How to read an industrial hygiene report
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.
Most people read an industrial hygiene report, the written record of an air monitoring survey, for two things. One is the number, and the other is whether it says "below the legal limit", meaning OSHA's permissible exposure limit (PEL). Then the report goes in a drawer.
On its own, the number is the hardest part of the report to defend. Everything that gives it meaning sits in the supporting sections, which cover how the sample was collected, what the pump was doing, who the result represents and what you are required to do next. This page is a guide to reading those sections.
What the report has to contain
| Item | Why it matters |
|---|---|
| Rationale for the similar exposure groups (SEGs) | Why these workers were sampled and who they represent. Without it, nobody can tell whether the results cover 6 people or 60 |
| Shift length used for the time-weighted average (TWA) | An 8-hour TWA on a 12-hour operation is a different claim from an 8-hour TWA on an 8-hour shift. Under standards with formulas that lower the limit on longer shifts, this decides the outcome |
| TWA and dose for each worker | Named or coded for each person, not a group average. Compliance is judged person by person |
| Sampling parameters: flow rate, and calibration before and after | Without both calibrations, nobody can verify the sample volume. See below |
| Analytical methods cited by number | NIOSH 7300, OSHA ID-121, and so on. "Analyzed by ICP" names a type of laboratory instrument, not a method |
| Laboratory accreditation | The report should name the laboratory and its accreditation, with the lab's ID. A result from a lab that is not accredited is open to challenge |
| Field notes written at the time | What ran, when, at what rate, with which controls (such as ventilation) operating, and any stoppage. Written on the day, not reconstructed later |
| Comparison against both the action level and the PEL | Two different thresholds that trigger two different sets of obligations |
| Employee notification requirement, and the deadline | Most standards set one. The report should state the deadline, not leave you to find it |
| What the results require next | Frequency of periodic monitoring, medical surveillance, respirators, controls |
Flow rate and calibration
An air sample result is the amount of a substance on a filter divided by a volume of air, and the volume is the flow rate multiplied by the run time. If the flow rate used in the calculation is wrong, the result is wrong by the same proportion, and nothing later can correct it. A pump whose flow was off by 15% for the whole shift, with nobody noticing, gives a result that is wrong by about the same amount.
Calibration confirms the pump's flow rate, which is why it has to be recorded at both ends of the shift:
- Pre-shift calibration: the pump was set correctly before it went on the worker
- Post-shift calibration: it was still at that rate when it came off
A pump that drifts is a different case from a pump that is off by a constant amount. The difference is worth understanding, because the effect of drift is often overstated. Where a pump drifts over the shift, the sample volume is calculated from the average of the pre-shift and post-shift flow rates.
So a pump that started on rate and finished 15% low does not give a result that is 15% wrong. The average rate is roughly 7 to 8% off, and so is the result, which makes the error about half the drift.
That is still enough to move a result across a limit, and you can see it only because someone checked at the end. A report with no post-shift calibration record asks you to assume the pump held its rate. It also gives you no basis for the averaging in the first place.
Standard practice is to void a sample where the pre-shift and post-shift calibrations differ by more than a small allowed tolerance. You cannot apply that rule to data you do not have.
→ How to calculate an 8-hour TWA
A five-question check of the report you already have
Pull the report out of the file. This takes 10 minutes.
1. Can you find the flow rate and both calibration records for every sample? If the post-shift calibration is missing, the sample volume is unverified, and there is no way to check whether the pump held its rate.
2. Does it state how many hours the shift was, and confirm the TWA was calculated against that? If the report gives a TWA with no shift length anywhere in it, ask. On an operation with longer shifts, this can reverse a conclusion.
3. Does it name the laboratory, the accreditation, and the method by number? You need all three. A lab name alone does not prove accreditation, and "ICP analysis" does not name a method.
4. Does it compare results to the action level as well as the PEL, and say what each one triggers? Under most substance-specific standards, exposure at or above the action level triggers periodic monitoring and training.
Medical surveillance comes with one more condition that reports often leave out. The standards for lead, silica and hexavalent chromium set a threshold before medical surveillance is required. For lead, exposure must be at or above the action level on more than 30 days in a 12-month period. For silica and hexavalent chromium, the threshold is 30 or more days, and cadmium and benzene also use 30 or more days. "Below the PEL" answers none of that.
5. Does it tell you your deadline to notify workers and when your next monitoring is due? If it stops at the results table, the report has given you data. It has not given you what you actually needed, which was to know what to do.
If the report fails two or more of these, it will not carry you through a contested inspection. It may still be perfectly accurate, but you have no way to show that, and in an enforcement case or a lawsuit that is the same problem.
What tells you little either way
Length. Page count depends mostly on how much regulatory text is attached. What matters is how much of the report is about your workplace.
Graphs. A bar chart of results against a PEL line is presentation. The evidence sits in the calibration log and the field notes.
A confident tone. A report that flags a limitation, such as a lost sample, an atypical day or a group that needs more samples, is giving you information. A report that flags nothing may have nothing to flag, but it is worth asking.
What we do
A defensible report leads with results for each worker against both the action level and the PEL, and it states the shift length used in the calculation. It carries the full calibration record and the field notes. Where the data does not support a conclusion, the report should say so, and say what it would take to support it.
If you have a report from another firm, send it to us for a second read.
→ Similar exposure groups · → Area or personal samples · → Keeping exposure records · → Sample reports
Send us your last report. We'll tell you what's missing.
Ron Fleming has done this work himself since 1993. You talk directly to the person who does the job. No account managers. No call center. Our prices are well below the competition. Tell us what is going on and we will send a written scope of work and a price, at no cost.
Talk with us. No cost.References
The regulations and standards behind this page, with links to the official text, so you can check the source yourself.
- OSHA 29 CFR 1910.1025(j)(1)(i) Lead
- OSHA 29 CFR 1910.1026(k)(1) Chromium (VI)
- OSHA 29 CFR 1910.1053(i)(1) Respirable Crystalline Silica