Objective: To estimate the prevalence of hearing loss among community-dwelling older persons according to clinical criteria and to develop a brief self-report screening instrument to detect hearing loss.
Design: Survey.
Setting: National probability sample of noninstitutionalized older persons.
Participants: A total of 2506 persons aged 55 to 74 who participated in the National Health and Nutrition Examination Survey.
Main outcome measures: Hearing loss as defined by Ventry and Weinstein (VW) criteria and by the High Frequency Pure-Tone Average (HFPTA) scale.
Results: Hearing loss by VW criteria was present in 14.2% and by HFPTA criteria in 35.1% of those surveyed. The prevalence increased with advancing age and was higher among men and those with less education. A logistic regression model identified six independent factors for hearing loss by VW criteria: age > or = 70 years (adjusted odds-ratio (AOR) 2.7, 95% confidence interval (95% CI) 1.6, 4.4), male gender (AOR 3.0, 95% CI 1.9, 4.8), < or = 12th grade education (AOR 3.8, 95% CI 1.8, 7.7), having seen a doctor for deafness or hearing loss (AOR 8.9, 95% CI 5.3, 14.9), unable to hear a whisper across a room (AOR 3.2, 95% CI 2.0, 5.1), and unable to hear a normal voice across a room (AOR 6.2, 95% CI 2.6, 14.9). A clinical scale based on the logistic model had 80% sensitivity and 80% specificity in predicting hearing loss using VW criteria and 59% sensitivity and 88% specificity in predicting hearing loss using HFPTA criteria.
Conclusions: Hearing loss, as defined by two clinical criteria, is common and can be screened for accurately using simple questions that assess sociodemographic and hearing-related characteristics.