A recently-published study on the relative efficacy of three commonly-used hearing aid circuits is exactly the kind of project that consumers have long been asking for.
The study, published in the October 11, 2000, issue of the Journal of the American Medical Association, not only investigated a topic of great relevance to consumers, but it did it with the largest number of subjects (360) and with the most impeccable research design that I've seen with any hearing aid study up to this time.
The National Institute on Deafness and other Communication Disorders NIDCD) and the Veterans Administration (VA) jointly sponsored the study. It took place in eight audiology departments located within VA facilities around the country, all of whom followed exactly the same research protocol.
What makes this study so unique is that it incorporated a "double-blind" research protocol. That is, neither the audiologist who administered the tests nor the subjects could identify the hearing aid circuit being investigated, since all circuits were incorporated in the same case and because different audiologists programmed the device and tested the subjects.
This is a crucial issue in an experimental design because we know, from many other areas including previous hearing aid research, how powerfully suggestion and expectations can influence a person's judgements. It is difficult to eliminate unconscious bias from creeping into studies in which investigators have some kind of interest in the outcome. This pertains to industry-sponsored research or to studies in which the researchers may inadvertently communicate to sublects their estimation of what device or feature is really "superior". A double-blind research design eliminates this potentially confounding factor.
The three hearing aid circuits investigated different ways of processing input speech signals. In the first way (peak clipping - PC), the hearing aid works in a linear fashion up to the point it reaches maximum output. That is, every increase in input is matched to a similar increase in output, until the maximum amplified sound the hearing aid can deliver is reached. At this point, the "peaks" of the input signals (usually the intense vowel sounds) are "clipped" and converted to noise. This is basically what happens when one turns the volume control up on a cheap radio; rotating the volume control past a certain point increases the noise but doesn't make it any louder The "saturation" point has been reached. This is basically the method of limiting the output of hearing aids most commonly used for many years, but less so now.
The second circuit works the same way as the peak clipping one, except that at the maximum output the "gain" (degree of amplification) is reduced and thus the saturation point is never reached. Essentially, this is a type of automatic gain control circuit called "compression limiting" (CL). It is activated only when the actual amplified output gets close to the output limits of the hearing aid. When this occurs, an electronic circuit automatically reduces the amount of amplification to compensate for the increase in inputs.
The third circuit, Wide Dynamic Range Compression (WDRC) is a feature that I have described in previous issues of this Journal. Essentially, it works throughout the operating range of a hearing aid, by amplifying soft input sounds relatively more than loud input sounds. Any hearing aid that does not include a volume control incorporates some sort of WDRC. Lately, this has been considered the most desirable way of amplifying input sounds of varying loudness levels. It's one thing, however, to base one's clinical practices on theoretical expectations and anecdotal reports; it is another, to make clinical recommendations based on firm research evidence. Which is what this study was designed to accomplish.
The study incorporated both objective tests of speech perception and the subjective impressions of the subjects. Each subject tried each circuit for three months, enough time for them to form definite subjective impressions. The results showed that all three circuits significantly improved a subject's performance compared to the unaided condition. In other words, hearing aids do indeed help, an unsurprising conclusion for any of us who consistently wear them. As was pointed out to me by one of the participants, however, this is exactly the kind of data that healthcare and insurance administrators keep asking for before they will even begin to consider approving hearing aid coverage in insurance plans. Hopefully, the results of this study will deprive the "nay-sayers" of at least one of their ostensible reasons for denying hearing aid coverage.
In terms of the major research question, it was found that differences in objective scores between the three circuits were relatively minor. Generally, it appeared that the wide dynamic range compression (WDRC) and the compression limiting (CL) circuits were slightly superior to the peak clipper (PC), with little consistent advantage between the WDRC or the CL.
Not surprisingly, the objective benefits that people received from the hearing aids were greater when the speech input levels were softer. Subjectively also, people generally preferred the quality of the WDRC and compression limited (CL) circuits to the peak clipping (PC) circuits, with perhaps a slight edge for the WDRC. At the conclusion of the study, the sublects were asked to rate which circuit they personally preferred. The highest rankings were achieved by the CL circuit (42 percent), followed by the WDRC (30 percent) and the PC (29 percent).
Considering all the results, one detinite conclusion is that it is not desirable tor hearing aids to limit the maximum sound output by using peak clipping. (One caveat, not covered in this study, is that many people with the most profound hearing losses prefer just this type of hearing aid.) Insofar as the other two circuits are concerned, the WDRC tends to be rated a bit more comfortable at soft and loud input levels, even though overall the CL circuit received the highest overall rating.
Because the differences between the WDRC and the CL circuits were generally minor, the authors of the study suggest that dispensers should consider cost benefit factors in deciding on a particular circuit. This conclusion also implies that multimemory hearing aids, where people can switch between the circuits depending upon the listening conditions, may be a desirable option. This study, in my estimation, should represent the beginning of a continual series of similar studies. Not only are there currently many other potentially important circuits included in the current generation of hearing aids, but new ones seem to be arriving at a bewildering rate.
They are being marketed to consumers as "superior" simply by virtue of their technical sophistication. This is undeniable, but technical complexity, no matter how impressive, does not necessarily translate into better hearing. This has to be directly and carefully investigated. The NIDCD/VA study offers a model of how such studies should be conducted in the future.
From Hearing Loss, May/June, 2001.
Acknowledgement: This column is supported, in part, by GRANT #133E980010 from the U.S. Department of Education, NIDRR, to the Lexington Center.
Mention of products or companies by the author does not Indicate SHHH endorsement, nor should exclusion suggest disapproval. Since everyone's communication problems and needs vary, SHHH suggests consulting with your hearing health professional.