Hearing-aid users died less often in large cohorts. Here is the catch

On 13 October 2026 The Lancet Healthy Longevity published a pooled analysis led by researchers at the US National Institute on Deafness and Other Communication Disorders. They combined individual records from three population studies: the Blue Mountains Hearing Study in Australia (1997–2007), the AGES-Reykjavik study in Iceland (2002–06) and the US National Health and Nutrition Examination Survey (1999–2006). Together that is 13,851 adults with a mean age of 61.8 years, 55% of them women. Hearing was not self-rated: it was measured with an audiometer, using the average hearing threshold of the better ear and the severity grades used by the World Health Organization.

The researchers then followed who died and from what. Their models accounted for age, sex, education, body mass index, smoking, previous angina or heart attack, diabetes, high blood pressure, stroke and a history of noise exposure. A separate Norwegian study of 50,277 adults, the Nord-Trøndelag Hearing Loss Study (1996–98), was used to check whether the pattern held in another population. According to the authors, it did.

The risk rose with each grade, starting from mild loss

Compared with people with normal hearing, the risk of death from any cause was 19% higher with mild hearing loss (hazard ratio 1.19; 95% CI 1.10–1.29), 55% higher with moderate loss (1.55; 1.42–1.70) and 56% higher with moderately severe or worse loss (1.56; 1.38–1.76). Death from cardiovascular disease followed a similar pattern: 1.18 (1.02–1.37), 1.43 (1.22–1.67) and 1.30 (1.05–1.60).

The mild grade is the detail worth noticing: in this analysis even mild loss was already associated with higher mortality.

In these observational cohorts, current hearing-aid use was associated with a 9% lower adjusted risk of death from any cause (hazard ratio 0.91; 95% CI 0.83–0.99) and a 15% lower adjusted risk of cardiovascular death (0.85; 0.73–0.99).

A 9% difference in an observational study is not a 9% promise

These are associations in observational data, not the result of an experiment. Nobody was assigned a hearing aid by chance, and people who get one and keep wearing it may differ from those who do not in income, access to care and attention to their health. Statistical adjustment reduces that difference but cannot remove what was never measured. For the hearing-aid result, the upper bound of both confidence intervals is 0.99, so the effect is close to the line of no difference.

The published summary also leaves questions open: it does not say exactly which group hearing-aid users were compared with, or how often and for how long the aids were worn. The underlying studies collected their measurements between 1996 and 2007. Funding came from Macquarie University, the US National Institutes of Health (NIDCD), the Icelandic Heart Association, the US National Institute for Occupational Safety and Health and the Norwegian Institute of Public Health; one author reports a salary co-funded by Cochlear Ltd, a hearing-implant manufacturer.

An earlier US analysis pointed the same way

In 2024 the same journal published a follow-up of 9,885 US adults from NHANES (1999–2012), 1,863 of whom had hearing loss on audiometry. After a median of 10.4 years, among people with hearing loss, those who used hearing aids regularly had a lower risk of death than those who had never used them (hazard ratio 0.76; 95% CI 0.60–0.95). Occasional users did not differ from never users (0.93; 0.70–1.24). Only 12.7% of adults with hearing loss used aids regularly. Both analyses draw partly on the same NHANES survey, so they are not fully independent confirmations.

ACHIEVE measured thinking, not survival

The strongest test of hearing treatment so far is ACHIEVE, published in The Lancet in 2023. It randomly assigned 977 adults aged 70–84 with untreated hearing loss either to audiological counselling with hearing aids or to health-education sessions, and followed them for three years. Overall, cognitive decline did not differ between the groups. A prespecified analysis found that the effect differed between older participants at higher risk of cognitive decline and healthier volunteers, which suggests a possible benefit in the higher-risk group. No significant adverse events were attributed to either intervention. ACHIEVE did not test whether treating hearing loss extends life.

Hearing has already moved into the prevention conversation elsewhere: the 2024 report of the Lancet Commission on dementia lists hearing loss among 14 potentially modifiable risk factors for dementia.

What would turn the association into evidence

The authors themselves call the possible link between hearing aids, other hearing rehabilitation and life expectancy an important area for future research. Settling it would take a randomized trial with deaths and major illness as outcomes, long enough follow-up, and a record of how much the devices are actually worn.

Until then, the new analysis supports a narrower point: measured hearing loss, even mild, is a marker of risk that is easy to detect with a short hearing test. If you or someone close to you has trouble following conversation, or has an unused hearing aid in a drawer, it is a reasonable thing to discuss with a doctor.

This material is for information only and does not replace a consultation with a doctor. Nothing in it is a recommendation to start, stop or change treatment.

Sources

• Li CM, Hoffman HJ, Engdahl B, et al. Hearing loss, hearing-aid use, and mortality risk: individual participant data meta-analysis of population-based studies from Australia, Iceland, and the USA and a replication study from Norway. Lancet Healthy Longev. 2026. PMID 42869063

• Choi JS. Hearing loss, hearing aids, and longevity: from association to intervention. Lancet Healthy Longev. 2026. PMID 42869064

• Choi JS, et al. Association between hearing aid use and mortality in adults with hearing loss in the USA: a mortality follow-up study of a cross-sectional cohort. Lancet Healthy Longev. 2024. PMID 38183998

• Lin FR, et al. Hearing intervention versus health education control to reduce cognitive decline in older adults with hearing loss in the USA (ACHIEVE): a multicentre, randomised controlled trial. Lancet. 2023. PMID 37478886

• Livingston G, et al. Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission. Lancet. 2024. PMID 39096926