- THE MAGAZINE
Q: I have been contacted by a laboratory that says it can teach me how to sample for mold and will help me by interpreting the results. Do you think it is a good idea for laboratories to interpret the results of samples collected by others?
A: Before I answer your question, let me explain the assessment process. The steps in an investigation or assessment include gathering information; formulating hypotheses; testing the hypotheses and, finally, making recommendations. In this scenario, you are being considered the investigator or assessor.
After you have gathered the available information from your client and from the building inspection, you begin to formulate hypotheses about what is occurring in the building. Next you design a sampling strategy or plan. The sampling plan may consist of the numbers of samples needed, types of samples, types of sampling equipment, the media to use and the locations to sample. Each sample should be used to help answer a question or test the hypotheses.
Since you are the assessor, only you know what the hypotheses are and what questions you are trying to answer. It would then follow that only you can interpret the laboratory analysis of your samples.
"ACGIH Bioaerosols: Assessment and Control" makes the following comments:
5.2 Sample Collection
When developing sampling strategies, IHs, EHPs, and IEQ consultants must consider possible sources of error, iden-tify the desired precision and accuracy of the measure-ments to be made, and specify the degree of confidence they need to interpret the results (Gross and Morse, 1996). These decisions are coupled with the practical questions of where, when, and how to measure biological agents as well as how many samples to collect, as discussed in this section. Investigators base their decisions on their training, experience and the approaches other investigators have used successfully in similar settings.
TABLE 5.2. Fundamental Steps When Collecting Environmental Samples to Test Hypotheses
TABLE 5.3. Points to Consider When Designing a Sampling Plan
As you can see, a great deal of thought, based upon training and experience, goes into setting up the sampling plan.
Now, let's answer your question. In an effort to attract new customers and support them, some laboratories are offering to train individuals on how to sample the indoor environment and then offer to interpret the results. The problem is that many of these individuals are learning how to collect a sample without spending the months or years necessary to learn the what, when and where to sample, as well as how to interpret the sample results.
Let's look at a parallel situation. You wouldn't go to a laboratory medical technician to find out why you are sick. Instead, you visit the doctor. The doctor performs an investigation, which includes a medical history and a physical evaluation. The doctor then forms a hypothesis and orders the types of test to be performed. A nurse may collect blood or other data and submit them to the laboratory for analysis, but the results are interpreted by the doctor. Only the doctor has all the pieces of the puzzle plus the expertise necessary for interpreting the data and providing a diagnosis. If the data is inconclusive, the patient is re-evaluated and another hypothesis is tested until the diagnosis is made or the problem is solved.
The same is true with any other form of laboratory data. The interpretation can only be made in the context of the sampling strategy. It's not what you know that gets you in trouble; it's what you think you know, but don't.
The U.S. Environmental Protection Agency has made it clear how important the context of an investigation is for the interpretation of samples. In their document "Mold Remediation in Schools and Commercial Buildings," the EPA states, "Sampling should be done only after developing a sampling plan that includes a confirmable theory regarding suspected mold sources and routes of exposure." They confirm the importance of the investigator having the necessary expertise by stating, "Sampling for mold should be conducted by professionals with specific experience in designing mold sampling protocols, sampling methods, and interpretation of results." as well as, "For someone without experience, sampling results will be difficult to interpret. Experience in interpretation of results is essential."
One reason for using the scientific method for investigations is to help identify experimental errors. This is important in helping to improve the validity of the interpretation. Independent laboratory analysis is often useful in helping to reduce bias or preconceived ideas on the part of the investigator. If the sample results from an investigation are not consistent with the field observations and history of the building, then the situation needs to be reassessed. The laboratory provides an independent analysis of data that either supports or contradicts the investigator's hypothesis.
A hypothesis is an explanation accounting for a set of facts that can be tested by further investigation. It usually refers to the understanding of a situation before experimental work has been conducted. Laboratory data provides only one small part of the interpretation of the building conditions. It is one way of helping to confirm observations and to make the invisible become visible. A negative result doesn't prove the absence of a contaminant. It may mean that additional evaluation and the formation of a new hypothesis are necessary in order to quantify the problems.
The EPA has made some recommendations regarding the performance of mold investigations. They state:
In the computer world there is a saying: "Garbage in, garbage out." These warnings from the EPA say essentially the same thing. If the investigation and sampling plan aren't adequate, then the interpretation will be faulty. Only an experienced investigator can make the determination of the meaning of the data.
The scientific method works best when controlled conditions are observed. Unfortunately for most building-related inspections, there are numerous variables that prevent the extraneous factors from being controlled or eliminated. This is one of the primary reasons interpretation must be performed by a knowledgeable person that understands the limitations of the methods that are being used and the inherent variability of the specific indoor conditions.
Results from the laboratory may have laboratory error or limitations introduced during the analysis. Our company has experienced a number of situations where our quality control procedures have identified problems with laboratory analysis. Numbers have been transposed, oil droplets or latex particles have been mistaken for penicillium/aspergillus organisms on spore traps, and samples have been mixed up. These types of problems can often be identified or at least suspected when the investigator interprets their results in light of their quality control procedures and understanding of the scientific and laboratory process. This check-and-balance is lost if laboratory personnel interpret results.
Recognizing this, most quality laboratories will have a disclaimer for the results they provide that states the interpretation of the data they report is up to the person that conducted the fieldwork. The laboratory is only responsible for reporting what they observe in the samples. Laboratories and investigators that do not adhere to these considerations have greater risk of misdiagnosing the building.
Laboratories training people to sample rarely train them in the recognition of problems with the data provided by that laboratory. A laboratory interpreting the results, in fact, creates an inherent conflict of interest by removing a very important check-and-balance provided by the investigator.
A final consideration is liability. If you perform the investigation, you will be taking responsibility for any interpretations based on the results. It is unlikely you will be able to get errors-and-omissions insurance coverage on the interpretations issued by a laboratory. Mold litigation is common, and a short course in sampling with a laboratory interpreting the results will not protect you against the associated liability.