Share this article on:

Open‐canal fittings and the hearing aid occlusion effect

MacKenzie, Douglas J.

doi: 10.1097/01.HJ.0000286218.69092.dd

Douglas J. MacKenzie, AuD, is an Associate Professor in the Department of Communicative Disorders & Sciences at the State University of New York at Geneseo. Correspondences may be addressed to Dr. MacKenzie at

It is only natural for consumers to be somewhat skeptical of marketing campaigns that promote a particular product or feature as “new and improved.” As a comedian once quipped, “What were they selling us before, old and lousy?” Given the less-than-stellar empirical evidence and personal accounts relative to some past “breakthroughs” in amplification technologies, it is understandable why patients and clinicians might be leery of new products hyped as “innovative,” “revolutionary,” or “the next big thing in hearing aids.”

Now entering the arena is one of the latest trends in the industry: open-canal (OC) fittings. These miniature behind-the-ear (BTE) instruments, some of which utilize a thin sound tube and soft, vented silicone eartip, and others that incorporate a thin wire connected to a receiver located in the canal, are being marketed extensively as a viable option for patients with mild to moderate high-frequency losses. This is a population that for various reasons has been either reluctant to pursue amplification or unsuccessful when fitted with more traditional technology.

The latter outcome is often due to, among other reasons, sound quality issues associated with the hearing aid occlusion effect. In fact, one of the primary benefits touted by manufacturers in their marketing of OC instruments is alleviation of occlusion-related problems.

Back to Top | Article Outline


Hearing aid users frequently complain of the unnatural sound quality of their voice and other internally generated sounds such as chewing and swallowing. One of the most common complaints, particularly among patients with normal or near-normal low-frequency hearing, is that their voice sounds “hollow,” “muffled,” or as if they were “talking with their head in a barrel.” Although such complaints sometimes result from sub-optimal hearing aid settings, they also may be associated with significant occlusion created by the hearing aid shell or earmold.1–3

During vocalization, bone-conducted energy results in vibration of the mandible and soft tissue located in close proximity to the external canal. This in turn causes vibration of the canal's cartilaginous walls, producing energy that is subsequently transferred to the volume of air within the canal. When the ear canal is occluded, much of this energy is trapped, causing an increase in the sound pressure level delivered to the tympanic membrane and, ultimately, to the cochlea.

For some closed vowels, occluding the external ear using a shallow insertion depth can result in levels of 100 dB SPL or greater within the canal.4 This energy is centered primarily in the low frequencies, with the peak of the occlusion effect typically occurring in the range of 200 to 500 Hz.5 The magnitude of the occlusion effect varies among individuals. Typical values are around 12 to 16 dB, but in some cases may be as great as 25 to 30 dB.1,6

Patient dissatisfaction resulting from the occlusion effect can lead to inconsistent hearing aid use or outright rejection.4 Results of a large-scale study by Dillon et al., revealed that 27.8% of patients experienced problems related to the quality of their voice.7 Consistent with these findings, the MarkeTrak VII report revealed that only 70% of surveyed hearing aid owners were “satisfied” with the sound quality of their own amplified voice, while 11% were “dissatisfied” and the remaining 19% “neutral.”8 These data show that although satisfaction with own-voice sound quality has increased 15% compared over the MarkeTrak VI results, the percentage of “dissatisfied” has remained relatively stable, suggesting that there is still a need for improvement in this area.9

Back to Top | Article Outline


Clinicians traditionally have employed two major lines of attack in dealing with patient complaints relative to the occlusion effect. One strategy has been to extend the earmold or shell into or near the bony portion of the ear canal. Although deep fittings have been shown to be effective in reducing the occlusion effect by minimizing cartilaginous vibrations, some patients resist them due to comfort issues.4,10 The second, and more commonly employed strategy, has been venting.11 Large-diameter vents, variable vents, and IROS configurations have been used for decades to address user complaints of unnatural voice quality by enabling low-frequency, patient-generated sounds to escape from the ear canal.

Clearly, then, the concept of OC fittings is not new. However, a major challenge to opening up the ear canal has been the threat of acoustic feedback that compromises usable high-frequency gain when relatively large vents are utilized—a prerequisite in most cases for reducing the occlusion effect to a level where the patient's own voice sounds more natural or at least tolerable. Given the various limitations inherent in deep fittings and venting, it's no wonder that the “Don't worry, you'll get used to it” approach has been such a popular fallback strategy in tackling this problem.

Fortunately, the advent of active adaptive feedback-cancellation circuitry has enabled hearing aid manufacturers to develop a new breed of OC instruments that offer increased venting capabilities, yet can produce appropriate high-frequency gain for many patients without major feedback concerns. The question still remains, however, whether or not this “new and improved” technology is truly effective in alleviating the hearing aid occlusion effect.

As Mueller reports, OC instruments have many potential patient benefits, but elimination of the occlusion effect is the most highly rated.12 Moreover, Johnson reports that this benefit also is the most highly rated by practitioners dispensing these instruments.13 To date, though, few studies have evaluated the occlusion effect with OC instruments.

Most past studies have involved a single hearing aid model. For example, a study of an OC instrument conducted by Kuk and his colleagues found no occlusion effect below 700 Hz.14 Kiessling et al. also found no significant difference between OC and unoccluded probe-microphone responses for a specific instrument, with comparable subjective ratings of own-voice naturalness.15

To further investigate the occlusion issue with OC fittings, the current study was conducted to measure objectively and subjectively the hearing aid occlusion effect for OC mini-BTE tube-fit systems offered by three different manufacturers: GN ReSound, Phonak, and Siemens.

Back to Top | Article Outline



The participants in this study were 20 normal-hearing adults (10 male, 10 female) ranging in age from 20 to 59 years (mean = 28 years). All participants exhibited normal otoscopic findings and immittance results, including an absence of significant negative middle ear pressure.16

Back to Top | Article Outline

Open-canal tube systems

The GN ReSound Air, Phonak Fit'nGo, and Siemens Life OC tube systems were selected for evaluation. Each system is fitted to a mini-BTE product, and consists of a thin sound tube, flexible retention strand, and soft silicone eartip (“dome”) containing four to five slotted vents. The tubing length and eartip diameter were selected according to the instructions provided by each manufacturer using the measurement gauge and accessories supplied in their respective fitting kits. A compatible mini-BTE hearing aid was connected to each of the three tube systems to promote stability and ensure proper positioning relative to the pinna and head, and to assure that measurements taken were representative of use conditions.

Back to Top | Article Outline


Participants were scheduled for individual 30-minute sessions. An otoscopic examination preceded the collection of data.

Participants were seated in a sound-treated audiometric test room and the occlusion effect assessment protocol outlined by Hawkins and Mueller was followed.17 An Audioscan RM500CP probe-microphone system was used to objectively measure the magnitude of the occlusion effect. Before each session, the measurement microphone of the system was calibrated as specified by the manufacturer.18 For each measurement, the tip of the probe tube was positioned in the left ear canal according to the manufacturer's recommendations.18 This generally resulted in the tip of the probe tube being located approximately 4–5 mm from the surface of the tympanic membrane.

The selection order of the three tubing systems was counterbalanced across participants. The BTE hearing aid coupled to the OC tubing was not activated during any of the measurements, since the occlusion effect is not a consequence of the aided signal. Participants were instructed to vocalize the closed vowel “ee” at a sustained level of 80 dB SPL. Vocal level was monitored with a Brüel & Kjaer Type 2235 precision sound-level meter mounted on a tripod and positioned 12 inches from the participant's mouth. The examiner monitored the sound-level meter to determine when the target intensity level was attained. Once participants were able to sustain production of the vowel within + 2 dB of the target level for a period of approximately 3 seconds, the sound pressure level in the ear canal was recorded.

To measure the occlusion effect objectively, the Audioscan system was configured to conduct a spectral analysis of the subject's voice. Specifically, with the system set to “basic REM” mode, the examiner initiated the Quickscan sweep with the stimulus level set to the off (0 dB) position. This configuration allowed for real-time spectral analysis within the participant's ear canal as he or she vocalized.

Measurement of the unoccluded ear canal response was recorded first and stored as an “unaided” test. Subsequent measurements for each of the three OC tube systems were stored as individual “aided” tests. Special care was taken to maintain a consistent probe-tube position for each condition. The decibel difference between the unoccluded and OC responses for each of the three manufacturers marked the magnitude of the occlusion effect as a function of frequency.

In order to evaluate subjective perceptions of own-voice sound quality, participants rated the naturalness of their voice for each of the three OC systems. Specifically, participants were instructed to read aloud the Rainbow Passage and rate the naturalness of their voice on a 10-point scale ranging from 1 (“extremely hollow”) to 10 (“extremely natural”). Since participants for this study were not experienced hearing aid users and therefore unlikely to be familiar with the “hollow” descriptor used on the rating scale, they were instructed to read the passage aloud, first with their ear canals unoccluded and then fully occluded using E-A-R® foam earplugs, to establish reference points for each end of the rating scale.

Back to Top | Article Outline


Response curves representing average ear canal sound pressure levels from 200 to 8000 Hz for the unoccluded condition and the three OC tube systems are shown for female and male participants in Figure 1. Inspection of the curves reveals minimal separation between the unoccluded ear canal response and responses for “occluded” conditions for the three OC systems for frequencies below 1000 Hz—the region where the hearing aid occlusion effect is typically observed. In the higher frequencies, a slight downward shift in the resonant peak was observed with the OC tips compared to the unoccluded response.

Although the purpose of this study was not to directly compare residual ear canal resonance—and these are not traditional REUR/REOR measurements—these findings for the higher frequencies are consistent with other studies that have compared REUR/REOR data for OC fittings.18–20

Figure 2 shows average occlusion effect values at 250, 500, and 750 Hz for female and male participants. Average values did not exceed 2.1 dB for any of the three frequencies.

It is interesting to note that in some instances ear canal sound pressure was actually slightly greater for the unoccluded condition. This may be due in part to minor changes in probe-tube location from one measure to the next or the fact that a non-calibrated signal (the subject's own voice) was used to assess the magnitude of the occlusion effect. Despite efforts to control for vocal intensity, slight variations in vocal pitch were occasionally observed that may have been a factor when calculating the degree of occlusion effect.

Average occlusion effect values were calculated for each subject by subtracting the three-frequency average (250, 500, and 750 Hz) for the unoccluded condition from the three-frequency average for each of the three OC responses. A 3x2 mixed model analysis of variance (ANOVA) was performed with gender as the between-subjects factor and manufacturer as the within-subjects factor. The results revealed no statistical significance for the main effects of manufacturer (F2,54 = .21, p = .81) or gender (F1,54 = 2.45, p = .12) and no significant interaction effect (F2,54 = .22, p = .80).

These findings suggest that OC configurations are effective in minimizing the magnitude of the hearing aid occlusion effect for both men and women. In fact, among the 20 adults who participated in this study, the greatest magnitude of occlusion effect observed at any frequency between 200 and 1000 Hz was 6 dB. That is significantly less than what has been observed in previous studies using traditional earmolds or shells.1,6

For example, Figure 3 shows the results of this study compared with an earlier study, using the same equipment and test protocol, in which the effects of venting on the occlusion effect were examined.6 Because there were no significant effects for the independent variables of manufacturer or gender in the present study, all data were collapsed to facilitate comparison.

Results of the subjective ratings of naturalness for the three OC systems are displayed in Figure 4. Most participants did not notice significant degradation in the naturalness of their own voice while talking with each of the three systems positioned in the ear canal. Average ratings across participants were 9.45 (SD = 0.83), 9.35 (SD = 0.81), and 9.45 (SD = 0.76) for the GN ReSound, Phonak, and Siemens OC systems respectively, suggesting that participants perceived their own voice as being quite natural. Male participants tended to rate the naturalness of their own voice slightly higher than female participants across manufacturers. The lowest rating given by any participant was “7.” This rating was elicited from a female subject for one of the three eartips.

A mixed model analysis of variance (ANOVA) revealed no significant differences in perceptual ratings for the main effects of manufacturer (F2,54 = .10, p = .90) and gender (F1,54 = .02, p = .89), and no interaction effect (F2,54 = .10, p = .90). In addition to preserving naturalness, many participants commented on how comfortable the OC eartips felt in the canal.

Although these findings are encouraging, and are consistent with recent patient survey data with OC fittings,21,22 it is important to recognize that own-voice sound quality issues unrelated to the occlusion effect may still arise with OC fittings, as they do with traditional fittings. As previously mentioned, sub-optimal hearing aid settings may negatively affect the quality of the user's own voice.

Another possible cause that has received considerable attention is the potential for propagation delay, sometimes referred to as “group delay.”23 If digital processing time within the instrument is slow (>10 ms), unnatural sound quality may result from phase effects caused by the interaction of direct energy transmitted through the vented eartip and the processed signal delivered by the hearing aid. While nearly all current hearing aids are reported to have processing delays less than the critical value, clinicians must be adept at systematically differentiating and treating a variety of sources that can lead to complaints of unnatural own-voice sound quality.

Back to Top | Article Outline


With traditional occluding earmolds and shells, there is usually a significant difference in low-frequency ear canal sound pressure level—measured using the patient's own voice as the signal and the hearing aid turned off—when the ear canal is open as compared to when the earmold or hearing aid shell is in place. Typically, this difference reflects an increase in sound pressure level resulting from the hearing aid occlusion effect. Results of the current study, however, demonstrate essentially no occlusion when OC tube systems are utilized.

As mentioned earlier, some OC fittings use a receiver-in-canal approach. While these systems are also characterized as “open” and reportedly effective in reducing user perceptions of “hollowness,”24 it is not possible to generalize the present findings to this hearing aid style.

In addition to the acoustic benefits provided by open fittings relative to the occlusion effect, results of this study indicate highly natural perceptual ratings of own-voice sound quality. These findings were consistent across the three manufacturers for both male and female participants, and they suggest that OC fittings are an effective means of overcoming one of the major barriers to the acceptance of amplification: poor own-voice sound quality resulting from the hearing aid occlusion effect.

Back to Top | Article Outline


Special thanks are extended to GN ReSound, Phonak, and Siemens for providing product and technical support for this study.

Back to Top | Article Outline


1. Mueller HG, Bright KE, Northern JL: Studies of the hearing aid occlusion effect. Sem Hear 1996;17(1):21–32.
2. Kuk F, Ludvigsen C: Ampclusion management 101: Understanding variables. Hear Rev 2002;9(8):22–32.
3. Painton SW: Objective measure of low-frequency amplification reduction in canal hearing aids with adaptive circuitry. JAAA 1993;4(3):152–156.
4. Killion MC, Wilber LA, Gudmundsen GI: Zwislocki was right…a potential solution for the “hollow voice” problem (the amplified occlusion effect) with deeply sealed earmolds. Hear Instr 1988;39(1):14–18.
5. Mueller HG: Page Ten: There's less talking in barrels, but the occlusion effect is still with us. Hear J 2003;56(8): 10–16.
6. MacKenzie, DJ, Mueller, HG, Ricketts, TA, Konkle, DF: The hearing aid occlusion effect: A comparison of two measurement devices. Hear J 2004;57(9):30–39.
7. Dillon H, Birtles G, Lovegrove R: Measuring the outcomes of a national rehabilitation program: Normative data for the Client Oriented Scale of Improvement (COSI) and the Hearing Aid User's Questionnaire (HAUQ). JAAA 1999;10(2):67–79.
8. Kochkin S: MarkeTrak VII: Customer satisfaction with hearing instruments in the digital age. Hear J 2005; 58(9):30–43.
9. Kochkin S: MarkeTrak VI: 10-year customer satisfaction trends in the US hearing instrument market. Hear Rev 2002;9(11):17–26.
10. Pirzanski C: Diminishing the occlusion effect: Clinician/manufacturer factors. Hear J 1998;51(4):66–78.
11. Revit LJ: Two techniques for dealing with the occlusion effect. Hear Instr 1992;43(12):16–18.
12. Mueller HG: Open is in. Hear J 2006;59(11):11–14.
13. Johnson EE: Segmenting dispensers: Factors in selecting open-canal fittings. Hear J 2006;59(11):58–64.
14. Kuk F, Keenan D, Sonne M, Ludvigsen C: Efficacy of an open-fitting hearing aid. Hear Rev 2005,12(3):26–32.
15. Kiessling J, Brenner B, Jespersen CT, et al.: Occlusion effect of earmolds with different venting systems. JAAA 2005;16(4):237–249.
16. American Speech-Language-Hearing Association: Guidelines for screening for hearing impairment and middle ear disorders. Asha 1990;32(Suppl. 2):17–24.
17. Hawkins DB, Mueller HG: Test protocols for probe-microphone measurements. In Mueller HG, Hawkins DB, Northern JL, eds. Probe Microphone Measurements. San Diego: Singular, 1992: 269–278.
18. Audioscan: RM500 Series User's Manual. Dorchester, Ontario.
19. Yanz JL, Olson L: Open ear fittings: An entry into hearing care for mild losses. Hear Rev 2006;13(2):48–52.
20 Mueller HG, Ricketts TA: Open-canal fittings: Ten take-home tips. Hear J 2006;(11):24–39
21. Gnewikow D, Moss M: Hearing aid outcomes with open- and closed-canal fittings. Hear J 2006;59(11):66–72.
22. Taylor B: Real-world satisfaction and benefit with open-canal fittings Hear J 2006;59(11):74–82.
23. Groth J, Sondergaard MB: Disturbances caused by varying propagation delay in non-occluding hearing aid fittings. IJA 2004;43(10):594–599.
24. Otto W: Evaluation of an open canal hearing aid by experienced users. Hear J 2005;58(8):26–32.
© 2006 Lippincott Williams & Wilkins, Inc.