Putting Your Asbestos Lab Under The Microscope.

By

Background

Accurate analysis of asbestos bulk samples is one of the most important elements of a successful asbestos management program. Analytical results greatly impact response actions which in turn affect potential exposure to people, liability and costs. A facility owner who views this function as a simple task that any one can perform, could very well be in for a rude awakening. If you haven't tested your favorite lab recently for how well they analyze asbestos, as well as asbestos replacements, you could be leaving a lot open to chance. Putting your lab under the microscope is key to increasing your comfort level. The following discussion provides a current snapshot about asbestos lab services, some thoughts on how to know if your lab provider is proficient and ways to insure quality. The advice in this white paper is based on over 13 years of experience in dealing with hundreds of labs, most notably when I was the DuPont Company's Corporate Asbestos Program Manager.

A Little History

In recent years, labs performing Polarized Light Microscopy (PLM) of asbestos bulk samples have reduced in number as well as quality. In the mid to late 1980's, at the height of the Asbestos Hazard Emergency Response Act (AHERA) which required schools (Kindergarten through 12^thgrade) to inspect for asbestos, labs flourished. In 1988, there were a little over 1100 labs providing PLM service. Today there are 343 labs ¹ that are accredited to do asbestos PLM lab work. Many labs were forced to branch off into other areas such as lead and radon analysis to stay afloat. As such, there are far too few labs that are considered specialists in this area. During one of the National Voluntary Laboratory Accreditation Program's (NVLAP) round of proficiency samples, over 90 of the 343 labs (approximately 27%) failed. ² And that, in 1996, number does not include the labs who passed but still received points for analytical and procedural errors. More recently, in 1998, 20% of 311 labs missed a fairly simple asbestos floor tile sample in the spring round. As a follow up to this round, NVLAP sent out a ceiling tile sample that contained 4% asbestos and an anthophyllite (one of the 6 asbestiforms) samples. In this test, 29% of the 311 labs failed. And the one sample (the anthosphyllite sample) was a reference sample that every lab has. Needless to say, quality of results is an issue that owners really need to focus on in their asbestos management efforts.

Accredited Programs

Most laboratories who perform PLM analysis of asbestos bulk samples participate in an accredited program. In 1979, EPA began the Interim Laboratory Accreditation Program. This was an early attempt to bring some standards to a young industry and monitor proficiency. Its' successor was and still is the National Institute of Standards and Technology (NIST) NVLAP program. This was established in 1988. For a hefty fee ³ , this government agency performs in-field lab inspections every other year. However, every year, they send out 2 rounds of 4 proficiency samples for testing purposes. Also in 1989, the American Industrial Hygiene Association (AIHA), began sending 4 rounds of 4 proficiency test samples on a yearly basis to participating labs. In various asbestos regulations, EPA and OSHA both acknowledge the importance of selecting a lab that participates in at least the NVLAP program.

It's important for owners to note, however, that accreditation does not necessarily mean quality. Both NVLAP and AIHA ⁴ allow labs to miss samples and still remain accredited. To stay NVLAP accredited, a lab must not fail any 2 out of 3 successive test rounds. NVLAP uses a point system in their approach. Zero points equals excellence while 150 points equals failure. Suffice to say there are many labs with a history of points near the 150 failure mark analyzing samples. In other words, there are labs potentially providing owners with inaccurate results every day. If a lab provides either a false negative or false positive, owners are likely to pay the price in more ways than one. The price of potential exposure and a subsequent lawsuit is probably the biggest price to pay when a lab inaccurately says there is no asbestos present. Likewise, the cost of a false positive can trigger expensive asbestos work practices as well as expensive landfilling requirements. Treating non-asbestos as asbestos is a costly proposition.

Lab Services: Let The Buyer Beware

Owners are also likely to pay the price if a lab charges too much or too little for bulk sample analysis. In today's marketplace, you're paying too much per sample if it's over $15. Simarly, you're not paying enough if it's lower than $10 per sample. There is just no way to do a quality analysis of a bulk sample for less than $10.

There's a new phenomenon among labs that owners should be aware of. It's called "layered pricing." Some labs are promoting $5 and $6 samples. Some labs quote prices based on each layer found in a bulk sample while other labs only tell owners after the fact about their per layer pricing schedule. Make sure you understand how your lab charges for layered samples before you hire them. Since most Thermal System Insulation materials (TSI) come layered, it is likely you will run into this situation. The norm is to find at least two layers (a cover and base material). It is also fairly common to find 3 and 4 layers of materials on older piping and vessels in industry. There is no reason for any owner to be paying for bulk samples by layer, since a quality lab will always report and define asbestos containing layers within a total sample.

Bigger May Not Be Better

While there are some exceptions to this next statement, it has been my experience that the bigger the lab, the poorer the quality. Some laboratories with national reputations in environmental, health and safety issues just don't analyze asbestos that well. They might have a stellar record when it comes to groundwater or hazardous waste analysis, but are weak when it come to PLM analytical skills. No lab is any better than it's weakest microscopist. Again, not always the case, but I have found microscopists who have a geology background tend to be able to do PLM analysis of asbestos and other insulation materials fairly well. Also, those who have attended course such as McCrone Institute's Course 508A and their Advanced Course 508B or equivelent courses provided by other trainers, tend to be more quality oriented.

Also watch out for gimmicks from labs. There's one lab that recently offered frequent flier air miles for every dollar spent. Be careful of these types of labs known in industry circles as "K-Mart Labs" (no offense to K-Mart intended.) There are also several labs promoting their services on the internet. There's one particular lab that has been very successful in getting business across the USA by offering $5-$9 samples in order to drive competition out of business in various locales. Once this occurs, you can bet which direction prices go. Rumor has it this particular lab provider is attempting to become national in scope, become a publicly traded company for the sole purpose of being bought out. Also, beware of labs that hire aggressive telemarketing firms to do their bidding. Always make sure you speak to individuals who physically analyze your samples, not just someone trying to make a sale. There's no doubt it's a buyer's marketplace for PLM lab services today. However, owners are doing themselves a great disservice if they are too price conscious.

How To Find A Proficient Lab and Insure Quality

The late Dr. Irving J. Selikoff, the pioneer asbestos disease researcher, once told me "you have to inspect what you expect." This certainly holds true for labs performing PLM analysis of asbestos bulk samples. The best way to find a proficient lab is to test them yourself. In Figure 1, you'll find a list of questions that you should ask your lab. ⁵   Sending out a questionnaire like this is really the first step in the weeding out process. The true test of a quality lab is their people's detective abilities and skills in looking through a PLM scope. Strong consideration should also be given to sending your own blind test samples periodically. Over the years, this is the only foolproof way I have found to determine a labs ability to remain proficient.

Lessons Learned

When I was DuPont's Corporate Asbestos Program Manager, the company had at least 100 outside labs performing this service. Essentially every site used a favorite local lab. In 1989, we pared the list to about 30 potential candidates to become preferred suppliers. We then sent each of these 30 labs ten (10) blind test samples. The samples were typical of the types of materials found at DuPont locations … asbestos TSI samples, floor tile, transite, roofing, as well as several non-asbestos samples such Refractory Ceramic Fibers and Mineral Wool bulk samples. The samples were uniform from lab to lab. No parameters on cost, timing or reporting format were given in order to see differences in price, service and accuracy.

The results: Between $15 and $55 per sample, between 2 and 64 days turnaround, and; most importantly, between 20% correct and 100% correct. To our amazement, only 3 labs got all 10 samples correct.

Since leaving DuPont in 1991, we have performed similar tests for some of our clients, including a major oil company and a division of a Big 3 automaker. In Figure 2, you'll find results of a test performed in late 1994 among 16 labs in various parts of the country that received 7 blind proficiency samples. Only 4 of the 16 labs got all 7 samples correct. The point is you need to ask specific questions of your lab and you have to test them yourself. While both NVLAP and AIHA provide valuable services, owners really need to take this issue into their own hands in monitoring quality. This is critical whether you're performing a total site asbestos survey, pulling many bulk samples or just having a few analyzed.

In Conclusion

While finding an asbestos lab may be easier said than done, it is possible and worth the effort, especially in comparison to the consequence of errors. You may even have to send your samples to a lab in a different state. Facility owners should not select a lab because it is close by. Sure, there may be times when you'll need less than 24 hour turnaround on a sample, but with a little planning, there should be few samples that fall into this category. It really shouldn't matter where your lab is located. What matters most is quality. Accept anything less and you might be spending more time than you'd like in a courtroom some day.

¹ The number of labs may be slightly higher than 343. A few satellite offices of labs were grandfathered into the NVLAP program in the late 1980's.

² April 1996 NVLAP round M111996.

³ NVLAP charges labs an annual fee of $6300 during the year of an in-field visit. $4100 for other years.

⁴ AIHA uses a point system that ranks labs as proficient or non-proficient based on test round performance.

⁵ RFM can provide answers to the questions in Figure 1 as well as test samples for a fee.