Investigating Speech-in-Noise Outcome in Older Hearing-Aid Users Using Auditory Profiling

Mengfan Wu

Research output: ThesisPh.D. thesis

118 Downloads (Pure)

Abstract

According to the World Health Organization, nearly one third of all people aged 65 and above suffer from a hearing loss. Consequently, there is considerable demand for effective treatment methods such as hearing aids (HAs). Even though HA technology has advanced greatly during the past few decades, many users are dissatisfied with their HAs and some even rarely use them. Problems are typically largest in noisy situations, where HA users can differ greatly in terms of their speech-in-noise (SIN) outcomes.
Several studies have suggested that this large inter-individual variability is related to supra-threshold hearing deficits. Since HAs are typically fitted based on the audiogram only, HA users with pronounced supra-threshold hearing deficits are unlikely to receive effective treatment from them, which could lead to poor aided outcome. To develop more individualized HA fittings, the Danish ‘Better hEAring Rehabilitation’ (BEAR) project developed an auditory test battery for individual hearing loss characterization. This battery includes measures of audibility, loudness perception, binaural hearing, spectro-temporal modulation detection, and unaided speech recognition in noise. Based on the test-battery results of a large group of hearing-impaired individuals, four distinct auditory profiles were identified using a data-driven approach.
The overall purpose of the current PhD project was to extend this work by investigating if the identified auditory profiles can help explain the large inter-individual variance in aided outcome, and if suitable HA solutions can be proposed for them. To accomplish this, three studies were conducted.
The first study investigated potential interactions between the four auditory profiles and the perceptual effects of six simulated HA processing strategies in a complex listening scenario. The six HA processing strategies differed in terms of their directionality, noise reduction and amplitude compression settings. Speech recognition, perceived overall quality, and perceived noise annoyance were assessed with 49 older experienced HA users belonging to the different profiles. The results showed clear differences among the four profiles in terms of their aided speech recognition thresholds. However, no clear interactions between the profiles and the six tested HA processing strategies were found.
The second study followed up on the observed differences in aided SIN performance among the auditory profiles by exploring the influence of the noise scenario and the HA fitting. Three noise conditions and two SIN tasks were included. Thirty-one older experienced HA users were fitted with three modern HAs from different manufacturers using their recommended procedures. Consistent with the results from the first study, HA users with milder hearing deficits achieved better speech recognition than HA users with pronounced hearing deficits. Moreover, the SIN outcomes of HA users with pronounced hearing deficits varied with the tested HAs, suggesting that the choice of HA is particularly crucial for them.
The third study investigated if cognitive abilities can explain some of the variance in aided SIN outcome observed in the two previous studies. Thirty-nine participants completed the auditory test battery, tests of working memory and processing speed, as well as two measures of aided SIN outcome. Multivariate data analyses were then performed to explore the contributions of auditory and cognitive factors to aided outcome. While the results showed an association between auditory factors and speech recognition in noise, cognitive factors were unrelated to aided outcome.
In summary, SIN outcome differed substantially across HA users belonging to the four auditory profiles. HA users with pronounced hearing deficits consistently obtained poor SIN outcomes with current fitting strategies. For such users, the choice of HA seemed to matter the most with respect to SIN perception. In contrast, no influence of cognitive factors on aided SIN outcome was observed. Taken together, auditory profiling has the potential to help audiologists identify HA candidates who are at risk of obtaining poor aided outcome and, as such, can serve as a basis for more personalized HA treatment. To enable better aided outcome, supra-threshold hearing abilities should ideally be considered in HA fitting procedures.
Original languageEnglish
Awarding Institution
  • University of Southern Denmark
Supervisors/Advisors
  • Neher, Tobias, Principal supervisor
  • Fereczkowski, Michal, Co-supervisor
  • Schmidt, Jesper Hvass, Co-supervisor
Publisher
DOIs
Publication statusPublished - 20. Dec 2021

Fingerprint

Dive into the research topics of 'Investigating Speech-in-Noise Outcome in Older Hearing-Aid Users Using Auditory Profiling'. Together they form a unique fingerprint.

Cite this