Maximum power output remains an important parameter in hearing aid selection

Francis Kuk, PhD, is the director of audiology, and Petri Korhonen, MSc, (not pictured) is a research engineer, at the Widex Office of Research in Clinical Amplification (ORCA), Lisle, Ill; Lars Baekgaard, MSc, and Anders Jessen, BSc, are research engineers at Widex A/S, Vaerloese, Denmark.

Hearing aid acceptance in the United States has been accelerated by smaller hearing aid sizes. We have experienced this phenomenon with the completely-in-the-canal (CIC) devices; we are now experiencing the same cosmetic advantages with the new micro-size behind-the-ear (BTE) hearing aids coupled to a thin-tube or with a thin-wire, the latter commonly known as receiver-in-the-ear (RITE) or receiver-in-canal (RIC) hearing aids.

The small size of the BTE comes with a compromise. In a previous paper, Kuk et al¹ listed the lack of a telecoil, the absence of direct audio input (DAI) capability, and a smaller battery as the key factors that permit the miniaturization of BTEs. The placement of the hearing aid receiver (or loudspeaker) outside of the BTE case in a RIC model is not a key factor that allows BTEs to become smaller, but it is a factor nonetheless.

When a hearing aid receiver is placed outside the BTE case, its intended final placement and configuration impose a limitation on its physical size or dimensions. If the intended placement of the receiver is within the concha bowl, its dimensions may be larger than those that are intended for the wearer's ear canal. And, if the receiver is to be used in an open-ear fitting, its dimensions must be smaller than those that are intended for an occluded-ear fitting so that the ear canal may remain unoccluded when the receiver is in situ.

From a manufacturing perspective, if one were to choose a receiver that may fit the majority of ears, the dimensions of the receiver must be the smallest possible. A larger receiver will limit the number of wearers who may be able to use the RIC hearing aids. Therefore, it is no surprise that the majority of RIC hearing aids today use the smallest receiver available—so that a majority of wearers may be able to use them in both an open-ear mode and an occluded-ear mode.

TABLE 1. Size and MPO comparison of hearing aid receivers used in various styles of HAs.

TABLE 2. MPO and frequency response of various RIC hearing aids.

The Price of a Small Receiver

Larger receiver, higher MPO. While a small receiver allows it to be placed in more ears, a small receiver has the limitation of a lower maximum power output (MPO) or output sound pressure level (OSPL) at 90 dB SPL input (OSPL90) than a larger receiver.

The MPO/OSPL90 represents the highest output level that the hearing aid can deliver. A low MPO suggests that the hearing aid saturates with distortion and/or goes into compression limitation at a low output level; a high MPO indicates that high output can occur without distortion/saturation.

Table 1 summarizes the dimensions of typical receivers used in different hearing aid styles, along with their typical peak MPOs. It is evident that the MPO is somewhat related to the size of the receiver, with larger receivers associated with higher MPOs. It is also seen that the MPO in a CIC style hearing aid is the lowest; the MPO in a super-power BTE is the highest.

The receivers in most RIC are small. Table 2 shows the peak MPO and frequency response (on a 2cc coupler) reported by the major manufacturers of RIC hearing aids. One can see, because of the considerations discussed above, the MPO of the majority of RIC hearing aids are between a peak of 105 and 109 dB SPL. This means that any hearing aid output that exceeds this peak will be limited to that level.

Beware that the peak output level refers to only the peak frequency. The output level at other frequencies is lower than the peak level would indicate. For example, Figure 1 shows that, while the peak MPO is at 108 dB SPL (at 3000 Hz), the MPO at other frequencies (eg, 1000 Hz) is limited to only 101 dB SPL. These MPO values are comparable to the MPO seen in CIC style hearing aids.

FIGURE 1. MPO curve of a RIC hearing aid.

What's the Significance of a Low MPO?

Physically, a low MPO means that the maximum SPL that is produced by the hearing aid is limited. In most cases, an MPO of 100 dB SPL is low. Electroacoustically, a low MPO means saturation or compression limiting is reached easily with everyday sounds, including speech at a conversational level. Functionally, the MPO on a hearing aid is regarded "low" if the maximum output of the hearing aid is below the wearer's loudness expectation despite attempts to adjust the gain higher. This is especially true for those with a significant degree of hearing loss and when the input is above a conversational level.

The output of a hearing aid depends on the input level and the gain of the hearing aid. The MPO limits the output. This means that, as long as the output is below the MPO of the hearing aid, any changes in the input level or to the gain to the hearing aid will be accompanied by a corresponding change in output loudness that the listener should appreciate. Thus, for someone with a mild degree of hearing loss (who needs little gain), and/or when the input level is low, the impact of a low MPO may not be evident because the output level is below the MPO.

On the other hand, when the same hearing aid is worn by someone with a more severe degree of hearing loss (who needs considerable gain), or when the input level to the hearing aid is increased, the low MPO may not be able to produce a corresponding increase in the output because the output goes into saturation and/or compression limiting. The wearer with a more severe degree of hearing loss may complain that the hearing aid sounds "distorted," "muffled," "unclear," and "soft" even at a conversational input level.

FIGURE 2. Waveforms of output from three BTEs with MPOs of 108 dB, 124 dB, and 135 dB for Hearing Aids A, B, and C, respectively, for female speech in quiet presented at 58 dB SPL (top row) and 85 dB SPL (bottom row). The same flat 75 dB hearing loss was programmed in all three aids; different magnitude scales are used for the upper and lower rows, but the same sensitivity scales are used. Although there is very little difference between aids in the output at the lower input level of 58 dB SPL, the differences are fairly pronounced at 85 dB SPL.

The situation is illustrated in Figure 2. In this comparison, three BTE hearing aids with peak MPOs of 108 dBSPL (#A), 124 dBSPL (#B), and 135 dBSPL (#C) were used. All three hearing aids were programmed to the same flat 75 dBHL hearing loss with all the adaptive features deactivated during the recording. Female speech at 58 dBSPL (top row) and 85 dBSPL (bottom row) was presented. Output from the hearing aids was recorded via a Quest sound level meter (SLM). The same sensitivity scale on the SLM was used for all three hearing aids at the same presentation level. The magnitude scale of the display was adjusted for each input level, but it was the same scale for all three hearing aids.

At the 58 dBSPL presentation level (upper row), the output of all three hearing aids was roughly similar. This suggests that when the same moderately severe hearing loss was programmed in these three hearing aids of disparate MPOs, one can expect roughly the same output from the three hearing aids when the input level was low.

FIGURE 3. Hypothetical output from two hearing aids with identical gain settings but different MPOs.

The bottom row of Figure 2 shows the output at a high input level of 85 dBSPL. A large difference in output level was seen among the three hearing aids. The RMS output level of Hearing Aid A was 5 dB lower than Hearing Aid #B and 10 dB lower than Hearing Aid #C. For the wearer with the 75 dB hearing loss, the output of Hearing Aid #A may be unacceptably soft, muffled, and unclear. Hearing Aid #B should sound more acceptable than #A; but it would also be softer than #C. Of course, Hearing Aid #A should have never been fit to the patient with the 75 dBHL flat hearing loss.

The reduction in output level is explained in Figure 3. This is a hypothetical input-output (I-O) curve of two hearing aids set to the same gain characteristics but each with a different MPO. At an input level of 58 dBSPL, the output of both hearing aids is identical at 92 dBSPL. However, at a high input level (85 dBSPL in this case), the output should have been 115 dBSPL. Because of the lower MPO used in one hearing aid (#A), its output is restricted to a maximum of 105 dBSPL whereas the one with a higher MPO (#B) allows an output of 115 dBSPL. A reduced output level from the lower MPO is the culprit for the complaint.

Overcoming the loss of loudness?

One may think it possible to overcome the reduced output in a low-MPO hearing aid by turning up the volume control (or gain) of the aid. Unfortunately, such a limitation is impossible to overcome. Increasing the VC would, in theory, increase the output for all sounds. However, because the MPO of the hearing aid is fixed at a low level, the output intensity at high level inputs would stay the same. The complaint of "softness" or "muffled" speech would persist even though the loudness of soft sounds may improve.

One widely circulated suggestion is to insert the hearing aid (or earmold) deeper into the ear canal. This has the effect of reducing the residual ear-canal volume, increasing the effective SPL measured in the ear canal. This would overcome the limitation of a low MPO since the final output of the hearing aid—including the action of the MPO—would have been "amplified." This is similar to a vertical displacement of the input-output curve shown in Figure 3.

This suggestion has been advanced by some manufacturers of RIC hearing aids; they maintain that, in spite of the low MPO used in the RIC hearing aids, a more severe loss (ie, than the MPO might suggest) may be fitted with RIC hearing aids. It was argued that a deeper insertion of the RIC into the wearer's ear canal would compensate for the limited output from a low MPO.

At first glance, this seems like a reasonable approach. There are, however, two difficulties. First, the magnitude of the output increase is limited to the size of the ear canal volume reduction. In a simplistic sense, a halving of the ear canal volume would increase the SPL at the ear canal by 6 dB. Consequently, the output of Hearing Aid #A may become similar to that of Hearing Aid #B if the earmold can be inserted deeper so that the residual volume is only half of its original (or 1/4 of its original if it were to approximate the output of #C).

First, it should be pointed out that this is not always a viable option because of potential issues with physical discomfort from deeper earmold insertion.

FIGURE 4. Similar to Figure 2, this figure shows waveforms of output from Hearing Aids A (108 dB MPO), B (124 dB MPO), and C (135 dB MPO) to female speech in noise at a +15 SNR, with speech presented at 58 dBSPL (top row) and 85 dBSPL (bottom row). The same flat 75 dB hearing loss was programmed in all three aids; different magnitude scales are used for the upper and lower rows. This time, the hearing aids were equalized to the peak SPL of Hearing Aid #C to simulate deeper insertion into the ear canal. The result: Output is similar at the quieter (58 dBSPL) input levels, but the peaks and valleys of the input signal disappear in Hearing Aid #A at the louder (85 dBSPL) input levels.

Second, even if the deeper insertion provides a sufficient output, there is a potential that it may not be an optimal solution for listening in noise. The situation is illustrated in Figure 4 with the same three hearing aids that have the different MPOs. This time, the same speech signal used in the previous demonstration was presented within continuous speech-shaped noise at a SNR of +15 dB. To simulate the deeper insertion of the hearing aid into the ear canal, the outputs of the hearing aids are equalized to the peak SPL of Hearing Aid #C so all three have the same overall peak output level. (This step was necessary only for the 85 dB SPL input level.)

FIGURE 5. Hypothetical output from two hearing aids (A and B) with identical gain settings but different MPOs. Speech in the presence of a speech-shaped noise (+15 dB SNR) was used in the background.

As is the case in quiet, the output of all three hearing aids is similar (top row) at a 58 dB SPL speech input level. On the other hand, the waveforms of the three hearing aids are significantly different when the 85 dB SPL input level is used. After the equalization, the "peaks and valleys" of speech are clearly visible in the output of hearing aids #B and #C, but not in #A (which has the lowest MPO). Indeed, the highly modulated waveform that was seen at the lower input level is not visible anymore. Perceptually, the wearer of Hearing Aid #A with the low MPO would report increased noise level and "muffled" sound quality, even though its loudness may be adequate from deeper earmold insertion.

The reason for this is explained in Figure 5. At a high input level (in noise), the speech output exceeds the MPO of Hearing Aid #A, and its level is not as high as it could have been. Because the background noise is lower in input level than speech, it is amplified to the level of the MPO. Thus, no intensity difference is noted between the speech and noise output (or a poor SNR) for #A (with a low MPO).

For the hearing aid with a higher MPO (#B), both the speech and noise outputs are below the MPO of the hearing aid. Thus, the intensity relationship or SNR is maintained.

This means that, even though one may compensate for the reduced loudness from a low MPO by using a deeper-insertion hearing aid, such compensation does not fully overcome the negative effect of a low MPO. A SNR that is poorer than the original input may result, especially in a compression hearing aid that uses primarily fast attack and release times.

Identifying and Resolving MPO Issues

The problem with a low MPO can be manifested in several ways. Invariably, the complaints from the wearers are that conversational sounds "are not clear," "muffled," or "too soft," while perceptions of soft sounds are typically acceptable.

The most likely candidates to voice such a complaint are those whose hearing loss exceeds (or is in the upper limits of) the recommended fitting range of a device. In this case, check to see if the wearer's hearing loss exceeds the recommended fitting range or if it falls within the upper 10-20 dB of the recommended fitting range. If it does, choosing a hearing aid with a higher MPO might be the best solution.

Sometimes, an experienced hearing aid wearer may complain that the newly prescribed hearing aids are "not loud enough" despite the use of the same frequency-gain response between the new and previous hearing aids. Upon questioning them, one often finds that the insufficient loudness is primarily reported for conversational or loud input sounds, but not low input sounds. In recent years, this has occurred more frequently as experienced users upgrade to the newer, smaller hearing aids. If the wearer has been using hearing aids with a significantly higher MPO than the new hearing aids, the lower MPO could limit their perception of loudness, resulting in complaints. In this case, examine the MPO of the new devices and compare it to the MPO used in the previous hearing aids. Matching of the MPOs between the new and previous hearing aids may be necessary, especially for those with a hearing loss that reaches the upper limit of the fitting range.

In some cases, the MPO-related sound quality issues arise from our own doing. As part of the fitting, many protocols recommend measuring the individual loudness discomfort level (LDL) and setting the MPO below this LDL level. The rationale is to prevent loudness discomfort. The relative merit of this measure is beyond the intent of this article; however, it is conceivable that patients who present with an exceedingly low LDL will be provided a hearing aid with a low MPO. Thus, while one may avoid loudness discomfort issues in these patients by adjusting the MPO accordingly, poor sound quality may be an unwanted consequence.

Hearing Aid Selection and BTEs: Choosing Among Various "Open-ear" and "Receiver-in-canal" Options, by Francis Kuk, PhD, and Lars Baekgaard, MS. March 2008 HR. How to Measure and Demonstrate Four Key Digital Hearing Aid Performance Features, by David J. Smriga, MA. November 2004 HR. Understanding the ANSI Standard as a Tool for Assessing Hearing Instrument Functionality, by George J. Frye. May 2005 HR.

For patients with an exceedingly low LDL, it is worthwhile to reinstruct the patients and carefully validate the LDL measures. If the low LDLs are indeed valid, one may need to counsel the patients on the need of the low MPO and how that may impact sound quality.

Conclusions

The above discussion highlights the importance of the MPO in hearing aids. It suggests that, in choosing a hearing aid, one must read the MPO information on the hearing aid prior to its selection. It is important to select a hearing aid with a sufficiently high MPO such that the desired gain plus the input levels (ie, the desired output level) would not be limited by the MPO. The same considerations should also be applied to setting the MPO on a hearing aid in order to ensure maximum audibility and optimal SNR.

Reference

1. Kuk F, Keenan D, Baekgaard L. Speech in noise performance of a micro-size BTE. Hearing Review. 2007;14(10):60-64.

Correspondence can be addressed to or Francis Kuk, PhD, at .