Danish reading span data from 283 hearing-aid users, including a sub-group analysis of their relationship to speech-in-noise performance

Abstract

Objective: This study provides descriptive statistics of the Danish reading span (RS) test for hearing-impaired adults. The combined effect of hearing loss, RS score, and age on speech-in-noise performance in different spatial settings was evaluated in a subset of participants. Design: Data from published and unpublished studies were re-analysed. Data regarding speech-in-noise performance with co-located or spatially separated sound sources were available for a subset of participants. Study sample: RS scores from 283 hearing-impaired participants were extracted from past studies, and 239 of these participants had completed a speech-in-noise test. Results: RS scores (mean = 41.91%, standard deviation = 11.29%) were related to age (p <0.01), but not pure-tone average (PTA) (p = 0.29). Speech-in-noise performance for co-located sound sources was related to PTA and RS score (both p < 0.01, adjusted R-squared = 0.226). Performance for spatially separated sounds was related to PTA (p < 0.01, adjusted R-squared = 0.10) but not RS score (p = 0.484). We found no differences between the standardized coefficients of the two regression models. Conclusions: The distribution of RS scores indicated a high test difficulty. We found that age should be controlled when RS scores are compared across populations. The experimental setup of the speech-in-noise test may influence the relationship between performance and RS score.

Key Words:

• Reading span
• working memory
• hearing impairment
• speech-in-noise
• spatially distributed vs. co-located sound sources

Acknowledgments

The authors thank the participants and employees at Oticon A/S, who collected part of the data. A special thanks to Graham Naylor for his valuable comments on the manuscript and Thomas Behrens, Tobias Neher, Renskje Hietkamp, Karin Carvalho, and Louise Kragelund for their efforts in data collection.

Declaration of interest

The authors report no conflicts of interest.