Samsvar mellom audiologisk målt høreterskel og selvrapportert nedsatt hørsel i en populasjonsbasert studie: Resultater fra HUNT2
Abstract
Background: Self-reported hearing loss has been validated against measured hearing loss
across specific subgroups. However detailed validation studies on a large sample are limited.
Purpose: The aim of the present study was to assess the effect of a single question on the
identification of presence or absence of a hearing loss, compared against the gold standard of
pure tone audiometry. The effect of level of education and self-reported health on the adjusted
self-reported hearing loss was also studied.
Material and methods: The study consisted 42 663 participants from the Nord-Trøndelag
Health Study (HUNT2), aged 19 – 101 years (mean = 50.2, SD = 17.0). The question “Do you
have a hearing loss that you are aware of?” was compared to hearing thresholds measured
from pure tone audiometry. The performance on the question was assessed for sensitivity,
specificity, positive and negative predictive values. The association between self-reported
hearing loss and other variables were calculated, using binary logistic regression analysis.
Analyses were adjusted for measured hearing loss and age per year.
Results: The performance on question showed moderate to low sensitivity for hearing
impairment and performance was slightly better in younger than older adults. Results showed
reasonable specificity with minimal effect of age. The positive predictive values varied
between age groups and were moderate at best. The odds of reporting hearing loss increased
with degree of measured hearing loss. The odds of reporting hearing loss also increased with a
higher level of education for both female and male participants. Results showed significant
association between self-reported general health condition worse than “very good” and selfreported hearing loss, in younger and older female participants and for middle age male
participants, compared to the “very good” health group.
Conclusion: The single question performed quite poor to identify higher frequency hearing
loss. The question seemed to be able to identify a large proportion of people without hearing
loss. A substantial proportion of the sample self-reported hearing-loss was not identified by
the audiometric criteria. There was an effect of level of education and general health on the
performance of the question.
Relevance: Hearing difficulties are the most common cause of disability globally. Large scale
epidemiological studies often use self-reports to estimate prevalence of hearing loss. This
study aims to improve the knowledge of the association between hearing thresholds and selfreported hearing loss.