INTRODUCTION

Sampling for mold remediation is a rather broad topic. Therefore, the article is somewhat more limited in scope than what is implied by the title. The article discusses the need for sampling, distinguishes between “sampling” and “appropriate sampling”, discusses the sampling plan and how to estimate sample size, and then briefly discusses sampling during the various stages of the remediation process.

The initial assumption is that everyone agrees that sampling is required in the first place. However, this may not be the case. The following example illustrates why I believe sampling should always be part of a microbial investigation and remediation in public buildings.

Approximately 16 workers, all located in a particular suite of offices in a high-rise office building, reported having persistent symptoms that are often associated with exposure to mold. Management hired a consultant to perform an initial investigation. The consultant performed a visual inspection, and reported the offices to be clean, well maintained, and without any apparent signs of mold.

The symptoms continued, and a second consultant was hired. The new consultant performed a second visual inspection and collected a number of culturable airborne samples. This consultant did not detect any signs of contamination or visible damage either, and reported that the airborne samples had not detected any significant airborne contaminants. After two investigations, the credibility of those working in that area was being questioned by management.

I was the third consultant hired, and by the time I got on site, the workers were refusing to report for work. I performed a third visual inspection, including not only the offices, but also the air handling unit (AHU) serving the office suite. The visual inspection did not detect any signs of water damage or contamination. However, because of the reported symptoms, I decided to collect a total of 24 surface swab samples in the office suite and the AHU. Sixteen of the settled dust samples were positive for Stachybotrys chartarum, including both samples collected in the AHU. The surface dust samples clearly indicated wide-spread contamination in the offices, and implicated the air delivery system as the source. A simple, easily interpreted result, but it took three consultants to achieve it.

There are several significant lessons regarding sampling and remediation that are illustrated by this example. First, fungal spores are invisible to the unaided eye. A visual inspection can only detect gross damage and colonization, not surface contaminants. If the science of health & safety is one of prudence, then relying solely on a visual inspection to detect mold contamination may not be prudent.

Second, the office suite, the AHU, and the source contaminating the AHU (exterior wall cavities) clearly needed to be remediated, even though the office suite passed three visual inspections. If the appropriate sampling results had not been obtained, a significant source of mold contamination would have gone undiscovered. This would have resulted in a continuing personnel problem for management, and may have resulted in some workers losing their jobs. The continuing water intrusions would also have resulted in a further physical deterioration of the building envelope. Instead, the fact that the appropriate sampling was performed made everyone a winner.

Third, in my opinion, the question is not whether sampling is necessary, but what type of sampling is appropriate for the specific situation. In this example, the choice of sampling method was critical to identifying the problem. The selection of inappropriate sampling methods can result in false negatives (not detecting contaminants when present), obtained at significant expense, and providing little usable information.

THE SAMPLING PLAN

The focus of this article is on how sampling can be used to help define the scope of remediation. The remediation contractor needs to know which building components and systems are affected. In order to provide this information to the remediation contractor, the consultant first needs to develop a sampling plan, then collect and analyze the sample data, and finally, interpret the sample results and prepare a remediation plan, including a scope of remediation.

Consultants collect samples for a lot of reasons; and, they sometimes collect a lot of samples for no reason. This is especially true when investigating microbial incidents in public buildings. An experienced consultant will be able to explain why the sampling is required, how the sample data will be interpreted, and the questions the sampling is designed to answer. However, the first rule of sampling always applies: if you don't know what you are going to do with the data, or how you are going to interpret it, then don't collect the sample.

The preparation of a sampling plan is the first step in sample collection. The sampling plan identifies how the building will be partitioned for sampling. It also discusses in detail the sampling locations, sampling frequency, types of samples, and number of samples to be collected. The plan describes the questions to be answered, how a particular series of samples is expected to answer those questions, and explains how the data for each type of sample will be interpreted.

SAMPLE SIZE

Sample size always seems to end up being a compromise between what the consultant wants and what management can afford. However, there are guidelines that can be used to estimate what is a reasonable sample size. These guidelines are based on the concept of the similar exposure area (SEA).

A SEA is defined by the consultant, hopefully after applying a good bit of common sense. The definition reflects how the consultant intends to partition, or “stratify”, the building for sampling. The SEA is an area within the building where the occupants may be expected to receive similar exposures. For example, an SEA can be the entire building, or each floor, or each zone served by a different AHU, or a particular set of cubicles. As a final example, the windward side of a building may receive more wind-driven rain than the leeward side, and opposite sides of the building could be different SEA's.

Now, we can finally discuss sample size. The author tries to collect a minimum of 4 and a maximum of 10 samples per SEA, with an average of about 6 samples. Collecting fewer than 4 samples in a SEA results in confidence limits that are too wide, and limits the consultant's ability to interpret the data. Collecting more than 10 samples per SEA creates something called diminishing returns. Each additional sample then begins to cost more than the value of the information it provides.

Using this guideline, we can estimate a reasonable sample size that may be required to develop an effective and reliable scope of remediation. However, this is the estimated sample size for a particular type of sample, and should not be confused with the total number of samples that may be collected during the investigation. A number of different types of samples may have to be collected to adequately characterize a large public building.

Let's use a 16 story office building as an example. The building is located in a downtown “canyon”, and the West side of the building is frequently subjected to wind-driven rain. Wet spots are often observed on the drywall on the 10th floor on the West side of the building. A visual inspection reveals the presence of visible mold at various locations on the drywall in the affected areas. The consultant also determines that each floor is served by a different AHU.

The consultant has several options for defining SEA's. How they are defined will affect both the sampling plan and the project budget. In this case, the consultant decides (1) exposures on the10th floor may be different from those on the 9th and 11th floors, (2) exposures in cubicles along the West wall may be different from those along the East wall, and (3) fungal spores may have been drawn into the return air plenum, contaminating the AHU.

This “stratification” results in 5 SEA's: (1) cubicles along the West wall on the 10th floor, (2) other areas on the 10th floor, (3) cubicles on the 9th floor, (4) cubicles on the 11th floor, (5) exterior wall cavities, return air plenum and AHU, and (6) outdoor reference samples. Therefore, the estimated number of air samples, for example, required to characterize 6 SEA's is estimated to be between 24 and 60 samples, with 36 the expected number.

STAGES OF SAMPLING

So far, I have suggested that sampling for microbial contaminants, if appropriate, can be very useful in defining the scope of remediation. I have also briefly discussed how to partition a building for sampling, and how to estimate sample size. Now, let's discuss the various stages of sampling that occur prior to and during remediation. The stages of sampling may be referred to as characterization, remediation, and compliance sampling.

Each type of sampling generally has an identifiable objective. Characterization sampling has the objective of determining if mold is present, where it is located, and which matrices (air, settled dust, papers) contain elevated concentrations of mold. Finally, it should provide information on the potential causes of the mold contamination. Characterization sampling normally occurs within a multi-disciplinary environment, with input from a number of specialties.

It is not unusual to focus on the first source of mold that is discovered in a large public building. However, it is also not unheard of to discover additional sources of mold if one keeps looking.

Remediation sampling as sampling that has the objective of both defining the scope of remediation and implementing the remediation. This stage includes an in-depth survey of building components and systems. It includes the destructive testing of building components in order to identify which building components (exterior walls, carpets, furnishings) may be contaminated, as well as the physical boundaries of the contamination. This sampling is specifically directed towards identifying sources and developing the scope of remediation.

The objective of compliance sampling is to assure both occupants and management that contaminants have not been released from contained work areas. This typically involves airborne sampling inside and outside of containment. The bulk of compliance sampling usually occurs during destructive testing and remediation.

Finally, post-remediation or “clearance” sampling is performed. The objective is to assure the workers, management, and the remediation contractor that the remediated areas of the building are free of unusual types and/or concentrations of mold. If the sample results are determined to be acceptable, the remediated areas are released by the remediation contractor for re-occupancy.