The consensus among audiologists is that fitting two hearing aids offers greater benefit than fitting one only. Most forms of hearing impairment affect both ears, so fitting only one hearing aid seems not to make any sense in the great majority of cases. The exception would be where fitting two aids adds to someone's hearing difficulties in the form, for example, of binaural interference-where the input to one ear distorts the combined binaural signal. But such cases are rare.
The presumed benefits of bilateral fitting lie in improved directional hearing and related forms of binaural interaction, such as more audible signal (binaural summation), and improved hearing for speech in spatially separated noise. The latter is understood to be largely due to more effective functioning of the so-called head shadow effect. If a signal of interest (a target) is on one side of the head and an interfering noise is on the other side, a person listening with two hearing aids can take advantage of the fact that the aided ear on the same side as the target will be exposed to a better signal-to-noise ratio, because the head itself acts to diffract the higher frequency components of the contralateral noise.
Most laboratory evidence bears out these expectations about improved performance with two hearing aids, thus supporting the consensus for fitting two. This evidence is summarized in Dillon (2001). However, a recent report (Arlinger et al., 2003) concludes that there is no robust clinical field evidence supporting a claim for greater benefit from two hearing aids versus one.
New Assessment Scale
A project the author has been involved in began with the aim of developing a more broad-ranging self-report based inquiry into the functions of hearing in the world (Gatehouse & Noble, 2004). The motive for the project came from realizing that the laboratory-based procedures of the audiology clinic rest on the use of stationary listening conditions, and single, spatially stationary target speech signals heard in quiet or in standard forms of competing noise. While some real-world listening conditions are similar, on many occasions speech signals occur in more challenging conditions: as multiple inputs, or overlapping, and/or as changing in their spatial dynamics.
Only limited clinical attention has been given to spatial hearing, and that is typically confined to its directional component. But in everyday terms, spatial hearing comprises, and may indeed combine direction, distance, and movement discrimination, all of which are likely to be connected to dynamic communication settings in ways that have not been sufficiently investigated or understood. The experience of hearing in the everyday world involves identifying, recognizing, and segregating different components of the audible array. For that experience to be meaningful, even enjoyable, sounds have to be clear and natural, and listening and communicating should not be too much effort.
All these points may be self-evident once expressed, but the features described above have simply not been included in clinical assessments of hearing. As a first approach, we developed or modified from earlier sources a series of descriptive scenarios, attempting to capture as many as possible of the features of everyday listening outlined here. The resulting 50 items, framed as questions, form the Speech, Spatial, and Qualities of Hearing Scale (SSQ). Respondents are asked to rate their ability or their listening experience on each item on a 0-10 scale from poorest to best.
Examples of SSQ items include the following. For a speech item, "You are talking with one other person and there is a TV on in the same room. Without turning the TV down, can you follow what the person you're talking to says?" For a spatial item, "In the street, can you tell how far away someone is, from the sound of their voice or footsteps?" For a qualities item, "Do you have to put in a lot of effort to hear what is being said in conversation with others?" [The complete chart of SSQ examples is available online and is referred to as Figure 4.]
Clinical Applications of the New Scale
Interviews were conducted with a sample of 153 new hearing clinic clients, prior to any offer of a hearing aid. It emerged that the novel items-inquiring about challenging and dynamic contexts of listening and communicating, about the more dynamic features of spatial hearing, and about the attendant effort required for effective functioning-related more strongly than the more traditional ones to an independent self-estimate of everyday handicap. Handicap here refers to limitations on social life and emotional distress. This relationship remained even when degree of measured hearing loss was controlled.
Arguably, these more complex features of everyday listening rely on effective binaural hearing. The new scale could thus be useful in examining the question of benefit from two versus one hearing aid.
Two further clinic samples were accumulated, comprising people fit for the first time with either one hearing aid or two (Noble & Gatehouse, 2006). These were convenience samples, rather than cases randomly assigned to the two fitting profiles. Their compositions reflected clinic practice at that time, such that the group fit bilaterally had greater measured levels of hearing loss than the others.
We conducted a filtering exercise to derive three groups matched in terms of degree of audiometric loss, and age. This resulted in better-ear four-frequency average loss of about 55 dB in each group. The three groups were: those not yet aided (n=63), those with six months experience using one aid (n=69), and those with six months experience using two aids (n=34).
Results over the set of 50 individual SSQ items showed distinctly non-uniform outcomes across aid conditions among subsets of them. In order to show where contrasts between one versus two hearing aids, in terms of self-reported abilities, we grouped items into 12 subsets, four for each of the main parts of the SSQ.
In the Speech section, items are grouped into "speech in quiet," "speech in adverse conditions," "suppression of competing speech signals," and "monitoring multiple speech streams." The Spatial items are grouped into "direction," "distance and movement," "spatial fidelity" (sounds experienced as in expected locations) and "internalization."
The last category was represented by a single item inquiring about sounds in the world versus inside one's head. The item was included specifically to tap the issue of hearing aid listening producing an effect of reduced externalization of sounds, which is expected to be more pronounced with bilateral over unilateral fitting, and more with unilateral over unaided listening. The Qualities items are grouped into "segregation," "identification," "clarity and naturalness," and "listening effort."
Figures 1-3 show the average self-ratings over these 12 subsets of items, grouped within the three major SSQ subscales. The higher the graph bar, the higher the rating. In Figure 2, it may be seen that fitting one hearing aid offers benefit over unaided listening for speech hearing in quiet and in adverse conditions. It can also be seen that adding a second hearing aid offers no further benefit in these conditions. By contrast, a second hearing aid offers additional benefit over one aid in the more complex and challenging task of multi-streaming.
Figure 2 shows that two aids offer clear advantage over one in the discrimination of distance and movement, and help to some extent over unaided listening in the area of spatial fidelity. Two aids offer limited additional benefit over one in directional hearing. And, as expected, two aids deliver a penalty compared with one in rating of external sounds as experienced in the head rather than in the world.
In Figure 3 it may be seen that two aids do not deliver further benefit over one in the areas of segregation, identification, or clarity/naturalness, but do offer extra benefit over one in the area of listening effort.
All in all, it can be concluded that bilateral hearing aid fitting shows no signs of greater benefit over unilateral in domains of hearing that are the traditional objects of inquiry, namely, speech intelligibility in quiet and in various adverse (e.g., noisy) conditions. Rather, it is when the going gets tough-competing speech, rapidly switching or simultaneous streams-that two aids show advantage. In the spatial domain, the real benefit of two aids is in dynamic components of that function. As a pay-off, less listening effort is reported by users of two hearing aids.
The lack of reliable clinical field evidence for a benefit from two hearing aids is arguably due to the fact that inquiry has not gone to the novel areas of hearing function outlined here. Two aids offer benefit over one in difficult and dynamic listening contexts, and, of course, these are contexts that cannot be avoided, or, if avoided, represent a diminution of engagement with the social and physical worlds.
In order to achieve matched samples in the three conditions described here, milder hearing loss cases had to be removed. The findings described above may not generalize to people with levels of hearing impairment of lesser degree. In a study comparing cases whose better ear hearing loss was 10 dB lower than the present samples, no self-rated difference between one versus two aids was observed (Noble, Ter-Horst, & Byrne, 1995). In a recent review (Noble, in press), one conclusion was that preference for or persistence with bilateral fitting has typically been associated with greater measured and/or rated impairment and disability. Hence, it would be premature to conclude that such fitting will be universally adopted, even if audiologists so recommend it.