Prescriptive targets determined by a fitting formula are typically used as starting points when fitting a patient with hearing instruments. The challenge for clinicians has been to determine what adjustments will optimize the fitting for a particular patient in everyday listening situations.

A multimedia enhancement tool, the Beltone AVE™ system,1 and a Noah-based fitting software system, SelectaFit™, was developed to become the patient-facing component of a hearing instrument fitting system, allowing clinicians to more accurately and efficiently obtain information about their patients’ perceptions of various sounds. The system employs four measured speakers and a subwoofer in a controlled acoustic environment. The real-world sound recordings used in the fitting system were produced using a multi-array soundfield microphone in order to assure the most realistic sound quality possible. Environmental sounds, speech at various levels, and situations in which both speech and noise are present are all utilized during the fitting process. Adjustments based on patients’ perceptions of these real-world sounds can thus be made in the clinician’s office.2

Patients who experience aided real-world sounds before leaving the office should gain more realistic expectations about their amplification. It is possible that this can also enable them to accept a higher level of use gain. In addition, it was expected that subjects whose hearing instruments were fit using a multimedia fitting tool would require fewer adjustments before reaching final use gain than those who were fit without the benefit of this fitting tool.

The following article presents retrospective data collected during a two-year period on 26 subjects in order to examine the efficacy of such a fitting tool. All subjects were participants in unrelated research studies.

Methods
Subjects: Two groups of 13 subjects each were involved in the study. Subjects in Group A were fit utilizing the Beltone AVE.™ system; subjects in Group B were fit without the system. Audiometric data ranged from mild-to-severe sensorineural hearing loss, with similar distribution between the two groups (Table 1).

Freq (Hz) Group A
Avg Threshold
(dB HL)
Group B
Avg Threshold
(dB HL)
250 31.5 25.0
500 38.1 31.7
1000 46.5 37.9
2000 61.2 55.8
3000 62.8 68.3
4000 68.1 72.7
8000 71.8 61.4
Table 1. Average hearing loss for Group A and Group B subjects.

Approximately half of the subjects in each group were experienced hearing instrument users, while the other half were new users. Subjects in both groups were fit with the same style of hearing instrument (CIC) and circuit (two-channel WDRC programmable analog). A different manufacturing process was used to produce instruments for the two groups, but the difference had no effect on programming decisions or on the electroacoustic performance of the instruments. All subjects were fit binaurally.

Procedures: Although subjects in Groups A and B were participating in different research studies, there was no difference in the way patient concerns, programming decisions, outcome measures, or the determination of final use gain settings were handled.

Hearing instruments were initially programmed to settings prescribed by Beltone’s Adaptive Fitting Algorithm (BAFA), which generates targets for 50 dB, 70 dB, and 90 dB inputs.3 Group A subjects were fit following the delivery protocol in the multimedia fitting system. Group B subjects were fit following standard fitting procedures. Subjects’ subjective comments were elicited regarding their initial perception of sound quality, and adjustments were made accordingly. Use and care of the two-channel WDRC programmable analog instruments were explained to subjects. The unaided portion of the Abbreviated Profile of Hearing Aid Benefit (APHAB)4 was administered to enable investigators to assess perceived benefit with the hearing instruments.

After all the testing needed for the particular manufacturing process study was complete, each session continued as a routine hearing aid follow-up appointment. The patient and audiologist reviewed amplification experiences of the previous week. Adjustments were made to seek solutions to any negative experiences. Further counseling was provided as needed.

Real-ear measurements (REM) were conducted on all instruments to ensure that the patient was getting adequate gain in-situ, and to troubleshoot patient complaints. REMs were obtained on a Frye Fonix 6500-CX Hearing Aid Test System, with 50 dB, 70 dB, and 90 dB SPL inputs. Adjustments were made to the hearing instruments as needed.

At their final visit, subjects completed the aided portion of the APHAB questionnaire. Final use gain settings were determined based on both subjective and objective data. Sessions were spaced approximately one week apart for most of the subjects.

Final use gain for inputs at the three test frequencies was averaged for each group of subjects. This average was compared to the average prescriptive target gain, as calculated by BAFA, for each input level. The difference between target and actual gain was compared between the two groups, as well as the difference between initial use and final use gain.

The number of visits in which programming changes to the instruments were made was also tracked. (For the purpose of this analysis, sessions where research testing was performed but no changes made were not counted as follow up visits.) It was assumed that the number of these visits corresponded to the visits patients would have made to a clinician’s office for follow up in a traditional hearing instrument fitting.

Results and Discussion
The results for final use gain are displayed in Figures 1 and 2. “Initial Use Gain” refers to the frequency response at the end of the initial fitting; “Final Use Gain” refers to the use gain settings of the hearing instrument after approximately one month. As shown, the final use gain settings for Group A were closer to the prescriptive targets at all three input levels than for Group B. An independent samples t-test showed that the difference between the two groups was significant for all three input levels (p < .05).

figure f04-fig1-2guide.gif (7629 bytes)
Figure 1. Data for Group A subjects who used the multimedia fitting system. The blue lines represent the prescriptive targets generated by BAFA. The green lines represent the initial use gain obtained at the initial fitting of the instruments. The red lines represent the final use gain preferred after 1 month of wear. Note that, for Group A subjects, there is very little separation between initial use gain and preferred final use gain, meaning that little adjustment was made to the hearing instruments after the initial fitting appointment.

The finding that Group A subjects ended up with a higher average use gain than Group B subjects was not unexpected. Exposure to real-world sounds while still in the clinician’s office almost certainly leads to more realistic expectations for amplification. In particular, experience with potentially objectionable sounds can prepare patients for the fact that not everything they hear will initially be “pleasant.” For example, the multimedia fitting system allows the clinician to play sounds such as a newspaper rustling. If the patient has an adverse reaction, appropriate adjustments and/or counseling can be employed “on the spot.” This helps to eliminate the all too common situation where a patient states that their hearing instruments “sound fine” in the office, only to react adversely to the newspaper rustling, etc, once they are home. However, since final use gain settings were determined after only (approximately) one month of hearing instrument use, further research needs to be completed. This will enable us to determine the effects of a multimedia fitting tool with long-term use settings, and evaluate how the system impacts long-term hearing aid wear.

figure f04-fig1-2guide.gif (7629 bytes)
Figure 2. Data for Group B subjects who were fit without the multimedia fitting system. The blue lines represent the prescriptive targets generated by BAFA. The green lines represent the initial use gain obtained at the initial fitting of the instruments. The red lines represent the final use gain preferred after 1 month of wear. Note that, for Group B subjects, there is a greater separation between initial use gain and preferred final use gain than seen with Group A subjects.

The number of visits where programming changes were made to the hearing instruments was recorded and compared. Programming changes were made due to feedback, sound quality issues, or occlusion. The occurrence of these complaints was distributed relatively evenly between the two groups of subjects. Group A subjects required an average of 3.07 visits to reach final use settings, with a range of one to five visits. Group B subjects required an average of 3.92 follow-up visits, with a range of two to seven visits. These data suggest that final use settings can be reached with a multimedia fitting system in less time, averaging one visit fewer, than without such a system. Further investigation needs to be conducted to enable us to better understand which aspects of the multimedia system are responsible for this trend.

Subjective hearing aid benefit, as measured by the APHAB, was equivalent for the two groups of subjects (ie, there was no statistically significant difference). APHAB results for all subjects revealed benefit with amplification (aided versus unaided) for the EC, RV, and BN subscales.

Conclusion
Results demonstrate that final use gain settings for Group A subjects (those using real-world test signals) were closer to initial prescriptive targets than for Group B subjects. Furthermore, Group A subjects’ preferred final use gain was very similar to their initial use gain, indicating that few adjustments from the initial fitting were made to the hearing instruments. Group B subjects, however, showed a large negative difference (ie, final use gain was substantially less than initial gain) between initial and final use gain, indicating that more adjustments needed to be made to achieve satisfactory gain levels.

This is an interesting result in light of work done by Lindley et al.5 These authors found that subjects fit with a patient-driven (PD) protocol typically were fit with less high-frequency gain than those fit with an audiologist-driven (AD) protocol. Hearing instrument fittings utilizing a multimedia fitting tool would be considered PD; yet, the results of the present study indicate increased gain when compared to fittings done without such a system.

The system thus appears to bring the patient into an active role in the fitting process while not sacrificing as much gain as other PD protocols might. Since the multimedia system allows more precise adjustments to gain at the initial delivery, patients can achieve their preferred final use gain with fewer visits. Patients fit without such a system need to wear their hearing instruments longer to determine their likes and dislikes, while those fit with the system can give input based on loudness and quality judgments of specific listening situations at delivery, and adjustments can be made accordingly.

Patients who are fit with the multimedia fitting tool tend to reach final amplification settings in fewer visits than patients fit without such a tool. A multimedia fitting system may therefore have the potential to reduce the expense and inconvenience associated with repeated follow up visits.

The study suggests that using multimedia fitting software when fitting hearing instruments benefits both clinicians and patients. It allows clinicians to provide a higher standard of care by making the patient an integral part of the fitting process and to base their programming adjustments on a patient’s own perceptions of real-world sounds and sound scenarios. The ability to experience everyday sounds while still in the office helps patients set more realistic expectations for amplification—a big part of any successful hearing instrument fitting.

More research is needed to determine which patient populations will benefit most from a multimedia fitting tool, and how this technology can best be employed in a variety of clinical settings.

This article was submitted to HR by Jennifer Robinson, MS, Diane Russ, MA, and Bonnie Siu, PhD, who are research audiologists at Beltone Electronics Corp, Chicago. Correspondence can be addressed to HR or Jennifer Robinson, MS, Beltone Electronics Corp, 4201 W. Victoria St, Chicago, IL 60646-6718; email: [email protected].

References
1. Cox R, Alexander G. The Abbreviated Profile of Hearing Aid Benefit. Ear and Hear. 1995;16 (2):176-186.
2. Lindley G, Palmer C, Durrant J, Pratt S. Audiologist-versus patient-driven hearing aid fitting protocols. Seminars in Hear. 2001; 22 (2):139-160.
3. Meskan M, Robinson, J. A patient-focused approach to fitting hearing instruments. Hearing Review. 2000; 7 (12):52-55.
4. Russ D, Olsen G. Audio Verification Environment: How to present and assess real world performance in the office. Audiology Online. 2001, www.audiologyonline.com.
5. Russ D, Robinson J. Beltone’s Adaptive Fitting Algorithm: Combining loudness normalization and loudness equalization to achieve target gain. Audiology Online. 2001 www.audiologyonline.com