Military Hearing Tests and Audiograms Over Time

A test taken at recruitment, during routine surveillance, after an acoustic incident, at a release medical, by NHS audiology or by a medico-legal expert may each answer a different question. A recruitment audiogram can provide a baseline. Surveillance can show that thresholds changed during service. A later diagnostic audiogram can establish whether loss is sensorineural, conductive or mixed. A civilian test many years later shows what hearing is like now.

The short version

• Different military and civilian hearing tests serve different purposes. Read the series, not one chart.
• A recruitment test can provide a baseline before later service exposure — but it proves the starting point, not what caused any later change.
• Routine military surveillance is designed to monitor hearing acuity and detect deterioration.
• The BSA Surveillance Audiometry procedure requires air-conduction testing only, without masking, and says the frequencies tested may vary. Abbott records the same: it does not involve bone-conduction testing.
• Bone conduction is what distinguishes sensorineural from conductive and mixed loss, and Abbott calls it a crucial part of diagnostic audiometry.
• Bone thresholds are normally tested only from about 500 Hz to 2 kHz — so a shorter bone line than air line is normal practice, not an unfinished test.
• Not all occupational screening tests included 8 kHz. A missing 8 kHz value is *missing data* — not zero, not normal, and not a number to invent.
• Abbott rejected filling missing military 8 kHz readings from a later audiogram as "highly artificial".
• 0 dB HL does not mean "no hearing" — it is roughly the normal-hearing reference level, and young adults commonly record 0 or even negative thresholds.
• An imperfect old test is not automatically worthless, and one suspicious reading does not automatically contaminate a whole series.
• H1–H4 are military hearing-acuity and employability gradings — not diagnoses, and not compensation bands.
• A veteran does not necessarily have a dedicated "discharge audiogram". Do not assume one must exist.
• A modern diagnostic test cannot recreate what hearing was twenty years ago, and an old screening test cannot describe hearing today.

Why the timeline matters

Someone joins at 18 with good hearing, has several surveillance audiograms over ten years, leaves, and twenty years later an audiogram shows high-frequency loss. The question is not simply "does today's audiogram look like NIHL?" It is: what was hearing like on entry, did it change during service, which frequencies changed, was the change consistent between tests, was 8 kHz even measured, were there symptoms at the time, did deterioration continue only after service, and is there evidence of another cause?

Those questions can produce a very different answer from looking at one graph. That is exactly how Abbott approached Mr Craggs — weighing military tests from the late 1990s and 2001 against later civilian and medico-legal audiograms running to 2025.

1. Entry and recruitment audiograms

A recruitment or early-service audiogram matters because it may evidence hearing before much of the later military exposure occurred. It can show whether someone entered with normal hearing, existing loss, asymmetry or a high-frequency abnormality.

If there is no recruitment audiogram, the starting position is harder to establish — but that does not defeat a clinical diagnosis, an AFCS or WPS issue, or a civil claim. The earliest surviving military audiogram, service medical records, later serial tests, witness evidence and current expert assessment may all help. Do not invent a perfect normal baseline because the entry test is missing.

2. Routine surveillance — and what it was built to do

Military hearing surveillance monitors hearing while someone remains exposed to occupational noise. The IMEG Seventh Report describes routine surveillance of hearing acuity, early detection of noise damage, suitable hearing protection, and testing at time intervals dependent on age — plus additional clinical testing as required, "for example ahead of deployment, where there has been an irregular weapon discharge, or a hearing protection device has been dislodged".

So it is a surveillance tool. It is not automatically a complete diagnostic assessment — and the BSA procedure it runs on says so itself:

Abbott records the same distinction, naming the BSA Surveillance Audiometry 2023 procedure and stating that it does not involve bone-conduction testing, while the BSA 2018 pure-tone procedure is concerned with diagnostic-quality audiometry rather than screening.

And where bone conduction *was* performed, expect it across fewer frequencies. Abbott records that bone conduction should normally be performed from 500 Hz to 2,000 Hz, that testing below 500 Hz is not recommended, and that testing above 2 kHz is problematic. A chart showing air thresholds from 250 Hz to 8 kHz but bone symbols only between 500 Hz and 2 kHz is normal practice, not a test someone abandoned halfway.

3. Tests after an incident

Not every military hearing test happens because the calendar says one is due. A test close to an incident answers a different question from routine surveillance — it may document acute threshold change, new asymmetry, or deterioration from a recent previous test.

But an abnormal post-incident audiogram is not automatically proof of acoustic trauma. Timing makes it highly relevant; it still has to be read with symptoms, previous hearing, test quality, and whether the change persisted or recovered. See acute acoustic trauma.

Small variations between audiograms also occur. Audiometry is behavioural testing with real measurement variability, and Abbott itself contains disputes about apparently improving thresholds and whether particular changes were plausible. A difference between two tests does not automatically mean fraud or a faulty audiometer — read it against adjacent tests, age, frequency and the overall pattern.

4. Reading the numbers: zeros, blanks and 8 kHz

A blank, slash or missing value usually means that frequency was not recorded on that test. It should not be converted into zero, into normal hearing, into severe loss, or into the nearest later value. Leave genuinely absent information absent unless there is a sound evidential basis for something else.

This matters most at 8 kHz. Military noise can affect 8 kHz, and deterioration there can change whether a notch remains visible on later audiograms. But Abbott is explicit that "not all occupational screening tests, including some performed by the MoD, include testing at 8 kHz" — and the BSA surveillance procedure itself says the frequencies tested may vary. That creates a genuine evidence gap. It does not create permission to fill it.

5. Reading a series, and what to do with imperfect records

A single audiogram is a snapshot; several form a sequence, and a sequence can show stability, deterioration, recovery from a temporary change, asymmetry, high-frequency progression, when a notch first appears — and whether an apparent notch later disappears.

A noise notch can disappear over time as hearing worsens at nearby higher frequencies including 8 kHz. The absence of a notch on a later audiogram does not mean there was never military NIHL earlier. That is another reason not to discard earlier tests because a recent graph looks different — though equally, not every old notch proves NIHL, since notches occur for other reasons.

Old military records often have problems: the original graph missing, only summary values surviving, no calibration information, unusual symbols, absent frequencies. The correct response is not automatically to discard everything. In Craggs the original first military audiograms were unavailable and the surviving values were second-hand recitations; the judge said those deficiencies meant the results should be viewed with caution, but declined to treat them as non-existent where they formed part of a consistent pattern.

Nor does one bad result invalidate a series. A 1997 8 kHz value looked impossible against later reliable testing — but it was "barely legible and was scribbled out on the summary sheet", and the judge held the most obvious explanation was that it had been recognised as wrong at the time. That single erroneous figure did not lead him to discard the other military results.

Calibration and procedure affect weight, not simply validity. Counsel criticised the 2001 release-medical audiogram on the basis that crosses had been used for both ears where the conventional symbols differ, the calibration date was left blank, and there was no evidence the audiometer was within calibration. The Court engaged with that argument rather than either ignoring it or treating the test as worthless — and answered it partly on the basis that it was the *other* ear that mattered to the diagnosis.

6. Release and discharge records

Where downgrading to H3 or lower, or medical discharge, is under consideration, IMEG continues to recommend objective testing undertaken and interpreted by suitably experienced experts. That is a different proposition from every routine leaver receiving the same test.

7. What H1–H4 actually mean

H1–H4 are PULHHEEMS hearing-acuity grades used within military medical employability. The grade is based on threshold sums at 500 Hz, 1 and 2 kHz and at 3, 4 and 6 kHz. It describes hearing acuity — not the cause of hearing loss, a percentage disability or a compensation band.

That is all this page needs, because an H grade appears in a chronology as a summary of one audiogram rather than as evidence in its own right. Read: What Your Military Hearing Grade Means: H1, H2, H3 and H4 →

8. Civilian and medico-legal testing after service

Later assessment may come from NHS audiology, ENT, private audiology, occupational health or a medico-legal expert, and may contain far more diagnostic detail than an old military chart — bone conduction, masking, tympanometry, speech testing, clinical notes. Abbott describes the audiogram as quantifying "a patient's degree and type of hearing loss (sensorineural, conductive, or mixed)", which is what diagnostic testing is for.

But more detail does not make it more important for every question. What a later test tells you best is what the hearing is like at the time of that test. It cannot recreate hearing twenty years earlier.

Nor should the site suggest that proven noise damage can lie completely latent for many years and then simply appear. In Craggs the judge regarded it as far-fetched that a relatively short service exposure could produce no audiometric evidence at the time, none for roughly fifteen years, and then explain later NIHL. That does not mean every delayed diagnosis fails — earlier damage may have been undocumented, overlooked, or actually present on earlier audiograms. The chronology has to be examined rather than assumed either way.

A hearing-aid retailer's screening test cannot prove military NIHL by itself either. It may give useful current threshold information, but causation needs history, earlier records, exposure, age, alternative causes and diagnostic method. And civilian tests are not automatically better — quality depends on procedure, equipment, calibration, tester, environment and completeness.

Organising your own records

Put every test in date order and record what each one actually is. The point is not to calculate a claim yourself — it is to see missing periods, differing test purposes, when deterioration first becomes visible, and which test can answer which question.

Second-hand values copied into medical notes, an expert report or a summary sheet are weaker than the original graph — but not automatically worthless, as Craggs shows. And where two audiograms conflict, investigate the difference: genuine change, a temporary condition, testing variability, different equipment, calibration, procedure, recording error or masking differences are all possible. The correct response is not to pick whichever result helps.

Memory and audiometry are different kinds of evidence too. Someone may remember struggling in noise, tinnitus or temporary muffling while thresholds stayed relatively good — or remember no difficulty while a test shows deterioration. Both can matter. Neither should be forced to match the other. Our witness evidence guide covers the lay side.

Which audiogram is best?

Serial military audiometry can sometimes be more useful for historic causation than any modern test, because it shows hearing during the exposure period itself. That is why an expert should understand what each historical test actually was — a 1999 occupational screen should never be quietly relabelled as a diagnostic PTA with normal bone conduction when bone conduction was never performed.

The lesson from Abbott is not "pick the worst audiogram". It examined reliability, internal consistency, chronology, plausible threshold change, missing data, later diagnostic testing and the claimant's own symptom evidence — and where data were genuinely missing, it refused to invent them.

Common questions