The headline reason: hearing loss is the biggest fixable contributor to dementia risk we know of in mid-life, and an audiogram is the gateway to treating it. The test itself is cheap, painless, often covered by insurance, and takes under an hour every few years. It is also the only way a noise-exposed worker keeps a future hearing-loss claim alive, and the only way someone about to start cisplatin or aminoglycoside antibiotics knows what their ears could afford to lose. The catch is honest: a hearing test changes nothing on its own — it works because most people who learn they have a loss act on it, and the ~70% who don't get nothing.
An audiogram is three tests stitched together. The first is pure-tone audiometry: in a soundproof booth, the audiologist plays beeps at each of six pitches — roughly the low rumble of a foghorn up to the highest note a child can hear — once through headphones, then again through a small vibrator pressed against the bone behind your ear. Headphones test the whole hearing pathway; the bone vibrator skips your eardrum and middle ear and goes straight to the inner ear. If the two numbers match, any loss is in the inner ear or nerve. If the headphone number is worse, something is blocking the outer or middle ear — wax, fluid, a stiffened bone — and that block is usually fixable.
The second is speech testing. The audiologist reads or plays you words at decreasing volume and asks you to repeat them; the lowest volume at which you can still get the easy two-syllable ones half the time is your speech-reception threshold. Then they read single words at a comfortable volume and score how many you catch. This is the part of the test that captures the thing you actually care about — whether you can follow a conversation — and the number that comes out is the one a hearing aid is trying to move. It has one blind spot worth knowing: some people clear every threshold and still can't follow speech in a noisy room — hidden hearing loss — which only surfaces on a dedicated speech-in-noise test you have to ask for.
The third is tympanometry: a soft plug in the ear canal that puffs air and measures how your eardrum moves. It's the quickest piece, takes about 30 seconds per ear, and rules in or out fluid, infection, or eardrum perforation in a way the beep test can't.
The result is plotted on the actual audiogram — the chart with frequency across the bottom and loudness down the side, with each ear marked separately. The closer your line is to the top of the chart, the better; the steeper it slopes downward as you move right, the more high-frequency loss you have. The classic age-related pattern — presbycusis — is a gentle slide from left to right that starts in your fifties and steepens through your seventies. The classic noise-damage pattern is a notch carved out at 4,000 Hz — exactly where rifle shots, power tools, and concerts dump their energy Cunningham & Tucci 2017.
What the test actually changes
The dementia link is the biggest single reason this entry exists. The 2024 update of the Lancet Commission on dementia prevention named hearing loss the largest single changeable risk factor in mid-life — bigger than smoking, bigger than blood pressure, bigger than physical inactivity — and credited it with about 7% of all dementia cases worldwide Livingston et al. 2024. The original signal came from a 12-year follow-up of nearly 640 adults in Baltimore: every 10-decibel worsening of hearing thresholds — roughly the gap between "normal" and "noticing trouble in restaurants" — raised the risk of developing dementia by about a quarter Lin et al. 2011.
For years the obvious counter-question was whether treating the hearing loss did anything to the risk, or whether bad hearing and bad memory just travelled together. The ACHIEVE trial settled that, partly.
Two other downstream payoffs are smaller but real. Adults with untreated hearing loss are more likely to be depressed than adults of the same age who hear normally — a pattern that holds across 35 studies and ~147,000 people, mostly driven by social isolation Lawrence et al. 2020. And in the workplace: pure-tone audiometry has been the legally recognised reference test for noise-induced hearing loss for over forty years, and an employee without a baseline on file is, in practice, an employee without a hearing-loss compensation claim 29 CFR 1910.95.
The honest skeptic position: the US Preventive Services Task Force looked at the same evidence in 2021 and declined to recommend asymptomatic screening for adults 50+, calling the evidence "insufficient" — not because the test is bad but because the chain from "we found a mild loss" to "the patient bought a hearing aid and used it" breaks at roughly the rate of seven in ten USPSTF 2021. The Task Force is not arguing the test doesn't work. It is arguing that most people who learn they have a mild loss do nothing about it, which makes mass screening in primary care a poor use of public-health dollars. A reader who would actually act on the result is in a different position from the population the Task Force was rating.
When to get tested, and how often
The default adult schedule, drawn from professional audiology recommendations rather than primary-care screening rules: one baseline test by age 50, repeats every three years after 50, and every year after 70 or whenever something changes — a new ringing in one ear, a friend repeatedly asking why you have the TV so loud, a sense that restaurants have gotten harder. People with risk factors get on the schedule earlier: regular noise exposure (occupational, recreational, military), family history of early hearing loss, ototoxic medication on the horizon, diabetes, or a cardiovascular event.
What happens in the room: you sit in a small sound-treated booth, wearing headphones or foam earbuds. The audiologist plays beeps at one pitch and one volume at a time and asks you to press a button (or raise your hand) every time you hear one. They walk down in volume until you stop hearing the beep, then back up; that crossover is your threshold for that pitch in that ear. They repeat across six pitches per ear with headphones, then again with the bone vibrator behind your ear, then the speech words, then the air-puff tympanometry. Total time: 30–60 minutes. No needles, no medication, no recovery period.
Cost in the US is roughly $50–250 out of pocket at a private audiology clinic if you pay cash. Medicare Part B covers the diagnostic audiogram when a doctor orders it for a medical reason — including any of the changes listed above — with the usual 20% co-insurance after the annual deductible. Since 2023, Medicare also allows you to see an audiologist directly once a year for diagnostic hearing testing without a doctor's order. Commercial insurance is similar; most plans cover an ordered audiogram. Hearing aids themselves are a separate cost question — see the practicalities section.
If you work around noise
Different rules apply to anyone whose job exposes them to an average of 85 decibels averaged over an eight-hour shift — roughly the level of a busy city street, a running lawnmower, or a forklift. Construction, manufacturing, agriculture, oil and gas, music venues, military service: all in. Under US occupational safety law, the employer must arrange a baseline audiogram within six months of your first day on the job, repeat it every year for as long as you stay in the noisy job, and pay for both tests. The baseline has to be preceded by at least 14 hours of quiet — meaning either a day off or solid hearing protection in the run-up — so it captures your real hearing, not the temporary dip after a shift 29 CFR 1910.95.
The number the employer is watching is the standard threshold shift: a 10-decibel drop from your baseline, averaged across the three pitches noise damages first (2, 3, and 4 kHz), in either ear. Cross that line and your employer must notify you within 21 days, refit your hearing protection, retrain you, and in many cases log the case on the federal OSHA injury record.
Loud hobbies follow the same biology even if no employer is involved. Sustained recreational noise — gunfire without ear protection, motorcycle commuting, music played through earbuds above 80% volume for hours daily, regular concert attendance without plugs — produces the same 4-kHz notch on the audiogram as a factory job. If that describes you, run the standard adult cadence above and add a baseline now rather than at 50.
From audiogram to hearing aid
An audiogram is a measurement, not a fix. What it does — for the consequence most people in the catalogue's reader range care about — is tell you whether you qualify for a hearing aid and what kind.
The standard severity bands are simple. Average thresholds across the speech frequencies of 0–25 decibels count as normal. 26–40 is mild loss — the band where consonants start dropping in noisy rooms but conversation in a quiet kitchen is still fine. 41–55 is moderate — TV volume creeps up, group conversation gets hard. 56–70 is moderately severe; 71–90 is severe; over 90 is profound. Mild and moderate losses are hearing-aid territory; severe and profound losses get evaluated for a cochlear implant, a surgically placed device that bypasses damaged inner-ear hair cells entirely.
Until 2022 the only path to an amplification device in the US was an audiologist visit, a custom fitting, and a $4,000–6,000 pair of devices, usually not covered by Medicare. In October 2022 the FDA created a new category of over-the-counter hearing aids for adults 18 and over with perceived mild to moderate loss FDA 2022. These sell for $200–1,500 a pair, at pharmacies and online, with no prescription. The audiogram is still useful: it tells you whether you're in the mild-to-moderate range OTC devices are designed for, and rules out the conductive or asymmetric findings that need an ENT instead of an amplifier.
What people get wrong
"I'd know if I were losing my hearing." The most common belief, and the wrongest one. Age-related hearing loss starts at the high frequencies — the pitches where consonants live — and creeps down toward the speech range over years. The world doesn't get quieter; it gets mumbly. People with measurable loss describe it as "everyone mumbles," "restaurants are impossible," "my wife needs to face me when she talks" — never as "I can't hear." The average gap between qualifying for a hearing aid and owning one is nearly nine years, and most of that gap is people not knowing they qualify.
"Hearing aids are for old people." Two things have shifted in the last few years. Devices are now small, often invisible at conversational distance, and Bluetooth-connected to your phone the way wireless earbuds are. And the FDA's 2022 rule means a 40-year-old with mild loss from years of headphones-on-the-subway can buy a $300 device at a pharmacy without ever seeing a doctor FDA 2022. The stigma is doing real damage at this point — the strongest evidence on dementia delay involves people who actually wear the devices for years Lin et al. 2023.
"My doctor checked my hearing at my last physical." Almost never. The "whisper test" or rubbing fingers next to your ear catches severe loss only and misses mild and moderate loss almost entirely. The same applies to single-question screens ("are you having trouble with your hearing?") — they catch the people who already know. A real audiogram is a different test in a different room with different equipment.
"Screening is settled." It isn't. The US Preventive Services Task Force has not been able to recommend routine asymptomatic screening of adults 50+ because the chain from finding a mild loss to actually treating it breaks at population scale USPSTF 2021. Audiology and otolaryngology professional bodies disagree and recommend baseline plus periodic testing. Both sides agree the test itself is solid; they disagree about whether telling everyone to get it produces population-level benefit. For an individual reader who would actually act on a finding, this debate doesn't change the calculus much.
What happens if you keep skipping it
In the short run, nothing. Untreated mild hearing loss feels like normal aging — restaurants get harder, group dinners get tiring, you turn the TV up a notch, your partner repeats themselves twice a day. Nobody around you sees a problem yet. You don't either.
Over the next decade, that floor moves. You stop accepting invitations to noisy places because they're exhausting. Phone calls get shorter. Your circle gets smaller in ways you don't quite track to the hearing — it feels like everyone got busier. Your spouse starts answering for you in conversations because by the time you've parsed what was asked, the moment has passed. People you used to enjoy being around start telling someone else first. By the time you decide to do something about it, the social shape of your life has already reorganised around the loss Lawrence et al. 2020.
Over the decade after that, the cognitive bill comes due. The same midlife untreated hearing loss that made dinners exhausting in your sixties is — in the strongest current model — the largest single thing under your control that raises your dementia risk in your seventies Livingston et al. 2024. Not because hearing loss directly damages the brain, but because parsing speech when the signal is degraded is constant cognitive work, because withdrawing socially removes the daily novelty the brain needs, and because the cells in your inner ear and the cells in your hippocampus respond to a lot of the same vascular and metabolic insults. Adults who treated their hearing loss in the ACHIEVE trial slowed their three-year cognitive decline by nearly half — but only the ones who started at higher risk, and only after the devices were actually fitted and used Lin et al. 2023.
For noise-exposed workers there's a second timeline running in parallel. Each year of unprotected high-volume work without a baseline on file is a year of hearing damage that can't be claimed, can't be compensated, and can't be tracked back to the employer. By the time the loss is severe enough that you notice, the documentation needed to do anything about it doesn't exist 29 CFR 1910.95.
What changes if you treat it
Within a few weeks of fitting hearing aids — the practical outcome of an audiogram that finds a treatable loss — most people report the same set of small shifts. Restaurants stop being adversarial. You stop saying "what?" reflexively. The TV goes down two notches and your partner stops repeating themselves. The afternoon fatigue that you assumed was age turns out to be partly the cost of constant listening effort, and it eases off Lawrence et al. 2020.
Within a few months, the social shape starts to repair. The neighbours' barbecue stops being a thing you dread. You take phone calls you'd been avoiding. People who'd quietly stopped including you in plans don't even notice they've started again. The repair is partial — you don't get the years back — but the direction reverses.
Within a few years, in the higher-risk adults the ACHIEVE trial recruited, the cognitive trajectory measurably bends. Three-year cognitive decline cut by almost half compared with the control group — the only RCT-grade evidence for a behavioural intervention with that effect size in older adults at risk for dementia Lin et al. 2023. The effect was real in the high-risk cohort and not detectable in the lower-risk one, which is honest: if you're a healthy 70-year-old with no other dementia risk factors, treating mild hearing loss is unlikely to be the single thing that changes your trajectory. If you've already got cardiovascular disease, diabetes, lower education, or a family history, it might be.
For the noise-exposed worker, the payoff is more linear and shows up earlier: every standard-threshold-shift detection on the annual audiogram is a moment your employer is legally required to refit your protection, retrain you, and re-evaluate the noise source — a brake that gets pulled before the next year of damage 29 CFR 1910.95. For the chemotherapy patient, an audiogram during treatment that shows a meaningful shift can prompt an oncologist to drop the cisplatin dose or switch to a less-ototoxic agent — preserving hearing that, once lost to platinum, doesn't come back Kohrman et al. 2020.
How this goes wrong in practice
The most common failure: getting a clean audiogram, learning you have mild loss, and doing nothing. This is the failure mode the US Preventive Services Task Force was worried about — and it's also the reason the dementia-prevention payoff is conditional. A filed report doesn't slow decline; a worn device does. If you're not the kind of person who'd buy and wear amplification on a mild-loss finding, the test buys you a baseline and not much else right now.
The next most common: a primary-care office "checks your hearing" as part of an annual physical, finds nothing, and you take that as a real screen. It almost never is — a whisper test or a single screening question catches severe loss only and routinely misses mild and moderate loss. If you want an actual audiogram, it happens in a soundproof booth with calibrated equipment, run by an audiologist; it's not a five-second part of a physical.
The third: an audiogram surfaces a meaningful difference between the two ears, and nobody chases it. Asymmetric sensorineural loss — one ear measurably worse than the other on bone-conduction testing — is the audiogram's strongest reason to order an MRI of the inner-ear canal, looking for a vestibular schwannoma. Caught small, the tumour is monitored or treated with focused radiation and your hearing is preserved. Caught large, surgery becomes the option and hearing in that ear usually goes. Don't let an asymmetric reading sit in a file.
The fourth, specific to noise-exposed workers: a "baseline" taken without the required 14-hour quiet period in front of it. That baseline records your already-fatigued hearing as your normal — and from then on, every annual test compares against that lowered bar, masking real shifts. If your first day on a noisy job included an audiogram in the on-site occupational-health office that afternoon, ask for a repeat under proper conditions 29 CFR 1910.95.
Adjacent topics
Hearing-aid selection and fitting are downstream of the audiogram and a separate decision. Tinnitus — persistent ringing — often shows up alongside hearing loss and gets its own evaluation. Vestibular testing for balance disorders uses similar equipment but answers a different question. Newborn hearing screening is universal in the US but unrelated to the adult cadence here. Hearing protection — earplugs, earmuffs, in-ear monitors for musicians — is the upstream prevention story for noise-induced loss.
- — An audiogram is how age-related hearing loss gets measured and tracked.
- — If you're starting a known ear-toxic drug, a before-and-after audiogram is the test that catches harm early.
- — Ringing in your ears? The hearing test is step one, and it almost always finds the hidden loss feeding the noise.
- — If the test shows mild-to-moderate loss, you can now buy aids over the counter without a clinic fitting.
- — A normal audiogram doesn't rule out hidden hearing loss — ask for a speech-in-noise test too.
- — Hearing screening is the highest-value mid-life addition to a preventive checkup routine.
- — Untreated hearing loss is the single biggest fixable dementia risk — which matters most if you carry the gene that raises it.
- — If an audiogram shows early loss, hearing protection is the intervention that keeps the next test from looking worse.
Substance and claimed effects
The adult audiogram is a battery of audiometric tests — air-conduction pure tones from 250–8000 Hz, bone-conduction pure tones over the same range with masking when needed, speech-reception threshold (SRT) and word-recognition score (WRS), and tympanometry — that quantifies hearing thresholds, classifies any loss as conductive, sensorineural, or mixed, and grades severity from normal (≤25 dB HL) through profound (>90 dB HL) Cunningham & Tucci 2017. The entry covers the test itself plus the consequences the brief names: baseline timing and follow-up cadence; how an audiogram gates hearing-aid and cochlear-implant candidacy; its role in monitoring cognitive risk in mid- and late-life; and its mandated use in occupational hearing-conservation programmes. The downstream effects scored in meta — modest energy and focus gains (reducing listening effort once loss is treated), meaningful mood and longevity effects (depression risk, dementia-risk reduction), and a real but conditional health-quality-of-life lift — flow from the test enabling those interventions, not from the test in isolation.
Evidence by addressing question
mechanism
Pure-tone audiometry measures the minimum sound-pressure level a listener detects at fixed frequencies in each ear, plotted as decibels Hearing Level (dB HL) versus frequency. Air conduction (via supra-aural or insert phones) tests the entire pathway: outer canal, tympanic membrane, ossicles, cochlea, eighth nerve, central auditory pathway. Bone conduction (via a mastoid oscillator) bypasses the outer and middle ear and tests cochlea + neural pathway only. A gap of >10 dB between air and bone thresholds at a given frequency — the air–bone gap — localises the lesion to the conductive system (cerumen impaction, middle-ear effusion, otosclerosis, ossicular discontinuity); equally depressed air and bone thresholds point to sensorineural pathology (cochlear hair-cell or auditory-nerve damage) Cunningham & Tucci 2017.
Speech audiometry quantifies functional hearing. The speech-reception threshold is the lowest dB at which the patient repeats two-syllable spondee words 50% of the time; it should agree with the pure-tone average (PTA) at 500/1000/2000 Hz within ~10 dB — disagreement flags non-organic loss or retrocochlear pathology. The word-recognition score presents monosyllables at a comfortable level above SRT; scores that are worse than the audiometric pure-tone loss would predict (so-called "rollover") are a sign of retrocochlear disease, classically a vestibular schwannoma. Tympanometry plots middle-ear admittance against pressure: a Type-A trace is normal, Type-B (flat) indicates fluid or perforation, Type-C (peak shifted negative) indicates eustachian-tube dysfunction.
The mechanistic story behind why audiograms matter as a periodic test rests on the slow, asymmetric biology of presbycusis: hair-cell loss in the basal cochlea begins in the third decade, accumulates at roughly 1 dB/year at 4–8 kHz, and produces a downward-sloping high-frequency sensorineural pattern that the listener does not detect until consonant discrimination collapses around the moderate threshold. By that point the person has typically lost 40 dB at the most critical speech frequencies and waited an average of ~9 years before action. A baseline audiogram in mid-life and serial repeats catch the slope long before the subjective failure.
evidence
Diagnostic accuracy. Pure-tone audiometry is the reference standard for hearing-loss detection; all other methods (whispered voice, finger rub, single-question screens, app-based tests) are evaluated against it USPSTF 2021. Test–retest reliability under standard ANSI/ISO booth conditions is ~5 dB; clinically meaningful change is conventionally a 10-dB shift averaged across three speech frequencies (the threshold OSHA uses for a Standard Threshold Shift) 29 CFR 1910.95.
Prevalence and undertreatment. NHANES 2011–2020 data put audiometric hearing loss at ~22% of US adults overall, ~44% of those aged 70–79, and ~65% of those over 70 Humes 2023. Fewer than one in three candidates uses a hearing aid; mean delay from candidacy to first device is ~9 years.
Dementia risk. The Lancet Commission's 2024 update lists hearing loss as the single largest modifiable mid-life dementia risk factor, attributing 7% of population-level cases to it — more than any other factor at that life-stage Livingston et al. 2024. A pooled meta-analysis estimated a 37% relative increase in incident dementia for adults with audiometric hearing loss versus normal hearing; the original signal came from Lin's 12-year follow-up of the Baltimore Longitudinal Study of Aging cohort (HR 1.27 per 10-dB worsening of better-ear PTA) Lin et al. 2011. The ACHIEVE trial — the only large RCT of treatment — randomised 977 adults aged 70–84 with untreated hearing loss to a structured audiology intervention (audiogram + hearing aids + counselling) versus health-education control and found no overall cognitive benefit in the full cohort over three years, but a 48% slowing of cognitive decline in the pre-specified ARIC sub-cohort at higher dementia risk Lin et al. 2023. The Commission update reads this as the strongest causal evidence yet that treating hearing loss reduces dementia incidence — at least in the population that needs it most Livingston et al. 2024; a 2025 secondary analysis of ACHIEVE further confirmed the high-baseline-risk effect Reed et al. 2025.
Mood. A 35-study meta-analysis of ~147,000 adults found audiometric hearing loss associated with higher prevalence of clinically significant depressive symptoms; effect sizes are modest but reproducible and reverse partially with amplification Lawrence et al. 2020.
Occupational evidence. OSHA's hearing-conservation rule rests on decades of cohort data linking 8-hour time-weighted-average (TWA) exposures ≥85 dBA to permanent threshold shifts at 3, 4, and 6 kHz — the classic "4-kHz notch." Baseline audiograms within six months of first exposure and annual repeats are the legal mechanism for detecting and acting on a Standard Threshold Shift (a 10-dB worsening averaged at 2, 3, 4 kHz from baseline in either ear) 29 CFR 1910.95.
practice
USPSTF (2021): "I" statement — insufficient evidence to recommend for or against routine screening of asymptomatic adults aged 50+ via primary-care channels. The "I" turns on screening strategy and downstream uptake, not on the test's accuracy USPSTF 2021.
American Academy of Family Physicians: baseline audiogram in adulthood, repeat every decade, every three years after 50, more often with risk factors (noise exposure, family history, ototoxic drugs, comorbid cardiovascular disease).
American Academy of Audiology / American Speech-Language-Hearing Association: baseline at 50, periodic thereafter; immediate audiogram for any asymmetric or sudden loss, tinnitus, ear pain or fullness, or before ototoxic chemotherapy.
OSHA (29 CFR 1910.95): baseline audiogram within six months of first occupational exposure ≥85 dBA TWA, preceded by ≥14 hours of quiet (or hearing-protection use); annual repeats; mandatory employer response on STS — employee notification within 21 days, hearing-protector refit and retraining, possible OSHA 300-log recordable case 29 CFR 1910.95.
Ototoxic drug protocols: American Academy of Audiology position statement calls for baseline audiogram before cisplatin/carboplatin or aminoglycoside courses, monitoring every cycle or weekly during therapy, and a follow-up at 3 and 6 months post-treatment. Detection of a meaningful shift can trigger dose reduction or drug substitution where therapeutically possible Kohrman et al. 2020.
protocol
The default adult schedule: a baseline audiogram around age 50 (earlier with risk factors), then every 3 years after 50 and yearly after 70 or after any subjective change. People with occupational noise exposure ≥85 dBA TWA are on the OSHA cadence (baseline within 6 months of hire, annual thereafter) regardless of age. Pre-chemotherapy, pre-aminoglycoside, and pre-loop-diuretic-course audiograms are added on a per-treatment basis.
The standard adult test is performed by a licensed audiologist in a sound-treated booth, runs 30–60 minutes, and consists of: (1) otoscopy, (2) tympanometry + acoustic reflexes, (3) air-conduction pure tones 250–8000 Hz each ear, (4) bone-conduction pure tones with masking as indicated, (5) SRT and WRS bilaterally. Extended high-frequency testing (9–16 kHz) is offered at some clinics for early noise-exposure detection.
contraindications
None. The test is a passive measurement at safe sound levels; no medical contraindication exists. The relevant cautions are practical: active ear infection, cerumen impaction, or recent loud-noise exposure (≥14 hours of quiet recommended before a baseline test for occupational purposes) can confound results and warrant rescheduling or repeat. Behavioral testing is unreliable in advanced cognitive impairment; auditory brainstem response testing is the workaround.
misconceptions
Three recurring errors. (1) "I'd know if my hearing was failing." Presbycusis erodes the high frequencies — consonants — slowly; the listener experiences "mumbling" or "noisy restaurants," not absence. The self-perception lags audiometric loss by years. (2) "Hearing aids are for old people / not worth the money." The average candidate-to-uptake delay is ~9 years; cost ($1,500–8,000 a pair pre-OTC) and stigma drive the gap. The FDA's 2022 OTC ruling created a $200–1,500 self-fit category for adults 18+ with perceived mild-to-moderate loss, sidestepping both barriers FDA 2022. (3) "Screening is settled." USPSTF disagrees with most professional audiology bodies on whether asymptomatic primary-care screening helps — this is real expert disagreement, not paperwork.
failure-modes
Common ways a periodic audiogram fails to deliver value: a baseline taken without the OSHA-required 14-hour quiet period overstates the worker's true threshold and weaponises the document against the employer in compensation claims; a primary-care "whisper test" or single-question screen substitutes for and replaces a real audiogram, missing 30–50% of cases compared with audiometry; a finding of asymmetric SNHL is logged but not followed with gadolinium-enhanced MRI, missing vestibular schwannoma; mild loss is detected but the patient declines amplification on cost or stigma grounds and the audiogram becomes a filed report rather than an intervention gate.
stakes
The stakes of skipping periodic audiograms are mostly invisible until late. Untreated mid-life hearing loss carries the largest single modifiable dementia population-attributable fraction in the Lancet model — 7% of cases Livingston et al. 2024. It elevates depression risk meaningfully Lawrence et al. 2020, accelerates social withdrawal, and predicts faster cognitive decline in observational cohorts Lin et al. 2011. For noise-exposed workers without a baseline, an occupational claim becomes nearly impossible to win 29 CFR 1910.95. None of these costs is reversible at the point the listener finally seeks help.
payoff
The payoff is conditional: an audiogram by itself improves nothing; it makes intervention possible. The ACHIEVE trial established that fitting hearing aids in adults at higher dementia risk slows three-year cognitive decline by ~48% relative to control Lin et al. 2023. Smaller-scale evidence shows reduced depressive symptoms Lawrence et al. 2020, restored social participation, and reduced listening fatigue. The benefit lands earlier in noise-exposed workers (STS detection prompts protective action before threshold shift becomes permanent) and earlier in chemotherapy patients (audiometric monitoring can trigger dose changes that preserve hearing).
out-of-scope
Tinnitus management, vestibular testing, paediatric newborn hearing screening, and the auditory brainstem response are adjacent but separate. Hearing-aid selection, fitting, and follow-up are downstream of the audiogram and warrant their own entry.
The credibility range
Optimist case
The audiogram is the cheapest, most evidence-backed diagnostic in the catalogue's adjacency. The Lancet Commission identifies hearing loss as the largest modifiable mid-life dementia risk factor; ACHIEVE provides the first RCT-level signal that treating it slows decline in those at risk Lin et al. 2023Livingston et al. 2024. The test takes 30 minutes, costs $50–250 OOP (or is Medicare Part B covered when ordered for diagnostic purposes), carries zero harm, and is mandated by OSHA and recommended by every audiology body for adults 50+. A baseline that future audiograms can compare against — analogous to a baseline lipid panel or DEXA scan — is unambiguously high-yield, especially when occupational noise or ototoxic medication is on the horizon.
Skeptic case
The USPSTF reviewed the same evidence and landed on "I" — insufficient evidence to recommend routine asymptomatic screening — because the chain from "asymptomatic primary-care audiometry" to "patient-relevant outcome" depends on a treatment uptake step that fails 70%+ of the time USPSTF 2021. ACHIEVE's primary outcome was null in the overall cohort; the 48% benefit was in a pre-specified sub-group and survives subgroup-analysis scrutiny only because it was pre-specified, not exploratory Lin et al. 2023. The depression association is small Lawrence et al. 2020. The cascade — audiogram → diagnosis → hearing aid → daily use → cognitive benefit — has at least three high-attrition steps. Mass screening of asymptomatic adults may identify many mild losses whose owners decline intervention and gain nothing.
Author's call
The USPSTF "I" is about mass primary-care screening as a policy, not about whether a self-directed mid-life baseline audiogram is worthwhile for an individual reader who would act on a finding. The catalogue's reader self-selects for action; for that reader the calculus is closer to the optimist case. The entry leans toward a strong recommendation — get a baseline by 50, repeat every three years — while being honest that the dementia-prevention pay-off is conditional on actually using hearing aids if needed, and that the strongest evidence is in higher-risk older adults rather than universal application. The occupational-screening and ototoxic-monitoring use cases are uncontested.
Stakeholder and incentive map
- Audiologists and ENT clinics — recommend earlier and more frequent baselines; revenue from device sales has been a longstanding tension that the FDA OTC rule explicitly aimed to disrupt FDA 2022.
- Hearing-aid manufacturers — incentive aligned with screening recommendations; pre-OTC market structure (bundled fittings) sustained $4,000–6,000 average device pricing.
- OSHA and workplace insurers — mandate baselines in part to manage employer liability for occupational hearing loss claims; without a baseline the claimant wins by default 29 CFR 1910.95.
- USPSTF and primary-care guidelines — institutionally conservative on screening when downstream-treatment uptake is poor; the "I" rating reflects screening-trial methodology more than disagreement with audiology.
- Lancet Commission on dementia — strong public-health framing of hearing loss as modifiable dementia risk, which has shifted the discourse toward earlier intervention Livingston et al. 2024.
- Medicare — covers diagnostic audiograms ordered for medical reasons but not hearing aids themselves; this funding asymmetry is the structural driver of low uptake among the highest-risk age group.
Population variability
- Age. Prevalence rises from ~5% in 18–29-year-olds to ~65% in adults 70+ Humes 2023. The baseline-at-50 recommendation reflects the inflection point of presbycusis incidence.
- Sex. Men have higher prevalence and earlier onset than women, partly explained by higher cumulative occupational and recreational noise exposure.
- Race. NHANES data show lower prevalence in Non-Hispanic Black adults than in Non-Hispanic White adults at every age — a robust but incompletely explained effect (melanin in stria vascularis is the leading mechanistic hypothesis).
- Occupational exposure. Construction, manufacturing, agriculture, military, and music industries carry order-of-magnitude higher risk; OSHA cadence supersedes the general adult schedule 29 CFR 1910.95.
- Comorbidities. Diabetes, cardiovascular disease, chronic kidney disease, and head injury history all elevate baseline hearing-loss risk and warrant earlier and more frequent testing.
- Cognitive-decline risk. ACHIEVE's effect modification suggests the cognitive benefit of treating hearing loss is strongest in adults already at elevated dementia risk (cardiovascular comorbidity, lower education, ApoE4 carriage); the test's downstream value is highest in this subgroup Reed et al. 2025.
Knowledge gaps
The single most consequential gap is whether the cognitive benefit of treating hearing loss generalises to lower-risk and younger populations (ACHIEVE's signal was confined to the high-risk sub-cohort). A second open question is whether extended high-frequency audiometry (9–16 kHz) — which detects noise damage and synaptopathy earlier than the standard battery — should be routinely added to mid-life baselines; current evidence supports it for noise-exposed workers but not yet for the general adult schedule. A third is whether the OTC hearing-aid pathway delivers comparable cognitive benefits to the audiologist-fitted devices used in ACHIEVE; no RCT has tested this yet. Finally, the natural-history data on rate of presbycusis progression are sparse for adults aged 30–50, making the optimal mid-life cadence (every 3 vs every 5 years) more convention than evidence.
Scope match against brief. The brief named baseline timing, follow-up cadence, components, hearing-aid candidacy, cognitive-risk monitoring, and occupational screening. All six are covered. The mechanism section unpacks components; protocol carries baseline timing and cadence; practicalities carries hearing-aid candidacy and the OTC pathway; evidence + stakes + payoff carry cognitive-risk monitoring; the audience section is the OSHA / noise-exposed-worker treatment of occupational screening.
Category call. Placed under hearing. screening was a close alternative — the substance is structurally a periodic screen — but hearing already has its own category, and a reader looking for "should I get a hearing test" will look there first. Flag for editor: if hearing stays sparse, consider whether some entries belong in screening with a label-based cross-link.
Rating difficulties.
longevity 3andfocus 2are the hardest calls. The audiogram itself is just a measurement; the downstream cognitive / mortality effect is gated on hearing-aid uptake, which fails ~70% of the time. The spec (entry.md§1a,meta.md§5a step 7) instructs scoring the substance + its meaningful consequences holistically, so the score reflects the conditional cascade. Both could be argued one point lower if the reviewer wants the score to reflect "test alone, treatment-naive." Flag for review.- ACHIEVE's primary outcome was null overall and only positive in the pre-specified high-risk sub-cohort. The article is honest about this — the
longevitypitch reads "biggest changeable mid-life dementia risk" (a statement about the loss, not the test), and theevidencesection reports both the null overall and the 48% subgroup effect. evidence: 4rather than 5 because the screening-vs-diagnostic distinction matters: the test itself is unimpeachable, but the case for routine screening in asymptomatic adults is still contested (USPSTF 2021 "I"). A 5 would imply universal guideline alignment, which is not the case.controversy: 2reflects the USPSTF / AAA-AAO-HNS disagreement on screening strategy, not the test itself.
Audience scoping. No audience field set on meta — the substance applies to every adult, with the OSHA carve-out and the high-risk-for-dementia subgroup handled inside the article rather than via meta scoping. The bulk of consequence concentrates in adults 50+, but the OSHA cadence and OTC eligibility start at 18.
Contraindications. None. The test is passive and harmless.
Excluded on purpose.
- Hearing-aid fitting and selection — downstream, deserves its own entry.
- Tinnitus — adjacent, common alongside loss, separate evaluation and management.
- Pediatric and newborn hearing screening — different population, different test battery, different regulatory frame.
- Auditory brainstem response (ABR) — the workaround for behavioural testing in cognitively impaired or noncompliant patients; not the substance.
- Extended high-frequency audiometry (9–16 kHz) — mentioned in the research dossier as a knowledge gap; not yet standard of care for general adults, so kept out of the protocol section.
Future-link candidates. Hearing aids (fitting, OTC vs. prescription), tinnitus, vestibular schwannoma, cochlear implants, hearing protection (earplugs, IEMs), Lancet Commission modifiable dementia risks (umbrella entry), ototoxic-medication monitoring. The article references several of these in out-of-scope and would cross-link once those entries exist.
Separate-entry candidates. Vestibular schwannoma (warrants its own short know entry given how often the red flag is missed); hearing protection during noise exposure (clear do entry, distinct from the test).
Honest hedge in highlights. The highlights paragraph names the conditional explicitly ("the ~70% who don't get nothing"). This is the spec's "high-effort entries earn their candor" rule applied to the screening-uptake reality.
Adult Audiograms
Out-of-pocket cost is $50–250 per test, often covered by Medicare Part B or commercial insurance when ordered for diagnostic reasons; the OSHA workplace audiogram is employer-paid. Repeated every 3–5 years, total cost is trivial.
A single 30–60 minute clinic appointment every few years; no preparation beyond ~14 hours of quiet for occupational baselines. Trivially low effort.
Pure-tone audiometry is the reference standard for hearing-loss detection (USPSTF 2021), regulatorily codified for occupational use (OSHA 29 CFR 1910.95), and the gateway for the only RCT-supported intervention reducing dementia incidence in higher-risk older adults (Lin et al. 2023). The 'I' rating for asymptomatic primary-care screening applies to a screening-strategy question, not to the test's diagnostic validity.
Audiograms gate the only RCT-supported pathway to reducing dementia incidence in adults at elevated risk: in the ACHIEVE trial hearing intervention slowed 3-year cognitive decline by 48% in higher-risk older adults (Lin et al. 2023). The 2024 Lancet Commission lists hearing loss as the largest modifiable mid-life dementia risk factor, attributing 7% of population cases to it (Livingston et al. 2024). Score reflects the substance + cascade, conditional on treatment uptake.
Detecting and treating a missed conductive cause (cerumen, effusion) or fitting amplification for previously-undetected loss produces a real but modest day-to-day quality-of-life lift within weeks — reduced listening fatigue, easier conversation, less mishearing. The test itself does nothing; the lift is gated on a treatable finding and uptake.
Untreated hearing loss steals working-memory bandwidth for speech parsing; ACHIEVE-style intervention measurably improves cognitive performance in the higher-risk subgroup over 3 years (Lin et al. 2023). The gain is meaningful but, like longevity and energy, gated by uptake of amplification — the audiogram is the gateway, not the lever.
A 35-study meta-analysis links audiometric hearing loss to higher rates of depression in older adults; the relationship partially mediates through social isolation and loneliness, and amplification reduces depressive symptoms in observational and small-RCT data (Lawrence et al. 2020). Effect sizes are modest.
Untreated hearing loss imposes daily listening effort that produces measurable fatigue; correcting it lifts that load. Effect is real but small and entirely dependent on fitting amplification once a loss is found — the audiogram itself contributes nothing.