8 hours Ago
A multi‐institution research team including Columbia, Vanderbilt, Johns Hopkins and others, report that nearly one in three new dementia cases in older adults is associated with hearing loss, based on statistical modeling of population data. Dementia prevalence is expected to triple globally in the coming decades. Efforts to reduce its burden have increasingly focused on modifiable risk factors.
Hearing loss, affecting more than two-thirds of older adults in the United States, has emerged as a potential correlating condition. Previous estimates of dementia associated with hearing loss in the United States have ranged from 2% to 19%. Variability across studies has been attributed in part to differences in how hearing loss is measured.
Many studies use self-reported hearing loss, which tends to underestimate clinically significant impairment in older adults. Whether dementia risk can be reduced by treating hearing loss remains unclear. In the study, "Population Attributable Fraction of Incident Dementia Associated With Hearing Loss," published in JAMA Otolaryngology—Head & Neck Surgery , researchers performed a prospective cohort study to calculate the population-attributable fraction of incident dementia associated with hearing loss in older adults and to investigate differences by age, sex, race, and method of hearing loss measurement.
Data from 2,946 community-dwelling adults aged 66 to 90 years without dementia at baseline were analyzed in the Atherosclerosis Risk in Communities Neurocognitive Study. Participants were enrolled across four states. Data collection occurred between 2011 and 2019, with one hearing assessment conducted at visit six (2016–2017).
Hearing loss was measured both objectively using pure tone audiometry and subjectively via self-report. Dementia diagnosis was determined using a standardized algorithm incorporating neuropsychological testing, informant interviews, and medical records. Clinically significant hearing loss measured by audiometry was present in 66.
1% of participants. During the follow-up period, 9.9% of participants with hearing loss developed dementia, compared to 4.
7% of those with normal hearing. An estimated 32.0% (95% CI, 11.
0%–46.5%) of dementia cases were statistically associated with any audiometric hearing loss, based on population-level modeling. Mild hearing loss (26–40 dB) had a population attributable fraction of 16.
2% (95% CI, 4.2%–24.2%), and moderate or greater hearing loss (>40 dB) had a population attributable fraction of 16.
6% (95% CI, 3.9%–24.3%).
Among participants aged 75 and older, the model indicates that up to about 31% of dementia cases may be linked to hearing loss. The 95% confidence range runs from -6 to 53%, indicating that the true figure could be far lower or even nonexistent. Similar, statistically uncertain patterns appeared for women (~31%, 6–47% range) and for white participants (~28%, -6–50% range).
Overlapping confidence ranges offered no statistical difference between men and women or between races. Self-reported hearing loss was associated with a lower, though not significant, risk of dementia. No population attributable fraction was calculated for self-reported hearing loss due to a hazard ratio below 1.
Findings indicate that different methods of measuring hearing loss can lead to substantially different estimates of dementia burden. Future research may benefit from prioritizing objective hearing assessments to improve estimates of hearing-related dementia risk. Of note, the study did not track hearing or hearing loss over time.
What is described as hearing loss within the study context was lower than normal hearing ability measured at a single point , before any signs of dementia. More information: Emily Ishak et al, Population Attributable Fraction of Incident Dementia Associated With Hearing Loss, JAMA Otolaryngology–Head & Neck Surgery (2025). DOI: 10.
1001/jamaoto.2025.0192 © 2025 Science X Network.
