Age-related hearing loss is an independent risk factor for dementia, with even mild hearing loss nearly doubling risk (Livingston et al., Reference Livingston2017). Observational studies find that wearing hearing aids may protect those with hearing loss from excess cognitive decline (Maharani et al., Reference Maharani, Dawes, Nazroo, Tampubolon and Pendleton2018). This suggests that treating hearing loss may deliver substantial reductions in dementia risk. People with mild cognitive impairment (MCI) who are at increased risk of dementia are typically diagnosed in their seventies (Sachdev et al., Reference Sachdev2015), when age-related hearing loss is also common (Scholes and Mindell, Reference Scholes, Mindell, Craig, Fuller and Mindell2015). A trial of hearing aids in those with MCI may clarify whether hearing aids can delay or prevent dementia. As more than 40% of new hearing aid users do not use them consistently or even at all, such a trial may be impractical because of low recruitment for hearing aids and poor retention (Aazh et al., Reference Aazh, Prasher, Nanchahal and Moore2015).

We aimed to find out if a trial in those with MCI and untreated hearing loss would be able to recruit and retain participants. We also looked at whether standard UK NHS pathways for hearing loss would be sufficient to ensure that potential participants have hearing aids prescribed within 3 months and that these are maintained long term.

We recruited 17 people aged ≥60 years with ICD-10 diagnosis of MCI, from an NHS memory clinic in London, over 3 months. Twelve participants (12/17 = 71% [95% confidence interval (CI) 49–92%]) had hearing loss after a home-based assessment with a positive HearCheck screening result and confirmed in all cases as at least mild hearing loss by an audiologist though portable audiometry (pure tone average (0.5–4 kHz) > 25 dB HL). We sent the audiometric results to the participant’s general practitioner (GP), with advice to refer to NHS audiological services (which have a 12-week target for fitting hearing aids) if appropriate. At 12 weeks follow-up, three participants had been referred by their GP for NHS audiological assessment (25% [95% CI: 1–49%]). Of those, two received hearing aids (2/12; 17% [95% 1–38%]). At 2 years follow-up, one out of eight contactable participants was still using their hearing aids (12% [95% CI: 1–35%]).

Our study found that a high proportion of MCI participants attending an NHS memory clinic had untreated hearing loss. However, we also found that relying on the standard NHS audiological pathway, even after providing primary care with an audiological assessment consistent with hearing loss, resulted in low treatment uptake and low long-term adherence in people with MCI. This suggests that a trial of hearing aids in MCI could recruit participants, but that the usual audiological pathways would not be sufficient to find out whether they reduced cognitive loss. This has informed the Treating Auditory Impairment and Cognition pilot trial’s intervention to include a home-based hearing assessment and hearing aid fitting, with a study audiologist and a research assistant providing motivational and practical advice and support over 3 months (TACT; https://www.ucl.ac.uk/psychiatry/tact-trial).