Smart Sampling

Industrial hygienists are targeting their sampling more carefully than in the past as priorities and resources change.

Reflecting on his active practice as an industrial hygienist in the early "70s and "80s, Steve Levine, CIH, professor of industrial hygiene at the University of Michigan, remembers how sampling used to be done. "Back then, the danger to life and health from air contaminants was immediate and egregious. You walked into a plant, and workers were practically falling over. The rule was that when you sampled, you were trying to analyze the cause of evident health effects, so you did as many samples as you could."

In general industry, the practice of sampling is less frequent because most hazards of the past have been identified and controlled. However, sampling is continually rising in the construction industry because contractors are more aware of the ever-changing hazards on construction sites.

The identification of many exposures has led to a decrease in staff among many companies in general industry. Previous knowledge of processes have prompted industrial hygienists to use strategies to design their statistical analysis, rather than simply sampling everything. Likewise, in construction, less blanket sampling is being done because tasks differ from day to day. Although methods may differ among industries, all industries continue to sample because of the return on their investment -- the safety and health of workers.

Different Numbers

Today, sampling is still done to determine the causes of serious health effects, but that is not the only reason.

"Companies are sampling today for their quality programs, such as ISO 9000, for liability reasons and for health and safety issues," said Debbie Dietrich, corporate industrial hygienist and vice president of sales and marketing for SKC Inc., Eighty Four, Pa.

The reasons for sampling are not the only difference. In some industries, the numbers of samples taken have also changed. Alice Farrar, CIH, senior vice president and national director for Clayton Environmental Consultants Inc., Kennesaw, Ga., said less sampling is being done in plants because most industries have already characterized their major workplace exposures.

"The sampling we do today for clients is done to ensure that companies are maintaining exposures below acceptable levels and to characterize any new exposures from any new operations or changes in operations a company has made," said Farrar.

Howard Cohen, associate professor of occupational safety and health at the University of New Haven, agreed that there is less sampling in some industries because many exposures have been well-categorized. "Direct reading instrumentation has become more prevalent than in the past, and it keeps improving; so historically, the number of samples that have been submitted to labs for analysis have been going down," said Cohen.

The OSHA Salt Lake Technical Center has seen a drop in the number of samples submitted for analysis in the past 10 years.

"This year we received about 25,000 samples, compared to our all-time high a few years ago of about 40,000 samples," said Keith Motley, the center"s deputy director.

Reflective of this decrease in submitted samples is the number of staff at the center. "In the "70s, there was a lab staff of about 100 people. Today, the staff is made up of 65 technicians," said Motley.

Different Industries

In most manufacturing facilities, exposures do not change. Because there is less change, many industrial hygienists in general industry have been able to focus efforts more on analyses of exposures rather than identification.

For example, at Ford Motor Co., the process of monitoring has changed. "In the past, we conducted a detailed analysis of every operation to identify exposures; however, in recent years, we have cut back on those analyses. We rely on historical information taken over a period of time to maintain exposures," said Hank Lick, manager of industrial hygiene at Ford.

Identification and control of major exposures has also led to the decrease in IH staffs in some general industry facilities. For example, there are seven industrial hygienists available in the field to conduct sampling for the 100 Ford locations across the United States. In the past, said Lick, there was twice that number.

Ford samples for approximately 60 items in its routine sampling table. Lick said his company used to do more blanket sampling of all facilities. Now, program sampling is more selective. "Most of our sampling is done more out of employee concern or because of the implementation of a new process," said Lick.

In contrast, some experts believe processes in smaller industries still aren"t sufficiently monitored. Automotive body repair is one example of a small business where workers are exposed to a large number of chemicals with no baseline data to categorize the exposures.

"Usually, these operations are conducted by small employers with only five or six employees. These smaller employers just don"t have the money it takes to characterize these exposures," said Cohen.

Sampling frequency may have slowed in general industry, but the opposite is happening in the construction industry.

Some observers believe sampling is on the upswing in construction because contractors are more aware of continually changing hazards on a construction site.

"In general industry, the exposure today is the same as the exposure tomorrow. That is not the case in construction. Here, the hazards are changing every day," said Donald J. Garvey, senior construction industrial hygienist with St. Paul Fire and Marine Insurance Co., St. Paul, Minn.

Doing air and exposure monitoring is also increasingly more important because many contractors are including it in their contracts.

Garvey said there are still some fly-by-night contractors who are less concerned about things such as monitoring, but large contractors have reputations to maintain. That reputation is built, in part, on a quality safety and health program.

"It is in a contractor"s best business interest to go into a bid and say, "Look at the safety and health program we have. We do air monitoring, fall protection and all the things needed to keep workers safe and healthy,"" said Garvey.

In contrast to companies in general industry that have limited IH staffs, some large construction contractors are increasing their staff to conduct monitoring tests by developing safety and health departments.

"In the past, it was rare to see a contractor with a safety and health director, much less a department. Now, we work with contractors who have 10 to 12 safety people who are specifically concerned with the safety and health of those working on a site," said Garvey.

A Different Strategy

Whether sampling is done to analyze the health effects on employees, for OSHA compliance or for a company"s quality program, sampling strategies are better designed than in past years.

In the 1970s, said Levine, "We didn"t need to use a strategy because as soon as you walked into a plant, the health effects were so obvious that you were simply documenting serious situations."

Today, sampling strategies involve more planning.

Defining the purpose of the measurement is the first step in developing a sampling strategy.

Susan Raterman, CIH, a consultant with The Raterman Group in Chicago, first develops a sampling strategy for a project through an analysis of the potential problems. She develops a hypothesis by doing a building walkthrough and conducting interviews.

"Setting up your strategy should be determined by discovering why you are sampling. Ask questions such as, Is sampling being done for OSHA compliance? or Is sampling being done to give a client an idea of contaminants in their building?" said Raterman.

Similarly, Ford"s sampling strategy is developed through an exposure assessment. Lick said that this could be as simple as a plant walk-through or as detailed as a complete look at every operation in a given facility.

"We take a look at operations and classify them into job families. Then we look at all of the qualitative indicators in the area that we are sampling, such as the kind of ventilation. Then we come up with a hypothesis and sample to come up with an answer," said Lick.

In the construction industry, where a worker"s day could be made up of a variety of tasks, Garvey said a concept called task-based monitoring is used.

"In the old days, things were different. For example, if a worker ran the Bobcat for three hours, then the jack hammer for two hours and then did some grinding for a couple hours, you would just set the monitor on him and let it run the whole day," said Garvey.

This process worked, according to Garvey, because it gave industrial hygienists knowledge about the worker"s exposures when he performed each task. However, this became a problem when the worker"s tasks changed the next day, so task-based monitoring became more popular.

"Instead of taking one big sample, task-based monitoring allows individual measurements to be taken during each task. Then, through simple mathematical manipulations, it is easy to predict what the exposure may be for any combination of those operations," said Garvey.

Although all processes are different and a single sampling strategy cannot be established for all situations, Farrar said several factors should be taken into consideration in all types of sampling situations:

Where to sample. If the purpose of sampling is to evaluate a workers" exposure, it is necessary to collect samples at, or as near as practical to, the worker"s breathing zone. If the purpose is to define a potential hazard or to obtain data for control purposes, samples would normally be collected in the vicinity of the source.

What to sample. Material safety data sheets are the first source of information an industrial hygienist should consider when determining what to sample, followed by container labels. Byproducts, intermediates and final products formed from raw materials should also be looked at as a source of potential hazards.

Whom to sample. Estimation of the potential health hazards to employees requires that an initial judgment be made. Farrar said this determination is usually a simple one based on factors such as chemical, physical and toxic properties of the substance being processed, the size of the workplace, the amount and type of ventilation, and the proximity of the employee to the source of contamination.

When to sample. Plants located in areas where large temperature differences occur during different seasons of the year should be sampled during summer and winter months. It is also important to account for the fact that a plant may have more than one shift. If this is the case, samples should be collected during each shift.

How long to sample. The volume of air sampled and the duration of sampling is based on the sensitivity of the analytical procedures or direct reading instrument, the estimated air concentration and the OSHA standard, threshold limit values or other exposure guidelines used for the particular contaminant.

How many samples. Farrar said there is no set rule to determine the number of samples that are necessary to evaluate a worker"s exposure. The number of samples taken depends on the purpose of the sample.

Number of Samples

As a consultant and lecturer on the topic of sampling, Dietrich said the No. 1 question she is constantly asked is, "How many samples do I need to take?" Her best advice is a reference book, Strategy for Assessing and Managing Occupational Exposures, published by the American Industrial Hygiene Association (AIHA). "This is the most thorough reference to follow when determining sample amount. It has flow charts and specifically gives guidelines on the appropriate number of samples," said Dietrich.

Dietrich pointed to a key concept in the book called, "similar exposure group," that explains the proper sample amount. She said that generally, once this group has been identified, less than 10 samples will typically allow for the identification of their exposure levels.

Farrar said professional judgment is still the predominant driving force behind the selection of the number of exposures monitored. However, she noted that in the realm of statistical sampling strategies, sufficient numbers of random samples must be taken to perform statistical tests.

In this case, Farrar also advocates use of the AIHA guide. "If you are going to develop a defensible sampling strategy, you have to get into the statistics of sampling in order to have a valid study. The best way to do this is to familiarize yourself with the existing AIHA protocol for establishing a sampling strategy and then apply it to your particular industry," said Farrar.

TAGS: Archive
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.