Audiovisual speech reception in noise and self-perceived hearing disability in sensorineural hearing loss

Multiple regression analyses were performed on a corpus of data from 56 sensorineurally hearing-impaired subjects. In addition to tonal thresholds, the corpus included simple, quantitative self-ratings for disability and results from speech reception tasks in auditory-only, audiovisual, quiet and noisy conditions. The aim was to evaluate to what extent the predictive power of disability formulas based on the audiogram can be enhanced, when measures derived from speech reception tasks are added to tonal thresholds as supplementary predictors. The better-ear speech reception threshold proved to be a strong exclusive predictor: if speech reception thresholds were introduced into the prediction, the contribution of better-ear pure-tone averages became redundant. Speechreading ability always significantly enhanced the prediction's accuracy. Tonal thresholds or speech reception data combined with lipreading ability scores can explain considerable amounts (42 per cent and 54 per cent, respectively) of the variance in disability ratings. The conclusion is that auditory-only, audiogram-based prediction schemes must not be abandoned, but can be tailored to individuals on the basis of realistic audiovisual speech reception performance.