One-Round IH Sampling Cannot Defend a Controlled Exposure Call
← Back to Blog Industrial Hygiene

One-Round IH Sampling Cannot Defend a Controlled Exposure Call

Why a single round of industrial hygiene sampling fails AIHA's Bayesian decision analysis test, and what a defensible multi-round assessment actually looks like for a small manufacturer.

by iSi Environmental

If you have ever told a plant manager “we sampled once and we were under the action level, so we’re good,” and that statement made it into a written report — you have already paid a cost. You just have not been billed yet.

That is the realization frame for this post, and it is not rhetorical. The American Industrial Hygiene Association’s published acceptance criterion for a “controlled” exposure is that the 95th percentile of the similar exposure group’s distribution sits below the occupational exposure limit with at least 70% confidence. That criterion is mathematical, not aspirational. You cannot calculate a 95th percentile from a single measurement. You cannot calculate a confidence interval, a geometric standard deviation, or a Bayesian posterior over AIHA’s exposure categories from n=1. The single-round assessment generates data; it does not generate a defensible decision.

The dollar math is the part small manufacturers tend to learn the hard way. A serious OSHA citation runs $16,550 per instance as of 2026, and a willful or repeated citation runs up to $165,514 per instance under 29 CFR 1903.15 (https://www.osha.gov/penalties). iSi’s median project industrial hygiene engagement is $4,050. A multi-round AIHA-grade exposure assessment with six to ten samples per similar exposure group typically prices in the $4,050–$15,000 range. Compare that to a multi-instance citation stacked across two or three workers and two analytes, and the cost ratio is usually somewhere between 5x and 40x in favor of doing the assessment correctly the first time.

This post explains why one round of sampling fails the decision analysis test, what AIHA’s Bayesian framework actually requires, and what a defensible exposure assessment looks like for a small business that is not staffing a full-time industrial hygienist.

What Does OSHA Actually Require for Exposure Assessment?

OSHA’s general air contaminants standard, 29 CFR 1910.1000, sets permissible exposure limits and specifies that PEL determinations come from breathing-zone air samples (https://www.osha.gov/laws-regs/regulations/standardnumber/1910/1910.1000). The standard does not, however, prescribe a minimum number of samples. There is no “OSHA says you must take six.” The duty is functional: the employer must demonstrate that the assessment was representative of the exposure being characterized.

The OSHA Technical Manual, Section II, Chapter 1, lays out the agency’s expectations for sampling strategy. Inspectors are directed to consider whether the employer’s monitoring covered the relevant tasks, shifts, and worker populations (https://www.osha.gov/otm/section-2-health-hazards/chapter-1). When OSHA evaluates a contested citation involving exposure assessment, the inspector evaluates the adequacy of the assessment as a separate question from the measured numbers. The first question is not “what did you measure?” — it is “how do you know your measurement is representative?”

A single sample taken on one worker, on one shift, on one day, almost never answers that question. Industrial hygiene exposures are lognormally distributed with geometric standard deviations commonly between 1.5 and 3.0 (NIOSH Occupational Exposure Sampling Strategy Manual, Pub 77-173, https://www.cdc.gov/niosh/docs/77-173/default.html). On a GSD of 2.5 — typical for many manufacturing operations — the 95th percentile sits roughly 4x the geometric mean. A single sample at 25% of the OEL gives you essentially no information about whether your 95th percentile is at 50% of the OEL or at 200% of the OEL.

That ambiguity is the citation gap.

What Does AIHA’s Decision Analysis Framework Actually Require?

AIHA’s “A Strategy for Assessing and Managing Occupational Exposures,” now in its 4th edition, is the field’s standard reference for how to structure exposure assessment defensibly (https://www.aiha.org/education/marketplace/strategy-book-4th-edition). The framework rests on three concepts:

  1. The Similar Exposure Group (SEG): workers grouped by task, process, or job function whose exposures share a common distribution.
  2. The exposure profile: a statistical characterization of that group’s exposure distribution — central tendency, variance, and an upper tolerance limit.
  3. The exposure category: a label assigned to the SEG based on where its 95th percentile sits relative to the OEL.

The Bayesian decision analysis (BDA) implementation, formalized by Hewett, Logan, Mulhausen and others in 2006, defines five categories (https://pubmed.ncbi.nlm.nih.gov/16998991/):

  • Category 0: true 95th percentile less than 1% of the OEL — negligible exposure
  • Category 1: less than 10% of the OEL — highly controlled
  • Category 2: less than 50% of the OEL — well controlled
  • Category 3: less than 100% of the OEL — controlled
  • Category 4: at or above 100% of the OEL — poorly controlled

Decisions are made on the probability distribution across these five categories given the available data. AIHA’s published threshold for an acceptable, controlled exposure is that the probability the SEG sits in Category 4 is below 30% — equivalently, the upper tolerance limit (UTL95/70) sits below the OEL with 70% confidence (AIHA Principles of Good Practice, Section 2: Exposure Assessment Strategies — https://aiha-assets.sfo2.digitaloceanspaces.com/AIHA/resources/Get-Involved/Pages-from-AIHA-Guideline-Foundation-Principles-of-Good-Practice_Section2.pdf).

That is the criterion an AIHA-trained industrial hygienist is operating against. It is also the criterion an opposing expert witness will operate against if a citation reaches contest.

How Many Samples Does the Decision Analysis Actually Need?

AIHA’s published guidance is explicit. A baseline exposure profile for a SEG should rest on six to ten samples, with n=6 the practical threshold where the 70% upper confidence limit on the 95th percentile becomes inferentially meaningful. AIHA goes further: under the AIHA Exposure Control Category decision rules, an SEG should default to Category 4 (“poorly controlled”) if there are fewer than six measurements and a single measurement exceeds the OEL. The thin-data SEG does not get the benefit of the doubt — it gets assigned to the worst category until additional data refute that assignment.

This rule is the one that surprises small-business owners. It is not “we got an over-exposure on one of three samples — let’s call it borderline.” It is “we got an over-exposure on one of three samples — under AIHA’s own decision rules, that SEG is Category 4 until we generate the data to prove otherwise.” The default position protects workers; it does not protect employers who declined to invest in a proper assessment.

Bayesian analysis can technically operate on a single sample. The Hewett 2006 paper notes that BDA can compute a posterior distribution from n=1. What it cannot do is produce a posterior that is dominated by your data instead of by your prior. With n=1, the posterior is essentially the prior with a faint tilt. That is statistically real, but it is not what a reasonable safety manager — or a regulator — means by “we assessed the exposure.”

What Does an OSHA Inspector Actually Do With Single-Round Data?

The honest answer is: the inspector reviews the assessment for adequacy, not just for results.

OSHA’s Sampling and Analysis program guidance directs inspectors to evaluate whether the employer’s monitoring program is representative of actual exposures (https://www.osha.gov/sampling-analysis). The Industrial Hygiene Field Operations Manual (Navy IHFOM, widely referenced as a procedural template) instructs IHs to compute representative TWAs and to consider geometric standard deviation when characterizing a workforce’s exposure (https://www.med.navy.mil/Portals/62/Documents/NMFA/NMCPHC/root/Industrial%20Hygiene/IHFOM_CH3_220909.pdf).

In a contested-citation context, the employer’s expert is typically asked one or more of the following:

  • What is the geometric standard deviation of the SEG?
  • How was the SEG defined, and on what basis?
  • How many samples support the controlled-exposure determination?
  • What is the upper tolerance limit on the 95th percentile?
  • What is the Bayesian posterior probability that this SEG is Category 4?

A single-round assessment cannot answer those questions. The inspector and the agency know it cannot. The defense collapses before the measured concentrations are even discussed. The citation that follows is not technically about the exposure — it is about the inadequacy of the assessment that was supposed to characterize the exposure.

This is the realization moment for the small-business buyer. The penalty is not “we exceeded the PEL.” The penalty is “we did not know enough to claim we did not exceed the PEL.” The dollar exposure is the same either way.

What Does a Defensible Multi-Round Assessment Cost?

Sampling cost has three layers: planning and judgment (the IH’s professional time), field collection (sampling pumps, media, IH on-site time), and laboratory analysis (per-sample, per-analyte). A representative cost stack for a single SEG, single analyte, six-sample baseline assessment at AIHA-accredited lab pricing typically runs:

  • IH planning, sampling, write-up: $2,500–$5,000 depending on site size and analyte
  • Lab analysis: $30–$120 per sample depending on analyte (six samples typically $180–$720 total)
  • Decision analysis report (IHSTAT or IHDA-AIHA output, signed interpretation): bundled with the IH engagement

Total project cost typically lands at iSi’s median IH engagement figure of $4,050 for a single SEG, single analyte assessment, and scales to roughly $8,000–$15,000 for a multi-SEG, multi-analyte program covering an entire production area.

That is the budget that buys defensibility. The “one round, three samples, $1,200” budget that some firms quote buys data without buying the decision. That is not a discount — it is a deferral of cost into the citation that gets written when an inspector eventually reviews the assessment.

What Should a Small Manufacturer Do?

Four steps, in order.

Step 1 — Define your similar exposure groups before you collect anything. SEGs are defined by task, process, or job classification. A welder in fabrication and a welder in maintenance are usually different SEGs. Defining SEGs after sampling is a defensibility problem; defining them first is the methodology AIHA specifies.

Step 2 — Plan for six to ten samples per SEG, per analyte, in the baseline assessment. This is not gold-plating. It is the floor for a Bayesian decision analysis that an opposing expert cannot dismantle.

Step 3 — Run the data through IHSTAT or IHDA-AIHA, both free AIHA tools, and document the decision-probability distribution. The deliverable is not the measurements; it is the decision the measurements support, with the probability attached (https://www.aiha.org/public-resources/consumer-resources/apps-and-tools-resource-center/aiha-risk-assessment-tools).

Step 4 — Establish a re-sampling cadence based on the assigned exposure category. AIHA’s framework is iterative. A Category 1 SEG (highly controlled) needs less frequent re-monitoring than a Category 3 SEG (controlled, but borderline). The strategy is not “sample once and be done” — it is “sample, decide, re-sample on a defensible cadence based on what the data showed.”

For a small manufacturer without a credentialed industrial hygienist on staff, this is where iSi’s industrial hygiene consulting fits. iSi’s IH consulting service is built around multi-round AIHA-grade exposure assessment — SEG construction, baseline sampling at the n=6+ threshold, IHSTAT/IHDA-AIHA decision analysis, written report, and a re-sampling cadence sized to your operation. The median engagement runs $4,050; comprehensive multi-SEG programs run $8,000–$15,000 depending on analyte count.

Compare that to one willful citation at $165,514 under 29 CFR 1903.15 — or even three serious citations at $16,550 each, stacked across workers and analytes — and the cost-of-doing-the-assessment-correctly answer becomes structural, not optional.

The Bottom Line

The OSHA standard does not tell you how many samples to take. AIHA’s decision analysis framework does. An n=1 assessment cannot produce a defensible “controlled exposure” determination because the methodology requires variance, and variance requires more than one observation. Under AIHA’s own decision rules, an SEG with fewer than six samples and one over-OEL reading defaults to Category 4 — not because the data prove poor control, but because the data are too thin to refute it.

The cost of doing the assessment correctly is roughly $4,050 to $15,000. The cost of not doing it correctly is roughly $16,550 to $165,514 per cited instance, plus corrective-action expense and operational disruption. The math points one direction.

If you have made a controlled-exposure call on a single round of sampling — for any analyte, any SEG, anywhere in your operation — that decision is sitting in a file somewhere as defensible as you have made it. iSi’s IH consultants build the multi-round assessment that lets you defend that file when the inspector arrives. To start: call iSi at (316) 264-7050 and ask for an AIHA-grade exposure assessment scoped to your similar exposure groups.

Sources

Tags: