Dr. Ross on Hearing Loss
Consumer Satisfaction is Not Enough
by Mark Ross, Ph.D., Harry Levitt, Ph.D.
Consumer satisfaction has emerged, in recent years, as the "Holy Grail" in all sectors of the hearing aid industry (Kochkin, 1992; 1993; 1994; Williams, l993; Jedidi & Estelami, 1993). It is being used as the major criterion to judge the success or failure of all aspects of the hearing-aid dispensing process. It is the concept most often employed to justify or criticize dispensing methods, practitioner qualifications and industry practices. Of course successful hearing aid fitting require our client's perception that our intervention has significantly reduced the impact of the hearing loss in their lives. This is clearly a necessary condition for assessing the results of our efforts (Weinstein, 1993).
What we would like to argue here is that while client satisfaction is an important criterion for evaluating the intervention process, it is not the only criterion. There are many factors contributing to the success or failure of a hearing aid. Different clients use very different criteria in judging whether they are "satisfied" with their hearing aids. At one extreme, some clients have unrealistic notions of what a hearing aid can do, expecting (or hoping) that a hearing aid will essentially eliminate the psychosocial, vocational, and communicative consequences of their hearing loss. As a result, much of our energy as practitioners is devoted to helping these clients develop realistic expectations regarding the potential benefits of hearing aids. This is a group we counsel "down", trying to get them to accept less than the ideal solution they had anticipated (but not, however, by implying that the client should accept less than what is possible to achieve).
Will these people be satisfied with their hearing aids? A lot depends on how well we have succeeded in reducing their expectations to a realistic level. If our counseling has been effective then they may well be content with the help provided by their hearing aids. If they still harbor unrealistic notions of what a hearing aid can do, then, no, they will not be satisfied with their aids, regardless of any objective improvement in their psychosocial, vocational or communicative performance.
At the other extreme, there are consumers who judge the success of the intervention process by comparing their listening skills with the hearing aid to their ability to hear in the unaided condition. For these individuals, just about any amplification device, no matter how poor, is going to improve their functional auditory capabilities to some extent. These are the people who are likely to be satisfied with their hearing aids, even if the instrument that they purchased provides far less objective benefit than could have been achieved.
Many of these highly satisfied consumers are often euphoric about their hearing aids, at least for the first few months, after which reality tends to moderate somewhat the initial euphoria (Weinstein reference). But they typically still agree that their hearing aids help them, and that they are "satisfied" with them. It is evident, however, that the question itself, "are you satisfied?" implies "compared to what?" And, in their case, the benchmark for comparison is their markedly deficient unaided hearing.
While some measure of overall satisfaction is clearly necessary, it is important to recognize the limitations of this measure and how easily it can be misused. Not only is this measure ill defined (hearing-aid satisfaction means different things to different people) but the measure is too dependent upon situational variables and too influenced by a host of personal factors (such as expectations, communication demands, life style, etc.) to be the sole-measure of success.
Another very important issue is raised almost all the time by our clients who ask the question "Is this the best hearing aid for me?" We typically fudge this question since we really do not know the answer. It's not enough for either the dispenser or the patient to be "satisfied" with a hearing aid; we also need some assurance that, out of the myriad of possible electroacoustic combinations, the one selected is in fact the best. We are not arguing that out of all the possible electroacoustic combinations there is only one that is just right for each hearing aid user. We are, after all, referring to hearing aids and not to marriages made in heaven! Still, if we cannot assume that there is only one "correct" fitting possibility, we also cannot assume the obverse, that all possibilities are equally good. Widely divergent signal processing schemes will not result in the same speech perception scores, either for a particular person or across people with a range of hearing losses.
This question of the "best aid" has animated and perplexed us since the earliest days of audiology (Carhart 1946). It is possible to provide a reasonable answer to this difficult question. Unfortunately, many of us have chosen the easy way out by using the highly malleable concept of satisfaction" as the sole criterion of success in dispensing hearing aids. It is as if the multitude of hearing aid choices and impressive technological developments in themselves validate the hearing aid selection process, precluding the need to ask this very fundamental question. "Satisfaction" is fine and dandy, but we also need some assurance that the selected hearing aid will deliver a signal equal or superior to that possible with other hearing aids (or other signal processing schemes in the same hearing aid). After 50 years, this is still the relevant clinical question, one that is usually avoided in the typical clinical setting.
As a case in point, consider the 1993 study by Gatehouse. In this study, he measured the speech perception scores of people who wore hearing aids dispensed by the British national health system for at least 12 months. By this time, they had acclimated to their hearing aids as evidenced by a plateau in their speech perception scores. Presumably they were deriving some benefit from the aids, else they would not have been routinely using them. If asked, many would have undoubtedly expressed "satisfaction" with their aids, since their only basis for comparison was the unaided condition. He then fitted them with hearing aids which provided a more suitable high frequency response. Initially, there was no difference in speech perception scores, but over a period of three months the superior performance of the newer hearing aids became increasingly manifest. Here we have a situation where "satisfaction" and performance diverge. One can be "satisfied" with a hearing aid and yet not achieve the maximum amount of benefit from it.
Years ago when we had far fewer electroacoustic possibilities than we have now, and the range of hearing impairments for which a hearing aid was considered appropriate was much smaller, the method of selecting the "best" hearing aid was much simpler. The goal was to find a hearing aid within a small subset of available hearing aids that provided the highest speech discrimination scores. We could test a person with 3 or 4 hearing aids, each somewhat different than the others, and make a case, albeit somewhat shakily, that we have recommended the "best" one for them. It was also possible to employ a reliable procedure to make this decision, given a sufficient number of recorded stimuli, but this was an occurrence more noted by its absence than its presence. Now, with an almost infinite number of processing schemes and hearing aid types available, this procedure is no longer a realistic option. The primary purpose of any hearing aid is to improve a hearing-impaired person's access to acoustic events. Our explicit goal must be a hearing that optimizes speech comprehension for commonly encountered listening situations. Other, less obvious listening conditions that deserve special attention for certain individuals are listening to music and the reliable reception of alerting signals.
The fitting problem is compounded by the many different types of hearing aids as well as the vast range of signal processing schemes. Is it realistic to compare, for example, a single band CIC aid with a two band, multi-memory BTE hearing aid? The people who express preferences for these different types of instruments may differ in their communicative priorities, and maximizing speech perception scores may not be as important as cosmetic considerations for some. But they can hardly make an informed choice if they do not have an opportunity to compare the objective performance of the two types of instruments.
Unfortunately, speech perception tests now are being used more as a validating tool than as a criterion for selection, particularly (but not only) for ITE and ITC hearing aids. For these aids, the ear impressions and audiometric data are sent to the manufacturer who sets the electroacoustic parameters, perhaps with input and suggestions from the dispenser. When the hearing aids are returned, the dispenser may then administer speech perception tests (holding aside the issue of specific test conditions). But then what? What is acceptable performance? To what are these scores being compared? In this procedure, we have neither absolute or relative goals. That is, we have no speech perception target to compare the obtained score to, and neither do we have comparisons with variations in other fitting possibilities or hearing aids.
If the dispenser feels that the scores are too low, based on earphone measures or on clinical intuition based on experiences with similar patients, then he or she may try to modify the response and test again. Perhaps the scores go up and perhaps not, but unless aided performance is well below expectations, it is unlikely that the dispenser will return the aid for another one. Even this limited degree of flexibility is not possible with many CIC aids. Are we to assume that some kind of "optimal" amplification pattern has been reached on the first try? This is doubtful and impossible to confirm. The likelihood is that the obtained scores will appear "satisfactory" to the audiologist (and that from this, the judgment that the patient will also probably be "satisfied"). In our opinion, this is a recipe for professional mediocrity, with neither absolute or relative standards being applied.
Some audiologists try to avoid this dilemma by using a standardized "prescription" technique. Modifications from the standard seem to be based on the "does this sound better" method. Others may use loudness scaling procedures and select aids (in cooperation with the manufacturer) with output characteristics that compensate for abnormal loudness growth.While these techniques have appealing face-validity, they still do not answer the question whether speech perception scores have been maximized. And still other audiologists use other theoretical and practical considerations in making a specific selection, but where is the evidence that the right choice has been made?
The concept of what is "best" or what is "optimum" is a subtle one. As pointed out by Levitt (1993) there is no single set of electroacoustic characteristics that is optimum for all listening conditions (e.g., speech in quiet, speech in noise, music in quiet). A practical solution is to find the optimum electroacoustic characteristics for each of the 3 or 4 most commonly encountered listening conditions. It is unlikely that the same set of characteristics would apply to all of these conditions, but it is possible that a practical compromise can be reached. The goal would be a set of electroacoustic characteristics that is relatively good, if not optimum, on each of these most commonly encountered listening conditions. The advantage of a multiple-memory hearing aid should be obvious. The above approach is probably the only way in which we can honestly answer the crucial question "is this the best hearing aid for me?" The answer can be "Yes, for the listening conditions you will encounter most of the time". It is not the perfect solution but it is one that is both realistic and attainable. Finding the optimum electroacoustic characteristic for a given listening condition, however, does require an approach in which both objective performance and subjective preferences of the various options can be rapidly and reliably attained.
There are, then, approaches to the problem which we believe viable and which should be actively explored. The recent introduction of programmable hearing aids under computer control offer possibilities not previously available to practitioners (Engebretson, Morley, & Popelka, l987). Researchers have had access to this technology for well over a decade (Levitt, 1982) and hearing-aid fitting strategies using this technology have already been developed for the less complex programmable hearing aids.
We suggest that it is time that the question of the "best" hearing aid be put back on top of our professional agenda, rather than evading the issue by choosing a convenient version of "consumer satisfaction" as the sole measure of success. More often than not, lack of dissatisfaction (e.g., whether or not the hearing aid is returned) is used as the measure of success. This is hardly a criterion to inspire confidence in our professional competence. Professionals concerned with addressing rather than evading the central problem of maximizing hearing aid effectiveness need to pay greater attention to these advances and should take the initiative in implementing new approaches that address this central problem. In brief, hearing aids are still supposed to help people hear better and not just feel better.