The baseline audiogram is the single most important record in a hearing conservation program: every Standard Threshold Shift determination for a worker’s entire career is computed against it, and OSHA 29 CFR 1910.95(m)(2) requires it to be retained for the duration of employment. When a baseline cannot be found — after a vendor change, a facility move, a flood, or simply decades of filing drift — the damage is double: STS detection for that worker stops working, and the employer carries a recordkeeping gap that surfaces at the worst possible moments, during an OSHA inspection or a workers’ compensation claim. The correct response is not improvisation. It is a documented investigation, recovery where possible, and a properly re-established baseline where not.
A food processor with 180 noise-enrolled workers switched mobile van providers and, eight months later, ran its first annual cycle on the new system. Twenty-three workers had no baseline on file — the old vendor had held them, the contract had lapsed, and nobody had requested an export. Fourteen baselines were eventually recovered by written request; nine were gone. The nine workers got documented investigations and new baselines. The lesson the EHS manager took away: the loss happened at the vendor transition, but the discovery happened a year of unprotected STS comparisons later.
What a Missing Baseline Actually Breaks
OSHA’s STS methodology compares each annual audiogram to the worker’s baseline at 2000, 3000, and 4000 Hz — never to last year’s test. That design exists precisely to catch slow, cumulative loss that year-over-year comparison hides (the full comparison rule is explained in Baseline vs. Annual Audiogram). Remove the baseline and three things fail at once:
- STS detection stops: without a reference, no shift can be computed, notified within the required window, or evaluated for the OSHA 300 Log.
- The record series loses evidentiary value: in a hearing loss claim, an annual series with no baseline cannot show what the worker’s hearing was at hire — which is the heart of an apportionment defense.
- A recordkeeping deficiency accrues: 1910.95(m)(2) requires the record for the duration of employment, and 1910.95(m)(4) with 29 CFR 1910.1020 gives the worker access rights to a record you cannot produce.
The distinction that matters going forward is between an unexplained gap and a documented, remediated one. You cannot always undo the loss. You can always control which kind of gap you have.
Step 1: Run a Documented Investigation
The moment a baseline is discovered missing, open a written investigation record for that worker. Treat it like an incident investigation, because that is what it is — a record integrity incident.
- Fix the discovery date: when and how the gap was found (annual cycle, digitization project, access request, audit).
- Reconstruct the custody chain: who tested the worker at hire — an in-house program, a mobile van vendor, an occupational health clinic, a predecessor employer? Where did that custodian’s records go?
- Define the search scope up front: list every plausible location before searching, so the record shows a systematic effort rather than an improvised one.
- Log every step with dates: each archive searched, each custodian contacted, each response received.
If the gap surfaced during a broader legacy-records project, fold the investigation into that effort — the gathering inventory described in the historical digitization guide doubles as the investigation’s search log.
Step 2: Attempt Recovery From Every Custodian
Baselines presumed lost are recovered more often than employers expect, because the record frequently exists in a custody location nobody checked. Work the list in rough order of yield:
- Previous testing vendors and mobile van providers: the highest-yield source. Vendors commonly retain test data for years after a contract ends. Make the request in writing, identify the worker and approximate test dates, and ask for structured data plus the original report — the full request playbook is in How to Recover Audiometric Records From a Previous Testing Vendor.
- HR personnel files: pre-employment and baseline audiograms are routinely misfiled into general personnel records rather than the audiometric file.
- Occupational health clinics: if the hire-date physical included a hearing test, the clinic may still hold it under medical-records retention rules.
- Offsite storage and scanned archives: boxed records from facility moves, and shared-drive folders from past scanning efforts.
- Predecessor entities: for workers who arrived via merger or acquisition, the baseline lives in the predecessor’s program — a fragmentation pattern covered in Consolidating Audiometric Records After Mergers, Acquisitions, and Facility Closures.
A recovered baseline should be validated before it is restored to service: legible thresholds at the required frequencies for both ears, a test date consistent with the worker’s hire, and attribution you can defend (name plus a second identifier). A recovered record that passes validation fully restores the original reference — the best possible outcome.
As Professional Supervisor, the re-established baselines I worry about are the silent ones — where a later annual was quietly promoted to baseline with no note in the file. When I review a series like that years later, the worker’s early-career thresholds have simply vanished from the comparison, and any shift in those years is invisible. A file that says “original baseline unrecoverable after documented search; new baseline established under 1910.95(g)(5) conditions on this date” is one I can defend clinically. A silent substitution is one I cannot.
Step 3: Re-Establish the Baseline Properly
When the investigation concludes the original baseline is unrecoverable, the worker needs a new one — established with the same rigor as a hire-date baseline:
- Quiet period: at least 14 hours without workplace noise exposure before the test, or hearing protection worn for that period, per 1910.95(g)(5).
- Valid test conditions: calibrated audiometer, compliant ambient levels, qualified examiner — the record content requirements of 1910.95(m)(2) apply in full.
- Explicit designation: the new test is recorded as a re-established baseline effective on its date, with a pointer to the investigation record — not backdated, not blended with the missing original.
- Professional oversight: have the audiologist or physician supervising your program review the designation, especially if the worker’s surviving annuals suggest a shift may have occurred before the reset.
Be honest about what a reset does: it protects the program going forward, but shifts that occurred before the reset will not appear in future STS math. That is a real limitation — and it is exactly why the documentation in the next step matters.
Step 4: Document the Gap So It Stays Defensible
The investigation file is what converts a violation-shaped hole into a managed exception. Keep, permanently, in the worker’s audiometric record:
- The gap statement: which record is missing, the period it covered, and when the gap was discovered.
- The search log: every source checked and custodian contacted, with dates, plus copies of correspondence with former vendors or predecessor entities.
- The outcome: what was recovered, what was not, and the basis for concluding the record is unrecoverable.
- The corrective action: the re-established baseline’s date and conditions, and any surviving partial history retained alongside it.
This is the file an OSHA compliance officer, a records-access requester, or defense counsel sees years later. It demonstrates the two things that matter: the loss was not concealed, and the remediation was competent.
Preventing the Next Lost Baseline
Nearly every lost-baseline story traces to one of three events: a vendor transition without a full export, a facility move or closure, or decades of unmanaged paper. All three are custody failures, and all three are preventable:
- Never leave sole custody with a vendor: maintain your own complete copy of every audiogram, and export everything in structured form before any transition — the core warning of Thinking of Changing Your Audiometric Testing Vendor in 2026?
- Digitize legacy paper before it fails: fading originals become unrecoverable on their own schedule; see Fading Paper and Thermal-Print Audiograms.
- Index by employee in a managed system: a baseline that lives in a structured, backed-up, employee-indexed archive with retention controls does not get lost to a filing error or a departed administrator.
If you are unsure how many baselines you are actually missing, that is answerable without a contract: Soundtrace’s historical records audit takes your archive in any format, digitizes and indexes every record by employee, and surfaces exactly which workers lack a locatable baseline.
- OSHA 29 CFR 1910.95 — Occupational Noise Exposure
- OSHA 29 CFR 1910.1020 — Access to Employee Exposure and Medical Records
- OSHA 1910.95 Appendix F — Age Corrections to Audiograms
- CDC/NIOSH — Noise and Hearing Loss Prevention
- NIOSH Criteria for a Recommended Standard: Occupational Noise Exposure (98-126)
Frequently Asked Questions
OSHA 1910.95(m)(2) requires retention for the duration of employment, so an unproducible baseline for a current worker is a recordkeeping deficiency. In practice, the difference between a citable failure and a managed exception is the documented investigation and a properly re-established baseline.
Not informally. A silent substitution contaminates every future STS comparison and hides any shift that occurred before it. If the original is truly gone, establish a new baseline properly — quiet period, valid conditions, explicit designation — and document why.
Prior testing vendors and mobile van providers, HR personnel files, occupational health clinics from the hire-date physical, offsite storage and scanned archives, and predecessor entities for acquired workers. Log every source and outcome — the search record is part of the defense.
It resets the reference going forward, so pre-reset shifts will not appear in future STS calculations. That limitation is why the investigation documentation matters: it shows the reset was a last resort, not a way to bury an inconvenient history.
Discovery date, the custody chain reconstruction, every source searched and custodian contacted with dates and outcomes, correspondence copies, what was recovered, and the corrective action — typically the new baseline’s date and conditions. Keep it permanently in the worker’s file.
A recovered prior-vendor audiogram can fully restore the original baseline if it is legible, complete, and attributable to the worker. A previous employer’s audiogram may be used at the current employer’s discretion where timing requirements are met, though many professionals prefer a fresh baseline for a new employment relationship.
Keep your own complete copy of every record, export everything in structured form before any vendor transition, digitize fading paper before it fails, and store baselines in an employee-indexed system with managed retention. Vendor changes and facility moves are where baselines die — both are custody problems with custody solutions.
Find out which baselines you’re actually missing.
Send your historical records in any format and get every audiogram digitized, indexed by employee, and every missing baseline surfaced, with no contract required — the historical records audit from Soundtrace.
Start Your Historical Records Audit- Digitizing Historical Audiogram Records by Employee: Complete Guide
- How to Recover Audiometric Records From a Previous Testing Vendor
- Consolidating Audiometric Records After Mergers, Acquisitions & Closures
- Baseline vs. Annual Audiogram: What’s the Difference?
- Thinking of Changing Your Audiometric Testing Vendor in 2026?