Fitting hearing instruments deeply into the external ear canal reportedly produces desirable acoustic and cosmetic effects.1-11 These advantages include: an overall increase in sound pressure level (especially in the high frequencies), a reduction in acoustic feedback, improved cosmetics, a reduction in the occlusion effect, improved use of the phone, improved performance when listening in noise, wind noise reduction and a more secure fit. Although these advantages had been attributed to deep canal fittings, they have been suggested for completely-in-the-canal (CIC) instruments as well.12-20

If these advantages do indeed exist, is there a need for another approach to deep canal fittings? The following describes the problems and solutions posed by deep canal fittings relative to the new Songbird Disposable Hearing Aid (SDHA).

Rationale
In spite of the established market position of CIC instruments, the majority of problems the devices were promoted to circumvent still exist. In an unscientific reporting of a showing of hands at 50 dispenser (traditional and audiology) presentations in 1995 and 199621, attendees reported that most of the problems the CIC was promoted to circumvent, continued (Fig. 1). About 75% of dispensing professionals indicated that feedback was still a major fitting concern, as was occlusion (65%) and the device working itself out of the ear (30%). These numbers were confirmed recently22, suggesting the same problems continue.

The consequences of not completely resolving these problems have resulted in substantial remakes (15-20% by industry estimates) by manufacturers, patient dissatisfaction with the fitting experience and added costs for dispensing professionals and consumers.

Primary industry approaches to problem solving have included tapering and shortening the instrument tip to facilitate comfort and insertion, rolling off the high-frequency amplification to reduce acoustic feedback problems, and the use of different impression procedures to accommodate the fit, comfort, feedback and remake issues. All of these approaches appear contradictory to taking full advantage of deep canal technology. However, because the patient is “essentially” satisfied—especially with the cosmetic appearance and satisfactory telephone usage—it is frequently better to leave well enough alone. In other words, “going deep” is often too difficult or time consuming, and many CIC hearing instrument fittings are not “deep fits.”

A solution was sought that would provide a simple approach and circumvent many of the reasons deep canal technology was not being used. The goals were to develop a hearing aid that would: provide full contact in the bony canal without discomfort to reduce the occlusion effect, provide added gain in the high frequencies, be easy to insert and remove, significantly reduce acoustic feedback problems, fit deeply in the ear canal without the hassle of unusual ear impression techniques, be used on the telephone without acoustic feedback problems, be cosmetically acceptable, provide for security of fit and reduce the influence of cerumen on long-term hearing aid use. The hearing instrument was to be designed specifically for individuals with mild-to-moderate hearing losses.

To accomplish these goals, the designed instrument had to be of high quality (equal to or better than quality products on the market) and be simple, easy and effective to use. The attempt at finding a solution resulted in the concept of a disposable hearing aid of one size that would fit either ear for a high percentage of adults. Such a device was a novel approach to attaining deep canal advantages as a result of the following: by not requiring an ear impression, issues relating to procedures for making a deep impression would be bypassed; long-term problems of cerumen buildup associated with a deep fit could be minimized with a clean replacement hearing aid every 40 days; remakes of hearing aids for poor fit and feedback problems would be eliminated; and a standard tip could fit comfortably in the ear canal but still provide security of fit and full contact with the bony canal wall (to manage the occlusion effect) and an appropriate seal (for acoustic feedback control during mandibular movement).

The tip of the SDHA was designed with a 4 mm2 sound bore opening to maximize high-frequency signals and to minimize cerumen blockage and also to facilitate cerumen removal, if necessary. The tip, made of biocompatible soft materials, also fits deeply to maximize high frequency output SPL, and is designed to fit approximately 85-90% of ears with one size.

Field Study
Over the past four years, numerous studies have been conducted to determine the requirements for a disposable hearing aid that met the deep canal design goals for a “one-size-fits-most” device. The results that follow are from one study of a prototype of the final design of the SDHA (the exception being the use of a size 312 battery rather than the final custom battery) that evaluated fit and comfort issues following a two-week wearing schedule by adult patients.

Subjects: Advertisements to evaluate a new, disposable hearing aid were placed in local newspapers by audiologists conducting the field study in four cities: Boston, Philadelphia, Phoenix and Salt Lake City. This request for volunteers produced 84 interested patients (57 males and 27 females). Following telephone screening, 75 patients were invited to the four clinics for an otoscopic inspection of their ears and for pure-tone air-conduction and bone-conduction threshold measurements and/or tympanometry and other patient information. These patients produced the mean audiogram in Fig. 2. Patients meeting the specified audiometric configuration, hearing level requirements and ear size were accepted into the field study and fit with the SDHAs. The main reason for people not being fitted was that they did not fall within the predicted audiometric fitting range for the SDHA (Fig. 3). Other reasons that interested patients were not fitted included cosmetics, poor physical fit, acoustic feedback oscillation, excessive protrusion from the ear and because the appropriate electroacoustic prescription formats were not available at the time for the patient’s hearing loss. Ultimately, forty-nine patients (39 males and 10 females) were fitted in a monaural and binaural combination with 70 SDHAs. The results that follow consist of data from only 36 of the 49 (74%) because some subjects did not complete fully the questionnaires.

Fitting and evaluation methods: The Songbird Prescription Selector™ was used to select, validate and fine tune the fittings using a loudness growth protocol. Adjustments of the electroacoustic selection were made as needed with the Prescription Selector until soft sounds were soft and loud sounds were loud but not uncomfortable.23

The evaluation protocol called for obtaining unaided and aided Abbreviated Profile of Hearing Aid Benefit (APHAB)24, an in-clinic Fit and Comfort evaluation (available upon request from authors), Hearing in Noise Test (HINT)25, and insertion gain and/or functional gain. Some of the questions on the in-clinic Fit and Comfort questionnaire pertained specifically to how well the deep fitting uni-ear shell and tip design functioned. The APHAB first calculates the percentage of problems encountered by the subject in three listening environments: ease of communicating in quiet (EC), in reverberant conditions (RV) and in high background noise (BN). A fourth subscale rates the degree to which sounds are aversive or unpleasant.

Patients were sent home with their SDHAs and logged the hours worn and submitted other observations. After two weeks, they repeated the APHAB and filled out a final Fit and Comfort evaluation questionnaire. This form contained all of the questions that were in the initial Fit and Comfort evaluation, plus a few added questions. Thereafter, Market Facts, Inc., an independent survey agency, contacted the patients and asked questions about the entire experience.

Results: Mean signal-to-noise ratio (SNR) on the HINT for the subjects was -6.9 dB, indicating that, on average, the subjects could recognize the sentences correctly 50% of the time with the speech signal 6.9 dB SPL lower than the noise signal. HINT results are reported in greater detail elsewhere.26

The percent-problems scores on the four APHAB subscales may also be used to compare two test conditions, such as unaided versus aided, or two different hearing instruments. The result is plotted as the difference between the problems (in %) for the two test conditions on the four APHAB subscales. In the case of unaided versus aided percent problems, percent benefit is the result.

Although most of the participants in the study had not previously worn hearing instruments, eight of the subjects who had completed an initial and final APHAB evaluation were previous users. Fig. 4 shows a comparison of the mean APHAB scores with their own hearing instruments versus those with the SDHA. On average, these subjects had greater problems with their own hearing instruments than with the disposable aid. For example, on the Aversiveness subscale, the mean score with their own hearing instruments is over 25% greater than with the SDHA, indicating that their own hearing instruments made sounds more unpleasant than did the SDHA.

Since the APHAB is a standardized test, it is possible to compare scores obtained with it to different hearing instruments and different studies. Fig. 5a shows the APHAB benefit scores obtained with the SDHA against those reported by Valente et al.27 for a digital instrument. The mean benefit for the DSP instrument is about the same for listening in quiet (EC) and slightly greater in reverberant conditions (RV) and in high background noise (BN) than the SDHA. Since these differences are not greater than 10% on the first three subscales, they are not deemed significant according to Cox and Alexander.24 For the Aversiveness subscale (AV), the disposable aid, on average, made sounds less unpleasant than the DSP aid.

It is reasonable to assume that persons with greater hearing loss will have greater problems unaided on the APHAB and, thus, potentially receive greater benefit from wearing a hearing instrument.28 Therefore, comparisons between studies such as that in Fig. 5a might be made more viable by normalizing the hearing loss between the subjects in the studies. This procedure was acceptable to one of the originators of the APHAB.29 Fig. 5b reports the same data as Fig. 5a but with the percent benefit divided by the mean pure-tone average (PTA: 500 Hz, 1 kHz, 2 kHz) hearing loss for the two studies. Because the subjects in the Valente et al. study had greater hearing loss than those in the SDHA field study, a normalized APHAB benefit score for a SDHA is shown in Fig. 5b with the benefit now greater than that for the DSP instrument on the EC, RV and BN subscales.

Another way of equating hearing loss between studies is to divide by the percentage of problems instead of by the PTA hearing loss. The point to the data in Figs. 5a-b is not in showing that one specific instrument is superior/inferior to another; rather, the point is that the deep-fit SDHA results in APHAB scores that are (at least) comparable to one of the best high-performance instruments available.

Fig. 6 shows the distribution of initial and final Fit and Comfort evaluation ratings made by first-time users of hearing instruments (22) and also of individuals who were previous hearing instrument wearers (14), along with the mean ratings. For those patients who had worn hearing instruments previously, a portion of the Fit and Comfort questionnaire compared the SDHA to their own hearing instrument(s). If the disposable aid was rated the same as their own hearing aid for a particular question, a rating of 2 points was assigned. If it was rated better than their own hearing aid, 3 points was given. If their own hearing instrument was rated better than the SDHA, 1 point was assigned.

The ranking of their own hearing instrument(s) against the SDHA is shown in Fig. 7, which shows the means and standard deviations for the Fit and Comfort evaluations. The SDHA compared favorably with the 14 patients who owned hearing instrument(s).

Discussion
Data from Figs. 6 and 7 show that the disposable, one size uni-ear hearing device is very acceptable to patients following two-weeks use. Ease of insertion and removal, the number of attempts to manage insertion and overall handling ease are judged to be no different than for conventional CIC instruments (based on judgements of the four audiologists participating in this investigation since no published data exist with which to make a comparison).

The physical comfort (3.03) and soreness (3.44) out of a 5-point rating scale while wearing the SDHA are very acceptable for a device that is of one size and fits both ears. The fact that these ratings are as good as or better than ratings of their own instruments (Fig. 7) provides support for this observation. The slight increase in soreness and physical comfort following the two week trial was expected when compared with initial reactions, and seems to be no different from that experienced with CIC instruments. Fifty-two percent of the subjects wore the aids from 8-13.5 hours each day; and 30% wore them from 4-8 hours per day. These are relatively long use times when considering that the majority of patients had never worn hearing instruments previously.

Fullness or “stuffiness” ratings, a potential description of the occlusion effect, were acceptable (3.03 out of 5). Slightly more than 50% of patients reported that this experience was normal or that it was of little significance to them. In retrospect, the categories of “Normal,” “Very Little,” and “A Little” were not good descriptors for stuffiness in the questionnaire in that the set of responses were confusing to the patients. The lower-than-expected rating for this category could be explained by the fact that the instrument did not always penetrate the bony canal to the depth required to eliminate the occlusion sensation. In some ears, the plastic shell of the hearing aid prevented deeper penetration, leaving a residual volume large enough to contribute to this experience.

The fit of the SDHA in the ear, both medial and distal, is expected also to have some effect on the overall increase in SPL expected, especially in the high frequencies. A deeply recessed faceplate provides a good cosmetic fit and enhances the microphone pickup contribution because of its position in the concha/aperture of the ear. On the other end of the disposable instrument, a deep fit with a small residual volume is expected to provide a greater overall SPL. An evaluation of insertion gain compared to 2cc coupler measurements for the electroacoustic formats used did show the overall increase in SPL expected at some frequencies (Fig. 8). The lower frequencies (250 and 500 Hz) showed slightly poorer insertion gains than 2cc coupler gains. These reflect the inability of the low gain instruments used to overcome the “earplug” effect. The mid frequencies (1000 and 2000 Hz) showed slightly higher average insertion gains than 2cc coupler gains. In the higher frequencies, 3000 Hz shows a poorer-than-expected performance and requires additional evaluation. Of particular note, however, is the 6 dB or greater increase in insertion gain relative to 2cc coupler gain at 4000 Hz provided by the deep fit of the SDHA.

Acoustic feedback problems seldom prevented the disposable aids from being used. The low incidence of acoustic feedback problems indicated that telephone use was exceptional. This supports the fact that the ultra-soft tip is able to provide a seal in the bony canal. The acknowledgment that the aid is comfortable in the ear, does not slip from the ear and seals in the bony canal to reduce feedback is evidence that it has achieved a large part of its design goals for a deep-in-the-canal fitted hearing instrument—without taking an ear canal impression.

The disposable aid is designed to be cosmetically acceptable. An evaluation of “Excellent” is described if the faceplate of the device is at the aperture of the ear canal or deeper. “Acceptable” is defined as being similar to a mini-canal. “Not Acceptable” would have the device protruding from the ear canal by 5 mm or more. Results show that the disposable aid fit acceptably in 92% of the ears of this study and is in keeping with one of the desired goals of this device. The fit is generally between a CIC and a mini-canal instrument.

Reports of the clarity of sound are exceptional for the final fit rating (3.63 out of 5). Eighty-one percent of the patients reported the sound to be “Clear,” “Very Clear,” or “Extremely Clear.” Wind noise, although not evaluated specifically, did not show up as a complaint in any of the interviews. Cerumen management reduction due to the tip design and 40-day replacement cycle could not be investigated because the two-week listening experience was considered insufficient to make any comparisons.

Overall confidence that the SDHA is meeting its goals of achieving deep fit advantages are supported further by the APHAB results which are exceptional, especially in the “aversiveness” ratings.

Acknowledgments
Patient testing was completed in Boston by Deborah Dempesy, in Phoenix by Wayne Staab, in Salt Lake City by Geary McCandless and in Philadelphia by Karen Real. The authors thank Pat Santora for helping to summarize the data from the questionnaires and for analyzing and preparing the figures for the APHAB results.

This article was submitted by Wayne J. Staab, PhD and David Preves, PhD, vice president of R & D for Songbird Medical, Inc. Correspondence can be addressed to HR or David Preves, PhD, Songbird Medical Inc., 5 Cedar Brook Drive, Cranberry, NJ 08512; e-mail: [email protected]

References
1. Westermann S: The occlusion effect. Hear Instrum 1987; 38(6): 43.
2. Killion M, Wilber L & Gudmundsen G: Zwislocki was right: A possible solution to the "hollow voice" problem (the amplified occlusion effect) with deeply sealed earmolds, Hear Instrum 1998; 1, 14-18.
3. Orchik D, Cowgill S & Parmely J: Peritympanic soft hearing instrument fitting in high frequency hearing loss. Hear Instrum 1990; 41(11): 28, 30.
4. Staab W & Finlay B: Fitting rationale for deep fitting canal hearing instruments. Hear Instrum 1991; 42 (1): 6-10, 48.
5. Staab W: The peritympanic instrument: fitting rationale and test results. Hear Jour 1992; 45 (10): 21-26.
6. Staab W: Precipitous, high frequency hearing losses: a new solution. Hear Instrum 1993; 44 (4): 20-22.
7. Northern J, Kepler L & Gabbard S: Deep canal fittings and real ear measurements. Hear Instrum 1991; 42: 34-35, 53.
8. Agnew J: Acoustic advantages of deep canal hearing aid fittings, Hear Instrum 1994; 45 (8):22-25.
9. Garcia H & Staab W: Solving challenges in deep canal fittings, Part I. Hearing Review 1995; 2 (1): 34-49.
10. Garcia H & Staab W: Solving challenges in deep canal fittings, Part II. Hearing Review 1995; 2 (2): 30-34.
11. Giller R: Completely in the canal: a new style for a new clientele. Hear Instrum 1993; 44 (4): 24-25.
12. Bentler R: CICs: Some practical considerations. The Hear Jour 1994; 47 (11): 37-43.
13. Fortune T & Preves D: Effects of CIC, ITC, and ITE microphone placement on the amplification of wind noise. Hear Jour 1994; 47(9): 23-27.
14. Gudmundsen G: Fitting CIC hearing aids—some practical pointers. Hear Jour 1994; 47 (7): 10,45-48.
15. Meskan M: Fitting completely-in-the-canal instruments. Hearing Review 1994; 1 (7), 25-28.
16. Mueller G: 16 potential advantages of CICs. Hear Jour 1994; 47 (11):11.
17. Mueller G: CIC hearing aids: What is their impact on the occlusion effect? Hear Jour 1994; 47(11):29-35.
18. Preves D: Real-ear gain provided by CIC, ITC and ITE hearing instruments. Hearing Review 1994; 1 (7): 22-24.
19. Roesel G: CIC + WDRC = a logical combination. Hear Instrum 1994; 45 (8): 26-27).
20. Preves D & Leisses M: Some questions about CIC hearing aids. Audecibel 1995; Apr-June: 10-16.
21. Staab W: Problem solving of CIC hearing aid fittings. The 1997 Mid America Conference on Hearing, Louisville, KY, May 1997.
22. Staab W & Preves D: A new way to make deep canal hearing aid fittings work. American Academy of Audiology, Chicago, IL, March 16, 2000.
23. McCandless G, Sjursen W & Preves D: Satisfying patient needs with 9 fixed prescription formats. Hear Jour 2000; 53 (5): 42-50.
24. Cox R & Alexander G: The Abbreviated Profile of Hearing Aid Benefit. Ear Hear 1995; 16 (2): 176-183.
25. Nilsson M, Soli S & Sullivan J: Development of the Hearing in Noise Test for the measurement of speech reception thresholds in quiet and in noise. Jour Acoust Soc Amer 1994; 95: 1085-1099.
26. Preves D & Dempesy D: Speech recognition in noise field study results for a disposable hearing aid. Hear Review 2000; 7 (4): 34-38.
27. Valente M, Fabry D, Potts L & Sandlin R: Comparing the performance of the Widex SENSO digital hearing aid with analog hearing aids. Jour Amer Acad Audiol 1998; 9:342-160.
28. Fritz F: Personal communication, 1999.
29. Cox R: Personal communication, 1999.