Assessing speech recognition abilities with digits in noise in cochlear implant and hearing aid users.

OBJECTIVE The primary objective of the study was to investigate the feasibility, reliability, and validity of the Dutch digits in noise (DIN) test for measuring speech recognition in hearing aid and cochlear implant users and compare results to the standard sentences-in-noise (SIN) test. DESIGN The relation between speech reception thresholds for DIN test and SIN test was analysed to determine the validity of the DIN test. As linguistic skills were expected to make different contributions in these tests, their influence was analysed. STUDY SAMPLE Participants were 12 normal-hearing listeners, 24 hearing aid users, and 24 cochlear implant users. RESULTS The DIN test was feasible for more participants than the SIN test. Intraclass correlation coefficients showed high reliability. The standard error of measurement was smaller for the DIN test than for the SIN test. DIN test and SIN test were highly correlated (r = 0.95 and r = 0.56 for NH+ HA and CI users respectively). In the regression analysis no significant contribution of basic linguistic skills or personal factors was found. CONCLUSION In the assessment of speech recognition in noise of aided hearing-impaired listeners with hearing aids or cochlear implants, the DIN test is a feasible, reliable and valid test.