J Am Acad Audiol 2005; 16(07): 448-460
DOI: 10.3766/jaaa.16.7.5
American Academy of Audiology. All rights reserved. (2005) American Academy of Audiology

Fitting Hearing Aids to Adults Using Prescriptive Methods: An Evidence-Based Review of Effectiveness

Gustav H. Mueller
Further Information

Publication History

Publication Date:
07 August 2020 (online)

The use of a prescriptive fitting approach for hearing aid selection has been a common practice for the past 60–70 years. While there are prescriptive approaches that have been validated, in recent years it has become popular to deviate from these validated methods and use manufacturers' proprietary algorithms, which in many cases are significantly different. This research review was designed to examine if there was evidence supporting the use of specific gain requirements for hearing aid fitting. Specifically, the question that was asked was "Are there real-world outcome measures from adult patients that show a preference for the gain prescribed by a specific prescriptive fitting procedure?" Inclusion criteria were as follows: adult subjects, consistent technology (e.g., different prescriptive methods compared using same hearing aids), real-ear verification of gain, and real-world outcome measures. For this review, in addition to subjective responses, preferred use gain was considered a real-world outcome measure. The National Acoustic Laboratories' revised (NAL-R), revised for severe/profound (NAL-RP), and the National Acoustic Laboratories—Non-Linear 1 (NAL-NL1) prescriptive methods were used as a common reference, as they have been the most commonly studied methods with adults.

Eleven studies were identified that met the inclusion criteria. Eight of the studies supported gain similar to that prescribed by the NAL-R or NAL-RP methods; three studies supported prescribed gain less than the NAL-R or NAL-RP. There was no evidence that gain greater than that prescribed by the NAL methods should be used. The level of evidence was moderate, as the supporting studies were either Level 2 or Level 4, and the statistical power of the studies was low.