Rating criteria · 38 CFR §4.85 · §4.86 · §4.87
Hearing Loss: VA Rating Criteria
The exact rating criteria below are quoted from the Code of Federal Regulations as currently in force, not paraphrased. Compensation amounts come from the current VA rate tables.
By Mario Bailey, U.S. Air Force veteran · current Air National Guard member. Data updated 2026-07-24.
Plain-language guide
What this rating actually turns on
In a hearing claim the arithmetic leans on your better ear and quietly discounts the worse one. Under 38 CFR 4.85, your speech discrimination score and your puretone threshold average, the average of thresholds at 1000, 2000, 3000, and 4000 Hz, intersect on Table VI to give each ear a Roman numeral from I (best) to XI (worst). Table VII then combines the two numerals into a percentage, and the better ear drives the result: the rows are the better ear, the columns the worse one. A bad ear paired with a good ear rates low. This mechanical system dates to a final rule at 64 FR 25206 (May 11, 1999).
Before any of that, your hearing has to qualify as a disability at all. 38 CFR 3.385 requires a threshold of 40 dB or more at any frequency from 500–4000 Hz, or 26 dB or more at three of those frequencies, or a Maryland CNC speech recognition score under 94 percent. You can meet 3.385, be service-connected, and still compute to 0 percent on Table VII. That is a real and common outcome of the math, and a 0 percent rating still has value, it establishes service connection for later worsening.
A worked example
Because DC 6100 has no percentages written into it, the only way to watch a number get assigned is to run one through the tables. Treat these figures as illustrative, not as criteria: the real Table VI and Table VII exist only as grids of images in the eCFR, so put your own audiogram through them (or a VA hearing-loss calculator that reproduces them) rather than these.
Say a veteran has service-connected noise damage in the right ear only. The C&P audiologist measures a right-ear puretone average of 70 dB across 1000–4000 Hz and a Maryland CNC speech discrimination score of 64 percent. Those two numbers meet on Table VI at a Roman numeral for that ear (in this illustration, around VI). The left ear was never service-connected and independently tests near normal, so 4.85(f) assigns it numeral I, not its measured value. Table VII then combines the two, and because its rows are the better ear, numeral I in the better-ear row pulls the intersection all the way down to 0 percent, even though the right ear is genuinely damaged. That is the better-ear rule working exactly as designed, and the 0 percent still locks in service connection for the day the good ear starts to slip. Swap in two service-connected ears with real losses and the same walk lands somewhere up the scale instead.
What the C&P exam measures
38 CFR 4.85(a) is strict: the exam must be conducted by a state-licensed audiologist and must include a controlled speech discrimination test (Maryland CNC) and a puretone audiometry test, performed without hearing aids. Per M21-1, V.iii.2.B, cochlear implant users have the external processor removed for testing. When your CNC score comes in at 92 percent or less, the examiner must run a performance intensity function test, three repetitions, with the maximum score used. Only air conduction results go into the rating calculation; bone conduction is diagnostic only, even if an examiner says it better reflects your hearing. Thresholds above 105 dB are entered at 105. If the examiner certifies speech testing isn’t appropriate, Table VIa rates you on puretone average alone. The Hearing Loss and Tinnitus DBQ is not available for public use.
What to have in your file
- Every service audiogram, including entrance and separation exams. Per M21-1, in-service threshold shifts that don’t reach 3.385 can still support service connection later. Pre-1969 audiograms in ASA units must be converted to ISO/ANSI before comparison, an audiologist opinion handles that. Old whispered-voice tests are considered unreliable evidence either way.
- Noise exposure documentation. VA’s Duty MOS Noise Exposure Listing lets raters concede hazardous noise from your MOS alone; combat service gets conceded exposure on lay evidence.
- Private audiograms that meet VA’s standards. To count for rating purposes they need Maryland CNC speech testing, and if the score is 92 percent or less, evidence that performance intensity function testing was done. A private audiogram without Maryland CNC can support a diagnosis but cannot set your percentage.
Common mistakes
- Testing with hearing aids in. The regulation requires unaided results; aided results are unusable.
- Assuming bone conduction scores will rescue the rating. M21-1 flatly prohibits using them.
- Expecting a single bad ear to rate high. If only one ear is service-connected, 4.85(f) assigns the other ear a numeral I, unless the paired-organ rule in 38 CFR 3.383 applies: a compensable service-connected ear plus a non-service-connected ear that independently meets 3.385.
Worth knowing
38 CFR 4.86 covers two exceptional patterns. If all four rated frequencies are 55 dB or worse, the rater uses Table VI or VIa, whichever is higher. If you hear well at 1000 Hz (30 dB or better) but poorly at 2000 Hz (70 dB or worse), a steep ski-slope loss, the higher table applies and the numeral bumps up one step. Tinnitus is rated separately and combines with hearing loss under Note 1 to DC 6260. Total deafness can qualify for special monthly compensation under 38 CFR 3.350. Hearing loss is the sixth most prevalent service-connected disability overall, at 1,690,837 ratings, per the FY 2025 VBA Annual Benefits Report.
Rating criteria from the CFR
SOURCE: eCFR, 38 CFR Part 4 (issue date 2026-07-13, current through 2026-07-24) · retrieved 2026-07-28
Regulation text
38 CFR §4.85, Evaluation of hearing impairment.
(a) An examination for hearing impairment for VA purposes must be conducted by a state-licensed audiologist and must include a controlled speech discrimination test (Maryland CNC) and a puretone audiometry test. Examinations will be conducted without the use of hearing aids.
(b) Table VI, “Numeric Designation of Hearing Impairment Based on Puretone Threshold Average and Speech Discrimination,” is used to determine a Roman numeral designation (I through XI) for hearing impairment based on a combination of the percent of speech discrimination (horizontal rows) and the puretone threshold average (vertical columns). The Roman numeral designation is located at the point where the percentage of speech discrimination and puretone threshold average intersect.
(c) Table VIa, “Numeric Designation of Hearing Impairment Based Only on Puretone Threshold Average,” is used to determine a Roman numeral designation (I through XI) for hearing impairment based only on the puretone threshold average. Table VIa will be used when the examiner certifies that use of the speech discrimination test is not appropriate because of language difficulties, inconsistent speech discrimination scores, etc., or when indicated under the provisions of § 4.86.
(d) “Puretone threshold average,” as used in Tables VI and VIa, is the sum of the puretone thresholds at 1000, 2000, 3000 and 4000 Hertz, divided by four. This average is used in all cases (including those in § 4.86) to determine the Roman numeral designation for hearing impairment from Table VI or VIa.
(e) Table VII, “Percentage Evaluations for Hearing Impairment,” is used to determine the percentage evaluation by combining the Roman numeral designations for hearing impairment of each ear. The horizontal rows represent the ear having the better hearing and the vertical columns the ear having the poorer hearing. The percentage evaluation is located at the point where the row and column intersect.
(f) If impaired hearing is service-connected in only one ear, in order to determine the percentage evaluation from Table VII, the non-service-connected ear will be assigned a Roman Numeral designation for hearing impairment of I, subject to the provisions of § 3.383 of this chapter.
(g) When evaluating any claim for impaired hearing, refer to § 3.350 of this chapter to determine whether the veteran may be entitled to special monthly compensation due either to deafness, or to deafness in combination with other specified disabilities.
(h) Numeric tables VI, VIA*, and VII.
38 CFR §4.86, Exceptional patterns of hearing impairment.
(a) When the puretone threshold at each of the four specified frequencies (1000, 2000, 3000, and 4000 Hertz) is 55 decibels or more, the rating specialist will determine the Roman numeral designation for hearing impairment from either Table VI or Table VIa, whichever results in the higher numeral. Each ear will be evaluated separately.
(b) When the puretone threshold is 30 decibels or less at 1000 Hertz, and 70 decibels or more at 2000 Hertz, the rating specialist will determine the Roman numeral designation for hearing impairment from either Table VI or Table VIa, whichever results in the higher numeral. That numeral will then be elevated to the next higher Roman numeral. Each ear will be evaluated separately.
If your rating came back lower than this
A decision that ignores the criteria above is a frequent reason veterans are underpaid. Line your decision up against the levels on this page. If the evidence supports a higher level than VA assigned, you have three ways to challenge it, and the right one depends on why it went wrong:
- Something was missing from the record (an exam finding, a prescription, a flare-up statement, a nexus letter): add it with a Supplemental Claim.
- The evidence was already there and VA misread it: ask for a Higher-Level Review by a more senior rater, with no new evidence needed.
- A legal or judgment error: take it to the Board of Veterans' Appeals.
Line your decision up against the criteria on this page. Which review lane fits depends on why the rating is wrong: new evidence that was missing, a misread of evidence already in the file, or a legal error. All three are laid out in the decision reviews and appeals guide.
Before you file, recombine this rating with your others (VA math does not add ratings), check how long each lane is taking right now, and remember a VA-accredited representative or VSO will help you for free. You can also browse every other condition's criteria to make sure you are claiming everything you are owed.