PP432 - The Digits-in-Noise Test as an Outcome Measure for Hearing Aid Users

Abstract: This study evaluated the feasibility, reliability, and validity of the Brazilian Portuguese Digits-in-Noise Test (BP-DIN) as an outcome measure for hearing-aid (HA) users. Eighty-five participants (55 normal-hearing; 30 HA users) completed the BP-DIN and HINT-Brazil via loudspeaker at 0 azimuth. HA users were tested in the aided condition only. Same-day retest improvements were significant for HA users, though below the standard error of measurement. Excellent test–retest reliability was found for HA users, and BP-DIN strongly correlated with HINT-Brazil. Findings support the BP-DIN as a feasible and clinically robust measure of speech-in-noise performance.

Summary: Rationale: Speech-in-noise testing is an essential objective outcome measure in audiological care, as it quantifies functional communication performance, monitors rehabilitation progress, evaluates intervention effectiveness, and supports clinical decision-making. However, the scarcity of validated speech-in-noise tests in Portuguese limits their clinical use and prevents standardized outcome monitoring across services in Brazil. The Digits-in-Noise (DIN) test offers brief administration time, low linguistic and cognitive demands, and robust signal-to-noise ratio thresholds, making them highly suitable for routine clinical use.

Purpose: To investigate the feasibility, reliability, and validity of the Brazilian Portuguese Digits-in-Noise Test (BP-DIN) as a clinical outcome measure for hearing aid (HA) users.

Methods: This prospective observational study used a repeated-measures design (Ethics Committee Approval - Protocol No. 7.269.816). Eighty-five volunteers were enrolled and assigned to two groups: normal-hearing (NH) individuals (n =55, mean age 35 yrs, 47.3% males) and HA users (n = 30 mean age 71.8 yrs, 66.7% males). Hearing loss ranged from mild to moderately severe, with a predominance of moderate losses (66.7%). HA participants used mini-BTE (standard earhook 43.3%; thin tube 46.7% and RIC 10%). Average hearing-aid experience was 27 months, with a mean daily use of 7.6 hours. Prior to speech-in-noise testing, real-ear measurements were performed to ensure audibility and verify that hearing aids were fit to NAL-NL2 targets. The Hearing in Noise Test (HINT-Brazil) and the BP-DIN were administered in counterbalanced order via a loudspeaker positioned at 0° azimuth and 1 meter from the participant. Both tests used adaptive procedures to estimate speech reception thresholds (SRTs) in dB signal-to-noise ratio. HA participants completed all tests in the aided condition only. After a brief set of practice stimuli, each test was administered twice on the same day (test–retest). Additionally, 10 NH participants completed two further retests at 7 and 30 days after the first session.

Results: All participants were able to complete the BP-DIN and HINT-Brazil. For both groups and tests, better SRT values were obtained on the same-day retest, with statistically significant improvements for the HA group only (DIN: W=322; p=0.024; HINT-Brazil: W=341; p=0.027). Nonetheless, the magnitude of the difference was smaller than the calculated standard error of measurement (DIN = 0.90; HINT = 0.71). For the HA group, excellent intraclass correlation coefficients (ICC) were obtained for both tests in the test–retest comparison. In contrast, the NH group showed poor to fair ICC values, likely due to the homogeneity of the sample. A strong correlation was observed between BP-DIN and HINT-Brazil results (rho = 0.729; p < 0.001). For the HA group, socioeconomic and audiological factors (better-ear FPTA, hearing-aid experience, and daily use) did not influence BP-DIN SRT, whereas age and average pure-tone thresholds significantly affected HINT-Brazil SRT.

Conclusion: The BP-DIN administered in the sound field is feasible, reliable, and clinically robust for assessing speech-in-noise performance of HA users. Future research should examine its performance in individuals with more severe hearing loss and under test conditions using spatial separation of speech and noise.