Journal Logo

Article

Study finds use of Live Speech Mapping reduces follow-up visits and saves money

Cunningham, David R.; Laó-Dávila, Roberto G.; Eisenmenger, Barbara A.; Lazich, Richard W.

Author Information
doi: 10.1097/01.HJ.0000292491.05341.ac
  • Free

“Time is money.” Every clinician knows the truth of this axiom. We all strive to provide high-quality service in a limited time frame. The essential question is, how can we save time and maintain professional quality care? This report will describe a hearing aid fitting/verification technology that we began using in our practice some 18 months ago. This technology has reduced the number of post-fitting, follow-up visits and lowered our practice costs.

Before discussing our new approach to fitting and verification, we will describe how we used to accomplish this task. We routinely used standard probe-mic (real-ear) measurements to adjust electroacoustic parameters in order to reach targets prescribed by fitting formulas (NAL-NL1, DSL i/o, etc.). Typically, this involved measuring REAR (real-ear aided response) for soft, average, and loud inputs then running an RESR (real-ear saturation response) to be certain that the patient's UCL (uncomfortable loudness level) was not violated.

The problems with this strategy were twofold: (1) The measurements were made with either tonal or noise stimuli rather than actual speech, and (2) the patient did not seem to comprehend and fully appreciate the significance of these tests. We explained the simple black and white REAR/RESR printouts, but patients had some difficulty relating the data to the audibility of actual speech.

Using these same techniques to demonstrate the effects of multi-channel processing, multiple memories, or directional microphones appeared to leave patients more baffled than informed. Even if they could hear the differences produced by these features, they were not witness to them in a visual sense. Our measurements were precise, but they were not intuitive from the patient's perspective.

lIVE SPEECH MAPPING

We longed for the ability to use actual speech as a stimulus for real-ear measurements. We also wanted a more appealing way of depicting the results of various measurements so that our patients could not only hear, but also see the functions of certain high-tech circuit designs. As the old adage goes, “Seeing is believing.”

Enter Live Speech Mapping (LSM). LSM is accomplished during the REAR measurement. The stimulus is actual live speech. We read the “Rainbow Passage” as the speech sample. The examiner sits 1 meter from the patient as the passage is read with normal conversational effort. A VU (volume unit) meter on the fitting screen allows the talker to keep his/her voice at 65 dB SPL at the patient's ear. We may also ask the significant other to read the passage because he or she is usually the patient's most frequent communication partner.

Speech is an interesting and familiar stimulus for the patient and family. It is the signal that the hearing instrument is required to process. The REUR and the REAR produced by the speech signal are displayed in real time as a continuously upgraded peak ear canal SPL curve in the frequency range between 125 Hz and 8000 Hz (Figure 1). The REAR is overlaid on a shaded area that represents the modified long-term average speech spectrum for comfortable speech at 1 meter in quiet.

Figure 1
Figure 1:
Live Speech Map. U = uncomfortable listening level; O = patient's auditory thresholds in db SPL; shaded area = modified long-term speech spectrum for comfortable speech at 1 meter; lower curve = REUR for speech; upper curve = REAR for speech.

The top of the speech spectrum is a target for the peaks of speech produced while reading the Rainbow Passage. The long-term speech spectrum occupies a certain proportion of the normal listener's dynamic range. The shaded area in Figure 1 is modified to represent that portion of the hearing-impaired listener's dynamic range that will achieve loudness normalization for a variety of inputs. By comparing the patient's thresholds to the REAR values, we enable the patient to easily see how much of speech is audible. The clinician may adjust the aid while Live Speech Mapping is occurring because it runs over NOAH.

We often supplement the LSM with speech-weighted noise presented at soft (55 dB SPL), average (70 dB SPL), and loud (85 dB SPL) levels. We ask the patient to rate the loudness of these signals to confirm that these inputs will produce REARs that are “audible,” “comfortable,” and “tolerable.” Any programming changes that are made to the aid's electroacoustics are immediately visible to the patient.

Other important uses of LSM are demonstrating directional-microphone effects and compression features. In addition to measurements made at 0° horizontal azimuth, we routinely perform LSM from the side and/or rear of the patient as we toggle between omnidirectional- and directional-microphone selections. The frequency-dependent response reductions are easily visible (and audible) to the patient.

Wide dynamic range compression (WDRC) circuits are designed to provide gain that is inversely proportional to input. While LSM is running, we continue to speak at a normal conversational level as we walk away from the patient. The circuit will increase the gain (within limits) so that the REAR remains within the recommended range. The REAR stabilizes within the appropriate range for much louder speech presented near the patient's ear. In both cases, the patient is “witness” to the fact that the circuit is “ACTing” properly (making sound Audible, Comfortable, and Tolerable).

LSM is also useful for demonstrating a variety of other functions, including volume control adjustments, earmold modifications, multi-channel processing, multiple memories, and T-coil performance.

Precision and Presentation

LSM has reduced the number of follow-up “tweaking” appointments and our practice costs. We believe there are two reasons for this. One is precision, the other is presentation.

LSM allows us to place real speech—with its continuously fluctuating amplitude/spectral characteristics—precisely into the patient's preferred listening range (typically midway between threshold and UCL). Rather than relying on frequency-specific targets for soft, average, and loud inputs, LSM offers an intuitive intensity-frequency “range” into which amplified speech is delivered. It is easy for the patient to see that the dynamic characteristics of actual speech are presented at proper levels.

Our method for adjusting the manufacturer's “first fit” to achieve an aided response that is audible, comfortable, and tolerable is essentially the same whether we use tonal/noise stimuli or live speech. The only differences between LSM and traditional real-ear measurements are the type of stimuli used and the manner in which data are presented to the patient.

By “presentation” we refer to the way that the real-ear data are made meaningful to the patient. All of the measurements are carried out as the patient sits at the instrument console viewing real-time graphics on a large color monitor. The significant other is situated close to the patient and is also a witness to the process.

By the time the fitting/verification session is completed, the patient and significant other understand implicitly that the hearing instruments are tuned so precisely that further adjustments are rarely needed. Because we front-load our time and technology at the fitting appointment, we have significantly reduced the need for costly follow-up “tweaking” appointments.

Our Chart Review

We completed a retrospective chart audit at one of our office locations in order to tally the number of follow-up visits for adult patients fitted with and without the benefit of LSM. Aside from using or not using LSM in the fitting/verification process, all other variables were essentially constant: the site, the clinicians, the product choices, pricing, patient age, length of time from fitting to chart review, etc.

We randomly selected charts for review until we identified 35 patients who had been fitted with the LSM utility and 35 who had not. Nine charts had incomplete data and were excluded. The remaining 61 charts yielded 27 patients who were fitted with LSM and 34 who were not. Refer to Table 1 for the distribution of circuit type between the “without” and “with” LSM groups. These purchases represented a wide range of hearing aid manufacturers, styles, and prices.

Table 1
Table 1:
Number of purchasers, number of post-fitting (PF) visits, and number of post-fitting visits per purchaser for digital, analog, and all products (analog and digital) fitted without and with the use of Live Speech Mapping. Data include difference and percent reduction relative to the without condition.

Table 1 presents the number of purchasers, number of follow-up visits, and the number of follow-up visits per purchaser. Data are given for digital, analog, and all (analog and digital) hearing instruments combined. For digital users the total number of follow-up visits dropped by 50% when LSM was used in fitting/verification process. The mean number of follow-up visits per patient fell from 5.39 to 2.92 (a 45% reduction) for this same group of patients. The total number of follow-up visits fell 48% for analog product users when LSM was applied. Their mean number of follow-up visits per patient dropped from 3.43 to 2.47.

When all digital and analog purchasers were considered, we noted a 48% reduction in the total number of follow-up visits and a 36% reduction in the mean number of visits per patient when LSM was applied. By using LSM we “saved” nearly 2–1/2 (2.47) visits for digital purchasers and nearly one (.96) visit for analog purchasers. We saved 1–1/2 (1.51) follow-up visits when both levels of technology were examined collectively for all purchasers.

WHY THE DIFFERENCE?

What might account for the rather significant difference in the number of follow-up visits between the LSM group and the traditional group? The aid styles, prices, and circuit types were evenly distributed between groups. Just two of the authors (DRC and RWL) saw all of the patients and were equally likely to serve patients in both groups. It seems highly unlikely that a clinician-specific “halo effect” was operational.

Could the allure of a new piece of instrumentation (Otowizard) have engendered more interest and enthusiasm among the clinicians and patients alike? This is a distinct possibility. Perhaps inadvertently the clinicians devoted more time and energy explaining the significance of the measurements produced with LSM. It is also possible that these explanations, along with the visually appealing real-time color graphics, produced a more convincing presentation. This would be likely to persuade patients that their aids were adjusted so carefully that further modifications would be unnecessary.

Put in different terms, perhaps the LSM patients had more confidence in their products. This technology, which was purposely designed to make routine but sophisticated measurements more meaningful for patients, seems to have tapped into a previously unmet need—the need to make a standard clinical technique more understandable and, even, enjoyable for laymen and clinicians.

Cost Savings

Table 2 illustrates the potential annual cost savings to a hypothetical practice as a function of practice cost per hour and number of aids dispensed per year. The calculations are based on the data that we collected in the chart review reported in the preceding section of this paper. (See the endnote for an explanation of the formula that was used to generate these data.)

Table 2
Table 2:
Illustration of annual cost savings to the practice as a function of practice cost per hour and number of aids dispensed per year. Assumes typical post-fitting visit is 0.5 hours long, fittings are all binaural, and fittings are 40% digital and 60% analog. Data are dollars per year.

To use Table 2, find your own practice's cost per hour and the total number of aids that you dispense per year.

Locate the number of dollars that your practice might save if you applied LSM during the fitting/verification process. Note that there is a direct relationship between the number of aids dispensed, practice costs, and annual cost savings. By reducing the average number of follow-up visits by 2–1/2 for digital wearers and by one visit for analog purchasers, the practice saves valuable time—time that could be devoted to new hearing aid patients and other clinical/administrative functions.

A console version of the fitting/verification system discussed in this paper costs about $19,000. It includes not only LSM and a full range of probe-mic utilities, but also a clinical audiometer, ANSI test instrumentation, a video otoscope, a master hearing aid, a compact diskette player, a PC, and a Hi-Pro box. Based on the annual cost savings depicted in Table 2, a small practice with $100 per hour practice costs could amortize this equipment purchase in about 3 years, a larger practice in less than a year.

We want to make it clear that our group has no business relationship with the manufacturer of the instrumentation described in this article. Indeed, there may be other manufacturers whose equipment can perform actual speech processing in the probe-mic modality and display REAR in a similar fashion. Our intent is to share our practice's experience with LSM with our colleagues.

ENDNOTE

*Derivation of formula

Total savings per year results from the addition of the saving obtained from digital fittings and analog fittings.

Digital savings = .5APY x % digital x VS x HPV x CPH

Analog savings = .5APY x % analog x VS x HPV x CPH

Where APY is number of aids sold per year (assumes binaural fittings), VS is number of visits saved, HPV is number of hours per visit, and CPH is cost per hour.

Adding these formulas:

(.5APY x % digital x VS x HPV x CPH) + (.5APV x % analog x VS x HPV x CPH)

Applying the associative property of addition, the common variables are grouped as follows:

@p.5APY x CPH x ((% digital x VS x HPV) + (% analog x VS x HPV))

When variables are known, they can be substituted into the formula. Assuming that CPH is $100, digitals are 40% of total, analogs are 60% of total, VS is 2.5 for digital fittings and 1.0 for analog fittings, and HPV is .5 for both digital and analog fittings, the following formula results:

.5APY x 100 x ((.4 x 2.5 x .5) + (.6 x 1 x .5))

Multiply the numbers in parentheses

.5APY x 100 x (.5 + .3)

Add the numbers in parentheses

.5APY x 100 x .8 = final formula for determining cost savings depicted in Table 2.

ACKNOWLEDGMENT

We would like to thank Jeff McLaughlin, Gay Po, and Michael Po of the MedRx Corporation for their technical assistance. We also thank Melissa Tishok, AuD candidate at the University of Louisville School of Medicine, for her help in preparing this manuscript.

Copyright © 2002 Wolters Kluwer Health, Inc. All rights reserved.