Longitudinal Associations of Subclinical Hearing Loss With Cognitive Decline

Abstract

Several studies have demonstrated that age-related hearing loss is associated with cognitive decline. We investigated whether subclinical hearing loss (SCHL) or imperfect hearing traditionally categorized as normal (pure-tone average ≤25 dB) may be similarly linked to cognitive decline and risk of incident mild cognitive impairment (MCI)/dementia. Participants from the Baltimore Longitudinal Study of Aging were cognitively normal adults at least 50 years old with cognitive assessments from 1991 to 2019 and pure-tone average ≤25 dB measured between 1991 and 1994 (n = 263). The exposure was hearing based on the better ear pure-tone average. Outcomes were test scores in various cognitive domains. Multivariable linear-mixed effects models were developed to analyze the association between hearing and change in cognition over time, adjusting for age, sex, education, vascular burden, and race. Kaplan-Meier survival curves and Cox proportional hazards models portrayed associations between hearing and incident MCI/dementia diagnosis based on predefined criteria. Of 263 participants, 145 (55.1%) were female; mean age was 68.3 years (SD = 8.9). Follow-up ranged up to 27.7 years (mean = 11.7 years). Adjusting for multiple comparisons, a 10-dB increase in hearing loss was associated with an annual decline of -0.02 SD (95% confidence interval: -0.03, -0.01) in Letter Fluency. No significant relationships were observed between hearing and incident MCI/dementia. A relationship between SCHL and cognitive decline was observed for the Letter Fluency test. Further studies are necessary to determine where in the spectrum of hearing loss there begins to be an observable relationship between hearing and cognitive decline.