Treating hearing loss is on the shortlist of things that might delay dementia. The one trial built to test it came back null overall — and interesting in exactly one subgroup.
If you have read anything about hearing in the last few years, you have probably encountered some version of the claim that treating hearing loss protects the brain. It is one of the most repeated ideas in the category, and it sits on top of a body of research that is real, active, and considerably more careful than the headlines built from it. The distance between what the studies found and what the marketing says is where this article lives.
The short version: hearing loss is consistently associated with cognitive decline, and that association is strong enough that expert panels treat it as a risk factor worth addressing. Whether treating hearing loss prevents cognitive decline is a different question, and the best trial designed to answer it returned a mixed result. Both halves are true and both matter.
The Lancet standing Commission on dementia prevention publishes a periodic review of modifiable risk factors — conditions that are statistically linked to dementia and that could in principle be changed. Its 2024 report identified 14 such factors and estimated that addressing all of them across a population could potentially prevent or delay around 45% of dementia cases.[1]
Hearing loss is on that list, with a population attributable fraction of about 7% in the 2024 report — revised down from the 8.2% the 2020 Commission had assigned it.[2] That is a meaningful figure, and it is routinely misread. A population attributable fraction is a population-level estimate of how much of a condition might be avoided if a risk factor were entirely eliminated. It is not a personal risk number, and the authors of these analyses say so directly: a PAF “cannot be used to estimate individual risk.”[2]
The PAF also rests on an assumption that deserves stating out loud — that the relationship is causal and that the risk factor could be removed. Neither is settled for hearing. Age-related hearing loss cannot be eliminated from a population, and normal hearing cannot be restored by any current treatment.[2]
Association studies can only take you so far. To find out whether treating hearing loss changes cognitive trajectory, someone has to randomize people to treatment or no treatment and follow them. That is what the ACHIEVE trial did.
ACHIEVE — the Aging and Cognitive Health Evaluation in Elders study — randomized 977 older adults with hearing loss to either a hearing intervention or a health education control programme, and followed them for three years. Results were published in The Lancet in July 2023.[3] It matters what “hearing intervention” meant: hearing aids, a self-management “toolkit,” and ongoing instruction and counselling with an audiologist across the three years.[3] That is the prescription route at its most thorough. An over-the-counter device bought online and set up from a phone is not what was tested, and nothing below should be read as if it were.
The headline result, in the trial's own words: when both participant groups were analysed together, “the hearing intervention was not better than the health education control on slowing declines in thinking and memory abilities.”[4]
That is the primary finding, and it is the one least often quoted.
ACHIEVE recruited from two different populations. About 238 participants came from an existing cardiovascular health study and, as a group, were older, had more risk factors, and were declining faster. The other 739 were newly recruited community volunteers who were, broadly, healthier.[4]
In the higher-risk group, the picture was different: the hearing intervention reduced cognitive change by 48% over three years compared with the control[4] — in plain terms, participants in the intervention arm lost roughly half as much on the composite cognitive score as controls did over the same period. A later secondary analysis of the trial reached a consistent conclusion — that the cognitive benefit of hearing intervention varied according to participants' underlying risk of decline.[5]
This is a genuinely interesting result and it generates a reasonable hypothesis: that treating hearing loss may matter more for people who are already on a declining trajectory than for people who are not. It is also, in the strict sense, a subgroup finding — prespecified, which counts for a great deal, but still a smaller sample inside a trial whose overall result was null. The honest description is promising and unconfirmed, which is precisely how the investigators have described it while they continue following participants beyond the three-year mark.[4]
In the ACHIEVE randomised trial, a hearing intervention did not slow cognitive decline across the full study population over three years. In a prespecified subgroup at higher risk of decline, it was associated with 48% less cognitive change over the same period.[3][4] Reviewers of this literature note there is not yet convincing randomised-trial evidence that treating hearing loss reduces dementia risk.[2] Summarized from published research. General information, not medical advice.
Three explanations are usually offered, and they are not mutually exclusive.
Degraded sound takes more mental work to decode. The reasoning is that effort spent reconstructing speech is effort not available for encoding what was said — which would predict exactly the everyday experience of finding conversation in noise unusually tiring.
Difficulty following conversation makes social settings effortful, effort leads to avoidance, and social isolation is itself on the Lancet Commission's list of modifiable risk factors.[1] On this account hearing loss acts partly through a second pathway rather than directly. This pathway now has randomised evidence of its own: a 2025 secondary analysis of ACHIEVE found that over three years the hearing intervention group retained about one more person in their social network than controls, with small improvements in the diversity of their contacts and in loneliness scores.[6] The effects were statistically significant and, by the authors' own account, of uncertain clinical meaning — which is the honest size of the social-withdrawal evidence so far.
The explanation that most complicates the story: one or more common age-related processes may drive both hearing decline and cognitive decline, producing a strong statistical association without either causing the other.[2] If that is what is happening, treating the hearing would not be expected to change the cognitive trajectory — which is one reading of ACHIEVE's overall null result.
Nobody currently knows which of the three dominates, or in what mix.
“Associated with” and “protects against” are not synonyms. The gap between them is where most of the marketing in this category lives.
It does not justify buying a hearing aid to prevent dementia. No hearing device is approved, cleared, or demonstrated to prevent, treat, or slow dementia, and any product marketed to you on that basis is making a claim the evidence does not currently support.
What the evidence does support is more modest and, arguably, more useful: hearing loss is one of the larger modifiable risk factors on a well-constructed list; addressing it has immediate benefits for conversation and daily life, and measurable if modest ones for social connection[6]; and there is an unconfirmed but real signal that it may matter more for people already at higher risk of cognitive decline. Those are good reasons to take a hearing change seriously and get it measured. They are not a promise about your brain, and the researchers doing this work have been careful not to make one.
If you want the parallel version of this argument on the memory side — what the evidence actually supports for cognition after fifty, and what it doesn't — that is the subject of our Memory & Focus department, including what the largest trials found on ginkgo and why sleep outperforms every supplement studied.
And if you are earlier in the process than any of this, start with the measurement: the early signs of hearing loss and how a hearing test works.
General information, not medical advice. Hearing changes can have many causes — a licensed hearing professional or physician is the right person to evaluate yours. Talk to your doctor before changing any treatment or device you rely on.