Hearing Conservation Program (OSHA 1910.95)

A hearing conservation program is the set of noise controls OSHA requires whenever an employee's 8-hour average noise exposure reaches the 85 dBA action level under 29 CFR 1910.95. It has five parts: noise monitoring, audiometric testing, hearing protectors, training, and recordkeeping.

Noise is one of the most common and most preventable occupational hazards on a plant floor, and hearing loss from it is permanent. You do not get the hearing back. NIOSH estimates more than 22 million U.S. workers are exposed to hazardous noise on the job. The rule that governs how you protect them is OSHA's occupational noise standard, 29 CFR 1910.95, and it turns on two numbers most people mix up: the 85 dBA action level and the 90 dBA permissible exposure limit. This guide walks the whole program: when it triggers, what each of the five elements requires, how to read an audiogram, and how to size hearing protection so it actually works on the floor instead of just on paper.

When does OSHA require a hearing conservation program?

You must run a hearing conservation program whenever an employee's noise exposure equals or exceeds an 8-hour time-weighted average (TWA) of 85 dBA, the action level. That is different from the 90 dBA permissible exposure limit (PEL), the level above which you must also use engineering or administrative controls to bring exposure down.

The distinction matters because the two numbers do different jobs. Reaching the 85 dBA action level obligates you to protect people: monitor noise, test hearing, offer protectors, and train. Reaching the 90 dBA PEL obligates you to actually reduce the noise through feasible engineering or administrative controls, not just hand out earplugs. A rough field test: if you have to raise your voice to talk to someone an arm's length away, you are probably near or above 85 dBA and it is time to measure.

How does the 5-dB exchange rate work?

OSHA uses a 5-dB exchange rate, which means every 5 dBA increase in noise cuts the permissible exposure time in half. At 90 dBA you may be exposed for 8 hours; at 95 dBA only 4 hours; at 100 dBA only 2 hours. Louder means shorter, fast.

This is why a short burst of very loud work can blow through the daily dose even when the floor sounds quiet most of the shift. The permissible durations from Table G-16 of the standard are worth posting where planners can see them.

One note on the exchange rate: NIOSH recommends a stricter 3-dB exchange rate and an 85 dBA recommended exposure limit, because 3 dB reflects how sound energy actually doubles. OSHA's rule still uses 5 dB and 90 dBA as the PEL, so that is what you are cited against, but if you are designing controls to protect people rather than just pass an inspection, the NIOSH 85 dBA target is the safer design point.

What are the five elements of a hearing conservation program?

The program has five required elements: noise monitoring, audiometric testing, hearing protectors, employee training, and recordkeeping. Each one feeds the next, you monitor to know who is exposed, test their hearing to catch damage, protect them, train them, and keep the records that prove all of it.

Noise monitoring

You measure the noise to find every employee at or above the 85 dBA TWA and to pick the right protection. Where noise varies a lot across a shift, use dosimetry, a badge dosimeter worn for a representative shift captures the whole dose, not just a snapshot. Repeat monitoring whenever a change in production, equipment, or process could push someone over the action level. Employees have the right to observe monitoring and to be notified of their results.

Audiometric testing

Audiometric testing is the heart of the program because it is the only element that tells you whether hearing is actually being lost. Within six months of an employee's first exposure at or above the action level, you must establish a baseline audiogram, taken after at least 14 hours away from workplace noise. After that, a new audiogram every year is compared to the baseline, testing at 500, 1000, 2000, 3000, 4000, and 6000 Hz in each ear.

Hearing protectors, training, and records

You must make hearing protectors available to everyone at or above the action level, and require them for anyone at or above the PEL, for anyone who has had a standard threshold shift, and for exposed workers who have not yet had a baseline. Train exposed employees at least annually on the effects of noise, the protectors, and audiometric testing. Keep noise measurement records for two years and audiometric records for the duration of the worker's employment.

How do you stand up a hearing conservation program, step by step?

If you are starting from nothing, work the elements in this order so each step gives you the information the next one needs:

1. Assign an owner and a physician or audiologist. Name who runs the program and secure a licensed or certified professional to oversee audiometric testing. Ownership is the difference between a program and a binder.
2. Monitor the noise. Use sound-level surveys and dosimetry to find every job at or above the 85 dBA action level, and re-monitor whenever production, equipment, or process changes could raise exposure.
3. Establish baselines. Get a baseline audiogram for each exposed worker within six months of first exposure, taken after at least 14 hours away from workplace noise.
4. Select and fit protectors. Choose protectors whose derated attenuation gets each worker below 85 dBA, offer a choice of styles, and fit-check them on real ears.
5. Train, then retrain annually. Cover the effects of noise, how to fit the protectors, and the purpose of audiometric testing. Refresh it every year.
6. Run annual audiograms and act on shifts. Compare each yearly test to the baseline, and when a standard threshold shift appears, notify, refit, and record it if required.
7. Keep the records and review the program. Retain noise data for two years and audiograms for the length of employment, and review trends so you catch the exposures that are creeping up.

What is a standard threshold shift and what do you do about it?

A standard threshold shift (STS) is a change from the baseline audiogram averaging 10 dB or more at 2000, 3000, and 4000 Hz in either ear. It is OSHA's early-warning signal that a worker is losing hearing, and it triggers required follow-up.

When an audiogram shows an STS, you generally have to notify the employee in writing within 21 days, refit and retrain them on hearing protectors (or move them to protectors with greater attenuation), and, if the shift is work-related and also amounts to at least a 25 dB shift from audiometric zero, record it on the OSHA 300 log. You may age-correct the audiogram for presbycusis before deciding an STS occurred, and if a follow-up audiogram confirms the shift, that new level can become the revised baseline. Handled early, an STS is a chance to fix the exposure before the loss becomes disabling. This is the same "catch the leading indicator before the injury" logic that drives good near-miss reporting and any real ISO 45001 program.

How do you pick hearing protection that actually works?

Every hearing protector carries a Noise Reduction Rating (NRR), a lab number in decibels. The catch is that the lab number overstates real-world protection, so OSHA derates it: for A-weighted noise, subtract 7 from the NRR, then divide by 2 to estimate the protection you will actually get.

So a plug with an NRR of 29 does not give you 29 dB of protection on the floor. Under OSHA's method, (29 − 7) ÷ 2 ≈ 11 dB of real-world attenuation. If a worker is at a 98 dBA TWA, that plug lands them near 87 dBA, still above the action level. Fit and consistency matter more than the number on the box: a plug worn wrong, or pulled out to hear a coworker, protects nothing. That is why fit checks and training beat chasing the highest NRR, and why doubling up (plugs plus muffs) for very loud tasks does not simply add the two ratings together.

How does a hearing conservation program fit the rest of your safety system?

Noise rarely travels alone. The same tasks that expose people to noise often expose them to other hazards, so the smart move is to catch noise inside your existing hazard reviews rather than as a standalone audit. A thorough job safety analysis should flag noisy steps and push toward engineering controls, enclosures, damping, quieter tooling, because the same machine guarding project is often the right moment to enclose a noise source, too. Short toolbox talks keep protector fit and use top of mind between the annual training.

The part that trips plants up is the paperwork. A hearing conservation program generates years of monitoring data, audiograms, training rosters, and STS follow-ups, and OSHA can ask for any of it. Records that live in a filing cabinet or a spreadsheet on one PC are hard to trend and easy to lose. Harmony captures noise checks, protector fit-tests, and training completion as structured, timestamped data on the same floor system as your OSHA recordkeeping and quality checks, so a safety lead can see who is due for an audiogram or whose shift is trending toward an STS without digging through binders. See how one plant put its safety and quality records on one system or how the modules fit together. The standard is only as good as your ability to prove you followed it, and to spot the loss before it becomes permanent.