Abstract
To evaluate whether hearing aids improve speech recognition in noise and self-reported listening outcomes in clinically heterogeneous adults with subjective listening difficulties in noise. Prospective, uncontrolled, within-subject intervention study. Participants completed multidimensional baseline testing followed by an 8-week hearing-aid trial assessing speech-in-noise performance, listening effort, self-report outcomes, and individualised benefit. Sixteen native Australian English-speaking adults aged 18-60 (mean 39.8 ± 11.1) years with near normal hearing sensitivity and substantial self-reported listening difficulties. At baseline, participants had normal audiological results but heterogeneous auditory-processing profiles, with the most consistent deficits on speech-in-noise measures. Speech Reception Thresholds in noise with hearing aids (-14.2 ± 2.2 dB SNR) were significantly better than unaided (-11.4 ± 1.6 dB SNR), showing a 2.8 dB improvement (95% CI: 2.1-3.4), t (15) = 8.8, p<.001. Unaided performance explained 69.2% of aided performance variance, indicating a consistent relationship with some individual variability. Although aided benefit varied across participants, the present data did not permit reliable separation of measurement variability from true differences in treatment response. Self-report measures demonstrated reduced hearing handicap, improved perceived listening ability, and favourable hearing-aid use and satisfaction. Unaided spatialised-noise test scores did not change. Hearing aids were associated with statistically significant and clinically meaningful improvements in speech-in-noise performance in adults with subjective listening difficulties and heterogeneous auditory-processing profiles. Apparent variability in aided benefit should be interpreted cautiously, as it likely reflects both measurement variability and true inter-participant differences. The broad consistency in benefit was observed despite marked differences in participants' baseline characteristics across multiple dimensions.