Respirator Fit-Testing Records Under OSHA 1910.134: What to Document and Keep

Respirator fit testing records are what an OSHA compliance officer actually asks for at an inspection — and where most healthcare programs come up short. The fit test itself takes maybe fifteen minutes; the record it produces is the proof of compliance. Programs fail here not because they skip testing, but because the paperwork is incomplete, filed in the wrong place, or thrown out too early.

Fit testing is required for every tight-fitting respirator your staff wear, and there's a reason the rule insists on an annual retest rather than a one-and-done. A NIOSH study supporting the requirement estimated that the share of workers whose respirator no longer fits climbs from roughly 10% at one year to about 20% at two years and 26% at three (CDC/NIOSH Science Blog). Faces change. The record is how you prove you kept up.

Here's what belongs in that record, how long to keep it, and the trap that catches even well-run programs.

What must a fit-test record contain?

OSHA is specific. Under 29 CFR 1910.134(m)(2)(i), every fit-test record must capture five elements:

• (A) the name or identification of the employee tested;
• (B) the type of fit test performed;
• (C) the specific make, model, style, and size of respirator tested;
• (D) the date of the test; and
• (E) the pass/fail result for a qualitative fit test (QLFT), or the fit factor and strip-chart recording (or other recording of the results) for a quantitative fit test (QNFT).

Element (C) is where sloppy records live. "N95" is not a record. "3M 1860, size small" is. If a nurse fails on one model and passes on another, both facts and the specific facepieces belong in the file — that's the whole point of documenting make, model, style, and size. And note that (E) changes depending on the test method: a QLFT record stores a binary result, while a QNFT record stores a number. More on that distinction below.

One prerequisite that isn't part of the fit-test record but must exist before it: the medical evaluation. OSHA requires a medical evaluation to determine the employee's ability to use a respirator before they're fit tested or required to wear one (1910.134(e)(1)). Fit testing someone who was never medically cleared is a citation waiting to happen.

How often is fit-testing required?

Three triggers, from 1910.134(f)(2):

1. Before initial use of a tight-fitting facepiece;
2. Whenever a different facepiece is used — a change in size, style, model, or make; and
3. At least annually thereafter.

That second trigger trips more often than people expect. Switch a unit from the 3M 1860 to a Moldex when supply shifts, and everyone on that respirator needs a new test on the new facepiece — the annual clock doesn't cover it.

There's also a physical-change trigger. Under 1910.134(f)(3), an additional fit test is required whenever the employee reports — or the employer, PLHCP, supervisor, or program administrator observes — a change that could affect fit: facial scarring, dental changes, cosmetic surgery, or an obvious change in body weight. A nurse who lost 30 pounds since last spring is due for a retest now, not on their anniversary date.

Qualitative vs. quantitative fit tests

The two methods differ in what they measure, and that difference flows straight into your record.

Qualitative (QLFT) is a pass/fail method that relies on the wearer detecting a challenge agent — you either taste, smell, or react to it, or you don't. Appendix A recognizes four agents: isoamyl acetate (odor), saccharin (taste), Bitrex/denatonium benzoate (taste), and irritant smoke/stannic chloride (involuntary irritation response) (1910.134 Appendix A). Critically, QLFT may only be used for negative-pressure air-purifying respirators that need a fit factor of 100 or less — in practice, half-mask respirators like the N95 (1910.134(f)(6)).

Quantitative (QNFT) uses an instrument to produce a numeric fit factor. Appendix A recognizes three protocols: generated aerosol, ambient aerosol condensation nuclei counter (CNC, the common PortaCount setup), and controlled negative pressure (CNP). A QNFT passes when the fit factor is ≥100 for half facepieces or ≥500 for full facepieces (1910.134(f)(7)).

Whichever you use, it must follow an OSHA-accepted protocol from Appendix A — and compliance with those appendices is mandatory, not advisory (1910.134(f)(5) and (o)). So a QLFT record shows "pass," while a QNFT record shows "fit factor 178" plus the instrument's recording. Build your form to hold whichever the method produces.

How long must fit-test records be retained?

This is the point that surprises almost everyone.

Fit-test records only have to be kept until the next fit test is administered (1910.134(m)(2)(ii)). In an annual program, that's roughly one year — the current record supersedes the last one.

Now the trap. That short retention applies only to fit-test records. Medical evaluation records are a completely separate obligation. Under 1910.134(m)(1), they're retained under 29 CFR 1910.1020 — the access-to-medical-records rule — which requires employee medical records to be kept for the duration of employment plus 30 years.

Conflate the two and you fail in one of two directions: shred a medical clearance after a year (a serious recordkeeping violation), or hoard three decades of fit-test strips you never needed to keep. Keep them in separate systems, retained on separate clocks. Records must also be made available to affected employees on request.

What's at stake if the records aren't there

Incomplete or missing records are a frequent finding when OSHA inspects a respiratory-protection program, and the penalties aren't trivial. As of early 2026, the maximum civil penalties in effect were the amounts set for 2025: $16,550 per serious or other-than-serious violation and $165,514 per willful or repeat violation, effective January 15, 2025 (OSHA 2025 penalty amounts). Failure to abate carries a penalty of up to the same per-violation serious maximum for each day the condition goes uncorrected. An inspector who finds testing happening but records that don't hold the five required elements can still write it up.

One more thing for healthcare readers: OSHA state-plan states may impose stricter rules. California, for example, layers its Aerosol Transmissible Diseases (ATD) standard on top of 1910.134, with additional respiratory-protection and recordkeeping obligations for covered healthcare settings. Check your state plan before assuming the federal minimum is the ceiling.

FAQ

Does a fit test replace the medical evaluation? No. The medical evaluation must come first, before any fit testing or respirator use (1910.134(e)(1)). They're separate requirements with separate records and separate retention rules.

If an employee changes from an N95 to a different N95 model, do they need a new test? Yes. Any change in make, model, style, or size of the facepiece triggers a new fit test, independent of the annual cycle (1910.134(f)(2)).

Can I use qualitative fit testing for a full-face respirator? Not for the negative-pressure fit factor a full facepiece needs. QLFT is limited to respirators requiring a fit factor of 100 or less (1910.134(f)(6)); full facepieces requiring a fit factor of 500 must be quantitatively tested (1910.134(f)(7)).

Putting it into practice

Run the checklist for every test: all five data elements captured, the correct result type for the method, medical clearance on file first, and the next-due date set a year out (or sooner if the facepiece or the face changed). Store fit-test records on their one-year clock and medical records on the 30-year clock — never in the same bucket.

The annual-due date is the piece that slips first once a program grows past a couple hundred staff and lives in spreadsheets and reminder emails. A structured occupational-health system like carefoundryESC captures the five required elements per test, flags each employee's annual due date, and keeps fit-test records distinct from the 30-year medical file — so the record is ready before the inspector asks for it.