Audiometric testing and hearing loss records

Noise assessments, controls and protectors are all aimed at one outcome: workers' hearing stays the same over their working lives. Audiometric testing is how a program finds out whether that is happening. A baseline test records a worker's hearing before significant exposure, and periodic tests compare later results with it. Early, small changes, long before the worker notices anything, trigger follow-up that can prevent a permanent handicap. Taken together, the results of many workers show whether the program as a whole is working. This lesson covers how audiometric testing is done, how results are interpreted, what follow-up is needed and how hearing loss is recorded and compensated.

What an audiometric test involves

An occupational audiometric test is a pure-tone air-conduction hearing test. The worker sits in a quiet booth or room wearing headphones or insert earphones, and presses a button whenever a tone is heard. The tester, or an automated audiometer, presents tones at each test frequency, usually 500, 1,000, 2,000, 3,000, 4,000, 6,000 and 8,000 Hz, varying the level to find the quietest level the worker can hear about half the time. The results are recorded in dB HL for each ear and plotted on an audiogram, as described in Lesson 1. The test usually takes 10 to 20 minutes and should be combined with a short questionnaire about ear health, noise exposure at work and outside work, and protector use, and with a brief counseling session.

The quality of the test depends on several things. The room must be quiet enough not to mask the quietest tones, typically checked against maximum permissible background levels in each octave band. The audiometer must be calibrated: a daily functional check, a periodic acoustic calibration and an exhaustive calibration at longer intervals. The tester must be trained and must give consistent instructions. And the worker should have had a period without loud noise before the test, especially before the baseline, so that a temporary threshold shift does not distort the result.

The baseline audiogram

The baseline is the reference against which every later test is compared, so it must be good. A baseline taken right after a noisy shift may show a temporary threshold shift, making later results look better than they are and hiding real damage. A baseline taken with poor instructions or in a noisy room may show worse thresholds than the worker really has. Many programs take the baseline on a worker's first day, before any exposure, and some repeat it within a few weeks to confirm it. Pre-existing hearing loss found at baseline should be documented, explained to the worker and considered in protector selection.

Detecting a significant change

Each periodic test is compared with the baseline, and sometimes with the previous test, to detect changes. Some variation is normal: a test-retest difference of 5 dB at a single frequency is common. A program therefore uses defined criteria for a significant change, which triggers follow-up.

OSHA defines a standard threshold shift, or STS, as a change in hearing threshold relative to the baseline audiogram of an average of 10 dB or more at 2,000, 3,000 and 4,000 Hz in either ear. Employers may adjust for the expected effects of aging using the tables in Appendix F of 1910.95, and may retest within 30 days and use the retest result if it does not confirm the shift.

freqs = [500, 1000, 2000, 3000, 4000, 6000, 8000]

def shifts(baseline, current):
    return {f: c - b for f, b, c in zip(freqs, baseline, current)}

def average_shift_234(baseline, current):
    d = shifts(baseline, current)
    return (d[2000] + d[3000] + d[4000]) / 3

def hearing_level_234(current):
    c = dict(zip(freqs, current))
    return (c[2000] + c[3000] + c[4000]) / 3

worker = {
    "left":  {"baseline": [5, 5, 10, 10, 15, 15, 10], "annual": [5, 10, 15, 25, 35, 25, 15]},
    "right": {"baseline": [5, 5, 5, 10, 10, 15, 10],  "annual": [5, 5, 10, 15, 20, 20, 10]},
}
for ear, r in worker.items():
    avg = average_shift_234(r["baseline"], r["annual"])
    level = hearing_level_234(r["annual"])
    single = max(shifts(r["baseline"], r["annual"]).values())
    print(f"{ear:>5}: average shift at 2-4 kHz {avg:4.1f} dB, "
          f"largest single shift {single} dB, current average level {level:4.1f} dB HL")
    print(f"       10 dB average-shift criterion met: {avg >= 10}; "
          f"25 dB average level reached: {level >= 25}")

 left: average shift at 2-4 kHz 13.3 dB, largest single shift 20 dB, current average level 25.0 dB HL
       10 dB average-shift criterion met: True; 25 dB average level reached: True
right: average shift at 2-4 kHz  6.7 dB, largest single shift 10 dB, current average level 15.0 dB HL
       10 dB average-shift criterion met: False; 25 dB average level reached: False

In this example, the left ear shows an average shift of about 13 dB at 2,000 to 4,000 Hz and a current average hearing level of 25 dB HL, so it meets both the 10 dB shift criterion and the 25 dB level used in US recordkeeping. The right ear shows a smaller change that does not meet the shift criterion but should still be watched. No age adjustment has been applied; doing so would reduce the calculated shift somewhat for an older worker.

Follow-up after a significant shift

A significant shift is an early warning. It should lead to a set of actions aimed at stopping further loss.

Recording and reporting hearing loss

Under 29 CFR 1904.10, an employer must record a hearing loss case on the OSHA 300 log if an employee's audiogram shows a standard threshold shift in one or both ears, and the employee's total hearing level is 25 dB or more above audiometric zero, averaged at 2,000, 3,000 and 4,000 Hz, in the same ear. Age correction may be applied to the STS determination but not to the 25 dB test, and a retest within 30 days may be used. If a physician or other licensed health care professional determines that the hearing loss is not work-related or not aggravated by occupational noise, the case need not be recorded. Hearing loss cases are entered in a separate column of the log.

Workers' compensation for noise-induced hearing loss

State workers' compensation systems handle noise-induced hearing loss claims, with rules that differ widely in the degree of loss required, the formula for impairment and the treatment of age-related loss. Some states compensate only when the loss reaches a defined level in the speech frequencies, while others use a formula from the American Medical Association guides. Good baseline and periodic records help both the worker and the employer, because they show when and where hearing changed. Federal employees and longshore workers have their own compensation systems.

Confidentiality and counseling

Audiograms are confidential health records. The worker should receive a copy of each result and an explanation in plain language. The employer needs to know that testing was done, whether a significant shift occurred and what actions are needed, but not the detailed audiogram. Committees should see aggregate statistics only.

Counseling at the time of the test is one of the most effective parts of the program. A short conversation, showing the worker their own audiogram trend, explaining what any change means, checking protector fit and asking about leisure noise, connects the abstract risk to the individual. Workers who see their own notch beginning are much more likely to protect their hearing consistently.