January 4, 2024

expert reaction to study of hearing loss, hearing aid use and dementia

A study published in JAMA Otolaryngology–Head & Neck Surgery looks at hearing loss and hearing aid use and the risk of dementia. 

Prof Timothy Griffiths, Professor of Cognitive Neurology, Newcastle University, said:

“The study suggests an increase in the risk of dementia due to hearing loss that decreases in hearing aid users. Other ongoing studies include a large NIH-funded one in USA that address this using randomised trials as opposed to historical data on hearing aid use here. But if the hearing-aid effect is proven this supports models in which brain activity when we hear affects the disease process in dementia.”

Prof Martin Prince, Professor of Epidemiological Psychiatry, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, said:

“The clear potential for bias in these studies (including cohort studies), has never been properly highlighted.

“Epidemiological studies of the association between hearing loss and dementia are highly likely to be affected by Berkson’s bias. When classification of the exposure (hearing loss) and the outcome (dementia) are both dependent upon help seeking matched with diagnostic determination by health care services, then common underlying propensities may lead to overestimation of a true association or identification of an association where none exists within the populations from which the cases and controls, or cohorts have been drawn. The paper published in JAMA Otolaryngology – Head Neck Surgery (Cantuaraia et al) makes an original contribution to the evidence base, since it is a large long-term Danish historical cohort study, with a composite exposure definition (audiological assessment, and/ or hearing aid provision, and/ or hearing aid battery provision) and more or less complete ascertainment of clinical dementia diagnoses. Authors acknowledge the problem of unrecorded onsets of dementia, and hearing loss that has not led to help seeking, but do not fully acknowledge the potential for bias. The effect size for the association that they report is small, and much smaller than that reported from a Lancet Commission (Livingston et Al  Lancet. 2020) focused on a few small cohort studies conducted in the USA, where provision of rehabilitation measures (hearing aids) is much lower than in Denmark. This is consistent with the findings from the current Danish study that increased risk is concentrated among those with severe hearing loss and among those not receiving hearing aid rehabilitation.

“The potential for bias is such that little further elucidation of his association can be expected from epidemiological research, although long term systematic follow-up (with research as opposed to service diagnoses of dementia) of cohorts previously subject to audiologic assessment could be helpful. Randomised controlled trials of hearing aid provision are unlikely to be feasible. Under these circumstances, Mendelian randomisation may be useful. A paper published recently in JAMA Otolaryngology – Head Neck Surgery (Wang et al, JAMA Otolaryngology – Head Neck Surgery, 2023) suggests potential for this, with polygenic risk scores found to be associated with trajectories of hearing loss from childhood to midlife. These risk scores are likely also to be associated with significant hearing loss in later life, but not with help-seeking propensities. Therefore, an association between these polygenic risk scores and the onset of dementia in late life might provide the most persuasive evidence of a causal association between hearing loss and clinically apparent dementia. Or not, in the event of no association.”   

Prof Karen Steel FMedSci FRS, Professor of Sensory Function, Wolfson Sensory, Pain and Regeneration Centre, King’s College London, said:

“This is a very difficult topic because the data are hard to come by, but the authors have accessed the extensive public databases in Southern Denmark to conduct this analysis. It is important to consider that an association has been claimed, not a causal relationship, and it is still not robustly demonstrated that the use of hearing aids could delay onset of dementia; a prospective clinical trial that measured hearing ability at intervals and measured hearing aid use objectively as well as measures of dementia would be needed for this. As noted by the authors, one significant problem is that owning hearing aids is not the same as using hearing aids, as many users find them unhelpful so they sit in a drawer. The published study could not distinguish who was actually using their hearing aids. The other problem is the definition of the group with no hearing loss, because it is likely that many of the people included in this group did have a hearing impairment but had not acknowledged this or sought medical or audiological advice.  Nonetheless, the authors have carried out an interesting study that contributes to our growing understanding of the importance of good hearing to our general health as we age.”

Prof Jason Warren, Professor of Neurology and Consultant Neurologist, UCL Queen Square Institute of Neurology, University College London, said:

“This is a large well conducted study contributing to the growing body of evidence suggesting that hearing loss may promote dementia and treatment of hearing loss may delay or prevent it. However, studies of this kind should be interpreted with care – the brain changes associated with early dementia can affect hearing potentially some years before dementia is diagnosed and hearing loss (particularly if it is significant) acts as a ‘cognitive stress test’ that makes it more likely underlying dementia will be detected. In addition, people may be less likely to wear their hearing aids if they are developing cognitive problems. We should therefore be cautious about inferring that hearing loss causes dementia or that dementia can be prevented by wearing hearing aids.”

Prof Rob Howard, Professor of Old Age Psychiatry, UCL Division of Psychiatry, UCL, said:

“There has long been interest in whether treating deafness with hearing aids can reduce the risk of developing dementia or at least delay the onset of dementia.

“This is a large population register study, in which deafness was associated with a small (7%) increase in the risk of developing dementia and that this risk was slightly attenuated in people who used hearing aids. It is always important to remember that association does not equal causation and to think about the potential differences between older people with hearing impairment who acquire and persist with the use of hearing aids versus those who don’t. So far, no randomised clinical trial has shown that encouraging deaf people to use hearing aids significantly reduces the development of dementia. We should probably wait until we have the results of such trials before recommending this as a way to prevent dementia. However, for those who find them acceptable and useful, hearing aids bring many other positive benefits to the lives of people with hearing impairment.”

Dr Sarah Bauermeister, Associate Professor and Senior Scientist & Senior Data Manager, Dementias Platform UK, University of Oxford, said:

“This work contributes important findings to the field of understanding hearing loss and dementia. In agreement with our own work (Bucholc et al., 2021; 2022), the study suggests a longitudinal association between hearing loss and dementia risk, mitigated if the person using hearing aids.

“This study conducted on a large longitudinal population cohort with 14 years of follow-up provides support of a causal association between hearing loss and increased dementia risk. Only by using such large longitudinal studies can we really start to investigate these causative relationships which show change over time. We need to understand the biological mechanisms of the association and it is still possible that it is due to associated factors, such as social isolation.

“However, future clinical studies are required to understand the difficulties of wearing hearing aids in people with cognitive impairment and dementia, a focus of our own programme of work at Dementias Platform UK (DPUK), funded by ARUK. Also, the importance of regular hearing assessments across the adult lifespan cannot be underestimated. Destigmatising hearing aid use may also be an important outcome of regular testing.”

Prof Tom Dening, Professor of Dementia Research, University of Nottingham said:

“This is another interesting contribution to an important area of research. As hearing loss has been identified as a risk factor for the development of dementia, this has stimulated a lot of research into how hearing loss might cause dementia and also whether hearing interventions such as hearing aids might reduce this risk.

“The paper by Cantuaria and colleagues examines a large sample of over half a million people from the Danish population who attended public hearing clinics. They compared people with and without hearing loss in terms of the rate of incident (new cases of) dementia over an average follow-up period of nearly 9 years. The main finding was that hearing loss was significantly associated with the later development of dementia, though the risk (hazard ratio) was lower than has been reported in some other influential studies. The risk was greater in people with more severe hearing loss. The authors also found that people with hearing loss who did not use hearing aids has a 20% higher risk of dementia than those with normal hearing. People who used hearing aids for their hearing loss did not have such a high risk (6%, which was only just significantly above the risk for normal hearing). This suggests that hearing aid use may help to protect against dementia. This finding is consistent with other studies.

“The paper is strong because of the large sample size and the use of routinely gathered health data. Also, Denmark has the highest rate of hearing aid use among people with self-reported hearing loss, which makes it an interesting location for this study.

“The main limitation is that this is an observational study, so there may well be differences between people who do and do not wear hearing aids, for example the actual cause of their hearing loss or other medical and social factors. Whether hearing aids help prevent dementia remains an open question. In some ways, the group who do not wear hearing aids are more interesting. Why don’t they do so, and is this related to the type of hearing loss that they experience?”

Prof David Curtis, Honorary Professor at UCL Genetics Institute, University College London, said:

“This study repeats the frequently reported observation that having a hearing impairment is associated with subsequently receiving a formal diagnosis of dementia. This is consistent with the notion that hearing impairment may be an early feature of dementia or may share risk factors with some forms of dementia. It is not evidence that hearing impairment causes or exacerbates the progress of dementia. The fact that people who went on to be formally diagnosed with dementia were less likely to be using hearing aids would be readily explained by the fact that people in the early stages of dementia might derive less benefit from using hearing aids and would find their use more challenging. This study does not provide evidence that treating hearing loss with hearing aids would prevent dementia.”

Dr Leah Mursaleen, Head of Clinical Research at Alzheimer’s Research UK, said:

“This large, well-conducted study adds to the growing evidence that losing your hearing increases your risk of dementia. And this raises the possibility that a proportion of dementia cases could be prevented by correcting hearing loss through, for example, using hearing aids.

“Although the effects of using hearing aids are notoriously difficult to unravel, because researchers cannot be sure if people use them consistently, there is building evidence suggesting that using them can help reduce the impact of hearing loss on dementia risk. Further research is needed to help us better understand this link, and with around one million people now affected by dementia in the UK, and 12 million people estimated to have some type of hearing loss, this has never been more important.

“Like dementia, losing your hearing is not an inevitable part of ageing and early intervention is crucial. That’s why we are urging the government to include a hearing check in the NHS Health Check for those over 40. This could help millions identify hearing issues sooner, and potentially reduce the risk of dementia.”

Prof Gill Livingston, Professor of Psychiatry of Older People, University College London (UCL), said:

“This is good quality research from Denmark and an interesting paper. It is a longitudinal study so does not prove causality but fulfils criteria which suggest that causality is likely. It is strengthened by the large sample size of over half a million and using pure tone audiometry to have an objective measure of hearing loss. The follow up period was relatively long (mean 8.6 years, median 19 years) that one would ideally want and the relatively young age at baseline means that fewer people developed dementia (about 4%). The findings are in line with the rest of the literature and with what we would expect if hearing loss was causal. Those with more severe hearing loss had higher risk of dementia than those with mild hearing loss which is what would be expected if hearing loss caused dementia.

“There was a 20% higher risk for people who had hearing loss and no hearing aids (so uncorrected hearing loss) compared to 6% increased risk for those who had hearing aids. We do not know whether people who had hearing aids used them-in general a lot of people do not- and therefore the protection by hearing aids may be underestimated. Hearing aids in a drawer will not help. The authors only classified those who had hearing tested as having hearing loss and some of those classified as normal hearing because they were never tested may have been may have been misclassified as people often lack insight into their lack of hearing.  This would reduce the size of the  association of hearing loss with dementia.

“This is in line with one recent Norwegian study which followed up people for over 20 years, the time period meaning it was very unlikely that findings were due to cognitive impairment being mistaken for dementia  and also found that hearing impairment increased the risk of dementia (Hearing impairment and risk of dementia in The HUNT Study (HUNT4 70+): a Norwegian cohort study Christian Myrstad, Bo Lars Engdahl, Sergi Gonzales Costafreda, Steinar Krokstad, Frank Lin, Gill Livingston, Bjørn Heine Strand, Beate Øhre, Geir Selbæk 2023/12/4 eClinicalMedicine

“It also accords with a recent meta-analysis Yeo BSY, Song H, Toh EMS, et al. Association of Hearing Aids and Cochlear Implants With Cognitive Decline and Dementia: A Systematic Review and Meta-analysis. JAMA Neurol 2023;80(2):134-41. doi: 10.1001/jamaneurol.2022.4427) which found hearing aids were protective compared to no hearing aids.

“The authors consider confounders such as socioeconomic status, education and cardiometabolic disease and accounted for these in the analyses.

“This again adds to the evidence that using hearing aids, if hearing is impaired, is likely to help protect from dementia. The consistency of this evidence means that this is very important in the real world. Prevention is more important than cure and being able to hear also improves functioning and quality of life without toxic effects which drugs may have.”

Prof Gill Windle, Professor of Ageing and Dementia Research and Director of DSDC Wales Research Centre at Bangor University, said:

“This large Danish population study adds further evidence for the relationship between hearing loss and an increased risk of dementia, and raises an important public health message regarding the potential for hearing aids to mitigate this risk.” 

‘Hearing Loss, Hearing Aid Use, and Risk of Dementia in Older Adults’ by Manuella Lech Cantuaria et al. was published in JAMA Otolaryngology–Head & Neck Surgery at 16:00 UK time on Thursday 4^th January.

DOI: 10.1001/jamaoto.2023.3509

Declared interests

Prof Timothy Griffiths: No conflicts

Prof Martin Prince: No conflicts of interest to report

Prof Karen Steel: I have no conflict of interest with human studies of dementia and no plans to carry out any such study.

Prof Jason Warren: I have no conflicts to declare

Prof Rob Howard: No conflicts

Prof Tom Denning: I have no commercial or financial interests to declare. However, I am a person with hearing loss and a hearing aid user, so this area of research is of deep personal interest. My current view is that there are plenty of reasons to advocate greater use of hearing aids besides hoping that they may reduce my risk of future dementia, which would definitely be a bonus, of course.

David Curtis: No conflicts

Prof Gill Livingston: No declaration of interest

For all other experts, no reply to our request for DOIs was received.