How Much Hearing Loss Is Considered Deaf?

Most audiological standards classify hearing loss as “deaf” when the pure-tone average in the better ear exceeds about 70 to 90 decibels (dB), though the exact cutoff depends on who is doing the classifying and why. That number alone, however, only scratches the surface. The boundary between “hard of hearing” and “deaf” involves not just how loud sounds need to be for you to detect them, but how well you can actually understand speech, whether the loss affects one ear or both, and whether you or your community even frame deafness as a medical category at all.

The Decibel Scale and Severity Grades

Hearing ability is measured with a pure-tone audiogram, which tests how soft a sound you can detect at several different pitches. The results are averaged into a single number, the pure-tone average (PTA), expressed in decibels of hearing level (dB HL). A lower number means better hearing. Clinical research commonly breaks these averages into severity bands: normal hearing at 15 dB or below, slight loss from 16 to 25 dB, mild from 26 to 40 dB, moderate from 41 to 55 dB, moderately severe from 56 to 70 dB, and severe at 71 dB and above.¹ Some classification schemes add a “profound” category starting at 91 dB, which is where many clinicians draw the line for what they call deafness in a medical sense.

To put those numbers in perspective, a whispered conversation in a quiet room is roughly 30 dB, and normal speech at arm’s length is around 60 dB. If your PTA sits at 70 dB, you would struggle to hear someone speaking at a normal volume even in a silent room. At 90 dB and above, environmental sounds like a lawnmower or a blender might barely register, and unaided speech is essentially inaudible.

The World Health Organization uses a slightly different grading system with its own cutoffs, and some countries have adopted their own standards for disability determination. So when you see the word “deaf” in a medical report, a legal document, or a school evaluation, the threshold behind it can vary by 10 or 20 dB depending on the framework in use. The broad consensus, though, is that a PTA somewhere above 70 dB in the better ear marks the transition into what audiologists consider severe-to-profound hearing loss, the clinical territory most closely aligned with the word “deaf.”

Why the Audiogram Does Not Tell the Whole Story

A pure-tone test tells you the softest sounds a person can detect, but detecting a beep in a soundproof booth and understanding a sentence at a restaurant are wildly different tasks. Speech recognition depends on the ability to resolve fine frequency detail, so two people with the same PTA can have very different experiences with spoken language.²

Research on adults with severe hearing loss shows this gap clearly. In one study, people with severe impairment recognized single words correctly about two-thirds of the time on average, while those with severe-to-profound loss managed only about 39 percent. The researchers emphasized that cochlear implant candidacy should not rely solely on audiometric thresholds, because the audiogram alone misses how well someone functions with the hearing they have left.³ In practical terms, two people who both score 75 dB on a pure-tone test might have dramatically different abilities to follow a conversation, depending on where in the cochlea the damage sits, how the auditory nerve is functioning, and how their brain has adapted over time.

This is one reason clinicians increasingly use speech-in-noise tests and word recognition scores alongside the audiogram when making decisions about treatment. A person whose audiogram looks “merely” severe but who scores very poorly on word recognition may be functionally deaf in everyday situations, while someone with a profound audiometric loss but excellent use of residual hearing and lip-reading might manage reasonably well in quiet settings.

Medical “deaf” Versus Cultural “Deaf”

The lowercase “deaf” and the uppercase “Deaf” mean different things, and the distinction matters. Lowercase deaf is a medical description tied to audiometric thresholds. Uppercase Deaf refers to membership in a cultural and linguistic community that uses sign language and shares a distinct identity. For many Deaf people, being Deaf is not a deficit to be fixed but a way of life, and the boundary of belonging has nothing to do with a decibel number.

Research into Deaf identity consistently finds that community members frame themselves as a cultural-linguistic minority rather than as people with a disability. In a study of Deaf individuals in Turkey, participants rejected disability labels outright and pointed to their sign language as the primary marker of group membership.⁴ Similar findings appear across cultures. Deaf identity is developed through shared language, schooling, social networks, and the experience of navigating a hearing-dominant world, and it can exist alongside or in tension with medical frameworks like cochlear implantation.⁵

This means a person with a 50 dB loss who grew up signing and attending a Deaf school may identify as Deaf, while someone with a 95 dB loss who was raised oral and uses hearing aids may not. The audiogram does not determine cultural identity. If you are asking “how much hearing loss is considered deaf” to understand a medical classification, the answer is roughly 70 to 90 dB. If you are asking it to understand Deaf identity, the answer is that decibels are largely beside the point.

Single-Sided Deafness

When hearing loss is discussed in terms of severity thresholds, the assumption is usually that both ears are affected to a similar degree. But a significant number of people have normal hearing in one ear and severe or profound loss in the other, a condition called single-sided deafness (SSD). Research defines SSD as a pure-tone average of 25 dB or better in one ear and greater than 70 dB in the other.⁶

People with SSD can hear speech just fine in a quiet room, which sometimes leads others to assume they have no real problem. The difficulties show up in noisier, more complex environments. Losing input from one ear impairs your ability to localize sounds, to separate a voice from background noise, and to maintain spatial awareness.⁷ At a crowded dinner table, for instance, someone with SSD may catch almost nothing from the side of the deaf ear. The quality-of-life impact can be substantial, yet because these individuals technically have one “good” ear, they often fall through the cracks of disability classifications built around bilateral thresholds.

When Standard Tests Miss the Problem

There is also a category of hearing difficulty that does not show up on a standard audiogram at all. Hidden hearing loss is a recently described disorder in which a person has normal pure-tone thresholds but still struggles to understand speech, especially in noisy environments.⁸ The problem appears to involve damage at the synapses between hair cells and auditory nerve fibers, so the ear can still detect quiet tones but cannot transmit the full complexity of speech signals to the brain.

Standard audiometric evaluations are not sensitive enough to catch this. Researchers have found that ultra-high-frequency audiometry, which tests pitches above the standard range, may help identify people with underlying sensory damage that conventional audiograms miss.⁹ Hidden hearing loss complicates the question of “how much hearing loss is deaf” because it shows that you can have a perfectly normal-looking audiogram and still experience genuine, measurable difficulties with auditory processing. Someone with this condition would never meet any audiometric threshold for deafness, yet their day-to-day hearing experience may be far worse than their test results suggest.

Auditory Neuropathy Spectrum Disorder

Another condition that defies simple classification is auditory neuropathy spectrum disorder (ANSD). In ANSD, the outer hair cells of the inner ear function normally, but the signal breaks down somewhere between the inner hair cells and the auditory nerve. A standard test of outer hair cell function, called otoacoustic emissions, comes back normal, while the auditory brainstem response, which measures how well the nerve carries the signal, is abnormal.¹⁰ The result is a disconnect between how well the ear detects sound and how well the brain receives it. People with ANSD can have pure-tone thresholds that range from near-normal to profound, but their speech comprehension is typically much worse than the audiogram would predict.

This means someone with ANSD might technically pass an audiometric screening but be functionally unable to understand speech, or might show a moderate loss on the audiogram yet function as if the loss were profound. The condition highlights why clinicians who work with hearing loss tend to be cautious about using a single decibel number as a bright line for “deaf” versus “not deaf.”¹¹

When Does Hearing Loss Lead to a Cochlear Implant?

One practical reason people ask about deafness thresholds is to understand when a cochlear implant enters the picture. Cochlear implants bypass damaged hair cells and stimulate the auditory nerve directly, and they are typically reserved for people whose hearing loss is too severe for conventional hearing aids to help. The candidacy criteria involve both audiometric thresholds and speech recognition performance.

For adults, research suggests that a pure-tone average above roughly 57 dB and a monosyllabic word recognition score below 60 percent each serve as useful screening indicators for cochlear implant referral.¹² Those numbers are lower than what many people expect. You do not need to be profoundly deaf to benefit from an implant, and clinical guidelines have been gradually expanding candidacy to include people with more residual hearing than was once thought necessary.

For children, the thresholds tend to be a bit different. A scoping review of pediatric candidacy found that hearing levels in the moderately severe to severe range, roughly 65 to 90 dB, are generally supported as the audiological boundary for considering a cochlear implant, with the lower end of that range serving as the cutoff below which implantation is less commonly recommended.¹³ The stakes are higher in children because the window for language development is time-sensitive, which is why candidacy criteria can be more aggressive.

Hearing loss in older adults is independently associated with accelerated cognitive decline, and cochlear implants are often the only effective treatment for people with severe-to-profound loss, the group at highest risk. Yet very few of those who could benefit actually receive them.¹⁴ Barriers include cost, surgical risk perception, lack of awareness, and the lingering misconception that implants are only for people who hear nothing at all.

Prelingual Versus Postlingual Deafness

When the hearing loss occurs matters almost as much as how severe it is. Prelingual deafness, meaning loss that is present at birth or develops before a child acquires spoken language, affects the brain differently than postlingual deafness, which develops after language is already in place. Neuroimaging research has shown that compared to people who became deaf after learning to speak, those with prelingual deafness show reduced white matter integrity in brain regions tied to auditory and language processing, including the superior temporal gyrus and parts of the corpus callosum. The authors concluded that receiving early auditory stimulation before language acquisition may be more critical to white matter development than the nature of the stimulation itself or how long the ears have been without input.¹⁵

In practical terms, this is why newborn hearing screening programs exist and why early intervention is emphasized so strongly. A child born with profound hearing loss who receives a cochlear implant or begins sign language exposure in the first year of life will typically have very different language outcomes than a child whose loss is not identified until age three or four. The audiogram might be identical in both cases, but the functional impact diverges sharply depending on timing.

How Even Mild Hearing Loss Affects Children

Most discussions of deafness focus on severe and profound loss, but the effects of hearing loss begin well below those thresholds in children. A study of children aged 9 to 15 found that those with only slight to mild hearing loss, the mildest category, were about 50 percent more likely to be placed in a lower educational level compared to peers with normal hearing.¹⁶ A separate study of Aboriginal children in remote Australia found that even mild hearing impairment was associated with lower scores in writing and spelling, while moderate or worse impairment predicted lower scores in writing and numeracy as well.¹⁷

These findings matter because mild hearing loss in children often goes undetected. A child who can hear their name being called across a room may still miss soft consonant sounds in a noisy classroom, and the resulting academic struggles might be attributed to inattention or learning difficulties rather than hearing. The takeaway is not just about the definition of deafness but about the broader reality that even hearing loss well short of “deaf” can have meaningful consequences when it goes unrecognized.

The Economic Ripple Effect

Hearing loss does not stay confined to the ear. Its effects cascade into employment, income, and social participation. A large U.S. study found that after controlling for education, age, sex, and race, adults with hearing loss had nearly twice the odds of being unemployed or underemployed compared to their normal-hearing peers, and about 60 percent higher odds of having a low income.¹⁸ These disparities existed across the full spectrum of hearing loss, not only among people who would meet a clinical definition of deaf.

The mechanisms are not mysterious. Difficulty following conversations in meetings, missing phone calls, struggling in noisy workplaces, and the fatigue that comes from straining to hear all day can erode job performance and career advancement. Many people with moderate hearing loss cope by withdrawing socially rather than disclosing their condition, which compounds the isolation. Globally, the WHO estimated that the number of people with disabling hearing impairment rose from 42 million in 1985 to about 360 million by 2011, a figure that included roughly 7.5 million children under five.¹⁹ Given population aging and increased noise exposure, that number has continued climbing since.

Progressive Hearing Loss and Shifting Categories

Hearing loss is not always static. Many forms are progressive, meaning a person can move through several severity categories over the course of their life. In a study of a large family with a dominantly inherited form of deafness, researchers documented hearing loss that began with damage to the auditory nerve and inner hair cells and then progressed over 10 to 20 years to involve the outer hair cells as well, eventually producing a profound sensorineural loss with no measurable brainstem responses.²⁰ Someone in this family might start with a moderate loss in their twenties and arrive at profound deafness by middle age.

This progressive nature matters for the “how much is deaf” question because it means the answer can change for the same person over time. Someone who currently has moderately severe loss and functions well with hearing aids may eventually cross into the range where those aids no longer provide enough amplification and cochlear implant candidacy becomes relevant. Monitoring over years, rather than a single snapshot audiogram, gives a much more accurate picture of where someone stands and where they are heading.

Age-related hearing loss follows a similar arc on a slower timeline. Presbycusis typically begins with high-frequency loss in middle age and gradually spreads to lower frequencies. Most people with age-related loss never reach the profound range, but a meaningful minority do, especially if noise exposure, genetics, or ototoxic medications compound the process. For these individuals, the category labels on an audiogram are not fixed destinations but waypoints on a moving trajectory.