Directional Hearing Aids: A Case Study

Understanding speech in the presence of background noise is one of the greatest challenges faced by hearing aid users. As the signal-to-noise ratio (SNR) decreases, there is a predictable decrease in word recognition ability. Directional hearing aids can improve the SNR ratio, thereby enhancing word recognition under specific listening situations. In the following case study, both the Hearing in Noise Test (HINT)[1] and the Client Oriented Scale of Improvement (COSI)[2] were used to document the benefits of directional microphone technology. The HINT initially was developed to provide a reliable and efficient measure of speech reception thresholds for sentences in noise. The HINT stimuli consist of 250 digitally recorded sentences organized into 25 phonetically balanced lists of 10 sentences. The HINT masking noise is filtered to match the long-term average spectrum of the sentences. The test can be run adaptively or at a fixed SNR. Broadly speaking, on this test, a 1 dB improvement in the SNR results in a 10% increase in word recognition. The COSI is an open-ended, self-report tool that is completed on at least two occasions. On the first occasion, the patient lists in rank order the five listening situations he or she most wants improved by amplification. Patients are asked to be as specific as possible. These areas/goals can then be incorporated into the audiologic rehabilitation plan. At the follow-up appointment, the patients are asked to judge (1) the degree of change for listening in these situations attributable to the hearing aid(s) and (2) their perceived communication ability in each situation with their new amplification.

This case study involves a 75-year-old female who reported a subjective decrease in hearing sensitivity that had slowly progressed over the past 15 years. She indicated that 13 years prior she had worn hearing aids for approximately 1 year. These aids were lost and were never replaced. No further significant case history was reported. Audiological assessment revealed a sloping mild to moderately severe, sensorineural hearing loss, bilaterally (see Fig. [1]). Word recognition scores of 72% and 68% were obtained for the right and left ears, respectively, for NU-6 (Northwestern University, Test 6) word lists presented at the patient’s most comfortable listening levels of 85 dB HL and 95 dB HL. Uncomfortable loudness levels were obtained for speech at 110 dB HL in both ears. This patient appeared to be an excellent hearing aid candidate, and binaural amplification was recommended.

Figure 1 Audiogram. REC, recognition; RE, right ear; LE, left ear.

The COSI intake questionnaire was administered prior to odering the hearing aids were ordered. The patient’s top four priority areas of concern were: (1) watching Wheel of Fortune; (2) conversing in the car with her husband; (3) conversing with friends and grandchildren in restaurants; and (4) hearing her girlfriends. Note that these situations require listening in both quiet and noise. The patient was counseled regarding available hearing aid circuits and styles, as well as on the advantages of directional technology. Binaural, programmable, custom half-shell hearing aids with multimicrophone capabilities were selected. She was fit with two-channel, AGC-I (low kneepoint of 55 dB HL) hearing aids with manually switchable omnidirectional and directional microphone systems. A t-coil was installed in the left aid which provided left ear use of the telephone. If assistive listening devices were used in the future, the right aid would need to be retrofitted with a telecoil (see the article by Ross,[3] this issue). The hearing aids were programmed so a moderate input level signal met the NAL-R[4] target. Target gain and real ear saturation responses were verified using probe-microphone measurements.

Four weeks after the hearing aid fitting, the patient returned for follow-up. At this appointment, she completed the HINT for the first time and the follow-up COSI. The HINT sentences were administered at various SNRs until a percentage correct score was obtained between 30% and 50% and one between 50% and 70%. The sentences were presented at 0 degrees azimuth and the noise was presented at 180 degrees azimuth. Table [1] shows the percent correct scores obtained at the various SNRs tested for omnidirectional and directional hearing aid microphone settings. The COSI responses in Table [2] show that the patient rated listening in the two noisy situations as “better” and “much better” than without the hearing aids.

Table 1 HINT Scores at Each S/N Tested S/N Percent Correct Omnidirectional Mode Directional Mode + 5 75% 100% + 3 65% DNT + 1 60% 100% 0 47% 73% -1 39% 76% -3 18% 61% -5 DNT 31% -9 DNT 19% HINT, Hearing in Noise Test; S/N, Signal-to-noise ratio; DNT, did not test.

Table 2 COSI Responses SITUATION Degree of Change Perceived Final Communication Ability Worse No Change Slightly Better Better Much Better 10% 25% 50% 75% 95% Watching Wheel of Fortune X X Conversations in Car with Husband X X Restaurants - Dinner with Grandkids & Friends X X Having Conversations with Linda and Shirley X X COSI,Client Oriented Scale of Improvement.

In summary, the patient performed better on the HINT test when using the directional microphone mode than she did when using the omnidirectional microphone mode. This finding was reflected in her responses to the COSI in which she reported substantial hearing aid benefit for listening in noise. Although the HINT listening environment does not replicate a challenging real-world listening situation, it does provide an approximation of listening to spatially separated speech and noise as experienced in the real world. See the article by Walden et al[5] in this issue that describes the limitations of clinical testing of directional microphone advantages. The incorporation of the COSI into the counseling process allowed confirmation that the measured benefit was perceived by the patient as an improvement for listening in noise.

In this case study, the performance and benefit of directional technology were quantified in a clinical setting. The benefits of using the HINT and COSI approaches to validate improvement in listening in the presence of background noise and in patient satisfaction with directional technology were demonstrated. Through the HINT evaluation process, the patient demonstrated improvement in her ability to hear in noise. Although the HINT test environment/protocol does not replicate a challenging real-world listening situation, the controlled clinical environment can be used to insure that the directional microphones are providing benefit in ideal situations (e.g., when the signal of interest is in front of the listener and the noise is behind the listener in an environment with very low reverberation). Incorporating the COSI into the counseling process confirmed the subjective improvement noted with this amplification in the patient’s difficult listening situations. This improvement was demonstrated in the degree of change as well as in the final ability to hear categories of the COSI. Using the outcome scores as a counseling tool, the patient left with an assurance that the directional microphone technology she invested in was providing help in the “noisy” listening situations she had described as problematic prior to the fitting.

REFERENCES

• 1 Nilsson M, Soli S D, Sullivan J. Development of the Hearing in Noise Test for the measurement of speech reception thresholds in quiet and noise.  J Acoust Soc Am. 1994;  95 1085-1099
• 2 Dillon H, James A, Ginis J. Client Oriented Scale of Improvement (COSI) and its relationship to several other measures of benefit and satisfaction provided by hearing aids.  J Am Acad Audiol. 1997;  8 27-43
• 3 Ross M. Telecoils (T-Coils): issues and relevancy.  Semin Hear. 2005;  26 000
• 4 Byrne D, Dillon H. The National Acoustic Laboratories’ (NAL) new procedure for selecting the gain and frequency response of a hearing aid.  Ear Hear. 1986;  7 257-265
• 5 Walden B E, Surr R K, Cord M T. Real-world performance of directional microphone hearing aids.  Semin Speech Lang. 2005;  26 70-77