Hearing aids for mild to moderate hearing loss in adults

This is an update of an existing review. The main clinical intervention for mild to moderate hearing loss is the provision of hearing aids, which amplify speech in addition to environmental sounds. Hearing aids are routinely offered and fitted to those who seek help for hearing difficulties. Since the previous 2017 review, further randomised-controlled trials (RCTs) have been published. This review updates the certainty of the evidence and assesses two additional outcomes: mental health and cognition.

To evaluate the benefits and harms of hearing aids in adults with mild to moderate hearing loss.

Cochrane Information Specialists searched CENTRAL, Cochrane ENT registry, MEDLINE, Embase, Web of Science, ClinicalTrials.gov, ICTRP and additional sources for published and unpublished trials. The date of the final search was 2 August 2024. We have also included published existing and ongoing studies up to 18 May 2026.

We included RCTs and cross-over trials that investigated the effect of acoustic hearing aids on adults with mild to moderate hearing loss. We did not include cluster-RCTs. We included studies where the control comparison was passive (waiting list, no intervention) or active (e.g. placebo hearing aids, education programmes, assistive listening devices, auditory training). We excluded studies with interventions delivered in a group setting.

The critical outcomes were hearing-specific health-related quality of life (participation as the key domain) and the adverse effect, pain. Important outcomes were health-related quality of life, listening ability, mental health (loneliness as the prioritised subdomain; other subdomains were depression, anxiety, and social isolation), cognition (working memory as the prioritised subdomain; other subdomains were memory, immediate attention span, set shifting, inhibition, language, processing speed, visuospatial function, and brief measures of global cognitive performance), and the adverse effect, noise-induced hearing loss. We show only prioritised outcomes in the synthesis of results section below and in the summary of findings.

We used the Cochrane tool for assessing risk of bias (RoB 1), and assigned a judgement about the risk of bias for each included study.

We synthesised results for each outcome using random-effects meta-analyses of the standardised mean differences (SMDs). Where this was not possible, we synthesised results using a narrative approach. We used GRADE to assess the certainty of evidence.

We included 16 RCTs involving 2261 participants; we added 11 new studies to this update. The studies were conducted in the USA, Europe, Brazil, Hong Kong, and Australia, and were published between 1987 and 2025. Participants had mild to moderate hearing loss. The average age ranged from 58 to 83 years. Study duration was 4 weeks to 3 years.

All studies except one had high or unclear risk for performance and detection bias because blinding was inadequate or absent. Most studies had low risk for selection, attrition, and reporting bias. For hearing-specific health-related quality of life (participation), there is likely to be a large difference favouring those wearing hearing aids over the control comparison (SMD -1.25, 95% confidence interval (CI) -1.63 to -0.87; 8 studies, 1683 participants; moderate-certainty evidence, downgraded for risk of bias). The evidence for pain is very uncertain. Six studies (1186 participants) monitored adverse effects. Of these, one study reported two instances of pain or discomfort (one participant stopped using hearing aids due to pain while wearing them; one stopped using hearing aids due to ear dryness requiring treatment). Certainty of evidence was very low, downgraded for risk of bias and extreme imprecision. For health-related quality of life, there may be a small difference favouring those wearing hearing aids over the control comparison (SMD -0.27, 95% CI -0.46 to -0.09; 4 studies, 1558 participants; low-certainty evidence, downgraded for risk of bias and indirectness). For listening ability, there is likely to be a large difference favouring those wearing hearing aids over the control comparison (SMD -1.28, 95% CI -2.41 to -0.15; 5 studies, 622 participants; moderate-certainty evidence, downgraded for risk of bias). For loneliness, there may be little to no difference between those wearing hearing aids and the control comparison (SMD -0.12, 95% CI -0.25 to 0.01; 2 studies, 907 participants; low-certainty evidence, downgraded for risk of bias and indirectness). For working memory, the evidence is very uncertain about the effect of hearing aids compared to the control (SMD 0.51, 95% CI -0.21 to 1.23; 3 studies, 910 participants; very low-certainty evidence, downgraded for risk of bias, indirectness and inconsistency). Of the six studies that monitored for adverse effects, none reported noise-induced hearing loss (1186 participants; evidence certainty not graded).

The available evidence suggests that hearing aids likely improve hearing-specific health-related quality of life and listening ability in adults with mild to moderate hearing loss, with a large beneficial effect. Hearing aids may also improve general health-related quality of life. This evidence is compatible with the widespread provision of hearing aids as first-line clinical management in those who seek help for hearing difficulties. The review did not provide evidence that hearing aids were effective at improving loneliness or working memory, nor that hearing aids resulted in adverse effects. To improve the certainty of evidence and ascertain whether the effects of hearing aids vary according to demographics (e.g. age, sex, degree of hearing loss, socioeconomic environment, race and ethnicity), greater consistency is needed in outcome measures used. Longer-term, placebo-controlled studies may be more sensitive to potential effects on mental health and cognition, but it may not be ethically justifiable to withhold hearing aids long term.

The review had no dedicated funding.

Registration (2015) PROSPERO: CRD42016043834 Protocol (2015) DOI: 10.1002/14651858.CD012023 Original review (2017) DOI: 10.1002/14651858.CD012023.pub2.