An untested hearing problem may explain a third of new dementia cases

Hearing loss measured with an actual audiometric test, not the kind someone reports about themselves, may be tied to as much as 32% of new dementia cases over eight years, according to a Johns Hopkins-led study published in JAMA Otolaryngology–Head & Neck Surgery. The same study found no meaningful dementia link at all when hearing loss was based on self-report alone. Most older adults, it turns out, are poor judges of how well they actually hear.

Self-reports miss the risk that hearing tests catch

The distinction matters because most large health surveys ask people to rate their own hearing rather than testing it, and this study found that gap changes the entire result. Audiometric testing, a hearing exam administered with calibrated equipment rather than a questionnaire, showed a strong statistical association with future dementia diagnoses. Self-reported hearing status showed essentially none.

“This provides strong evidence… that a large fraction of population-level dementia risk could be associated with hearing loss,” said Jason R. Smith, the study’s Johns Hopkins-based lead author, in comments reported by HearingTracker. Douglas Beck, an audiologist not involved in the research, framed the everyday problem more simply: “Hearing is just perceiving or detecting sound; listening is making sense of sound,” he said, describing why a person’s own decline can go unnoticed until someone else points it out.

Researchers have floated a few explanations for why an untreated hearing problem and dementia risk might move together rather than one simply causing the other outright. One holds that a brain working harder to decode degraded, muffled sound has fewer cognitive resources left over for memory and attention. A related idea points to the social withdrawal that often follows untreated hearing loss, since people who struggle to follow conversation tend to participate in less of it, and reduced social and auditory stimulation over years is itself associated with cognitive decline.

ARIC-NCS: the 2,946-person cohort behind the number

The 32% figure comes from the Atherosclerosis Risk in Communities Neurocognitive Study, a National Institutes of Health-funded cohort that has followed a group of older adults since the early 1990s. Researchers analyzed hearing and dementia outcomes for 2,946 participants examined between 2011 and 2019, tracking new dementia diagnoses over an average of eight years. Dementia status was determined through standardized cognitive testing and clinical review rather than relying on participants’ own sense of whether their memory had declined, the same testing-over-self-report approach the study used for hearing.

Fourteen researchers from Johns Hopkins’ Bloomberg School of Public Health and School of Medicine worked on the analysis, published in the June 2025 issue of JAMA Otolaryngology–Head & Neck Surgery. Smith and his co-authors wrote, in a summary distributed through EurekAlert, that treating hearing loss might delay dementia for a large number of older adults, while cautioning that the study cannot prove a causal effect.

Most older adults misjudge their own hearing

Participants misjudged their own hearing ability in more than 70% of cases, comparing self-reports against the audiometric results collected during the same study visit. That mismatch is the practical reason the study’s authors emphasize testing over self-assessment: a hearing loss someone cannot perceive in themselves does not show up in a self-report, but it still shows up on an audiogram, and, in this analysis, it still carried dementia risk.

Age-related hearing loss typically develops slowly enough that people adapt without noticing, one reason hearing aids remain underused even when a measurable loss is already present. Untreated hearing loss has separately been linked to a higher risk of falls and to symptoms of depression in older adults, so the dementia finding adds to a list of consequences tied to a condition that a simple hearing test, rather than a conversation about symptoms, is what actually catches.

A second, smaller study points the same direction

Smith led an earlier, smaller analysis in 2023 using a different data set, the National Health and Aging Trends Study, and found a lower population attributable fraction: about 17% of dementia cases tied to moderate or greater audiometric hearing loss, or up to 18.7% for any degree of loss, in a 2023 paper in The Journals of Gerontology. The newer ARIC-NCS estimate runs roughly double that older one, a gap the authors attribute to differences in cohort age, follow-up length and how hearing loss was categorized, not to a single more dramatic new finding. The ARIC-NCS sample also skewed slightly older on average and followed participants for a full eight years, longer than the single-visit design behind the 2023 estimate, which may explain why it picked up more of the long-run relationship between hearing and cognition.

Neither study shows that treating hearing loss prevents dementia, only that measured hearing loss and diagnosed dementia move together in these cohorts. Smith’s team says a hearing test, not a self-assessment, is still the only way to know which side of that statistic an individual older adult is actually on.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview