Predicting hearing aid uptake in older persons: Do new predictors and multivariable prospective models hold the key to success?

Activity: Lecture / PresentationInvited talkAcademic


Background: Reviews identified various factors predictive for hearing aid uptake in older adults: self-reported hearing, hearing loss (pure-tone average; PTA), age, and attitudinal beliefs about hearing aids. Nonetheless, various other factors were either scarcely examined or show mixed results (e.g., stage of behavior change, coping). Moreover, studies seldom included multivariable models and prospective data, limiting the findings’ validity. By using multivariable prediction models based on prospective data, this study aimed to examine what are new relevant predictors of uptake and what is their predictive strength relative to that of predictors well-established in the literature.
Methods: Sample: 377 older persons (55 years and over) who presented themselves with hearing loss (PTA1,2,4 kHz≥35 dB HL) to an ENT-specialist (n=110) or a hearing aid dispenser (n=267). Shortly after participants learned their hearing aid eligibility they filled out a questionnaire. After five months participants’ uptake status was determined: did (1) /did not (0) enter hearing aid trial. Multivariable logistic regression analyses (corrected for source of recruitment) were applied to determine whether predictors established in the literature were significant predictors in our sample. Subsequently, reclassification tables indicated whether new factors significantly added predictive strength. New factors included stage of change, coping, discrepancy in views (participant and care professional) on hearing aid necessity, speech-in-noise recognition, self-efficacy, partner status, educational level, occupational status, social network size and participation, comorbidity, self-rated health, cognition, and mastery.
Preliminary results Of the established predictors, viewed hearing aid benefits (Odds Ratio, OR=1.19), social pressure (OR=1.15), PTA1,2,4 kHZ (best ear) (OR=1.05), self-reported hearing (OR=1.02), and stigma (OR=1.06) were significant predictors of uptake (Nagelkerke R2=0.60; Area Under the Curve=0.91).
Conclusions: Preliminary results support the high predictive ability of well-known predictors of hearing aid uptake. The results of further analyses will be presented and should show whether new factors are able to add predictive ability.
Period2 Jun 2016
Event titleHearing Across the Lifespan 2016
Event typeConference
LocationCernobbio, ItalyShow on map


  • hearing help seeking
  • prediction modeling
  • hearing loss