Skip to content
What Causes What
MECHANISM? CONDITIONAL

Hearing loss → Cognitive performance

Higher Hearing loss moves Cognitive performance only in some conditions. Read the mechanism.

In 977 older adults with untreated hearing loss, three years of hearing aids did not slow a global cognition score versus health education; a prespecified ARIC slice of 238 people at higher risk did.

Randomized trial · Contested · N=977

Supported by an experiment or a tightly identified design. Read the caveats.

Mechanism

Lin, Pike, Albert and colleagues assigned 977 adults aged 70-84 with untreated hearing loss and no substantial cognitive impairment to hearing aids plus counselling or a health-education control (NCT03243422). Four US sites. Two recruitment streams: 238 from ARIC, 739 healthy volunteers. Mean age 76.8. The primary 3-year change in a global cognition factor score was not different: −0.200 versus −0.202, difference 0.002 (−0.077 to 0.081), p=0.96. A prespecified split by recruitment source was. In ARIC, hearing intervention cut 3-year cognitive change 48 percent versus control (difference 0.191, 0.022 to 0.360). In the volunteer stream the difference was −0.061 (−0.151 to 0.028). Control participants in ARIC declined 2.7 times faster than volunteers. The signed object is untreated hearing. Aiding hearing is less hearing loss. The total-cohort arrow is a null. The ARIC arrow is a slowdown. Yu, Proctor, Soni and colleagues pooled 50 cohorts (1,548,754 people). Adult-onset hearing loss tracked incident dementia (HR 1.35), MCI (1.29), and unspecified cognitive decline (1.29). Each 10 dB tracked a 16 percent rise in dementia risk. Vascular dementia did not. That association does not raise the stamp. Grenier, Berr, Goldberg and colleagues scored audiometry and a five-test battery in 62,072 CONSTANCES adults aged 45-69. Mild and disabling loss tracked global impairment (OR 1.10 and 1.24). Hearing-aid use versus unaided disabling loss did not (OR 0.94). Cross-section. It does not raise the stamp.

Caveats

ACHIEVE is unmasked. The control is health education, not a sham aid. The paper names 977 people. The main contrast is the total-cohort 3-year score, which missed. The 48 percent figure lives in 238 ARIC participants, a prespecified sensitivity, not the primary. Do not read ARIC as the main sign. The volunteer stream was younger, healthier, and slower to decline. Practice effects on first-time testing are an open story. 88 percent White. Four US sites. Cognitive impairment as a time-to-event endpoint was not different. Yu is cohort association. Residual confounding by vascular disease, education, and who gets an audiogram remains open. Grenier is one-time audiometry in midlife, not a trial of treatment. Blood lead and ferritin are other biology. This page is a hearing score, not age. This is a journal entry, not medical advice and not a reason to buy a hearing aid.

Effect

qualitative. ACHIEVE ITT n=977: 3-y global SF −0.200 vs −0.202, diff 0.002 (−0.077 to 0.081), p=0.96. ARIC n=238: 48% less change, diff 0.191 (0.022 to 0.360). De novo n=739 ns.

Nodes

Biology & endocrinologyHearing lossPsychology & behaviorCognitive performance

Named confounders

Cite this page

What Causes What. “Hearing loss → Cognitive performance.” https://whatcauseswhat.org/edges/e-hearing-cognition (atlas updated 2026-09-02).

https://whatcauseswhat.org/edges/e-hearing-cognition

Papers