Standard threshold shift: what it is and when to record it

This page explains what the rules and common practice call for. Every site is different, and what a survey or a filing needs at your facility is something we work out with you after we see it.

A standard threshold shift (STS) is one test. A hearing loss case you must record on your OSHA log is a different test, and it is harder to meet.

A worker can have a confirmed STS (a measured drop in hearing) and trigger every follow-up duty in OSHA's noise standard. That same worker can still, correctly, not appear on your OSHA 300 log.

The two get confused in both directions. Some workplaces record too few cases, while others log every STS on the 300 log and push their rate up for no reason.

What counts as a standard threshold shift

OSHA's noise standard defines it as a change in hearing threshold (the quietest sound a person can hear), relative to the baseline audiogram. The change must average 10 decibels (dB) or more at 2000, 3000 and 4000 hertz (Hz) in either ear. The baseline audiogram is the worker's first hearing test.

In short, the calculation uses three frequencies, averaged, in either ear, and it is measured against the baseline, not against last year's audiogram.

The audiograms themselves cover more frequencies than that. The noise standard requires testing at 500, 1000, 2000, 3000, 4000 and 6000 Hz, with each ear tested separately. The STS is calculated from only the middle three, but the others help explain what is going on, and they belong in the record.

Worked example: an STS that is not recordable

This example is for the right ear. All values are hearing levels in decibels (dB HL), where a higher number means poorer hearing.

First, average the three frequencies for each audiogram. Then subtract the baseline average from this year's average.

Shift = 21.7 − 10.0 = 11.7 dB. That is 10 dB or more, so this is an STS and all the follow-up duties in the noise standard apply.

Is it recordable? OSHA's recordkeeping rule sets a second, separate test, and a case is recordable only when two things are true. There is a work-related STS, and the worker's total hearing level is 25 dB or more above audiometric zero, averaged at 2000, 3000 and 4000 Hz, in the same ear. Audiometric zero is the zero point of the audiogram scale.

The current total hearing level in that ear averages 21.7 dB, which is below 25, so the case is not recordable.

Worked example: an STS that is recordable

The shift is the same size, but the starting point is different.

Shift = 36.7 − 25.0 = 11.7 dB. That is an STS, the same size as in the first example.

Total hearing level = 36.7 dB. That is 25 dB or more above audiometric zero in the same ear, so the case is recordable on the 300 log, assuming the STS is work-related.

The shift is identical and the recordability decision is the opposite. What differs is where the worker started.

The age correction trap

Age correction is an allowed adjustment for the hearing loss expected from aging alone (called presbycusis).

OSHA's recordkeeping rule allows age correction when you determine whether an STS has occurred. You may apply it to the shift calculation.

You may not apply it when you check the 25 dB level. The total hearing level test is run on the uncorrected audiogram.

Read those two together, because the uneven treatment is deliberate. Age correction can take a shift out of STS territory, but it cannot take a hearing level below 25 dB.

Audiometric software that applies the correction everywhere will give wrong answers on recordability, so check what your provider's system is doing before you rely on its recordable flags.

What an STS requires

The physician exception does not reach the written notice. The exception opens the paragraph of the standard that holds four steps, lettered (A) through (D), and it covers only those four:

• Fitting or refitting and retraining in hearing protectors
• Referral for a clinical audiological evaluation or otological examination
• Notice where an otological examination is needed for a problem unrelated to protector use
• Required protector use by workers not otherwise obliged to wear them

Those four may be set aside if a physician or other licensed health care professional finds that the shift is not work-related, or has not been aggravated by occupational noise. That finding has to actually be made and documented. You cannot assume it because the worker has a loud hobby.

The written notice carries no such condition. You owe the worker written notice within 21 days of the determination on every confirmed STS, including one that a physician later attributes to something other than work.

The hearing protector target changes too. For workers in general, the noise standard requires protectors that reduce exposure to an 8-hour time-weighted average (TWA) of 90 dB or below. For a worker who has had an STS, it requires 85 dB or below.

That is a real change in what you provide, not a paperwork change. It may mean a different protector, plugs and muffs worn together, or a change in the worker's assignment. It is one of the few places in the noise standard where one person's audiometric history changes the requirement for that person.

It helps to be clear about what this target is and is not. The 85 dB target concerns hearing protector attenuation, so you meet it with personal protective equipment (PPE). Engineering and administrative controls are a separate duty, triggered by the levels in Table G-16, OSHA's table of permissible noise exposures. That duty is never set person by person from audiometric history.

→ How to select hearing protectors

The 21-day clock

The 21 days run from the determination, not from the test date. When the audiogram is repeated to confirm the result, or a professional reviews it first, the clock runs from the point when that determination is made.

Two 30-day limits put a cap on that. They are what stop "we hadn't determined it yet" from running on indefinitely:

The determination can therefore properly wait on a confirming retest, but only on that, and only for 30 days.

In practice this catches workplaces that use a mobile testing service. The van comes in March, the results package arrives in May, and someone reviews it in June. By then the retest window has closed, and the 21-day notice is being counted from a determination that should have been made months earlier.

Agree on three things in advance, in writing, with the provider:

• Who makes the determination
• When results are sent
• How a retest gets scheduled inside 30 days

What to check in your audiometric testing program

We do not perform audiometric testing or review audiometric testing programs, because that work is outside the four fields we work in. These points are still worth checking in your own program:

• Do baseline audiograms exist, and are they dated inside the deadline the standard sets?
• Are annual audiograms compared against the baseline, not against the year before?
• Is age correction applied to the shift calculation only?
• Did STS notices go out inside 21 days?
• Were workers with an STS moved to the 85 dB target?
• Is the noise data behind all of it still on file? The noise standard only requires noise exposure measurements to be retained for 2 years, so the survey that would explain a shift is often already gone by the time anyone looks for it.

Measuring the noise is our part.

→ Hearing conservation programs · → Access to exposure and medical records · → Noise surveys · → Noise dosimetry