Hearing is rated by table lookup, not by judgment

Almost nothing else in the schedule works like this. There is no "moderate" or "severe" to argue about — §4.85 turns two measurements into a Roman numeral for each ear, and those two numerals produce one percentage. It is mechanical, which cuts both ways: it is hard to argue with, and it is easy to check.

The exam has to be the right exam

A state-licensed audiologist, a puretone audiometry test, and a controlled speech discrimination test (Maryland CNC) — tested without hearing aids. A word-recognition score from some other test cannot be quietly substituted into the table.

The four frequencies, and the one that does not count

Each ear's puretone threshold average is the sum of the thresholds at 1000, 2000, 3000 and 4000 Hz, divided by four.

500 Hz is not in it. It appears in §3.385, which defines when hearing loss is a disability for VA purposes at all — a different question, answered with a different test. Mixing the two is the most common error in home calculations.

The lookup, step by step

1. Each ear's average, as above.
2. Table VI — intersect that ear's Maryland CNC score with its puretone average. Out comes a Roman numeral, I through XI.
3. Table VIa uses the puretone average alone. It is available when the examiner certifies the speech test is not appropriate — language difficulties, inconsistent scores — or when §4.86 calls for it. It is not a free choice whenever it happens to give a better numeral.
4. Check §4.86 for each ear separately, before going further. See below.
5. Table VII — the better ear on the horizontal row, the poorer ear on the vertical column. The intersection is your DC 6100 evaluation.

Step five is where another common mistake lives. You do not get a percentage for each ear and then combine them under §4.25. Both ears produce one number.

The exceptional patterns, where one word changes the answer

The difference between those two rows is a single elevation, and it is worth checking against your own audiogram. Paragraph (b) requires nothing in particular at 3000 and 4000 Hz for the pattern — though those frequencies still count toward the average.

Why a real hearing loss can rate zero percent

Take a right ear averaging 35 dB with 92 percent discrimination, and a left averaging 55 dB with 80 percent. Table VI gives I and IV. Neither ear meets an exceptional pattern. Table VII, better ear I against poorer ear IV, gives 0 percent.

That is a documented, service-connectable hearing loss at a noncompensable evaluation. It is not a mistake and it is not a denial of service connection — a 0 percent rating can coexist with established service connection, and it preserves the record for later worsening.

This is the hardest thing about DC 6100 to hear, and it is better to learn it from a page than from a decision letter.

If only one ear is service connected

§4.85(f) ordinarily assigns the non-service-connected ear a Roman numeral of I for Table VII — not its measured numeral. §3.383 has a special paired-organ provision that can change this. Where deafness is severe enough, §3.350 special monthly compensation may also be in play.

Tinnitus is separate

The note under DC 6260 expressly allows recurrent tinnitus to be evaluated alongside hearing impairment — unless the tinnitus is already supporting another evaluation. See tinnitus.

Checking it yourself

Keep the audiogram, the actual Maryland CNC report and the table lookup together, so you or a representative can reproduce the result. Then compare, in order: the four thresholds, the average, the Table VI numeral, whether §4.86 applied, and the Table VII cell.

The official tables are printed in the published CFR — Tables VI and VIa on the third page, Table VII on the fourth. Use those rather than an unlabeled number from any calculator, including a well-meaning one.

Reviewed October 6, 2026