Pediatric audiologists rely on evidence-based procedures when fitting hearing aids to their young patients. Although the science is concrete and clinically feasible, there are some practical topics of relevance that may not have been covered in a formal publication. This article aims to address some frequently asked questions related to procedures used for pediatric hearing instrument fitting.
Getting Started
How are auditory brainstem response (ABR) threshold estimates used in hearing aid fitting?
It is possible to prescribe and fit a hearing aid to a young infant based on data obtained from an ABR assessment. However, it should be noted that the thresholds estimated from the ABR are typically higher when compared to behavioral thresholds. In fact, they can be as much as 20 to 30 dB higher, depending on the frequency (Stapells, Gravel, & Martin, 1995; Stapells, 2000). This is a significant difference in terms of intervention. Therefore, a correction is applied to the ABR threshold estimates to better predict the behavioral threshold. This corrected ABR is sometimes referred to as Estimated Hearing Level (eHL) to distinguish it from a typical ABR referenced in Normalized Hearing Level (nHL). Some ABR systems have the correction embedded in as part of its calibration. For those that do not, a correction must be applied prior to calculating the hearing aid prescription.
How do I know if the ABR system has the correction embedded in it?
Unless the corrected ABR data from the system has a reference other than nHL (i.e. eHL or HL), there is no clear way to determine if the correction has been embedded into the system's calibration. If the clinician fitting the hearing aid has not conducted the ABR, it is important to communicate with the clinician who conducted the ABR to determine what type of data the ABR system is producing (i.e. HL, eHL, or nHL). If the clinician who performed the ABR is also doing the fitting, a conversation with the person who calibrated the ABR system is needed. With either scenario, it is important to know whether the correction has been applied or not. Given the large value of the correction, it is crucial that it be applied, and only once. Under- or over-amplification can result if the correction is applied twice or not at all. Therefore, it is ultimately the clinician's responsibility to understand what type of data the ABR system is producing in order to determine if a correction is required.
How are the corrections applied to the ABR values?
The frequency-specific corrections are subtracted from the ABR values. Since previous research has shown that ABR threshold estimates (nHL) are typically higher than behavioral thresholds (HL), the correction value is subtracted to provide a better estimate of behavioral thresholds (eHL) for the purposes of hearing instrument fitting. The schematic in Figure 1 describes this calculation, which is also represented on the accompanying audiogram.
Figure 1. Schematic of correction factors applied to ABR (nHL) threshold data to estimate behavioral thresholds for the purpose of hearing aid fitting. On the audiogram, the green plus signs indicate the ABR threshold value in nHL, and the red circles indicate the corrected value in eHL. It is important to note that the subtracted corrections improve the initial thresholds estimated by the ABR.
The correction values are frequency-specific and are impacted by how the ABR system is calibrated, as well as the stimulus and collection parameters used. For some prescriptive algorithms (i.e. DSL m[i/o] v5), the clinician can 1) enter data in nHL and a default correction will be applied, 2) enter data in nHL and apply clinic-specific corrections, or 3) enter the data in eHL. The third method assumes that a correction has already been applied and the algorithm will not apply one. For more information on how these data are used in DSL m[i/o] v5, refer to Bagatto, et al., 2005.
The Real-Ear-to-Coupler Difference (RECD)
It can be a challenge to measure the RECD on a young child. What are some strategies for measuring an RECD with this population?
- Pre-measure the probe tube (15 - 25mm from intertragal notch) and measure the coupler portion of the RECD before approaching the child for probe tube insertion. This will save time and cooperation needed from the child.
- If the child has a personal earmold, use it to obtain a more customized insertion depth by running the probe tube along the bottom part of the earmold. Be sure the medial end of the probe tube extends approximately 3 to 5 mm from the sound bore of the earmold. Mark the probe tube where the most lateral portion of the earmold meets the tube (see Figure 2). Insert the tube until the mark meets the intertragal notch.
Figure 2. Measuring the RECD probe tube using a customized earmold - If you have some clear wrap or soft surgical tape, you can connect the probe tube to the earmold and insert the unit simultaneously (see Figure 3). To the child, this may seem like you are merely inserting his earmold.
Figure 3. The probe tube connected to the earmold using clear wrap. - Toddlers may be curious about what you are doing at their ear. It is helpful to have an assistant hold a mirror in front of the child while you insert the probe tube to reduce the child's head movements. He/she will be able to see what you are doing by looking into the mirror.
- If there is a cord clip on the probe microphone module, clip it to the shoulder opposite the ear you are working on. This will allow the probe module to lie snugly against the child's cheek. Movement of the module will be reduced and the child will be less tempted to reach up and grab the cord (see Figure 4).
Figure 4. The probe microphone cord is clipped across the child's shoulder to minimize cord movement and child interest. - The accuracy of measured RECD values is important because it will impact the accuracy of audiometric data collected with insert earphones, prescriptive targets and 2cc coupler verification procedures (see additional RECD FAQs for more details). Always troubleshoot your measurement to ensure that the probe tube has been inserted appropriately. A negative RECD value means that the individual ear is larger than a 2cc volume. A negative RECD will more frequently be seen in the low frequencies because of slit leaks, myringotomy tubes, or eardrum perforations. Shallow probe tube insertion will influence the high frequencies by "rolling them off" more quickly. Bagatto (2001) offers troubleshooting suggestions for RECD measurements.
Strategies for measuring RECDs on toddlers may not work for young infants. Infants are usually being cradled in the caregiver's arms or in a stroller or car seat. Access to the ear is a challenge due to this positioning and the close proximity of the ear to the shoulder. A variation of strategy 2 of the previous question has been evaluated and shown to be feasible in young infants (Bagatto et al., 2002). The strategy involves connecting the probe tube to a small eartip or personal earmold using plastic wrap or soft surgical tape (see Figure 3).
The probe tube should extend approximately 2 to 4 mm beyond the sound bore. Unfortunately, the plastic wrap used for protecting hearing aids from moisture (Moisture Guard) is no longer available. Therefore, soft surgical or first aid tape cut in very thin strips is recommended. Soft surgical tape should be used as opposed to regular tape as this may cause a sharp edge when applied.
Figure 5. Insertion depth guideline when probe tube is inserted independently.
If inserting the probe tube separately from the eartip is preferable, the insertion depth guideline determined in the Bagatto et al. (2002) study is 10 mm from the opening of the ear canal.
What is the difference between a foam tip and an earmold RECD?
The main difference will be noted in the high-frequency region. An RECD measured with a foam tip will show a reduction in values starting around 3 or 4 kHz. When an earmold from a toddler or adult is used in place of a standard foam tip, the increased length of tubing on the earmold will cause a reduction in RECD values around 2 kHz (see Figure 6). This change in values may not be present when measuring with a young infant's earmold as the tubing may be the same length as a standard foam eartip.
Figure 6. Comparison of RECD values when using a foam tip versus a custom earmold. Note the high-frequency difference in values.
It is important to be able to distinguish that the change in high-frequency values is due to increased tubing length and not shallow probe tube insertion. However, if the clinician used the customized probe tube insertion depth strategy outlined previously (Figure 2), then a roll-off in high-frequency values will likely be due to increased tubing length from the earmold.
In many clinical situations, audiometry is conducted using foam eartips coupled to insert earphones, and an RECD is required for the hearing aid fitting. The RECD is then measured using the child's personal earmold. What are the implications of this strategy?
RECDs are used in several stages of the hearing aid fitting process. One is to convert audiometry measured in dBHL or dBeHL to dBSPL at the eardrum. This provides a more accurate description of hearing thresholds, and the values are used to calculate the prescription in some formulae (i.e. DSL m[i/o] v5). Another use is in the verification stage. For the pediatric population, many times it is not feasible to conduct actual real-ear measures of the hearing instrument. With the RECD, clinicians can obtain a prediction of real-ear hearing instrument performance based on 2cc coupler measurements. Therefore, it would be ideal to apply RECD measures obtained with different coupling (i.e. foam eartip and earmold) for accuracy at all appropriate stages of the hearing aid fitting process.
If you measure the child's hearing with a foam eartip and the RECD with his/her personal earmold there will be some error in the dBSPL (ear canal) threshold measurements (by the amount that the real-ear measurement obtained with the foam eartip differs from the earmold). Accurate predictions of real-ear performance will be obtained if the earmold is coupled to the HA-1 (ITE) coupler for coupler-based measurements. Additional research needs to be conducted in this area. Manufacturers' implementation of RECD measurement procedures could be modified to allow end-users to choose from two couplers (HA-1 or HA-2) for the coupler part of the RECD procedure and from two real-ear coupling options (foam tip or earmold) to better accommodate the possible choices.
In the meantime, if the RECD was measured with the earmold and audiometry was conducted with a foam eartip, there will be some discrepancy in the high-frequency region in the converted SPL thresholds. Preliminary data analyses show that RECD values obtained using an eartip are approximately 5 dB larger in the high frequencies compared to RECDs measured with the child's earmold. This difference will transfer to the calculation of targets since they are calculated from the SPL thresholds. The discrepancy should only be present at the initial stages of the child's hearing aid use because subsequent audiometry can be conducted with the child's personal earmolds. Until the application of RECDs with different coupling is included in fitting approaches, an RECD with the child's personal earmold should be applied throughout the fitting process, even if the audiometry was conducted with foam eartips.
How often should the RECD be measured?
It is recommended that the RECD be measured whenever a new earmold is obtained. Infants and young children require earmolds frequently due to the rapid growth of the outer ear. Since the size of the ear canal has changed, an RECD measurement is needed to capture the associated changes in ear canal acoustics so that they can be applied to the hearing instrument fitting.
Please note that with new earmolds, it is important to trim the tubing to the length that it will be when worn with the hearing aid prior to conducting audiometry and measuring the RECD. In addition, the RECD should be measured any time there is a change in middle ear status in one or both ears.
Is it necessary to measure the RECD on both ears?
If the physical appearance of the outer ear and the middle ear status is the same for both ears, the RECD measurement from one ear can be used for the other ear. There is some data to suggest that if these conditions are met, the RECD values should be similar for both ears (Tharpe, Sladen, Huta, & Rothpletz, 2001).
What modifications to the procedures, if any, need to be made when the earmold has a vent?
For many pediatric hearing aid fittings, the earmold will be too small to accommodate a vent. However, a vent larger than about 1 mm will affect the low frequency region (i.e. negative values). There are two stages in the fitting process where venting can have an affect. In the assessment stage, when conducting audiometry or measuring an RECD with a vented earmold, the vent should be plugged on the medial end. A fully occluded ear canal will ensure that the behavioral thresholds and RECD values are accurately represented at 250 and 500 Hz. This will ensure accurate conversion from HL to SPL for the purposes of calculating the hearing aid prescription.
In the verification stage, the effects of the vent cannot be fully accounted for unless the performance of the hearing aid is measured in the real-ear. With many pediatric patients, coupler-based or predicted real-ear measurements are the preferred method for verification. With this strategy, the impact of the vent will not be completely accounted for. Attaching the earmold to the BTE and coupling it to the HA-1 coupler for predicted verification is a way to account for some of the impact of the vent. However, it may be a challenge to securely attach a soft earmold to putty. Therefore, when fitting a hearing aid with a vented earmold, some error will be introduced in the low-frequency region until more appropriate procedures are developed.
How is a myringotomy tube or eardrum perforation accounted for when applying RECD?
Myringotomy tubes or eardrum perforations will cause negative values in the low-frequency region (i.e. -5 to -10 dB). If this is the case, the negative RECD values should be used for the conversion of audiometry to SPL and for verification. This is because the status of the eardrum impacts the conduction of sound for both audiometry and hearing aid use. In addition, the hearing thresholds in the low frequencies are likely to be worse due to the abnormal status of the eardrum. The combination of higher hearing thresholds and negative RECD values will result in more prescribed gain for average speech, which is what is appropriate for the patient at this time. When the status of the ear changes, the RECD should be re-measured and any necessary modifications to the electroacoustic characteristics of the hearing aid should be made.
What if measuring the RECD is not possible?
In applications of DSL v5, there are age-appropriate predicted RECD values. These have been updated since DSL v4.1. The updated values are for foam tip and earmold coupling and are provided in one-month age increments. However, it is important to note that while these values are more accurate than using average adult RECD predictions, large variability exists in the data making the predictions less than ideal. For example, at some frequencies, the differences range from ±12 dB. Therefore, it is recommended that individual measurements of RECD be obtained whenever possible. In cases where it is unobtainable, more precise RECD predictions are now available. (See Bagatto et al., 2002 and Bagatto et al., 2005 for more details.)
The Hearing Aid Fitting
Other than applying evidence-based prescriptive targets, what else should be considered when selecting hearing aids for an infant or young child?
Seewald and Ross (1988) state that there are at least three important components to the hearing aid selection and fitting process for infants and young children: (1) be attentive to the quality and quantity of auditory input received by the child so that the speech and language development can by maximized in a naturally occurring way; (2) the basic premise of the initial hearing aid selection is that all initial electroacoustic characteristics are tentative and (3) hearing aid selection must be viewed as an ongoing process that is part of a complete habilitative program.
Keeping these points in mind, there are several non-electroacoustic items that should be considered when selecting a hearing aid for an infant or young child. One is to ensure the behind-the-ear (BTE) hearing instrument has a pediatric-sized filtered earhook. This will allow for better retention on the child's ear and the filter will reduce peaks in the hearing instrument response that could compromise the fit to targets. Direct audio input is also an important feature as use of an FM system is likely to occur in the future. Also, locking mechanisms for the volume control and battery door are crucial. For most situations, the volume control should remain in one position in the hearing aid. Locking it will ensure that accidental increases or decreases in the amount of volume do not occur while the child is wearing the aids. This could result in over- or under- amplification for a period of time. Locking the battery door compartment keeps the battery from inadvertently coming out and being swallowed by the child. Hearing aid batteries are toxic; therefore, this is a critical safety precaution. Finally, deactivation of advanced features (i.e. directional microphones, noise reduction, and multiple memories) is useful, especially at the early stages where advanced technologies may not be applied.
When should advanced signal processing be used / not used for infants or children?
At this point, there is very little research regarding the application of advanced technologies with infants or children. One main thing to consider is whether or not the caregiver and/or the child can reliably apply the technology in the appropriate situations. Since incidental learning is a large part of how the child learns to attach meaning to sound, reducing noise or sound from certain locations may not be desirable. Advanced technologies should be considered on a case by case basis with careful consideration of the child's listening needs and the caregiver's understanding of the technology being applied.
What is the most accurate and practical way to verify hearing instrument performance for infants and young children?
Performing predictions, simulations, or coupler-based real-ear measurements have been shown to be an accurate and practical way to verify hearing instruments for the pediatric population (Seewald et al., 1999). Infants are not able to participate in real-ear measurements of hearing aid performance. Young children are not likely to sit still and quiet while facing a speaker for the duration of time it takes to make adjustments to the hearing aid at various levels and frequencies. By applying the RECD, the performance of the hearing aid can be predicted in the real-ear by making measurements in a controlled test-box environment. And when compared to actual real-ear measurements, it is accurate to within ±2 dB on average (Seewald et al., 1999). When the child is old enough to remain quietly seated in front of a loud speaker for real-ear verification, this strategy can be applied.
Acknowledgments
This work has been supported by the Canada Research Chairs (CRC) Program, The Ontario Rehabilitation Technology Consortium (ORTC), The Canadian Language and Literacy Research Network (CLLRNet), and the Masonic Help-2-Hear Foundation of Ontario. The contributions made by Richard Seewald and Susan Scollie are gratefully acknowledged.
References
Bagatto, M.P. (2001). Optimizing your RECD measurements. The Hearing Journal, 52(32), 34-36.
Bagatto, M., Moodie, S., Scollie, S., Seewald, R., Moodie, K., Pumford, J., & Liu, R. (2005). Clinical protocols for hearing instrument fitting in the Desired Sensation Level Method. Trends in Amplification, 9(4), 199-226.
Bagatto, M.P., Scollie, S.D., Seewald, R.C., Moodie, K.S., & Hoover, B.M. (2002). Real-ear-to-coupler difference predictions as a function of age for two coupling procedures. Journal of the American Academy of Audiology, 13, 407-415.
Seewald, R.C., Moodie, K.S., Sinclair, S.T., and Scollie, S.D. (1999). Predictive validity of a procedure for pediatric hearing instrument fitting. American Journal of Audiology, 8(2), 143-152.
Seewald, R.C. & Ross, M. (1988). Amplification for young hearing-impaired children. In M. Pollack (Ed.), Amplification for the Hearing-Impaired, 3rd edition (pp. 213-271). Orlando: Grune & Stratton.
Stapells, D.R. (2000). Threshold estimation by the tone-evoked auditory brainstem response: a literature meta-analysis. Journal of Speech-Language Pathology and Audiology, 24(2), 74-83.
Stapells, D.R., Gravel, J.S., & Martin, B.A. (1995). Thresholds for auditory brain stem responses to tones in notched noise from infants and young children with normal hearing or sensorineural hearing loss. Ear and Hearing, 16(4), 361-371.
Tharpe, A.M., Sladen, D., Huta, H.M., & Rothpletz, A.M. (2001). Practical considerations of real-ear-to-coupler difference measures in infants. American Journal of Audiology, 10(1), 41-49.
