Assessing exposure

Respirator selection depends on a number: the concentration the worker would breathe without protection. That number comes from an exposure assessment. Without it, selection is guesswork, and guesswork in respiratory protection tends to fail in one of two ways: a respirator that is too weak, or an expensive, uncomfortable respirator that workers will not wear because it is far stronger than needed. This lesson explains how exposures are assessed, how air sampling and direct-reading instruments work, how to calculate a time-weighted average and how to deal with uncertainty.

The exposure assessment process

An exposure assessment is a structured process, not a single air sample. Professional occupational hygiene practice, reflected in guidance from bodies such as the American Industrial Hygiene Association and the occupational hygiene associations in Canada, follows a cycle that looks like this.

Similar exposure groups are an important idea. Sampling every worker every day is impossible, so hygienists identify groups whose exposures should be alike, such as all the welders on a production line, and sample enough of them to describe the group. If one worker in a group turns out to be very different, the grouping is wrong and needs revisiting.

Personal air sampling

The reference method for comparing exposure with a TWA limit is personal breathing-zone sampling. A small battery-powered pump worn on the worker's belt draws air at a known flow rate through a collection device clipped near the collar, within about 30 centimeters of the nose and mouth. After the sample, the collection device goes to an accredited laboratory, which measures the mass of contaminant collected.

The concentration is calculated from the mass collected and the volume of air sampled.

Here \(m\) is the mass in milligrams, \(Q\) is the flow rate in cubic meters per minute and \(t\) is the sampling time in minutes. Pumps are usually set in liters per minute, so remember that 1,000 liters equal 1 cubic meter. Pumps must be calibrated before and after each sample with a primary flow calibrator; if the flow changes by more than about 5 percent, the sample may be invalid.

Calculating an 8-hour TWA

A full-shift sample is ideal, but workers often move between tasks and sample media sometimes need changing during the shift. The TWA is calculated by weighting each concentration by its duration.

Always divide by 8 hours for an 8-hour limit, even if the worker worked less, and account for any unsampled time. If the worker was exposed during unsampled time, you need a reasonable estimate; if the worker was genuinely unexposed, for example in a clean lunchroom, that period counts as zero.

Direct-reading instruments

Direct-reading instruments give results in real time. They are essential for some decisions, such as whether a confined space is safe to enter or whether an area is above an immediately dangerous level, and they are very useful for finding sources and peaks. Common types include multi-gas monitors with oxygen, flammable gas, carbon monoxide and hydrogen sulfide sensors; photoionization detectors for many organic vapors; electrochemical sensors for specific gases; and optical aerosol monitors that estimate particle concentrations.

Direct-reading results are usually not a substitute for laboratory-analyzed personal samples when comparing with TWA limits, but they are often the best tool for short-term limits, ceilings and emergency decisions. Instruments must be bump-tested before each day's use and calibrated as the manufacturer specifies, and users must be trained to understand cross-sensitivities. A carbon monoxide sensor that also responds to hydrogen, for example, can give misleading readings near battery charging.

Estimating exposure when you cannot measure first

Respirators often have to be chosen before sampling results are available, for example on the first day of a new task or for a one-off maintenance job. In that case the assessment relies on other sources: previous sampling from similar tasks, published data, trade association guidance, manufacturer data and simple models. When the estimate is uncertain, the respirator choice should be conservative until data are available.

One simple model estimates the steady-state concentration in a well-mixed room from the rate at which a contaminant is released and the ventilation rate.

Here \(G\) is the generation rate (for example in mg/min) and \(Q\) is the effective ventilation rate (m³/min). Real rooms are not well mixed, and concentrations near the source can be many times higher, so this model gives a lower bound for the worker at the source. It is useful mostly for ruling things in, not out.

Dealing with variability

Exposures vary a great deal from day to day, even for the same worker doing the same job. Occupational exposures typically follow a lognormal distribution: most days are moderate, but a few days are much higher. A single sample tells you little about that distribution. A single result at 40 percent of the limit does not mean the worker is always below the limit; on some days the worker may be well above it.

Professional practice compares a statistic, such as the estimated 95th percentile of the group's exposures, with the limit. Many programs use these rules of thumb: if the 95th percentile is clearly below the limit, exposure is acceptable; if it is near or above the limit, more controls are needed. Statistical tools for this are freely available, but the key idea for supervisors is that a few results near the limit signal a problem, not a pass.

Using results for respirator selection

For respirator selection, the number that matters is not the average exposure but a reasonable upper estimate of the concentration the worker could face during the task. Programs commonly use the highest measured value for a task, an upper percentile, or a modeled worst case. Selection must also consider short-term peaks, because a respirator chosen against the TWA may be overwhelmed during a peak that is 10 or 20 times higher.

Exposure assessment is not a one-time event. Reassess when materials, processes, equipment or controls change, when workers report symptoms or odors, when medical surveillance shows a problem, when limits change, and periodically even when nothing seems to have changed. Keep records: who was sampled, what task, what controls and respirators were in use, the result and the laboratory report. Those records support selection decisions and may be needed decades later for occupational disease claims.

In the United States, 29 CFR 1910.1020 gives workers and their designated representatives the right to access exposure and medical records, and exposure records must generally be kept for 30 years. Substance-specific standards add requirements to notify workers of monitoring results within a set number of days.