are very good that we will grossly over-fit the child.
Academy: And so we need to measure the hearing aid’s output in a cavity that resembles the child’s ear canal, or preferably, in the ear canal itself.
Seewald: Exactly. Therefore, RECD is just a mathematical correction that allows us to know what we’re doing. It allows us to measure the actual SPL in the infant’s own ear canal.
Academy: Right, and then we can say if the child has the following thresholds, the SPL needs to so many dB in order for the child to hear the sounds, and then X, Y and Z decibels loud with respect to gain across various input levels, so as to appropriately set compression to maximize the child’s dynamic range, without exceeding LDL.
Seewald: That’s it. That is all there is to it. And as you know, we always prefer to have RECD measures on each child and each ear, but if that’s not possible, we have new norms in DSL 5.0 based on the child’s age in months, which are usually in the right ballpark. In other words, they are averages based on age, and they are rarely accurate for a particular child, but they are much closer than the 2cc coupler or average adult values that are often applied in the manufacturer’s fitting software. So if for some reason the RECD cannot be measured, we have estimated values based on the child’s age. However, using averaged RECD data may still be 12 to 15 dB off, and so very often, averaged data is just not an excellent solution. That is, if you are using averages, you are starting the fitting protocol with a substantial source of error.
Academy: What about ABR data? How is that applied with DSL?
Seewald: We base our ABR-based fittings on some of the work by Dave Stapell and colleagues. So we make frequency-specific corrections from nHL to estimated hearing levels (eHL), which gets us to the audiogram, and then we apply the RECD to go from HL to SPL in the ear canal.
Academy: And I’m pretty sure you need a few different data points, such as tone bursts at 500, 1,000, 2,000 and 4,000 Hz?
Seewald: Yes, frequency-specific threshold estimates are required for accurate hearing instrument fitting in babies. If all we have is a click-based ABR, we just won’t proceed with a fitting. It’s just not enough information on which to base a hearing aid fitting.
Academy: Yes, that makes sense. What about ASSR?
Seewald: Well the ASSR values are already in the eHL scale, so then you apply RECD and the DSL software system will then predict the SPL at threshold across the measured frequencies.
Academy: What do you recommend for audiologists starting to use DSL?
Seewald: I always recommend that of you have not yet performed RECD measurements, it’s best to not start on a child! Let me say that again, it’s best to learn the technique with other willing and able adults, so you learn to place the probe mic and get familiar with the hardware and software. It’s really fast when you know how to do it, but learning on a child makes it unnecessarily difficult, so start with adults. We measure RECD in some 90 to 95 percent of the babies we see and it usually takes about 3 to 4 minutes from start to finish. It used to take a few minutes for the stimulus, and now it just takes a few seconds once the equipment is in place. So it keeps getting easier.
Academy: Richard, thanks so much. It’s been a pleasure speaking with you. Thanks so much for your time and for sharing your knowledge.
Seewald: My pleasure, Doug. Thanks for your interest.
Richard C. Seewald, PhD, is a distinguished university professor and the Canada research chair in childhood hearing, at the National Center for Audiology, The University of Western Ontario, London, Ontario, Canada.
Douglas L. Beck, AuD, Board Certified in Audiology, is the Web content editor for the American Academy of Audiology.
For More Information, References and Recommendations:
Bagatto, MP (2001) Optimizing your RECD measurements. Hearing Journal, 54: pp. 34-36.
Bagatto, MP, Scollie, SD, Seewald, RC, Moodie, KS & Hoover, B (2002) Real-ear-to-coupler difference (RECD) predictions as a function of age for two coupling procedures. Journal of the American Academy of Audiology, 13(8): 407-415.
Seewald, RC & Scollie, SD (2003) An approach for ensuring accuracy in pediatric hearing instrument fitting. Trends in Amplification, 7(1): 29-40.
Scollie, S, Seewald, R, Cornelisse, L, Moodie S, Bagatto, M, Laurnagaray, D, Beaulac, S & Pumford, J (2005) The Desired Sensation Level Multistage Input/Output Algorithm. Trends in Amplification, 9(4): pp. 159-197.
Bagatto, M, Moodie, S, Scollie, S, Seewald, R, Moodie, K, Pumford, J & Liu, KP. (2005) Clinical protocols for hearing instrument fitting in the Desired Sensation Level method. Trends in Amplification, 9(4): pp. 199-226.
Seewald, R, Moodie S, Scollie, S & Bagatto, M (2005) The DSL Method for pediatric hearing instrument fitting: Historical perspective and current issues. Trends in Amplification, 9(4): pp. 145-157.
Bagatto, MP, Seewald, RC, Scollie, SD & Tharpe, AM (2006) Evaluation of a technique for measuring the real-ear-to-coupler difference (RECD) in young infants. Journal of the American Academy of Audiology, 17(8): 573-581.