Hearing impairment: A modifiable risk for dementia? By Dr Piers Dawes

• The identification of hearing impairment as a ‘modifiable risk factor’ for dementia assumes that hearing impairment is a cause of cognitive decline.
• So far, the available evidence has not demonstrated a strong causal impact of hearing impairment on dementia. There is evidence that common factors cause both hearing loss and dementia, and that cognitive problems impact on hearing.
• Research must continue to build an evidence base for the benefits of hearing loss prevention and treatment with a view to developing recommendations for policy change

A recent review of dementia prevention, intervention and care in the Lancet identified hearing impairment as the largest potentially modifiable mid-life risk factor for dementia. There are increasing numbers of people with dementia due to aging populations. There is therefore intense interest in modifiable risk factors in preventing or delaying the onset of dementia and minimising the personal and financial costs of dementia.

But the assertion that hearing impairment is a modifiable risk factor for dementia rests on the strong assumption that the association between hearing impairment and dementia risk is due to a causal impact of hearing loss on dementia.

All of the studies that reported hearing impairment associated with dementia risk are correlational. It is not possible to establish if there is a direct causal link between hearing impairment and dementia from correlational studies. It may be that hearing impairment contributes causally to dementia risk. But it is also possible that the association is the other way around (i.e. cognition impacts hearing) or that a shared factor impacts both hearing and cognition.

Hearing impacts cognition

There are various ways that hearing may causally impact cognition. First, hearing impairment may increase ‘listening effort’, meaning that people have fewer cognitive resources available for other tasks. People with hearing impairment may appear to have a cognitive deficit, when the underlying cognitive system is OK. A problem with this explanation is that associations between hearing and cognition remain even when cognition is tested with visually-based tasks.

A second possibility is that cognitively taxing chronic listening effort has a permanent adverse impact on cognition. This is an interesting idea, but it doesn’t seem very plausible: doing cognitively taxing things (like crosswords or Sudoku) is linked to preserved cognition, not cognitive decline. Brain training games stress cognition, but none of these programs has ever reported cognitive decline as an adverse side effect that I am aware of.

Lastly, hearing impairment may impact indirectly on cognition. Hearing impairment may lead to social isolation, depression and/or reduced self-efficacy, and these things may lead to reduced cognitive performance and increased dementia risk.

Cognition impacts hearing

Hearing is cognitively taxing. One must selectively attend to and sustain attention on the sounds of interest, avoid being distracted by irrelevant sounds, keep updating information and make links to prior information in memory. Reduced cognitive capacity therefore adversely affects listening. Someone may appear to have poor ‘hearing’ when their listening difficulties are due to cognitive impairment.

All of the studies that showed an association between hearing and dementia risk used hearing tests that are cognitively taxing. Some of the apparent association between hearing and dementia risk may be because hearing tests are sensitive to cognitive impairment. (One way of avoiding confounding with cognition on hearing tests would be to use objective physiological tests of hearing function).

A common factor(s) impact both hearing and cognition

Both hearing loss and dementia are complex, multifactorial conditions that share causes including inflammation and oxidative stress, hormonal and vascular changes. Hearing loss and cognitive decline/dementia share the same environmental/lifestyle risks and some genetic susceptibility factors. Aerobic exercise, healthy diet, non-smoking, moderate alcohol consumption and favourable early childhood development are all linked to both better hearing and cognitive function (see, for example, my earlier Policy@ Manchester blog on prenatal and early childhood developmental factors and cognitive function). The size of the impact of these early life factors on sensory/cognitive function is large enough to entirely account for the reported associations between hearing and cognitive function in adulthood. Hearing loss is therefore a marker of brain health.

Note that all three possibilities above are not mutually exclusive, and all three may be true to some extent.

Creating the evidence-base for effective policy

One way of testing whether hearing impairment is a modifiable risk for dementia would be to do a randomised controlled trial of hearing aid use, with incidence of dementia measured over several years. But such a study would need to be very large and long running. There are substantial practical, scientific, financial and ethical problems with such a study.

An alternative is to model long term cognitive outcomes associated with hearing aid use in longitudinal observational studies that have data on hearing, cognition, incident dementia/cognitive impairment and hearing aid use statistically controlling for differences between hearing aid users and non-users. I am aware of five studies that have done this to date:

We found no impact of hearing aid use on cognitive decline or incident dementia over 15 years in 666 US adults with hearing impairment. In an on-going analysis of data from the large UK Biobank study, we found no impact of hearing aid use on cognitive decline in 5,000 people with hearing impairment over 2-4 years. In another study currently under preparation, we found no impact of hearing aid use on trajectories of cognitive aging modelled over 11 years in ~10,000 participants in the English Longitudinal Study of Aging. One small US study reported a positive impact of hearing aid use on one out of three cognitive tests over 20 years.

We need more studies to test the impact of hearing aids and other hearing interventions on dementia risk and other cognitive outcomes. The evidence so far does not suggest a robust clinically important effect of hearing aids in reducing cognitive decline or dementia risk.

Note that the benefits of treating hearing impairments on reducing dementia risk do not necessarily need to be clinically significant (i.e. relevant at the level of the individual). Because hearing impairment is very highly prevalent, small gains in reducing dementia risk by effective management of hearing impairment could translate to a substantial reduction in numbers of people with dementia.

One possibility is that hearing aids (and other hearing interventions) do not have a substantial impact on cognitive decline, but may reduce dementia by reducing functional impairments. Dementia is diagnosed on the basis of cognitive decline plus sufficient functional impairment such that a person struggles to cope in daily life. Hearing impairments cause functional impairments and may exacerbate the impact of cognitive impairment on function. Hearing aids are very effective in reducing functional impairment.

Moving from theory to practice

I and my colleagues at the Manchester BRC believe that effective prevention and treatment of hearing impairment will reduce developmental, social and economic costs of hearing loss and this is why research into the wider effects of hearing impairment is so important.

At present there are no robust data that support a significant change in health policy in relation to hearing-related interventions to prevent dementia. I think that we need to understand the interactions of hearing impairment with cognition and functional impairments as well as the benefits of treating hearing impairment within the overall context of aging. This approach requires taking into account other sensory impairments (particularly vision impairment), and physical, psychological and environmental changes with age.

In order to affect policy, we need health economic data to demonstrate the cost benefits of effective identification and management of hearing impairment. We also need workable public health strategies to prevent, identify and treat hearing impairments to in order to improve trajectories of well-being in older age.