Scientific Meeting on Drug Testing Alternative
Specimens and Technologies
Date: September 9, 1998

TABLE OF CONTENTS

General Topics - Mr. Robert Stephenson
Congressional Hearing - Mr. Joseph Faha
Opiate Testing - Dr. Donna Bush
Hair Testing - Dr. Donald Kippenberger
On-site Testing - Mr. David Evans
Sweat Testing - Mr. Neil Fortner
Oral Fluids - Dr. Sam Niedbala
Hair Testing in the Workplace - Dr. Ray Kelly
On-site Testing - Dr. Robert Willette
SAPAA Survey - Mr. Robert Schoening
On-Site Oral fluid Testing - Mr. Tom Foley

Proceedings (8:40 a.m.)

MR. STEPHENSON: Good morning. This is the open session of the Drug Testing Advisory Board.

MR. STEPHENSON: I would ask those of you who have joined us today as observers as well as Board members to sign-in at you earliest convenience. I hope each of you picked up an agenda on the way in. I would like to introduce Joe Faha, who is with our agency. He is head of our SAMHSA legislative affairs office. I have asked him to give us an overview of a Congressional hearing that we participated in recently and to summarize the process we are about, some of which you are going to hear about today.

Mr. Faha provided comments regarding the hearing that occurred on July 23 before the Commerce Committee and the House Subcommittee on Oversight and Investigations.

Note: Mr. Faha’s comments have been deleted since the documents related to this hearing are included as a separate item on our website.

MR. STEPHENSON: Thank you. The issues related to that hearing and the process were very informative for me, too. It was one of the most interesting and rewarding opportunities, as a federal employee, to go before Congress and actually be able to provide information that was received objectively by members of two different political parties, and to review the issues related to the science in what I consider the proper respect for science.

I came away from that with a much restored belief in the overall process for how we inform ourselves to make public policy decisions. If we can do that well in all the other areas that we interact in, we have got a really great system. Now, it is up to us to provide the grist for that mill. We have got to do the work every day that produces the kind of opportunities and informative products that can be shared on the Hill, to help drive new, better national policies for drug detection technology. We made it clear during that hearing that hair was only one of several matrices and alternative technologies that we were looking at. We made it very clear in the written testimony that we submitted, that it was our intent to look at the entire process by which we valued and understood new technologies and specimen issues. We weren't simply scoring a one-time project of inclusion or not inclusion, but looking at the entire process of new technology.

We know, as you are going to see today, that there is an onslaught of new technology that isn't waiting for anyone. We need to be able to periodically assess it, and come to grips with issues that are new that we haven't dealt with before, and get it into a mainstream for inclusion, if it is appropriate to do so. In that context, we are looking at change in regulations. One of the segues that is appropriate here is, in that hearing, a second topic or brief discussion included the raising of opiate cutoff and testing procedures under the federal guidelines. They were discussed and there were issues that were raised. We believe that we have addressed them as well as can be done in the context of the science and underpinnings of large studies. All of those have been presented in previous Federal Register notices. It described why we did what we did, and what we propose to do.

At this point, I would like to have Dr. Bush make a brief presentation about where we are going with those testing cutoff concentrations at this time.

DR. BUSH: I would like to update you on our ambitious effort to change the opiate cutoffs.

If you recall, we are changing them in two ways. We are increasing the cutoffs for morphine and codeine, but adding the additional analyte through the GC/MS confirmation process, 6 acetylmorphine, the metabolite which is characteristic of ingestion of heroin, and which we do find in the urine.

Last fall we published a Federal Register notice with an ambitious goal of May 1 for implementation for these new cutoffs. Well, all roads lead to challenges and we found that both the manufacturers and the laboratories who would need to simultaneously implement these changes needed more time to make sure that they could do it in an accurate and reliable fashion. We sent a letter in February to our colleagues, informing them that, because of our desire to meet their needs for accuracy and reliability, and our desire for the accuracy and reliability, we said we would give it a short time, give it a breather before we established another date. We would renege on the May 1 implementation date, and that we wouldn't set another implementation date immediately. We would take a look and see how the manufacturers were working through their production issues, and how the labs were going to work through their confirmation procedures. Well, all is well with the world, because everyone is meeting their respective missions quite well.

We know now that there are sufficient quantities of immunoassay reagents that have been cleared by our colleagues. The kits have been cleared by our colleagues at FDA in an expeditious manner, and the laboratories have used these immunoassay kits recently in some PT challenges, and the labs are happy with them and have implemented them well. The manufacturers are happy that the labs have implemented them well. So, so far, so good, through that first step of immunoassay challenge to changing the cutoffs.

Now we have to evaluate the laboratory side of things. They had a multiple pronged approach that they had to take. Not only did they have to implement that immunoassay change, but then they had to look at taking two well-known and loved analytes -- that is, morphine and codeine -- and raising their cutoffs, changing their linear ranges, looking at their upper limits of linearity, doing all the things that good labs do in looking at accurate and reliable testing of those two compounds.

Then they had to set up new assays for 6-acetylmorphine, the metabolite characteristic of heroin. So, they had implementation challenges such as implementing their own in-lab blind quality controls, open quality controls, SOP changes, and they are well on that road to doing so.

They have received a couple of sets of performance challenges, depending on the particular lab's needs, depending on whether they were feeling particularly challenged by these changes or feeling very comfortable with these changes and how they performed on our proficiency testing specimens.

We are feeling very comfortable right now with how the labs are proceeding. There is a lot of communication between the laboratories concerning analytical challenges that they have met, maybe with a new derivatizing agent, things that you find with a new cutoff, looking at a new procedure and interference by other similar opiate compounds with the new analyte, 6 acetylmorphine.

All the lab life issues are moving along well. As we speak, I believe the laboratories will be receiving a set of opiate PTs sent out by our contractor, RTI, yesterday. This should give us our last necessary look at where the labs are with their implementation, how accurate and reliable they are, how precise, and what the quality controls look like.

We will be doing a lot of communication shortly with the laboratories after they send us back that information.

With all that said, we are looking strongly and clearly with December 1 as the implementation date. The clearance package for that Federal Register notice is in the clearance pipeline within our agency. It is still within the Center for Substance Abuse Prevention, and will be moving up to Dr. Nelba Chavez for her review and approval and discussion if necessary, shortly, and then on downwards to the Office of the Secretary, Donna Shalala, for her review.

I can't tell you when it is going to come out in the Federal Register, but it should be shortly, and with the implementation date of December 1. We are confident of that at this time. That is the long story for a very short process. The labs are getting their PTs today and we are moving ahead with December 1.

MR. STEPHENSON: It looks like we have kind of caught up here on time line. One of the things that we will briefly go over is what do we expect to have come out of this meeting today. Coming out of the Congressional hearings and looking at the process that we have been about for the last year and a half, this is an important time period.

We have identified a matrix relationship between factors that are required for reliable and accurate drug testing, that helps inform us as to how we are going to score any potential technology or specimen. We have requested and received information from the various industry components that support each individual technology.

We have gathered today a new set of information that is going to summarize, by those different specimen types, where we are at the current time.

When we are through, we are going to establish a process for small group activities that will take each of those alternative technologies or specimens and guide a working group made up of all of the members of that particular niche in the industry to the point that they can identify the critical issues that remain unresolved through the information that is going to be provided here today, to look at the issues of research that is required to answer those questions, and to work together to develop a strategy by which we can identify the actual research agenda. Then collectively, we will work on a process to try to obtain the funding and priority for action to actually get that research conducted at the earliest possible time.

This is going to be a challenge for some, because it is an area where some of the technologies are so new that you don't necessarily want to give away your trade secrets to your potential competition. At the same time, it is a challenge for us where we do have a lot of old established relationships, not all of them conducive to working well together, in which we are going to have to get beyond that, and work in an objective environment, to look at the issues, and to deal inclusively with all the potential players within an industry arena. We are going to do that. It is going to happen. We are going to deal with these issues in several ways that I think are informative.

We have made a commitment, and any time we have new information presented, we are going to literally have a transcription of that material developed as a part of the meeting, of this material today. From whatever is said verbally, it will be transcribed. We are trying to move into a paperless environment where many of the materials you will see will be presented in a PowerPoint environment. Those can then be provided to individuals who would like to look at them later on.

As we get these transcripts and these other materials available, as rapidly as possible, we will make them available on the internet, through our web page, so that they can be shared and commented upon.

We will provide the technical and logistics support to the small group activity meetings to see that we are able to carry on, so that it won't fall as a burden to anyone else outside our agency or outside our areas of assistance, so that we can actually get some work accomplished.

In preparation for the meeting today, have all of you gotten copies of the materials that were provided in the back? This represents the kind of process that we are about. You have in your hand the materials that relate to the activities, to the state of where we are right now. This is the way we are going to continue to proceed. Those of you that are present who wish to participate in any of the industry niches or activities and have something meaningful to contribute from your own personal academic training or practical experience as an individual working in the industry, let us know.

The individual coordinators will be making their presentations during the course of the morning. We will look forward to having some new faces and some new insights into this group process.

DR. KIPPENBERGER: I am the hair testing coordinator. In July, I responded to the draft document on the alternative matrices. In the submission, I incorporated all the input I had received. Obviously, there are some people that didn't give me anything. So, I could only incorporate what I received from the outside and my own internal feelings. In the last year, a lot of information has been given to the DTAB on the science. It appears now, since we are doing the draft document and all, that we are moving from science into a regulatory format. In this process, I have been working with Bill Thistle, who is our general counsel at Psychemedics. I have decided to let Bill present some of the regulatory information and answers to the matrices, and I will handle any of the science questions that might come up.

MR. THISTLE: For those of you who don't know me, just by way of background, my involvement with hair testing was originally as a client, back in 1989. I mention that to underscore the length of time that we have been coming here and doing some of the hair testing. I instituted a hair testing program at a casino hotel in Lake Tahoe for 3,000 employees in 1989. The success of that program was immediate. The number of employees opting to enter into EAP programs increased and user applicants simply applied at surrounding casinos. Those casinos then implemented hair testing, more or less in self-defense. Hair testing spread throughout that industry rapidly, and within a few years it became the norm to have a hair test in the hotel/casino industry, no matter what state you were in. Two urine testing labs in Las Vegas alone began to offer hair tests.

In 1994, MGM tested 10,000 applicants prior to opening. The wide acceptance and use in the casino industry spawned a myth that hair testing is confined to the casino industry. Actually, that is now a very small part of the hair analysis user group. It is used in a wide variety in a cross section of United States businesses, including some of the largest Fortune 500 firms in the country. Some of the largest corporations in the country -- General Motors -- uses hair testing. BMW, Michelin, Anheuser-Busch, Steelcase, EDS, Champion International, and over 1,300 more corporations that I am aware of use hair analysis. Some of them have been using hair analysis for over a decade. The largest police forces -- NYPD, Chicago, other police forces, and state police forces, use hair analysis.

There is a myth that hair analysis has not been held up in the courts. There have been a small number of challenges, due to some safety net aspects of hair analysis that I will get into later. Given the number of tests, the challenges are few. Since 1990, the use of hair analysis has been upheld in federal and state courts, by administrative agencies, as well as courts marshal. Last year alone, there was a New York State Supreme Court appellate division decision upholding the use of workplace hair analysis, as well as five military courts marshal, U.S. v Johnson, U.S. v Ramero, U.S. v Adams, U.S. v Barksdale, and U.S. v Adens, that upheld the use of hair analysis.

In 1997, a military appeals court upheld the use of hair analysis -- U.S. v Bush. Not only is it being upheld by the courts, it is being used by the courts as part of probationary and diversionary programs.

Hair analysis has been enthusiastically embraced as an extremely effective pre-employment test. It has longer windows of detection, and it is far more difficult to adulterate. Recently, in the latest U.S. Supreme Court decision on drug testing candidates for office, the testing in that case was struck down for several reasons, among those reasons the finding that scheduled urinalysis tests performed in private were largely symbolic. The public -- we think the public -- and corporations are not looking for symbolic gestures. They want effective testing that can identify users. Hair analysis does that.

Hair analysis is also used in random programs, post-rehabilitation, and periodic testing. Again, the long window of detection makes evasion much more difficult. There is, of course, greater deterrence when the longer period is utilized. Hair testing can provide useful information in post-accident and reasonable cause testing. I did see a submission by another lab that would prevent the use of hair analysis for these types of testing -- the latter two, post-accident and reasonable cause -- because it provides a historical picture of use and not immediate use or under the influence. I would point out that urine testing does not show immediate use or under the influence either. It merely shows use closer in time. Blood or oral fluid may be better for that type of information.

Hair testing does provide some information. Suppose, for example, you had reasonable cause to test someone; erratic performance, information from a source that they were involved with heroin. If you gave only a urine test that showed a morphine level of 2,000 but no 6-MAM, a not unusual event, what would you do? Under the guidelines, that person would be negative and they would get behind the wheel of an 18-wheeler. If a hair test revealed 6-MAM and prior historical heroin use, you could put that person in rehab. It is important information. It is information that an MRO, I think, would like to have. It is clear that the outcome of putting that person in rehab would be better for the employee as well as the other citizens that that employee would be involved with.

Would you use hair analysis as a primary post-accident test? Probably not, but I wouldn't use urine as a primary either. They each provide useful information that an MRO can utilize in making informed decisions. If you wanted to monitor a trucker's use over a cross-country trip, perhaps a sweat patch would provide useful information in that regard. The use of all these matrices is situationally dependent. I would not block the use of hair for post-accident or for cause testing any more than I would recommend the banning of urine testing for pre-employment testing where it is not optimal.

We want to provide the tools that employees can utilize to eliminate drug use. That is the purpose for our being here; that is the purpose for the regulations. With that grandstanding, I will go on to the matrix.

We have -- the first area of the matrix that listed anything to do with hair testing was B-1 and 2, collector training and certification. There are trained collectors now, as I mentioned before. A large number of major companies drug test. Some of them have factories, outlets, manufacturing, all over the globe. There are trained collectors that exist now. Training involves currently on-premises training, hands-on training, videos, pamphlets and, in the case of some laboratories, graded examinations to provide a certificate at the end. Collector training is certainly not much of an issue. It is listed on the grid. You think it is possible; we think we do it, and we do it. So, we can move on to the collection device.

FDA clearance; this is FDA clearance for the collection device itself. The current FDA proposal is that if the underlying test is recognized, the collection container is class I and exempt from pre-market notification. I would point out that hair testing containers currently in use have some remarkable consistency. Just to explain how the container works, for those of you who are not familiar, there is -- I realize that you can't see this from the back, but I will hold it up anyway for the people up front. There is a piece of foil. The hair is collected. The collector holds the sample -- this was explained, I think, in detail in April 1997 when we had the meeting. I will go over it real briefly here. The collector is holding the sample, the back of the head. The sample is placed in aluminum foil. That is placed into a collection envelope. There are the donor initials that it is their hair. By these initials, I certify that I am a test subject, that the sample contained in this envelope is my sample. It was collected close to the skin and a witness to the sample collector sealed the sample into the envelope. The sample is put into the envelope. The envelope is sealed with an evidence tape seal, and that is then placed into the typical tamper-proof plastic bag. That is also initialed off by both the collector and the donor and that is sent to the laboratory.

If I pull one from another lab, what we find is that it has aluminum foil. It has the same card. It has the exact same wording: By these initials, I certify that I am a test donor who provided the hair sample enclosed. The hair sample was taken from my head -- same thing; same envelope, same aluminum foil, same seal, same plastic bag, basically the same procedure. So, we have remarkable consistency there already. There is no problem in making collection devices standard.

In multiple testing, D-2, it is common practice to test more than one aliquot of a specimen as part of a safety net program where a new specimen is collected. The original specimen is also retested with the recollected specimen. Again, it is a P, possible. If we don't have an issue, we can move on.

Potential for split specimens, D-3. Currently, split specimens are being done. Different labs use different amounts of hair. It can create some cosmetic issues. The advantage of hair, however, is the ability to collect a new specimen if the first is challenged. It is not necessary to do a split specimen at that point in time. The new specimen can replicate the original time frame of the original test, assuming that it is done fairly close in time. Once you get the test results back, if you simply have a rule that is part of the submission -- I think we put within 72 hours you make a decision to have a new test done. The new sample virtually eliminates questions of specimen mix up or lab error. This is a huge advantage. It provides unprecedented fairness to an employee who makes those claims. For the lawyers in the group, we know that those challenges are the most common challenges made concerning a test, lab error, sample mix-up. That is the usual challenge. This takes hair testing a little bit -- it removes it a little bit from what we currently do with urine testing. I don't mean to say we don't have to be as careful with it. I think you do. It becomes more similar to a cholesterol test. If you have an abnormal result or someone says, that can't be my result, you can simply take another sample. You have repeatability. You can do it again and again to verify results.

D-4, stability and storage. Drugs in hair can be detected several hundred years later. For the technical people in the group, there is some hydrolysis, I understand. At least that is what the science guys tell me. Basically, you can detect the drugs several hundred years later. We are not looking to store hair for several hundred years. We are looking for about a year. That is fairly easy to do. Specimens containing drugs will, on retest, provide statistically equivalent results to the first test, even after months of storage at room temperature. The one-year time frame should not be an issue.

Deterring tampering and adulteration. Adulteration of hair specimens at the time of collection is simply not an issue. As I mentioned briefly before, the specimen is obtained in full view of the collector. It is directly controlled by the collector and placed in a collection envelope in full view of the donor. The donor acknowledges that the specimen was obtained from them, that it was cut close to the scalp, and it was sealed in their presence. The donor is never left alone with the specimen. There is no break in the chain of custody. With urine testing, frankly, there is no chain of custody. You have a mechanism to prevent employees from claiming the employer tampered with the sample. You don't have a true chain of custody. The chain of custody is broken at its inception. You leave the donor alone with the sample. If you were to leave the donor alone with the same sample in the laboratory, you would invalidate the sample in a urine test, and we call it the chain of custody. With the hair analysis, again, the collector holds the sample. To achieve that similar security in a urine collection, you would have to have the collector hold the penis into the cup. That is not likely to happen. It is an extremely secure method of collection.

Transportation of specimen, this is basically the same as urine. They are usually sent Fedex or some other carrier. It is identical and there is no biological hazard associated with hair specimens. Actually, there are probably hair specimens around you, on the floor and everyplace else. We accept that with hair. We don't accept that with urine, but we do accept it with hair.

G-2, again, it is another possible. The hair specimen is stable at room temperature. It is being stored now at room temperature. We are all storing it at room temperature right now; most of us are storing it at room temperature. In the lab itself, specimens are maintained in secured areas with chain of custody procedures for removal, the same as urine.

G-3, again, another possible. Collection under close supervision prevents substitution of specimens. We went over that. You are not going to have substitutions. It is not going to happen. The collector is there. It is not going to happen.

The adulteration of a specimen is difficult. Substances placed onto hair are removed with extensive wash procedures. Methodologies that digest hair with 100 percent extraction efficiency from the carotene matrix require major destruction of the hair to remove all drugs. This destruction is readily identifiable. Other techniques, while not removing all the drugs from the hair, the more effect the hair treatment can have on that. It is just, the less effectively you extract the drug, the more effectively attempts to adulterate it can become. Whether they are very effective or not depends on the methodology. Again, hair testing is not in a vacuum here. If you look at the ease of adulterating some matrices -- for example, urine -- it is much harder and much less effective to adulterate hair. As an example, we did have a case a year ago where an officer stripped his hair completely, redyed it so that it would look normal, failed his hair test. He opted for the second hair test, the safety net, stripped his hair again, redyed it again within a two-week period and failed again. He was positive both times. It is not completely effective to do that.

G-4-A, initial test FDA cleared. Just as a history of the FDA clearance, the FDA recognition of screening assays was initially required by the military. At that time, in the early days of testing in the military, screening was not followed by mass spec confirmation. Today, all positives are a product of a confirmation test over which the FDA has not exercised any jurisdiction. It is common practice with urine labs to dilute or extend FDA-recognized assays. They are using adulterated assays that they then will have to subsequently prove continue to work accurate. But they are not the FDA approved assay at that point. While we recognize that the assay can be approved by the FDA, the question becomes, do we accomplish anything. Urine tests that commercially pre-existed, the 1976 device amendment, have been recognized without any review. Subsequent recognition for other urine tests is obtained on the basis of 510(k) review. The standard of review for other assays under 510(k) is a showing of substantial equivalent to the pre-existing, the unreviewed, assays. I can't say this any better than Dr. Guttman said last September when he was speaking. The good news is that a new product will not be worse than an old product that was on the market in 1976. Of course, the bad news is that it may be substantially equivalent to a product that wasn't anything to write home about that was on the market in 1976. I realize there is a good argument for that is the way we have always done it. Other than freezing the technology, I am not sure we accomplished much with the FDA review. Again, it is possible to do it. I am not sure with the mass spec confirmation that it means all that much any more.

Can HHS target analytes be detected? The following are detected by laboratory antibodies: marijuana analytes, uncharacterized marijuana analytes which accumulate preferentially in hair. This is not an HHS target analyte. Psychemedics believes this provides greater screening accuracy. Amphetamine, methamphetamine, D, MA, UNHALLOWED, codeine, morphine, PCP, cocaine, benzoylecgonine, cocaethylene, those are all the same as urine testing.

Does the testing use HHS established testing levels. Hair testing does not. They use lower levels. The procedure to determine cutoff levels was described in last year's presentation. That should be on file somewhere in DTAB. I am not going to go into it now. We went into it extensively, I think, last year. Acceptable performance around the cutoff, it should be the same as urine. Analyte screening assays use a standard curve with points below and above the established cutoff. Controls are run above and at the cutoff. Blind controls for non-users and known users are assayed with each patch. I am getting more into Don's area here; I can see that.

The ability to repeat the initial test. The screening test can be repeated on either the original digest of the specimen or on a new aliquot. Results of the original digest repeats are usually within plus or minus 20 percent. New specimen, the safety net that I referred to earlier, is also available. Again, I don't want to stress this too much. This is the advantage of hair analysis. You are not limited to a single test specimen. You have a drug history that is carried with you. The new sample can replicate the original time frame. You eliminate sample mix up, lab error, chain of custody claims. Again, it is not available with many other matrices, and it provides unprecedented fairness to employees.

Confirmatory tests, the same as urine, mass spec, used for confirmation of all workplace drug tests. That is described in other literature. We kind of know what it is and how it works. Acceptable performance around cutoffs, laboratories utilize assay controls at concentrations below and above the cutoff. Because of lower levels observed, criteria, we believe is that the control should be within plus or minus 25 percent of the target concentration. Cutoffs reflect drug use. Unlike urine, rough correlations of drug levels in hair with severity of drug use allow establishment of meaningful cutoff that correspond to use, or cutoffs that represent something not less than use. Field studies have shown that the cutoff levels do not suffer from the internal exposure problems that would be caused by massive amounts of poppy seed ingestion.

Again, this has been described in more detail in some of the earlier submissions.

There has been criticism that, because of the cutoff levels and the high cutoff levels, that hair analysis only gets moderate to heavy drug users. I don't think that is the case. If it is the case, that should tell you something pretty dramatic about urine analysis, if hair analysis is getting the moderate to heavy users and we are seeing far more positives than a urinalysis program.

The certified laboratory program, NLCP certification can certainly be extended to hair analysis. Currently, hair labs can be certified under CLIA, as high complexity labs.

Labs can be certified under numerous state laboratory programs. There are urine based programs -- CAP and SAMHSA (NLCP) -- that can also provide some indication that fundamental laboratory practices are in place, although these don't involve hair specifically.

External proficiency testing. The International Society of Hair Testing has begun a proficiency testing program. NIST, as we are aware, conducts a laboratory comparison program that provides samples from lab to lab. Clients have been submitting blind specimens that can be submitted through private providers. That is what currently exists, although there is no reason why a proficiency testing program cannot be extended for hair analysis that is currently available to urinalysis labs.

Lab inspection program, again, no different issues with hair than with urine. I suppose if you are doing both, you need to demonstrate that you either use separate equipment or that you have procedures in place to prevent carry over problems in that regard.

Blind specimens, blind material is available and can be used the same way for hair as with urine programs.

Certifying scientists, yes, same as urine. We can go through some of these things here. The ones that are Ps, again, are possible. They are, in fact, being done already and are really not issues.

Results reported by specific drug, the same as urine, although additionally, current reports include usually 6 MAM, cocaine, and cocaethylene.

Confidential and timely written lab reports, same as urine. Standard report form, basically anything you want. It can certainly meet the requirements established for form design.

Interpreting results. Interpretations are generally more certain than with urinalysis. Again, we can eliminate the lab error, sample mix up, chain of custody issues. There are issues as to what a negative urine test actually means, since it doesn't provide information on more than a few days. Does it actually mean you don't have some drug habit. Maybe, maybe not. We have repeatability. There are diminished passive internal exposure problems. Poppy seed will not cause a positive hair test. Hair permanently entraps 6 MAM. It has a much longer window, obviously, than a several hour urine window in that regard. Passive external contamination issues are addressed through a combination of cutoff levels, extensive washing, wash kinetic analysis, and metabolite measurement.

Do we have the actual drug present as well as metabolites? At times. The more extensive history requires inquiry into a larger time period. That is something that an MRO would have to get used to, and we have MROs that are doing hair analysis review now, and it simply requires an inquiry into longer time periods.

In terms of the interpretation, I would like to take a moment to address bias issues. As we are aware, anything that can affect a result has the potential to create a bias issue. With urine testing, MROs do not take into consideration that females may be generally smaller than men, and that a lesser quantity of cocaine can make them positive. We don't take into account that females may retain water and have a longer detection period. We don't take into account age factors and maybe a bias against older donors in urine testing. We don't differentiate between heavy people and lightweight people with urine testing. We don't adjust the urine test based on athletic activity. We don't note whether someone is in a wheelchair or whether they are a football player hydrating their system. We don't adjust for ethnic diets. The fact of the matter is, clients who have performed side by side testing with urine and hair tests on the same applicants can have data on thousands of comparisons and will find that there is no difference in the positive rates or the disparities in the positive rates in blacks and whites, which is the common claim brought forth for hair analysis. Clients who perform thousands upon thousands of these side by side tests find that in the large groups there are no differences in the disparity between blacks and whites in hair tests and urine tests. These clients generally, almost always, the race of the individual is blind to the hair lab. The race of the individual is blind to the SAMHSA urine lab that does these tests. The results consistently show that the disparity between blacks and whites is the same with urine and hair testing.

Is there disparity between blacks and whites in cocaine use? Yes, and we see that again and again. I see that with urine testing; I see it with hair testing. We see that it usually goes the other way with methamphetamine. It is probably a drug of choice question. The 1990 U.S. Postal Service, considered the seminal study on the benefits of drug testing in the workplace, we quote from it all the time, involved 4,000-and-some individuals, where the Postal Service hired the drug users. That study showed a six to one disparity, blacks to whites, in cocaine positives with the urine testing. There is disparity between blacks and whites with cocaine use. That has been demonstrated with urine testing, and it is the same with hair testing. I don't want to say this again. The results are the same, when you look at large populations. If you look at individuals, of course, you have individual biovariability that makes those results not comparable.

MRO training can be included in the existing programs. There should be a difference. The issues are the same and, in some cases, less. Alternative medical explanations, generally no different than urine. You don't have the poppy seed and other passive exposure issues.

Result related to dose/time response. Hair growth rate is approximately a half inch per month. The time period represents approximately 90 days and provides a historical pattern of drug use. As we said before, that information is useful in a number of circumstances, less useful in other instances, far more useful in many others; for example, in pre-employment.

Specimen contamination. Extensive decontamination procedures are employed to eliminate these issues. We don't necessarily need to remove all traces of drug. You need to have a mechanism in place to raise a flag that there is contamination, causing you to then do other things. Washes can be extremely effective. I have seen people criticize some of the wash procedures, and they do part of them. You can't do part of them. You have to do them or you have to not do them. You can't say, well, we did the Psychemedics’ procedure except. The except can't be there. You have to do it and you can effectively remove contamination or, in fact, identify hair that is contaminated and call it contaminated. There have been studies on this. Dr. Mieczkowski did a study that utilized the Psychemedics lab, on narcotics officers exposed on a frequent basis.

They bought drugs, they sold drugs, they were in the company of users. At times they admitted they taste tested drugs. Contamination was present. It was removed by the wash. Dr. Mieczkowski additionally inserted samples from known users and soaked hair in cocaine, as well as submitting known positives -- I am sorry, he submitted known positives, known negatives, as well as soaking hair in cocaine solutions. The positives were identified as positives, the contaminated were identified as contaminated, and the negatives were, of course, negative. A 1992 Canadian study by Dr. Caramm used early Psychemedics wash procedures that were published in 1989. They exposed hair to the equivalent of 5,000 lines of vaporized cocaine. That contamination was removed with a wash.

Most labs have some wash procedure. It has to be done. You can't simply ignore it because 95 percent of the time it is not an issue. It is that small percentage where it is greatly important. God is in the details in this regard. Alternatively, the devil is in the details, and I think it depends on whether or not you are striving for accuracy or striving to cut corners. I would say that you can't eliminate a wash procedure simply because you find that your method doesn't work well. If it is not working well, you need to do more; you don't do less. Contamination, just to bring us back to the fact that we are not dealing with a hair vacuum here, is an issue with all matrices, and should be considered in all determinations. There was just a recent study at Yale University, in Pediatrics Journal, July 1998. One in three infants brought to the emergency ward had cocaine in their urine, one in three. They were inner city children. There have been other studies about children being exposed and having cocaine in their urine. There have been studies concerning evidence handlers that have urine positives. There are studies showing positive levels after ingestion of minimal amounts of cocaine, in urine. There are studies on side stream marijuana smoke. As marijuana has gotten more potent, it becomes more of an issue and a consideration in the MRO deliberations.

All testing should be interpreted with caution. There are mechanisms with hair that provide extensive information, sometimes more so than that available with urine, to MROs to make these cautious determinations. In the regulations that Don came to me and we helped put together, we simply tried to mirror the urine language -- and I believe a copy of it is out there on the table. Wherever possible we tried to set minimums and avoid creating legal problems with the wording. For example, I think originally the phrase was in there, collected in a draft-free environment. It is a nice phrase to put in, but it invites legal challenge, for positive results based on extraneous issues. Yes, my client's test was positive, but by god, he felt a draft, so the procedures weren't followed completely. We have tried to eliminate that kind of language in there.

There will be some disagreement among labs as to what was put forth. I think, actually, there would be more agreement than disagreement, but I think there are differences in all labs with all matrices. We tried to be as inclusive as possible, but still retain elements that create accurate testing. Comments and calls were received as late as Saturday from toxicologists wanting to add or delete material. If your thoughts weren't included, it was because you didn't submit anything. Again, as this is an ongoing process, I think the opportunity is there for any testing lab out there that is doing hair, to present information to the coordinator and the coordinator can coordinate it. That is why we have coordinators.

In closing, I would just say that all the tools and technologies should be made available to employers to enable them to make the best informed decisions. Urine testing is a good tool. It is not optimal in all settings -- pre-employment, the detection of heroin and others. A hammer is a good tool but you can't build a quality house with just a hammer. All over this country, people are using additional tools, like hair testing, to eliminate drug use. It has proven to be effective. It has proven to be accurate, and it has proven to be reliable. You cannot have effective deterrence without effective detection, and hair analysis accomplishes both. Hair analysis should be made available for use in federally mandated testing, the same way it is available and being used, and has been used over the past decade in the private sector.

We appreciate the DTAB's efforts in this regard and we look forward to moving forward. Thank you.

MR. STEPHENSON: Thank you very much. At this time, we are going to take a break.

[Brief recess]

MR. STEPHENSON: Okay, before we continue with our next presentation, which was the one originally scheduled to be on our agenda today by David Evans, I have asked Dr. Walter Vogl to walk you through the documents that were on the outside table, to make sure that you understand what it is that you have in your possession and understand how to interpret the results.

DR. VOGL: When you entered the room, I hope you took a copy of each of the documents that was on the table. The first one that you should have picked up was the draft document that we had prepared. It was on our web site and we had solicited public comment. If you page through that document, you see that there are several areas where information is missing for the different alternative testing technologies. We had asked the coordinators who had assisted us for last year’s Board meetings to provide the missing information. Recently, we removed the draft document from the web site.

The second document is the factors for forensic drug testing that has been developed by the Board. It is the latest version and you should all have a copy.

The next document relates to sweat testing. Each of the coordinators submitted their infomration for the draft document on a disk. I extracted the new information each provided into separate files rather than repeat the entire 50 pages of the draft document for each alternative specimen. I tried to use the same format for sweat, oral fluid, and hair. If you look through these summaries, there is a parallelism that exists to help you see what information was submitted and where it belongs in the draft document.

Next, is a copy of the slides Neil Fortner will use for his presentation.

The on-site testing document, was not changed or extracted from a file.

The hair testing information was extracted from the document submitted by Don Kippenberger. You also have a copy of the slides for the hair testing presentation.

Next, hair testing information and slides submitted by Associated Pathology Laboratories.

A copy of the information for oral fluids.

Two submissions, one by Intoximeters. It is for a new on-site technology or on-site testing of a urine specimen using a new technology. To my knowledge, no one is here to present that. It was a copy submitted for the public who are attending today's meeting.

The last copy was submitted by LifPoint and it is a new technology for on-site testing of specimens.

Hopefully, this will help you understand what the documents represent.

MR. STEPHENSON: At this time I would like to introduce David Evans, who is representing the on-site testing industry component. He represents an association that has been formed of a number of the technology providers and manufacturers inside that particular environment.

MR. EVANS: Good morning. The approach that I am going to take this morning is to go over briefly some general concepts about on-site testing and respond to the matrix that was provided by HHS. I have some friends with me today who are going to cover some technical issues and some practical issues. After I speak, we are going to have Julie Murdoch come up. Julie is the director of forensic testing for Bensinger Dupont.

They are currently operating the largest on-site testing employment program in the United States. It is for the United States Postal Service. She has a lot of practical tips on how on-site testing can be implemented, at least in a quasi-governmental context with the U.S. Postal Service, after consultation with their attorneys and so forth, and it is operating really well. She will talk about how on-site testing can be done.

Next we are going to have Dr. Stewart Bogema. He is a senior toxicology associate with Bensinger Dupont. He also has his own company. He is president of Forensic Quality Control. He is going to talk about some of the quality control issues.

If there are any other legal questions, I would also like to get some help from Sandra Debow. Sandra Debow is an attorney from Boston, who is an expert in drug testing and on-site testing. With all this support, I think we will be able to hopefully answer any questions that you may have.

Let me tell you a little bit about the National On-Site Testing Association. We are a young organization, about a year and a half old. We are small, but we have a lot of fun in our meetings. We are composed of the manufacturers and the users of on-site drug and alcohol tests. We define an on-site drug test as a test that is easily portable and can be administered in a location outside a laboratory, such as a work site. We require that any manufacturer or distributor of an on-site test who joins NOTA adhere to our mission statement.

Our mission statement includes that we recommend that on-site drug tests, first of all, be cleared by the FDA for commercial distribution. We think that sets a basic quality standard. We also believe that any on-site test, any on-site drug test, should be able to meet standardized cutoff levels such as those found in the mandatory guidelines put out by HHS. We are currently, as an organization, working on developing certification standards for on-site test operators, and we are doing a lot on the state level. Last year we were lucky to be able to participate in passage of legislation in Alaska, Oregon and Idaho, all states that have passed laws either permitting or setting standards for the use of on-site testing.

Oregon and Alaska are real good models. They set certification standards for test operators and test operation, which we think are, in most cases, pretty good models. We have bills now that we are working on in Florida, New Jersey, Maryland, and New York. We are expecting to have other states going on-site pretty soon. Iowa also this year went on site, in that they removed their requirement for laboratories being involved in conditional tests for employment.

In going over the materials that we submitted to HHS, the approach that NOTA is taking on the issue of quality control, which seems to be a most important issue, is that we are going to take an approach similar to what the Federal Department of Transportation has taken when it comes to initial on-site alcohol tests. The quality control should be built in. The care that the test is properly administered should be built into the certification and training of the test operator. Now, our on-site tests also have an additional advantage. They have a built-in control with each test. There is something in each of our tests, at least the ones that NOTA recommends, that will provide some assurance that the test is operating properly.

Proper training of the operator, proper certification of the operator, an on-going training, updating process with that, built-in control with the test. Then we also recommend that there be quality control with each lot of test that a customer receives. When they get a shipment and it is a different lot, that lot should be tested, proficiency tested. We are also recommending proficiency testing at least twice a year, using an outside proficiency testing agency to do that.

Let's go over our response in more detail. That is sort of a summary. We only focused on the areas that HHS asked us to focus on. We assumed that the other areas, we had already answered all their questions on specimen collection and so forth, and there is no real difference between on-site testing and laboratory testing.

It is our position that training can be conducted by the manufacturer or the manufacturer's distribution representative or a third party. We have found that Bensinger Dupont, for example, is doing the training of test operators for the U.S. Postal Service. They are not a manufacturer, they are not a manufacturer's representative, but they are a third party. Julie is going to talk about the very detailed and very professional procedures that they have implemented. Training programs for test operators should have at least the following information: general information on drug abuse; specimen collection; chain of custody; specimen adulteration; interpretation of results; recording of results; and confirmation of presumptive positives.

We looked to what used to be called the National Association of Collection Sites, and I have forgotten the new name. Anyway, it used to be NACSA. They do have very good training programs, certification program for specimen collectors, and that is the type of model that we would be looking to.

Certification of test operators, we think, should be done. There are two states right now -- Oregon and Alaska -- that by law have set up certification programs for test operators. You can look to those states.

FDA clearance - absolutely. We recommend that any on-site drug testing cleared by the FDA for commercial distribution and, as I said earlier, that meet the cutoff levels established by HHS. Getting FDA clearance ensures that there is a quality manufacturing process. It assures that there is some regulatory control. In addition, we recommend that the tests have a control built in, which gives you an extra feeling of confidence.

We were asked to address the issue of split specimens. The answer to that is yes. With on-site testing, specimens can be split. We have no problem with that at all.

There were some questions that came up about deterring tampering and also looking at the issue of adulteration. Again, no problem with on-site testing. We believe that proper collection procedures are the key to the prevention of adulterated samples. With on-site testing, you can check for adulteration. It is a responsibility of the specimen collector. You can use temperature strips on the specimen containers, visual inspection of the sample. You can use a dip stick, all of which can detect adulteration and tampering. In addition, the tests that we recommend have a built-in control to show that the test is operating properly.

Transportation of specimens, there is no difference between the on-site test and any other urine test.

Can an on-site test be specific for a drug class? Absolutely. We use the same analytes at the same cutoff concentrations as laboratory based urine screens.

Documentation of on-site test results? No problem. We have developed a set of forms for positive test results, for negative test results. Again, you can either use a separate form, or we recommend you can also use a log book. Julie Murdoch will talk about that.

Results can be documented. Testing levels can be defined. Target analytes, again, no problem, nothing different about that than laboratory testing.

Is there an objective differentiation of positive and negative? We feel this is comparable with laboratory screening. On-site tests indicate positive and negative results and have a built-in control. If there is any doubt about the test, if it appears to not be clear, then it should be sent to a laboratory for confirmation, if there is any doubt at all.

I have already addressed the issue of proper controls. One of the issues that has come up is the analyst not knowing the donor’s identity. I guess people feel this gives an extra measure of security to the test. There are several ways that on-site testing can be established so that the test operator will not know the identity of the donor. We feel this is an issue for the employers to address. It is going to depend on the workplace, the needs of the workplace, and there are several ways that it can be done.

HHS asked us to address the issue of having the results verified by a second analyst. We feel that is not necessary. We feel that the built-in controls, training of the operator, certification of the operator, are all that is necessary.

The results, if there is a presumptive positive, should be sent to a laboratory for confirmation.

On the issues of quality control and quality assurance, our recommendation is that HHS certify test operators. A good model for this is the Federal Department of Transportation on-site alcohol testing certification program. I participate in that. I am a certified test operator myself. It seems to be successful. It seems to be working. There seem to be a minimal amount of problems with it, and we recommend that model.

There are several ways that quality control can be implemented: first, in the manufacturing process, operator training, and when each test device is run. Each test device, at least the types of tests that we recognize, are unitized. It is a separate testing device in a separate unit with its own built-in quality control. This is different from a laboratory test where, if something goes wrong, you can affect a whole run of the test. With an on-site test, there is only going to be one test that is going to be affected. We feel that the operator should keep a quality control log. The manufacturer's instruction in the on-site test package provide instructions for storage of the test and test operation. Operators should be examined in those. They should take an examination to show that they know what they are doing. Of course, they should follow the package insert approved by the FDA.

External proficiency testing, we recommend it. There are a number of companies and professionals who can perform this testing. We recommend that customers who purchase an on-site test have proficiency testing performed twice a year, using an external proficiency testing agency. Julie Murdoch is going to talk a little bit about how, in a practical sense, that is done.

A standard report form is possible with on-site testing. There is no problem at all, using a log book that would have the test results. It should have the specimen collection and test location, the dates of the collection and the test, the identification of the specimen collector and the on-site test operator, if they are different, the donor identification and the results for each substance tested, and the final disposition of the specimen.

The results should be either that it is screened negative or it is screened presumptive positive. We are not going to use the word positive test with an on-site test. We are going to use the word presumptive positive, because we recognize that the test needs confirmation.

Chain of custody is no problem.

As far as an MRO interpreting the results, we feel that the test results should be sent to the laboratory and those results should be confirmed by an MRO. As far as alternative medical explanations go, there is no difference between this and laboratory based testing. MRO training, although it would be helpful for MROs to get some training about on-site testing, in terms of interpreting the results, there is no difference, because they are going to be interpreting a laboratory confirmation test.

The last time we had a meeting here, we talked about whether there was an issue about conducting tests in front of a donor. Somebody raised this as a bubba issue. You know, you give a test to bubba. Is bubba going to punch you out if the result is positive. It just is not our experience that this happens. This is a false issue. It has not been happening with the DOT on-site testing. If you are concerned about it, then we recommend that you just do not tell the donor what the results were. You leave it up to management.

It is the exact same process that happens with a laboratory test. The laboratory doesn't tell the person. Management tells them or the MRO tells them. You are not going to have those confrontations if you don't want to. The U.S. Postal Service does let people know. They tell people when they are negative and then, if it is a presumptive positive, they say, the test needs further testing. I don't see any bruises on Julie's face. If you are concerned, you don't have to tell the donor the results. It has not been our experience that it is necessary.

I would like to bring up Julie Murdoch and then Stu Bogema. Julie is going to talk about the Postal Service testing program and show you how an on-site testing program can work.

MS. MURDOCH: Good morning. I think that Bensinger Dupont happens to be in a very unusual position. In 1996, Bensinger Dupont was asked by its client, the U.S. Postal Service, to look at the possibility of taking just its applicant testing program to an on-site program. The process actually started before I got to BDA, but became my baby shortly after I got there in August of 1996.

The U.S. Postal Service is the perfect client for doing this. One, it does about 220,000 applicant tests a year, so it is a big program. It is nationwide. The best thing is, it is not governed by the HHS guidelines, and it is not governed by state law. It is in the perfect position to do what it wants. There is the Constitution and stuff like that, so you have to kind of think about that, but the Postal Service is exempt from applicability of all state law, so we didn't have to worry about all of those issues. As it turns out, we did have to worry about them when obnoxious health inspectors would come to the collection sites demanding to know why we were going to be doing this, but as a legal matter, we didn't have to. It was the perfect client for doing that, in terms of the context of implementing the program. It was also a very difficult process because we are leading with our chin. There is no similar program anywhere in the world that we are aware of.

The Postal Service is very diverse. There are 100-some different districts and 11 different areas. The Postal Service has over 1,200 collection sites, over 5,000 collectors, and each of the districts has its own little fiefdom. When we first came on board, all of the contracting for drug testing services was done by district. We have since had that pulled up to a national level, but we had some contractual issues as well. The biggest thing was, we didn't know how to do this. You can't just go and -- are there any on-site testing device manufacturers here, by the way? Okay, I will be a little more careful. You can't just go to the manufacturer's package insert or to the internet and see these on-site testing devices and say, oh, that looks good; we will just put that in place. It was a great deal more difficult than that, and we took a very cautious approach because the Postal Service is a target of opportunity, deep pockets and all that sort of thing.

The last thing we wanted to do was to implement a program that would either undermine the current applicant drug testing program, subject it to some kind of challenge, bring the unions into play, because obviously these people aren't unionized when they are applying for a job, their brothers and their sisters and their mothers or whatever are current Postal Service employees and are unionized.

We didn't know the technology. We knew that all the on-site testing devices were using immunoassay, but what was the reliability of the devices. How user friendly were they. That was where we started. We started with a variety of different devices and Dr. Bogema, our toxicologist, developed a reliability study. We did side by side with the devices using spiked and clinical specimens. Eventually, both for liability reasons and also because of some use issues, we ended up focusing on the Roche test cup.

Before I go any further, I will say, I don't shill for Roche or any other company but my own, and I do not represent the U.S. Postal Service. Whatever opinions you hear up here are mine. I am authorized, however, to talk in public about this program.

We did focus on the test cup for a variety of reasons. The single biggest reason, from a practical perspective, is the district owners, occupational health nurse administrators that we had talked to about going to on-site testing, did not want to get involved with the dripping and the dipping, so we stuck with the tipping. They didn't want to do all that pipetting stuff.

Back in early 1997, when we were looking at this, even the pipette ones, there were very few that were one steps. You had to add buffer and reagent, and they were much more complicated. Obviously, technology is burgeoning. We did focus on the Roche test cup. Then we put that through a very severe set of reliability studies to make sure that it was doing what we wanted it to do.

Our primary concern, since this is a pre-employment program, was the possibility of false negatives. We did not want to hire individuals for the Postal Service who would have been screened out using a laboratory testing program. The term, going Postal, has a certain resonance in this arena. What we wanted to ensure was that the device that we selected was as reliable in terms of false negatives as laboratory screening. In fact, in our final statement of work -- Stuart is going to talk about our quality control program -- we imposed a five percent false negative rate, which is actually more stringent than the laboratories, or Roche.

Our first runs with the test cup were not as acceptable as we had hoped they would be. We worked with the manufacturer and now, the device we are quite comfortable with, despite some research indicating different results. Then we started the process of, okay, we have got this wonderful device, or a device. What now? How do we actually put it out there in the workplace and let collectors use it.

We looked at the current collection and testing protocol and also looked at what Roche had recommended for use of the test cup and we decided that we didn't like either of them. So, we developed our own. We do not, for example, collect the specimen in the test cup, which is the way the package insert states that the collection should be done. It is done in a separate cup that has a temperature strip on it and has an attached lid that snaps tight.

So, we developed our own protocol, a draft protocol, and then we did a side by side study of about 1,000 specimens at several different sites in New York and the Washington Metropolitan area. We asked people -- we told them we were doing the study. They had to consent to participating in the study or we wouldn't take their specimen. We took their specimen for the laboratory that was prepared, and an aliquot was reserved or a portion was reserved, and out of the presence on the donor, in that circumstance, the test cup analysis was run. We got very good results from that which I think you have presented, haven't you, Stuart? Dr. Bogema has presented it in at least one form.

One of the things that came up in the side by side, which took a couple of months, was that some of our practical procedures needed to be refined. We took a couple of months to do that, actually sat down with collectors who had done the side by side study, and we developed our own manual. It is a manual that includes information for collectors, information for donors, a log book which we use instead of the chain of custody form for most specimens. We have training materials, just about everything you can think of.

One of the things we decided up front -- and I strongly agree with David on this issue -- is we felt it was critical that all collectors be trained and certified to conduct this program using this device and these protocols.

That is what we did. In August of 1997, we did Postal Service television network training to 95 different sites, and conducted training. It took about two hours. At the sites, we had already pre-positioned the materials, test cups and spiked specimens. So, while we were doing the training, the people actually had that.

Then they took an examination, which takes about an hour, that we had developed. It involves both written, true-false questions, multiple choice questions, and a practical examination involving identifying -- we have digital images of test cup results, and also hands on, three spiked specimens, they have to accurately identify what the drugs are in those specimens.

All those examinations got sent to me and I got to grade them, but those were our trainers. We trained them not only to how to do collection, but also how to do training. Then, once they were certified, we sent out the training packages to us. They trained the next tier of trainers, who are now the ones who are going around and doing the training to the contract collection sites, and to new occupational health nurse administrators.

We actually started doing on-site testing in January of 1998. It started out slow, as organizations sort of got comfortable with the concept of doing on-site testing.

I guess we are now rolled out in just about every area. We have 11 areas, except on the west coast we have one that is a little bit behind. The last I have heard in terms of results coming back, I think we have done about 30,000 tests, between 30,000 and 40,000. That number may be inaccurate, and I apologize if it is.

I get calls, not as regularly as I used to about issues. The issue of state law applicability continues to come up.

I do get reports on every single specimen that was identified as needing further testing during the test cup and did not confirm at the laboratory. I get reports on all of those and any other anomalous result.

We did institute an adulteration check process during the collection process. We use the Adultacheck IV. Stuart can talk about some of the quality control we do on that as well as the Roche test cup. If it is abnormal on the Adultacheck IV, the specimen goes to the laboratory. If it is abnormal on the test cup analysis, the specimen goes to the laboratory.

We have preserved a pristine portion of the specimen. That part is poured into a bottle and sealed under a seal just like the regular procedures, put under chain of custody, and sent.

The issue of the donor seeing the test result and being involved in the analytical process has come up over and over again in these discussions. It is deja vu all over again. These were the same ones that we had when I was at the Department of Transportation and I was doing the alcohol testing. We wrestled with this issue with the Postal Service. Do we have the donor present when the analysis is run, the confrontation, whatever.

I felt very strongly -- this is my personal opinion -- that you get a better participation in the program, more confidence in the program, and less subject to challenge when the donor sees the analysis run and there is no question about what the result was. So far that has been the case. We have had no complaints. Believe me, all the collectors have my phone number, because it is in the training materials. If there was a situation that had occurred with a confrontation, I would have heard about it. I have not to date, and I don't expect to, frankly, just like it didn't happen in the alcohol testing arena.

We do have very stringent controls over the training of the collectors and over how the collectors run their program, how they actually do a collection, and what we do with the donor. If the donor doesn't want to sign, initial, the log book after the analysis is run, the testing process is stopped. If they don't want to initial the specimen bottle seal, the testing process is stopped. If they don't want to whatever, the testing process is stopped. So, we do exercise some control there.

As I said, it seems to be going fairly well. There are a number of legal issues. I don't really have time to go into them all. One of the things was -- and Dr. Costantino and Mr. Fortner are both here -- we had just let the Postal Service laboratory contract like two months before we rolled out on-site drug testing. That was an issue for the laboratories, needless to say, because we only send to the laboratories those that need further testing. We don't use the word positive at all, by the way. Further testing needed is it. That is all we use. So, we did have to renegotiate with the laboratories on how the pricing would be done, and they had notice in advance on that, that that would be coming. So, that was one issue.

As I said, so far it has rolled out nationally, but the pace is beginning to pick up. The Postal Service's real hiring period is starting now and they will hire probably 100,000 people between now and the end of the year for temporary hires. We do not, at this point, have any intention of trying to go -- the Postal Service doesn't, I should say -- more broadly with on-site testing.

This is a pilot program. It may turn out that the costs exceed the benefits, but I tend to doubt it. The primary benefit that the Postal Service was looking for was reducing recruitment costs. They were losing a lot of recruits between the time the recruit was given the drug test and was offered a job and the time the test result came back and was given to management and what not. Even with a negative, that is often two or three days. When you are talking about this population, they have gone to another job by then. The Postal Service thought if the hiring process could continue right there on the spot with the negatives, they would reduce their recruiting costs, and that would more than offset any additional cost for testing. So far, that seems to be the case.

Now, Stuart is going to talk about quality control. The quality control doesn't end with the training of the collectors. It is an ongoing process. We put in fairly stringent controls in the statement of work and are continuing to monitor that. I would like Dr. Bogema to talk about that, because he is a toxicologist and I am not.

DR. BOGEMA: The quality control program for the U.S. Postal Service on-site drug testing program basically consists of five parts. As Julie mentioned, there is a very formal procedure for training the collectors and individuals who do the tests at different sites around the country for the Postal Service, standard operating procedures as well as a training manual for that.

As part of that process, both the instructors and the trainees receive a set of specimens that they can use to practice with, to demonstrate the device, and also, for the trainees, a set of exam specimens that they test and report the results back to Bensinger Dupont and Associates, in order to be, in fact, verified that they have completed their training. They also receive the examination that Julie mentioned to you.

On the device itself, on an ongoing basis, random basis, we take a lot of devices and do a reliability study on that lot of devices when we first receive them, and then at three-month intervals for a year, to check the stability of the device, to check to make sure that they are meeting the specifications that have been set forth by the Postal Service and Bensinger Dupont.

On an ongoing basis, to ensure that the collector testers are doing the testing properly, we send out on an open blind PT basis, a set of specimens for them to test with the device and report the results back to Bensinger Dupont. This testing has two purposes. One is to check the tester themselves, to make sure that they are continuing to perform the test properly and interpret the results properly. Also, it is a continued quality control of the device itself, at all the different sites around the country that they are being used.

The random blind PT is being set up so that sites that do more testing are going to be tested more often for this proficiency testing survey. Sites that do less testing will be selected less often. The next step is that -- this program will be going into effect soon -- a percentage of negative specimens will be routinely sent from the sites to the laboratory for analysis at the laboratory, in the standard laboratory testing procedures, as they check on false negatives in the program. Finally, the last part of the quality control process, is that non-negative specimens, as well as abnormal specimens tested for adulteration and dilution go to the laboratory. They are deemed, further testing necessary.

Julie mentioned that she gets information about each specimen that is not confirmed at the laboratory. We then periodically go in and do an audit of those specimens to determine what the issue is, with why it did not confirm. About half of them are legitimate, either borderline specimens, specimens that were positive by screen that did not confirm because of a cross reacting substance. In some cases, they are positives from the laboratory that are overturned by the MRO.

The other half are specimens that screen completely negative at the laboratory. We then look at issues at a particular site, to see if there is an interpretation of a result issue. Some retraining has taken place. In many cases, these issues have been resolved.

Most recently, the first of August, we instituted a more comprehensive dilution and adulteration check within the on-site testing process using the chimera research and chemical Adultacheck IV dip stick. This stick tests for creatinine, nitrites, glutaraldehyde and pH. As part of the training and quality control samples, we now include specimens with low creatinines.

We include specimens that are outside the acceptable pH range and we include specimens that are adulterated with nitrites, both as part of the training process because the collector trainers have to be able to identify these specimens with the Adultacheck as well as with the quality control process, the blind PT process.

So, these are the elements that have gone into this continuing program to maintain quality within the on-site testing program. Any questions?

MR. EVANS: I have a question for Stuart and Julie. What do you do if the test result is not clear? Sometimes I have heard some criticism of on-site test devices that when you look at the read out, you can't really tell what the result is. What would you do in that circumstance?

MS. MURDOCH: Under the instructions, the way we have framed it, if there is any blue present, that is a negative result, and that is the way it is called. If there is any question that it is not clear, it is sent to the lab for testing.

DR. BUSH: Thank you very much. Up next, for a description to us of information concerning the sweat patch and testing for drugs of abuse with sweat is Mr. Neal Fortner. He provided us information which you have written copy, and he has got the power point slides here, which I am sure contain similar information for ease of discussion at this time.

MR. FORTNER: While Donna is going through and setting this up, you should have a couple of hand outs. One of them, as Walt talked about, is the update modification proposals for incorporation of sweat testing into the federal program. Then you have -- it is easy to read, it is labeled, Sweat. It is obvious, if there is any question on that. Then you have this power point presentation.

I would like to extend my appreciation to Donna, Walt and the members of SAMHSA and the Drug Testing Advisory Board.

What we are really going to do today is give you an update. When I got the call from Walt saying, look, we would like to go back and revisit that, it was going to be a very clear focus.

Here is the checklist. The checklist has, in essence, not changed since our meetings last year. Please address those issues where there is a ‘P’ saying it is possible, or where there is an ’I’ saying it is incomplete, saying we don't have enough information at that point. Appropriate with everyone going back to school -- my kids just started -- we are going to do checklists today. We are going to fill in the blank. Section B, I have sort of gone through and just labeled these like the checklist itself.

B-1 says, does a collection training program for collectors exist. The answer is yes. We have had a program since late 1996.

One of the things I asked Walt and Donna, I said, how much information do you want. How much paper. We don't live in a paperless society yet. The response was, please keep it to a minimum. If we want more information, if we want the volumes, we will ask for them. So, what you have got here is a very condensed version. Certainly, if members of DTAB want additional information or want to see the whole thing, I will be glad to send it to you. Some of it is quite large in volume, I think. Dr. Huestis can personally attest to that. Be careful what you ask for. Sometimes you will get it.

So, this program, which is a formalized program for collector training, is done on site. It is also a train the trainer program. It has now evolved to the point where it also uses videotapes. You can do teleconferencing with it, you can do phone conferencing, I guess, if you want, along with this program.

Pursuant to that, of course, there is a training manual. The training manual goes through not only application (patch) removal, it goes through a lot of other aspects with respect to some of the questions that are raised later on specimen integrity. It talks about how you ensure the sample in the sweat patch itself is still intact. How do you look for adulteration. What do you do. How do you ensure that the patch that you put on is, in fact, the patch that came back to you on the observer.

As part of that, we went to a certification program. So, this is a training program that says, we are going to give you an hour and a half to two hours, is the typical length of that program, to train a collector. Following that program, there is a written exam. The written exam then results in certification of that person as a trainer. This is the program that we have utilized under the U.S. federal court system for certification of collectors.

The answers to B-1, B-2, where they were listed as possible, are yes.

Section C goes on to the collection device itself. C-3 says, does the sweat patch collection device have an FDA clearance. The answer is yes. Hopefully you will be able to read what is in your handout. I tried to blow it up a little bit. In October of 1990, the sweat patch was developed by Sudermed Corporation, was cleared by FDA. This is, you know, the top third of that letter from FDA coming back as a collection device. That was the only issue that was listed as an incomplete in Section C.

Section D, are sweat patches subject to multiple tests. Well, yes. Currently, the approach is to use an immunoassay. Actually, it is an ELISA screening technique to the patch eluate. I wasn't going to get into any of the testing requirements. Those were explained in rather, I think, detail in our last couple of meetings that we had. Again, if we wanted to revisit that, we certainly could do that.

After the initial screening, which currently screens for the five NIDA drugs, if you will, the patches can then be subject to two confirmation tests using gas chromatography/mass spectrometry. Again, we won't go into that. The multiple testing was listed as incomplete.

Potential to split specimens: The answer to this, I think, is no, not in the formal sense that you think of a split specimen. I mean, one void is classically how it is run under the federal programs. That one single void of urine is then poured into two independent cups, independently sealed. There are analogies to that, in that the patch currently is a single device. So, if multiple devices wanted to be used, you could do that and you would have, in essence, a split sample.

It is possible to enlarge the size of what is the window frame and incorporate two patches underneath the same frame. There you would have a split sample from the same collection site.

Now, the FDA process that we will talk about briefly into this presentation did look at applying multiple patches to individuals on a variety of sites on the body, typically the upper arms, lower rib cage and back. These were involved in clinical trials and controlled dose studies. Those analyses show that there is no significant difference with respect to drug detection, with respect to wear site. So, you could apply multiple patches if you wanted to run a split sample program.

D-4, has stability and storage of drugs in the sweat patches been evaluated. Yes. I mean, there were a number of studies conducted under the FDA filings for this. What it entailed was taking patches that were worn, spiking drugs onto them, sealing them and looking at reference patches and then taking similar patches and sending them out through a variety of commercial couriers. So, we are looking at effects of seven days to 28 days on the storage ability, retainability, detection ability of those drugs and patches.

The other program that we looked at has to do with storage of positive samples following GC/MS confirmation. Those samples are currently stored in accordance with HHS guidelines at -20 degrees or more. They are stored for at least one year. We have patches going back farther than that. We have had the opportunity to go back and re-test them. Because they are in a methanol buffer solution, we have found very good stability of drugs. Certainly, the ability to go back and do retests six months to a year later certainly exists. The time periods on the wear studies and transportation at room temperature varies from seven to 28 days. The results indicated that the effective transport through commercial courier systems, mail, overnight express, local courier, showed no significant differences.

E-5. This one, I think, is also listed as an ‘I’, incomplete, more information. Can the patch be evaluated for specimen integrity. The answer is yes. As part of the specimen collector training program, collectors are trained in the observation and detection of tampering. The sweat patch, if you remember from the last couple of DTAB meetings, is covered by a polyurethane device. That device has an adhesive on it and you can detect tampering of the sweat patch itself by loss of integrity on that patch. Also, in instances where we have seen attempted adulteration of the patch, the pad itself physically discolors. So, it is capable of detecting whether the integrity is still intact.

Again, talking about tampering adulteration, it is tamper evident. The patches have a unique bar code, identifier, imprinted on the patch itself, such that that number is recorded when a patch is applied. When an individual comes back, at least 24 hours later at this point, that identifier is verified in the removal process. So, you can track integrity, track attempted substitution.

Are there any unique requirements for transportation? This goes under E-7, listed currently as a ‘P’ for looks like it is possible. The answer is no. These are currently sealed in forensic bags in accordance with appropriate chain of custody documentation. There is a unique chain of custody for each patch and they are currently being sent, transported, across the United States by a variety of courier systems.

Does short and long-term storage of the sweat patch -- we are skipping now. Section F is related to on-site testing, so this is G-2, ensure specimen integrity. The answer is yes, in here. Again, utilizing established chain of custody, forensic procedures, sweat patches are removed at a collection site by a certified observer. That individual ultimately seals the sweat patch inside a forensic bag, labels it with appropriate bar codes, very similar to the processes that currently exist under the federal testing program for urine specimens. There are initials of the chain of custody, of the tamper seal on the bag itself. Copies of the chain of custody are then subsequently submitted to the laboratory with the sweat patch. They are handled in accordance -- all the testing that we do in our laboratories are set up to follow the federal guidelines, whether they are federally mandated or not. The sweat patch is certainly just another one of those alternate fluid matrices and follows the same general type of handling, restricted access, chain of custody.

G-3, this is indicated as an ‘I’ for incomplete. Can identify adulterated, substituted specimens. Yes, the sweat patch does have this unique identifier on it, that is recorded when the patch is applied. When the individual reports back to the collection site for subsequent removal, you need to verify that identifier. They are unique. We have had instances of individuals trying to substitute that particular patch. It is readily detectable to do that at that time.

Some of the adulterants that we are seeing are where people are attempting, we believe, using a syringe underneath the skin or underneath the tegaderm, to insert items such as bleach, number one, and physically alter the patch itself. It discolors it. Depending upon the severity of the material, you can get loss of integrity of the membrane itself. If they go so far as to use concentrated bleach, if you remember from last year's presentations, the polyurethane dressing is a non-occlusive device, and it will trap anything larger than water dimer molecules. So, they end up with chemical degree burns underneath the skin. So, yes, it can certainly address those particular issues.

I think we talked about that. Specimen collector, this is an integral part of the training for looking at specimen integrity, looking at the identifiers on the patch, looking to ensure that the polyurethane adhesive is still intact around the outside of the pad.

G-4 talks about initial testing. Four and five are the testing process itself. It says, is this an FDA cleared test. If you look at the subsequent documents that are in there, this is part of a larger process. Actually, back in 1992, we undertook a process to submit sweat patch testing to the Food and Drug Administration. It initially started out as, could you go and do direct GC/MS on every sample. Yes. Demonstrate that the device does collect drugs. It then evolved to link it to an immunoassay screen. We settled on an ELISA technique. That technique, in addition to the clinical trials, controlled dosing, GC/MS, were ultimately, over a two-and-a-half year period, submitted to the Food and Drug Administration. In 1995, FDA cleared the first three classes -- cocaine, opiates and the amphetamines. Then, in 1996, marijuana and PCP. What you have here are just copies, again, of the top portions of the letters from FDA, which cleared this process under their programs. This is one of the documents that I have not submitted. They are rather large in size. Certainly, they are available to the Drug Testing Advisory Board. The summaries are probably 800 to 1,000 pages per drug. So, they are available. Be careful what you ask for.

We can go through. This is the next couple. So, you have PCP, marijuana, cocaine, opiates, methamphetamines.

4-B, do they detect target analytes. Yes, I had to answer yes, in that they do detect target analytes. However, as with some of the alternate matrices, we are not looking at the same compounds that you typically look for in urines. Sweat testing is really more akin to some of the testing that you would do in blood, or maybe in oral fluids, as it has a predominance to look for parent drug. If you look at the typical excretion pattern -- this was developed, with the exception of PCP, these were all based off of controlled dose studies.

FDA and NIH simply do not permit, or we couldn't get clearance, to administer PCP to individuals, although there was no lack of volunteers to participate in those studies. But there are some efficacies to deal with, and the Investigational Review Boards.

If you look at the class amphetamines, the predominant drug that you are seeing in sweat is methamphetamine. The ratio of methamphetamine to amphetamine is significantly different. We believe that part of this is involved in polarity of the drug, excretion of the drug across the skin tissue itself.

With cocaine, the predominant drug we see, again, is parent drug. Cocaine is, by far, the largest analyte that we detect, followed by benzoylecgonine and, in some instances, depending upon use, ecgonine methyl ester.

The opiates, you see the classical coding in morphines. The patch is rather unique. Because it is a storage device, it will detect parent heroin as well as major metabolite 6 acetyl morphine. This is one of the unique properties of the sweat patch itself.

PCP: it detects parent non-metabolized PCP. Marijuana: it detects parent drug. It does not detect, with a high degree of specificity, the carboxylic acid form.

What we are looking for in most of the patch analysis or the analysis of sweat is parent drug, versus metabolites which you typically see in urine samples.

4-C says, does it use HHS testing levels. Well, here the answer is no. With many of the alternate matrices, you really need to go back and look at what you are measuring and what is the cutoff to use. Those issues, I think, for many of these alternate fluids or samples, have yet to be resolved. We use slightly different cutoff levels, and that is an under-statement because they are significantly lower. But they are based on empirical data, and they are based on receiver operating characteristics following controlled dose studies. Typically, what you are going to see in the screening level for amphetamines, cocaine and opiates, is a testing level of 10 nanograms per milliliter of sweat eluate, sweat material. PCP is at 7.5 and marijuana is at 1.5.

4-D said, do you have, or can the assay demonstrate acceptable performance around the cutoff? This is in accordance with the classical plus or minus 25 percent.

What I did was, I went through. If you want to pick a worst case scenario, you don't look at within-day precision or within round precision. You look at day to day precision. These are numbers that are generated off our analysis. This represents day to day precision over a six-month period. These are our CVs. These are in accordance with plus or minus 25 percent, using the cutoffs that I previously showed for the initial screening assay. I think you can see that, even down at 10 nanograms, or down to 1.5 nanograms, we are looking at CVs that are below 10 percent. So, reproducibility around the cutoffs in this particular area.

4-E says, is there an ability to repeat the initial test for sweat. Yes. As referred to in D-2, you have, in essence, about two milliliters of sample to work with, after reconstitution, after pulling the drugs off the patch. The ELISA technologies typical with urine technologies, use very small 5 microliter samples. Then you are looking at a half to maybe one milliliter for GC/MS confirmation. Can you repeat the initial test? Yes, there is more than adequate sample to be able to do that.

5-A says, do you use mass spectrometry for confirmation. Yes, that is the only appropriate method to be able to get down to the levels that we are looking at here. With the exception of THC, all these assays are currently done using electron impact ionization. It is an EI mode. THC is done currently using negative ion GC/MS. You will see that in the cutoffs that we are using here you are looking at about 0.5 nanograms.

Again, there was a question in the checklist that said, demonstrate reproducibility, acceptability around the cutoff. This is 6 month's worth of day-to-day precision data off of our GC/MS runs. You see probably 11.6 percent is the highest CV that we looked at. THC is close, but again, we are looking at 0.5 nanograms per mL.

Do the cutoffs used for sweat patch testing reflect drug use? Yes. The FDA clearance procedure, as previously alluded to in the receiver operating characteristics analysis, looked at establishing cutoffs based on controlled dose administration, two drugs by volunteers. This is how we came and established what is the appropriate cutoff to use for both the initial testing and for confirmation testing.

Is there a certified laboratory program? We are now into section H-1 for sweat patch testing. Not at this time. However, PharmChem has worked with several other laboratories to help them develop testing procedures for sweat patch testing. We readily share that information. There are several laboratories involved in NIH/NIDA research grants, as well as some other commercial laboratories, who are interested in providing retest capabilities for the sweat patch. In those instances, we have provided them with quality control/quality assurance standards. If they ask, we will provide them with methodologies as well.

Is there an external proficiency program for sweat patch testing? Yes, we do have one that is developed. Now, it is not developed in the same format that the current NLCP program is developed, although that is certainly capable of being developed. We currently have a client who runs a double blind proficiency testing program on us for sweat patch testing. They use worn patches. They will send both negatives, and this client has the ability to spike patches with specific drugs. So, it is a spiked, double blind PT program.

Is there a laboratory inspection program for sweat patch testing? Not at this time. However, this is one of the possible categories that, with some refinement and if you look at the draft document that says sweat up at the top, it incorporates the same type of checklist items and programs that you would need to develop an inspection program in accordance with HHS, NLCP guidelines.

Blind sample testing, I think that there are a couple of different areas here. I previously described that there is an external double blind program. In accordance with our procedures, we also run an internal double blind program, using both negative and spiked patches, both the initial test and then on to confirmation testing.

Does a certifying scientist review lab results? In accordance with all our procedures, yes. I mean, it is a review of all the ELISA, immunoassay data, GC/MS, quality control, quality assurance, all chain of custody documents prior to the reporting of a result.

Are they reported by specific drug? If you look at these next sections, I-1, I-2, I-3, they are all going to say yes, see laboratory report. So, if we just skip through these, in today's technology, electronic transmission, you can generate just about any kind of laboratory report that you want. Certainly you could incorporate this laboratory report onto the custody and control form, if you want to exactly mimic the current federal program.

This is just an example of a standard laboratory report. It certainly contains all the pertinent information, screening levels, confirmation levels, test results, through the certifying scientist visit.

Does a program exist for training on the interpretation of sweat patch results? Yes, we have a formal program. The basis of that program is the scientific analytical data generated out of the FDA 510(k) filings. This is a program that is currently used to train existing clients with respect to interpretation of those. We use it quite a bit in the U.S. federal court system. We currently do not have a large number of clients in the workplace that are doing this, but this program is adapted, over doing it over to that side. It is about a four-hour certification program, just for the interpretation of results.

Does a program exist to look at alternative medical explanations of sweat patch results? Again, this is the same program that incorporates the data out of the FDA 510(k) as well as a number of other independent researchers, scientific articles that are published with respect to the testing of sweat. Sweat is not only used in the United States as an alternative matrix, but there is a fairly active group in Europe that uses sweat testing as well, to look at metabolism of a wide variety of drugs.

Does a program exist for MRO training? Yes. This is the same kind of wording as before. Training of medical review officers, in terms of interpretation of results, in terms of alternative explanations, all come down to summary review of the scientific data in the FDA 510(k)s.

Are sweat patch test results resulted to dose time response? I would have to say no. While you can go through and correlate, based off of the FDA controlled dose studies that were done, certain levels in a sweat patch, you can get some general ball park information. Remember that this device is typically worn for at least 24 hours up to 14 days. Therefore, it is a storage device. Multiple drug use or multiple drug exposures under that program increase the levels in the patch. So, it becomes very difficult to go back and say, this is a specific concentration related to a specific dose. Like urine testing, it is a positive/negative approach.

Can the sweat patch test be contaminated? Dr. Cone, I think the answer would be yes under some circumstances. Dr. Cone and his group have reported, to the best of my knowledge, one instance involving external contamination.

Certainly is the skin area is not adequately clean or if the individual who is applying the patch has access to drugs, particularly the parent drug, not necessarily metabolites in urine, for instance, and touches the pad, you can contaminate the patch.

There is an issue that has to be addressed in the collector training with respect to that. The levels that you typically see, if it is a contamination due to external drug application, are significantly higher than what we have seen in any of our controlled dose, clinical trials. Yes, there is an area. All samples have some degree of susceptibility. Safeguards need to be implemented, and you need to be aware of those to address those.

I think we are at the end of the check list. Hopefully this has provided the Drug Testing Advisory Board with sufficient information to address the Ps and the Is.

If it was left blank, my instructions were, don't address it, because we think we have enough information. Certainly any other information that the DTAB would like to review is available upon request.

DR. BUSH: If the board members have any specific questions at this time, could I ask that you address them now. We have just a couple of minutes, if there is any question from the audience.

PARTICIPANT: On the testing levels, what is your LOD or LOQ?

MR. FORTNER: LOD, LOQ. We handle that the same way that we handle our urine testing. It is empirical data based on either serial or spiked samples. For the amphetamines, cocaines, opiates and PCP, we are between one and two nanograms per mL. On the THC, we are at 0.05 nanograms for our LOD/LOQ, using GC/MS negative Chemical Ionization.

PARTICIPANT: (Question off microphone.)

MR. FORTNER: Actually, it looks at both. If you look at the ELISA, it is precision of the ELISA assay. If you look at the GC/MS, that is really a function of both extraction efficiency and instrument efficiency. It looks at a worst case scenario, encompassing all variables, on a day to day aspect, as opposed to within round.

PARTICIPANT: (Question off microphone.)

MR. FORTNER: Recovery of the drug or efficiency of recovery of the drug in the ELISA assay? No, those really look at precision reproducibility. They are semi-quantitative at best, so you have above and below control. We really treat them as positive, negative under that program. What you are really looking at is reproducibility of the absorbent type readings that you produce, and not true concentrations. While we run a standard curve, we do not run it in a quantitative mode. It is a semi-quantitative.

DR. BUSH: Thank you, Neil. I would like to ask Dr. Sam Niedbala to present the information on oral fluids.

DR. NIEDBALA: Let me start out again by thanking the Board for allowing me to speak on the use of oral fluids. If you noticed, everybody is being politically very correct this morning and saying oral fluids. So, we were looking for something a little bit more tangible and that everyone can grab hold of and feel comfortable in every situation, and oral fluid testing certainly feels that way. Actually, there is a purpose to that statement. Oral fluids are really not as simple as they may seem. Back in, I think it was 1992, there was a meeting in Florida in Panama City, organized by the New York Academy of Sciences, to take a look at oral fluids as a diagnostic medium. It was a really interesting meeting because there were lots of topics covered in the potential uses of oral fluids. One of the conclusions, I think, that overrode everything was that oral fluids become a medium by which you can do qualitative testing for the identification of the presence of materials. As a diagnostic fluid for use in things such as therapeutic drug monitoring, it definitely had limitations. I wanted to just spend a few minutes, again, bringing everyone up to date on oral fluids in general. I have one diagram out of an article talking about HIV testing using oral fluids, on which to base this initial discussion.

Also, I wanted to explain a little bit about my position. I am from STC Technologies. We are a manufacturer of diagnostic test kits. The things that I say are as neutral as possible in terms of the collection devices that are available potentially for use in testing oral fluids, as well as the technologies for screening those. I will try to keep my comments within the context, but I think it is fair for everyone to know where I am coming from, and what I do for a living, as far as drug testing is concerned.

As far as the oral cavity is concerned, it is really a complex area of the human anatomy. There is something which everyone perceives to be saliva, but there really are a number of components that exist within the cavity. First, the oral cavity itself is surrounded by a great number of capillaries that are circulating serum components, obviously. Many of those components do cross the membrane barrier into the oral cavity, and this dose include drugs of abuse. Much of the work in past history has focused on drugs of abuse in the cavity itself. There are only a few papers that really touch on the potential mechanisms which, for the chemists in the group, seem to all relate to the pK of the drug which, again, is related to the pH of the oral cavity at any one point in time.

There are some constants that I think we can lean upon, and that is also why, again, I will keep harping on the use of oral fluids rather than saliva. True saliva itself comes from glands located around the cavity itself. These parotid glands as well as sub-mandibular glands, contain mucins and other materials which we would characteristically think of as the sort of things that saliva is made of. To name just a few of the things that are contained in saliva, because the list is extensive, it is not a simple fluid. There are lots of proteins that help with digestive or hydrolytic functions, amylases, gluconaronadases, esterases, phosphotases, ribonucleases. There are some proteins that have anti-bacterial properties, lactoferin, lysozyme. There are anti-viral properties, cystatins and anti-fungal properties, histatins, that are all in saliva. There are other things that are mucins, which are characteristically what give saliva its sticky, bad appearance, but it is really not saliva that we are working with, I believe, when we are measuring for drugs of abuse. There are those components present, but all the collection devices, in some way, shape or form, seem to deal with the mucins themselves, and get rid of them as part of the preparation before testing.

Most of the language now used in the literature refers to these other materials as mucosal transudate. Now, you have to say that 10 times very quickly, mucosal transudate. Where that occurs in the mouth is right here. There is a certain amount of exudate that comes out from the provicular regions here, and a certain amount of these serum plasma components are collected in the buccal cavity. Most of the devices that are currently available for collection all revolve around putting a device like a lollipop stick into your mouth and putting it into that cavity, or moving it around the mouth until it is sufficiently full. There are a number of devices, some more sophisticated than others, that simply collect the fluids that are there, all the way to pads being treated with a salt solution that creates an osmotic gradient, which forces more of the serum components into the collection device. I am not reviewing those devices today. Really, what we are talking about is the mechanism of how to use these materials for drug testing.

Just a quick review. This is going to be kind of busy, but it is available for those who wish to have a copy afterwards. Just to take a look at some of the analytes that are of interesting, picking the top five and then looking at some of the references that are there, I want to give a perspective on looking at oral fluids in urine and blood, what the relative windows of detection are from the literature, and then what are some of the drugs that were detected. When you look through this, you see very quickly that the same materials that we are looking to detect in urine today are the same as can be found in oral fluids, I think THC being the most debatable. Some of our most recent work, as we are expanding some of the alpha and beta testing that we are doing with our products, suggests that with THC we are not only going to consider the parent, but we are also looking at ways to possibly detect at very low levels the THC carboxylic acid. This is a change from some of our thoughts about a year ago on THC, and I will cover this a little bit more as we go through things.

I wanted to give a general overview of what the literature says about oral fluid testing and our drugs of interest. In general terms, I think that there is a testing algorithm for oral fluids that can be relative to any of the devices that are out there.

First, oral fluids has both opportunities and dangers in drug testing. I think there are ways that there is clearly superior characteristics and ease of use that we could take advantage of for particular situations. I think in addition to that, some of the dangers are we have to recognize that oral fluid has its own potential, call it adulteration or lack of adequate specimen issues.

The collection devices that are on the market, not all of them have a way of you knowing that you have collected an adequate amount of sample to support your testing.

What we have developed over time is something that tests for IgG, again, a simple ELISA that goes along with our drug screening products, which we find works very well, and is established in the literature, that there is a minimum amount of IgG present in any human oral fluid specimen. That tells us, number one, that it is a human specimen and, number two, that there was a sufficient amount collected to support going on for testing. If you look at the algorithm, it is not unlike that of what we currently do for urine testing, in a very simplistic fashion. What we do right now is that when the tests arrive at the lab, we test it for IgG to make sure that it is adequate. If there is less than 0.5 nanograms per mL, then we go back and report it as inadequate. If it is sufficient, then we go on and screen it by EIA, going through the normal initial positive versus negative, on to confirmation. If it confirms, then we report it out as positive.

To get to the crux of things, then, what I did, again, in a simplistic way, was go through the factors that were listed for reliable workplace drug testing, and try to compare what is done for urine testing versus what could be done for an oral fluid specimen. You will see, in many cases, it is just the same. I think there are a few areas that we will focus on that really show the differences and the opportunity that this particular matrix may provide.

If we look through some of this, again, at the collection site, the preparations, security are both the same.

Privacy, no unauthorized persons allowed; collection shall be private. Private in the sense of a urine test versus an oral fluid test are two very different things. There are many cases where people, for instance, in the retail industry are uncomfortable taking someone in to get a urine specimen, whereas it is very simple to sit across the table, hand them something that is a pad, a lollipop on a stick, they put it in their mouth, it is put into a tube and sent off to the laboratory for testing. That is a very different dynamic and one where privacy becomes a relative thing, because it can be done in someone's office as a matter of doing business. In terms of the observed collection, you can do all the same things you do with urine. There is a secure chain of custody when something is to be sent back to the laboratory for analysis.

As far as part B, the collector, is concerned, the training requirement, this is one that has been interesting going last, because I get to hear what everybody else has to say on this particular issue. I would like to think there is a fair amount of consensus here. Number one, everybody feels the requirement for training is a good one and there are ways to address it. I feel the same way about oral fluid collection. There are models that already exist. The Epitope device, one that we are very familiar with in a lot of our work, has been put through intense scrutiny, because of its clearance by FDA for HIV testing. That has not been a trivial matter for the company itself. As a public entity, I know that in their trials over the years, they have gone through $48 million to bring that one to market. I am happy to say we are not doing that for drug testing, are we?

In doing that, they have developed some very sophisticated ways to make sure that the collector is qualified, has ongoing training, and that the specimen is adequate and well controlled. We can take a lot of those things and use them as models for drug testing. In addition to that, I think as far as the program itself, for certification, STC, as a company, lives in a couple of different worlds. One, we do regulated testing in the laboratory setting. We also have an on-site alcohol test that has been used for many years for DOT. We were part of helping develop some of the training modules that are used now for screening test technicians, or we start out with oral fluid test technicians. Those things have worked very well, and other comments this morning really support that, that those models exist.

The next part, part C, on the collection device. The collection container for urine is very well defined. For oral fluids, there is a number of devices, as I keep saying. Basically, there is some component of plastic with a cellulosic swab that is somehow treated or untreated, put into a secure collection container, and then sealed and ready to be sent to the laboratory. The same sorts of tape and other tamper evident security systems can be used with the oral fluid specimen.

Then, as you drop down to the other requirement in terms of FDA clearance, we believe that, again, FDA adds a great deal of value in terms of looking at these potential products. What we have done over time, as per our work with the sweat patch, is that the system of testing is really composed of two clearances by FDA. One is the collection device itself and the other is for the kit. So, what you really have is that this collection device cannot be used with some other test kit, and that this test kit cannot be used with some other collection device without allowing FDA to look at that and go through the normal clearance process.

I think it is a valid system for us to use and look to in the future. At this point in time, we have 4 out of 5 of the SAMHSA drug panel cleared. We are still working on PCP. As far as the impact of the device on the specimen, there have been many articles over time that have dealt with the issue of THC sticking to collection devices, et cetera. I think, again, in oral fluids this is in the area of work that needs to be taken on by each of the manufacturers.

As part of the validation of the collection device, each must look at this issue and report out to potential users what they must be ready to contend with. Cellulosic materials, as a matter of course, are very sticky to certain compounds. You just have to recognize that in the science and develop the products appropriately.

Part D, covering issues on the specimen itself, the oral cavity is our target. As far as multiple testing is concerned, the answer is absolutely yes. I think as Neil Fortner pointed out, ELISA can use very small amounts of specimen, leaving as much as possible for confirmation.

Split specimens, the way that we have made a recommendation on this is that you can collect two oral fluid specimens. One can be used for the initial screening and confirmation and one can be used for follow on testing in the future. We have asked this question of a number of potential users, laboratories, and no one seems to have an issue with it. I thought it would be, but apparently not.

As far as storage and stability, looking at oral fluids again, the data needs to be generated for all devices. I think it is a challenge for the manufacturers of both collection and test devices, which will be challenged by FDA, to show that the storage and the stability conditions and limitations are well documented for any one of these devices as a system.

In terms of the guidelines that the Board may put forth, I think that you can cover that saying that it is a requirement and should be well documented.

In terms of a collection procedure, in most instances, it is the same. The only real difference in the first part is that, for the chain of custody form, we need to modify and include an oral fluid specimen.

As you look at the tampering and adulteration, I don't see the same requirements in terms of direct observation as a backup, because you can really do that in the first collection step. You will also use an IgG for sample adequacy of some sort. IgG is a recommendation to begin with, that you know you have a valid specimen as part of the collection. There are a couple of things here. This is the first time I used the infamous to-be-determined criteria for accessioning specimens. There are a number of things that the laboratories, in the course of doing business, have well established for urine testing. I think that is essentially going to be the same for oral fluids. I left the to be determined, because there is some of the paperwork here that needs to be fleshed out, and didn't feel it appropriate in this particular presentation. It is just that it can be done, but needs a little bit more thought here.

In terms of short and long-term storage to ensure specimen integrity, again, I think this needs to be determined, rather than by the lab, by the manufacturer and then validated by the laboratory as part of the bringing forward of this technology.

Then the next part, in terms of identifying adulterated and substitute specimens, I think yes, for oral fluids as well as urine, there is technology available.

The checklist of an FDA cleared test, I think yes. Detect HHS target analytes, I put yes, because I am assuming that there will be one selected as part of this process.

The testing levels themselves, I think part of this is what Neil had said about the sweat patch, in that ROC analysis is what we have used for working up all the oral fluid tests. I think that is a valid scientific methodology that is supported in the literature, and it also supports that the cutoff is optimized to give you the best sensitivity and specificity -- clinical sensitivity and specificity. The rest of this, in terms of being comparable to urine testing, I think, are the same. Where you get down to some potential differences is the confirmatory test next.

GC/MS can certainly work with oral fluids. As we do more and more field testing, we are really starting to ponder whether GC/MS or GC/MS/MS could be the next generation. I think absolutely, positively, the secondary confirmation has to be that gold standard. The question is how golden is that standard going to be at this point in time. I think that at least from our perspective, the jury is still out on this. The science is being generated and, at a minimum, it is still GC/MS. Max, it is probably GC/MS/MS.

As far as cutoffs reflecting drug use, which I think is the next one, you have to remember that the oral cavity is used, in most cases, for administration. There is a period in time where you would see potentially someone smoking marijuana and you would see a very, very high level of the drug in the oral cavity. That quickly absorbs into the blood stream, but then in some cases redistributes back into the oral cavity at much lower levels. This is an important issue that I think still has to be worked out over time. Is there some reflection of current drug use? I think somewhere in there is a yes, but I am qualifying that as saying that really needs more work to really understand the ramifications of that and potential utility.

In terms of a quality control program, it is the same as any of the other fluids that are being used. It can be set up. I think it is the standard sort of exercise people are starting to go through as we look at lots and lots of different fluids.

That segues into a little bit more of the same in terms of certified laboratory program, external PT samples, inspection programs, blind samples. All those are yeses for this.

What is nice about the collection devices that are being used for oral fluids is that in many cases they come with a diluent into which the specimen is stored as part of the shipment back to the laboratory. So, you have the chance and a fluid to work with, that allows you to make up these PT samples, if that is what you should desire, and use them as part of an ongoing quality control program. Certainly those things are available.

The next part is on reporting. It is really the same model as urine, in that there are qualified scientists at the laboratories that are responsible for performing the work, for looking over the reports by specific drug, that there is confidentiality in those written reports, and that there is a standard way of doing this.

I think it really takes place in the same model as what we do now for urine.

Then as far as the MRO, the interpretation of urine versus oral fluids is essentially the same. There are alternative explanations for the presence of different materials, or different drugs within oral fluids. All of them can be explained based on literature, as well as scientific studies that are ongoing. The GC/MS is ultimately used as the dividing line between positive and negative.

I think that in terms of the challenge to the MRO, it is one of training, it is one of coming up to speed on this fluid, just as we have talked about the other alternatives to urine testing.

Just on a couple of miscellaneous issues related to dose/time response, you know, for the moment we are saying, not possible. We are really treating it as a qualitative result.

As far as the specimen contamination is concerned, that is where we are using the IgG as the adulteration or as the test to understand that you have collected an adequate amount of oral fluids.

That is running through the list, as far as what would be required to create reliable workplace drug testing using oral fluids.

There are a couple of issues that I just want to bring up in conclusion. Number one, I think out of all the potential alternative fluids, all of them, I am sorry, have made a certain amount of progress toward commercialization in non-regulated settings. Oral fluid testing has been used for many, many years in the insurance industry for cocaine and cotinine, but only recently have the drugs of abuse come to the forefront as a potential future application. I think the methods, even though we have gone through -- and I speak from my perspective -- we have gone through the FDA clearance process, we are very cautiously moving forward, doing and organizing many, many beta tests before we bring this to full commercialization.

I think this is important. There are thousands of specimens that need to be tested for us to clearly understand, first of all, the collection and all the issues around it, the screening and to be sure that all the right cutoffs are being used, that the confirmation tests have been worked out adequately, and that in every one of these steps, validation is the key word.

For those of you who are not manufacturers of medical devices, FDA has implemented new guidelines for the development of these products over the last few years. We really are held to a new standard right now, and it is a good one. It is one that forces us to the table to do a lot more documentation and testing before a product can be introduced. Carrying that banner forward, we look at oral fluids as a great opportunity for situations which I hope can be comparable in every respect to what we do currently for urine testing.

I guess it was last spring when we had this last meeting and I talked about oral fluids probably being very good for, you know, for cause situations, as an example.

Really, as we are doing many more tests in the field, we are seeing positive rates that equal that or urine. That suggests to me that we are not getting a lot of recreational users. We are really hitting on the people who are consistent abusers of drug. That is a good sign that says, I think the window is going to be a little bit broader for oral fluid testing over time, in that the utility is going to be much larger than I had initially thought.

Over the next year or so, we would be happy to help with any of the documentation and things that the board would need. As Neil said, any one of these submissions that we go through, that could be a thousand pages or more, and not wanting to kill any trees, we won't do that until we are asked.

We look forward to bringing to reality the drugs of abuse testing in oral fluids. Thank you very much.

DR. BUSH: Thank you, Dr. Niedbala.

[Whereupon, at 12:10 p.m., the meeting was recessed, to reconvene at 1:15 p.m., that same day.]

MR. STEPHENSON: I would like to begin with our afternoon presentations, which are going to be additional, alternative specimen presentations and information. It turns out that it looks like we have probably five presentations this afternoon. We have about an hour and a half to do it in. We also have a responsibility to have some discussion afterwards, public comments and follow-on questions. We do have a piece of business to conduct, which will be about the formation and how we are going to do business in terms of the small working groups on the different technologies.

I would like to suggest that each of the presenters at this time try to keep their comments as brief as can be done, consistent with what we have already heard this morning. Try to make a complementary piece to fit with what has already been presented where possible. I would invite members of the Drug Testing Advisory Board to ask questions as they feel the need, although I would suggest that we try to do that at the end. With that in mind, I would ask Dr. Ray Kelly to come forward.

DR. KELLY: I want to thank the Drug Testing Advisory Board for the opportunity to talk this afternoon. What we have tried to do is to kind of respond in detail to the draft guidelines. You have a copy of our response. In this talk, I am just going to try to hit the high points.

In terms of how hair testing can be used in the workplace, it is just a very ideal sample for pre-employment testing, for many of the reasons that you already heard presented. It also avoids common strategies that will help a user beat urine tests, such as temporary abstinence and urinary dilution. Because it covers a longer period, it is more cost effective for employers. When it comes to reasonable suspicion, generally you are interested in a sample that reflects the chronological time of the incident. A similar consideration applies to post-accident. Perhaps it isn't generally appropriate in that situation. Furthermore, it doesn't really define the time interval of use even as well as urine does. It was pointed out to me in the break, though, that in any of these applications it does provide a deterrence factor. There is no question that employees, if they are aware there is a hair testing program in place, they are less likely to use drugs, because it can turn up in any kind of situation. In regard to post accident, again, some of the same considerations apply where, again, a key question is, is somebody under the influence, depending on how a company's policy or even state law is written. Usually the issue revolves around whether somebody is under the influence. With regard to random testing, there is certainly some application there. We have clients that utilize it in that way, and it certainly seems to be a cost effective approach, again, by doing only four tests a year on an individual. Again, because it covers a longer time frame, you are more likely to detect users. In regard to return to duty, there can be a problem with having a positive result due to prior use, perhaps even before the incident that required people to leave duty. People can be gone from work for a period for a variety of reason. If they are gone due to previous drug use, then when they return and have a hair test, it is important to try to exclude the possibility that the test is positive from the prior use. For abstinence monitoring, it is an extremely appropriate application which we think, though, may increase the liability for employers, because there is a longer period between tests. Conceivably, it could enhance liability, because it would be a longer period before they were detected as a user. The chances of detecting them would be greater.

I wanted to say a little something about the necessity for privacy during hair collection. We have discovered that it is important to make sure that someone's hair hasn't been augmented.

In regard to the custody and control form, we recommend that perhaps a split sample not be collected initially because it is possible, as was brought up this morning, to collect a sample subsequently from the same individual, representing the same time period, which is something that just isn't available for urine. So, it makes sense to handle it in that manner. Using as a basis the present CCF, there is really not any change required for A, B, C and E. In section D, if you go along with our recommendation in regard to reasonable suspicion, then perhaps that could be excluded and pre-employment would be included, and random and follow up testing could be included. In step two, we recommend that certain things be added to indicate the site where the hair is collected from, whether it be head hair, which is typically what is recommended, or in some situations it will wind up being under-arm, chest or leg hair. The question of pubic hair always comes up. That is something our lab doesn't really recommend be utilized. The practical reason is that it can be exposed to urine and the quantities of drug in urine can vastly dwarf the amount of drug in hair. Although there is a substantially different spectrum of parent drug and metabolite found in those two samples, certainly the amount of metabolite that can be present in urine can overwhelm the amounts that would normally be present in hair.

Secondly, it gets away a little bit from one of the big advantages that hair has, which is that it is less invasive. You could actually do a collection and you wouldn't really be invading a person's privacy in quite the same manner that you do with a urine collection. But if you are going to collect pubic hair, I think that advantage kind of goes away. It might be appropriate to record the length of hair collected, and additional observations might be such things as whether the hair apparently had been treated recently, bleached, or whether there was any other notable factor about it.

Not any changes really with steps 3 and 4. Step 5, we could eliminate the reference to split specimens, while keeping it for urine.

In regard to analytes on the next slide, it is a lot of times appropriate to test for parent compounds as well as metabolites, which can easily be done in hair, since they are the predominant species present. For cannabinoids we would then recommend THC be tested for, as well as carboxy-THC. With opiates, codeine, morphine but also 6-acetylmorphine, which up until right now hasn't been done in urine, but is extremely appropriate with hair. With regard to cocaine and its metabolites, cocaethylene and benzoyloecgonine ought to be included as well. There are various other minor metabolites, which I don't know whether they should be tested routinely, but there should be a recognition that those things such as ecgonine, methyl ester and norcocaine could also be tested. Basically, no change with regard to amphetamines and PCP.

An overview of the collection, the first step, of course, is verifying a donor's identity. That would normally be done with photo identification. In our facilities in southern Nevada -- and I don't know whether the location has anything to do with this observation -- but it is just amazing how many people come in and don't present proper photo identification. Again, it is not so much employees. We do a lot of other testing for a variety of other types of agencies. We have found it necessary to develop an alternate means of establishing someone's identity. If they can't present a photo ID of any kind, we will just take a polaroid picture of them, have them sign it, have the collector sign it and date it. We send that back to the agency requesting the test, for that purpose. The exact timing of the collection is not as urgent, of course, as it is with urine. If you miss a few hours or a day or two with urine collection, you may have a completely different picture. With hair collection, that is not going to happen. In regard to removing garments, no garments really have to be removed, because the donor has no opportunity to interfere with the collection, if it is properly handled. Any opportunity that he has to adulterate the sample will have already happened by the time he comes for the collection. One is able to have a much tighter chain of custody process in which it basically goes from his or her head to the collector, and into the collection envelope, with no period of time when it is unobserved by the collector. So, it generates a perfect chain of custody.

The donor and collector need never be out of each other's sight, and therefore, the issue that sometimes comes up with urine collectors, whether to allow a donor to bring purses and valuables into the collection facility, isn't a problem. Washing of the donor's hands is, again, not necessary. If they have adulterated the sample, it has already happened prior to the collection, or their attempt to do so will have already happened. The laboratory procedures will serve to remove external contamination.

In regard to item F, the collector leaving the collection area, that is not necessary either, because the whole process can be completed in one operation.

It is almost impossible, in item I, to substitute a hair sample. I mean, it is virtually impossible. I can't think how to do it. That, again, is a problem with urine collections.

Moving on to the next slide, the collector would identify the donor, conduct the collection in a private area, free of drafts. I think the one benefit of doing it free from drafts is that hair does tend to blow around. You have to keep a tight hold on it, and it is helpful if there aren't drafts. There are just some prosaic items, like a chair, a counter, a tray of collection materials, and not a bunch of people walking through to interfere with the process. It is relatively straightforward. The collection process is then explained to the donor, including the potential for cosmetic damage. We have found with, particularly individuals who have very short hair or very tight hair cuts that there sometimes can be a visible wound left in the hairdo. That has proven, in our experience, to be of even more concern to males than to females, surprisingly enough. I already talked about pubic hair. The collector identifies the region of natural hair and has to rule out, at this point, hair pieces, extensions, wigs, and the like. Sometimes people have weaves and you have to ask them to take one of their braids apart so you can get some actual hair that grew out of their head.

Some problems come up that typically don't come up in urine collections, or even the collection of other samples that I have heard expressed today. Occasionally you have people with open sores on their head, bleeding or lice or other indications of disease. Our policy has been to not perform those hair collections. Our general rule is, if it moves, don't collect it. Again, some of you may be wondering what it is like in Southern Nevada. We don't have that many of these cases, but it has happened. What we will typically do now is refer that kind of situation to the ordering agency or employer and ask, you know, what they would like to have done. Typically, they will have a urine collected.

One of the very most common problems along this line is baldness. Some people are just naturally in that state. There are also medical conditions that can induce baldness. The former isn't a problem because usually there is enough hair someplace on the head to collect an appropriate sample. If somebody's head is completely bald or clean shaven, then sometimes body hair will suffice. There is a condition known as alopecia areata, which involves someone losing hair throughout their entire body. In that kind of situation, there has to be an alternative approach adopted. We do collect hair and urine specimens separately. I think that should be part of the guidelines, just to avoid any possibility for urine to get onto hair, since we know that drugs can be absorbed into hair from aqueous solutions.

Particularly, again, as I mentioned earlier, you do have high concentrations of metabolites. If that is what you are looking for, for example, in the case of the marijuana metabolite, it is vital that you not have any contact between the hair and urine. The packaging also should separate them cleanly.

The work area and the collector's hands are clean before collecting the hair. Gloves may or may not be used. We used them routinely in the collections. Then, the rest of it is set forth in the handout.

We do collect from the occipital crown, which is the top of the back of the head. That is, I think, an appropriate recommendation, as far as what the standard sample for the certified testing would be, and the amount and the ability to use multiple sites should be set forth. Of course, it is important to use the packaging material that was presented earlier, in terms of orienting the hair so that, once received in the laboratory, it can be easily aligned and you can easily establish which was the root end.

Transport to the lab is by standard methodology. Hair doesn't have a shelf life. With any kind of reasonable shipping interval and, for that matter, storage interval, it is unaffected, as far as we know, by extremes of temperature and weather that might be encountered during shipment, as long as it remains dry and separated from urine.

The next slide, some of the laboratory requirements, most of these are not going to change, whether you are doing forensic hair testing, or forensic urine testing or any other kind of forensic testing.

The same basic personnel requirements apply, the same security features of having restricted access areas for each type of process that is undergone by the sample, internal chain of custody and receiving process.

There are some slight methodological differences, but essentially the process is very similar than as it would be for urine. In fact, most of these present laboratory requirements in SAMHSA urine testing will serve for hair as well.

The only exception is that, in terms of facilities, it is appropriate for hair and urine processing to be separated, at least minimally, in the stage of accessioning and in the stage of extracting. It is our belief that you can use the same equipment, if appropriate, for initial testing and for confirmation as you use for other sample types, provided you have adequate blanks and controls and standards.

Moving along to equipment requirements, the utility of GC/MS/MS has been referred to earlier. It is essentially necessary to find very small amounts of marijuana and its metabolites in hair. It is useful, in the way we utilize it, for retests, because it has typically much lower limits of detection.

One of the things that will have to be developed as we proceed forward with certification is what represent appropriate confirmation criteria, by another method such as GC/MS/MS, as opposed to the standard ones used in urine. I know there are laboratories doing hair testing that utilize a single ion in, for example, positive chemical ionization, single quadrapole mass spectrometry, or single ions for MS/MS, single daughter ion with no ion ratios. That is one of the issues that will have to be sorted out in the working groups. Regarding SOPs, again, this similar criteria applies. You still need an SOP manual. It will still have to provide the same elements.

The testing levels are something that can be worked out. We have adopted a similar approach to what Neil was talking about earlier, with receiver operating characteristic curves for setting some of our cutoffs. Others are in process. This is something that can be chosen to be appropriate in terms of these new requirements.

In regard to how this initial testing process ought to work, we would recommend that FDA approve the initial testing methods. In many cases, they are already approved for another sample type. So, it should be a straightforward process. I understand that FDA has agreed to fast track the process. I think that adds a level of security and legal defensibility that is important.

As is the case now, authorization would be required to test for other drugs. The principle of initial testing, enunciated by Donna Bush in the distant past, once negative, forever negative, would certainly apply.

Again, if there is any initial test -- other additional initial test -- they would have to follow the guidelines as well.

For confirmation, what I have presented here, because we don't have general agreement yet, is just the confirmation cutoffs that we utilize. I do think we have to look at lower cutoffs for metabolites. Again, the finding is that parent drugs are mostly present, virtually 100 percent of the time, and are at higher concentrations, typically, than metabolites in hair. In some cases, metabolites are actually present at relatively low levels. It might be appropriate to have lower cutoffs for metabolites. Our findings for benzoylecgonine, for example, and cocaethylene follow this rule, and even for amphetamine, which doesn't differ too greatly in structure from methamphetamine. It is a more polar compound and is produced in lower amounts as a result of metabolism. So, it will be found in lower amounts in hair. I think these metabolites are critical to the interpretation. PCP, again, you wouldn't be able to find metabolites in the case of that drug. With cannabinoids, our experience is THC is commonly present, and carboxy-THC is present, not 100 percent of the time. I presented some of those data the last time I spoke to this group in September. Codeine and morphine and 6-acetylmorphine are self explanatory, apart from the fact that 6-acetylmorphine is going to be present in hair for a lot longer than it would be for urine.

In regard to the way of interpreting confirmation results, you are talking about once negative, always negative. You have to allow, in the planning of the program, for the use of not only single stage GC/MS, but GC/MS/MS, and you may wind up with slightly different criteria for those two methodologies. One of the things that could be looked at is to require the presence of metabolites in order to report a positive, although as I pointed out a moment ago, that will cause some known users to test negative.

In terms of storage of samples and data, not much change apart from the fact that you don't have to freeze hair. You don't have to store it any other way than at ambient temperature. It is not a problem storing it for one year and being able to get an equivalent result if you re-analyze it during that period, and again, maintaining data and samples in a limited access area.

In regard to QC -- I talked more about this last September -- there are several challenges, including very low drug concentrations in hair, some variable matrix effects that you can see from sample to sample, based on a sample's cosmetic history, the presence or the availability of limited specimens.

One of the things that donors do is complain about the cosmetic impact. For whatever reason, it seems that users complain more than non-users do. Again, parent drug will predominate with lower amounts of metabolites.

In terms of retesting, with urine, you collected a split sample and you are going to send out that split sample to another laboratory. If you do that with hair, you are going to make an even bigger cosmetic impact, which in some cases won't be noticeable and in some cases, will. Because you can go back to the same individual and collect substantially the same hair, which covers the same time interval -- and I think you do have to set some limits as to how long you will allow that process to happen -- a matter of a few days -- it is not as critical to collect a split sample on the front end, in my opinion. Then, if it becomes necessary, with the MRO's order, it could be done at a period of, I have said, a few days to a few weeks. You know, it may happen that the hair has been modified, but that is something that can be modified at the time the retest sample is received, at which point the original sample can be retested as well.

Occasionally, the entire specimen will be consumed in the analysis. That is why it is important to retain some amount of hair -- 3 or 4 hairs -- to express the original length and the original appearance of the first hair sample, for comparison with the new collection process. Again, our policy is to use a directed assay on retests. Typically, we utilize GC/MS/MS at our limited detection, in an analogous way to how it is utilized in urine as well.

Again, you would not have quantitative reporting to the MRO, or utilize the normal cutoff, whatever we have decided that ultimately is, but rather, the laboratory's limit of detection.

Some things we want to look at before we test and interpret a retest sample, in the collection process itself or when it is received in the laboratory, it could be noted whether the hair was the same length. Obviously, results could change if the drug was present in the distal portion of the hair sample that was tested. Then someone clipped their hair, knowing when they did actually use the drug. Thereby, you wound up with a new sample that didn't encompass that same time interval. It is important to be able to make those kinds of comparisons, to assess whether it has recently been cosmetically treated, and also to note whether it was obtained from a different site on the body, which can also alter interpretation.

In thinking about how the agency blind sample program could be established, it would seem to us that it would have to be powdered hair, because of the heterogeneity of hair that is well known. That interferes, in terms of the blind QC program. It interferes with making it truly blind. If you are used to receiving intact hair, and you receive a sample of powdered hair, you know what it is. There would have to be a process of producing these samples in a reliable way, and vendors would have to gain experience in doing that.

There are some programs that have hair testing proficiency samples. The Society of Hair Testing has been mentioned, the National Institute of Standards and Technology, and the State of Florida, I understand, is close to having such a program. Perhaps those agencies could offer advice to SAMHSA in putting together a proficiency program, and certifying them as to content and concentrations. We have a recommendation for how frequently this should be done, based on the fact that they would probably be a lot more difficult to make and expensive. They could be utilized at a frequency of 10 percent for the first 30 to 60 days that an agency was doing it, and then after that three percent, or a maximum of 100 per quarter.

Again, moving ahead to what happens when there is an error. Currently, a lab is required to pull frozen samples out of storage and re-analyze them all, urine samples. With hair, that may not be quite as practical, for a variety of reasons. If there is inadequate sample for that retest, it depends on what kind of time interval has passed.

An individual may have found another job or the original hair sample may have been consumed in analysis, and the passage of time may make a new sample collection not equivalent to the previous one. An alternative that we would propose is simply sending data on positives to RTI for evaluation, as something to think about.

Reporting and review by MROs, would be somewhat similar to the way it is done now, with the exception that a few additional analytes would be included. Otherwise, it would be confirmed positives only.

Since I am kind of getting to the two minute warning, I want to move ahead to a couple of issues at the very end, which are controversial, but in which I think there has been some recent progress, and that is hair color bias and external contamination. Those of you who have been attending these meetings for a while are aware that we have a somewhat different philosophy than the leader in this industry has, on some of these issues. I think that as we move forward in developing a certification program, these issues do lend themselves to being explored at the working group level, and some consensus arrived at.

With regard to hair color bias, there seems to be some evidence on both sides, and certainly Dr. Cone's work as well as Henderson and Harkey's work has suggested that there is a bias for melanin binding on the part of the cocaine, which may reflect itself in an ethnic bias. That conclusion is still subject to some controversy.

There is other evidence from melanin binding of weak bases such as PCP and methamphetamine. As far as I know, there is nothing yet on marijuana. Dr. Marilyn Huestis is the world's expert on marijuana. We might ask her after this session. On the other hand, there are some population studies, or studies of real people, real drug users, on relatively large numbers of people. It would appear to show that, if this thing does happen, that it doesn't have as many practical problems as what you would worry about. Again, I think we can resolve these issues using the working group approach.

In regard to external contamination, again, there is a major divide between, if you will, Dr. Kidwell on the one hand and his advocates, and Dr. Baumgardner and his proponents. This is an issue that, at some point, does need to be resolved through the use of the working group process.

There is one other thing that we tried to do preparatory to this meeting, and I don't think I am going to have time to present it. We tried to come up with a time line for implementation of this process with what we know about how the federal regulatory process works. Donna does have copies of that. I don't know if we actually duplicated them for the meeting or not. I am sure that if anyone was interested, they could get one from you, couldn't they? Thank you very much.

MR. STEPHENSON: Thank you. I think one of the things that is important to deal with is maybe sharing that time line inside the working group and see if that can become a blueprint for your own internal process.

DR. KELLY: Yes.

MR. STEPHENSON: Thank you very much; good comments. Next, we have a presentation by Dr. Bob Willette. Bob got his second round of on-site testing technologies for us.

DR. WILLETTE: The main thrust of on-site drug testing, particularly using non-instrumented drug testing devices, has been in the criminal justice arena, but it is expanding very rapidly in the workplace setting. As you heard this morning from Julie Murdock, certainly the U.S. Postal Service is the largest workplace program that I am certainly aware of using on-site devices. Criminal justice is still probably the biggest population, but because of the increasing expansion of this and the proliferation of devices that have come onto the market, the study that I presented last year was prompted by the U.S. courts.

These were the devices -- those appearing in white -- that were included in that study. Since that time, there are another 15 or more devices that are on the market. Under support from the Division of Workplace Programs, we extended that study to 15 additional devices. In addition to the 15 additional drug testing devices, we included the Adultacheck. For the study population, the specimens came from U.S. federal probation offices.

It turns out we got no positive nitrites, glucuraldehyde or abnormal pHs. Predominantly, these specimens are collected under observation. We were focusing really on creatinine to see if dilute samples could be detected on site.

Some of the devices overlap the first study, but have been changed in one way or the other. For example, in the first test we had just the 4 drug test panel in a couple of cases. In this round, PCP has been added to most of these devices now.

The other thing is that we have also looked at who actually manufactures these. These are 15, but these same products are available probably on another eight to ten names, because the manufacturer will make them for 4 or 5 distributors, all with their own unique name. We did have, in the study, 4 different versions that come in cup form. They all operate differently.

This particular brand name, this is out under 2 or 3 names. This happened to be the Drug Check. As soon as urine is added to the container, it starts wicking up, so the test starts right away.

This product has been around a little longer. It is the Rapid Drug Test Cup. You actually insert a test card into the specimen through a slot in the lid.

This is the Roche Test Cup. This is activated by turning the lid and tipping it, to allow the urine to get into the test chamber.

Newest one out is called the Genie Cup. The lid happens to be missing in this picture. It has a tamper evident lid. This test is activated by turning the lid, which depresses a plunger and allows the urine into the test chamber. So, they are all a little different.

These were the multiple test devices, 2 that come in this multiple dip stick form. These are all the cassette types with urine to be added.

Then we had 4 drug test dip sticks, and this is the Adultacheck, which is a dip stick. It is not clear in this picture. It has the 4 separate test pads, which are then compared against the color chart.

The study design we kept as close to the original study for some comparison. There were a couple of changes. One was, in fact, this was the correction I made. The specimens were selected from the initial laboratory screening, using actually, these are DRI reagents. In the first study, DRI is the current reagent used by the laboratory. The other change is, because we had cups which require larger volumes of sample, we used all frozen specimens, so we had sufficient urine to run 15 devices at the same time and at the same day. The instrumented device was the Emit EST using Emit reagents. Everything else is basically the same. We hired three laboratory technicians and followed the manufacturer's protocol.

We were limited to running one drug a day, when you have multiple test devices and single ones. So, we had a cocaine day, a marijuana day, and so forth.

In judging results, all devices in this study either exhibit a colored line or spot for negatives, and an absence of lines or spots for positives. In contrast to this, the Emit produces a numerical print out, where we are comparing to a cutoff established by using controls.

As in the previous study, we scored all results either positive or negative. In cases where there were faint lines or in some cases these were ghosts, where the bound drug is laid out on the strips. Even if you run water across it, you may see something. These were all scored as borderline positive or negative. In this initial data analysis -- we just finished this study last week -- in this data, all the borderlines were called negatives, which is the very conservative approach and what most of the company inserts recommend. This is just standard procedure.

This is the same data presentation as we did in the first study. In this case, the green bars are false negatives. The red bars are false positives.

I have sorted these in not necessarily random order, but because we are not prepared to disclose product names at this point, for the Board's purposes, it doesn't matter what the product name is. I think the thrust of this is to see how reliable -- what is the reproducibility and comparison between devices. These are actually ranked in increasing order of false negatives. I have maintained the lettering throughout the next drug. One interesting point, and I haven't had time to do further data analysis, three devices actually were provided to us in the methamphetamine and amphetamine format. They are either specific amphetamine or specific methamphetamine.

It is interesting to compare product A's methamphetamine with product A's amphetamine. In every case, where we have these pairs -- this is the product, methamphetamine, the product amphetamine.

The amphetamine device always has more false negatives. This population of drug users, federal probationers, is predominantly methamphetamine. As in the first study, we saw that the methamphetamine devices do better.

There are a couple of other products in here, and I should have put an A or an M on them in all cases. There are a couple of others that were amphetamine specific as well. These are now in the same order. PD is the perfect device. It has no false negatives or false positives.

I have identified the ETS because we are really looking at how the non-instrumented tests compare to an instrumented system. This is an on-site instrumented system. We did not compare any of the devices against the laboratory reagents, because what laboratory are you going to compare them against and whose reagents. The real study would be to compare these against all of the immunoassay reagents on the market because they all have different cross-reactivities. We really wanted to see, the ETS is the more widely used on-site, but as I showed on the first slide, there are two other on-site test systems that we didn't include, because they are not used within the federal court system.

We see cocaine generally tends to produce the fewest false positive results, with some devices with no false positives and varying degrees of false negatives. The best performance is with cocaine. It is probably the simplest analyte to deal with.

The opiates are the opposite, a lot of false positives, very few false negatives.

This is partially how well the cutoff is established, but the other problem -- it is not that significant, but there are at least half a dozen hydromorphone/dilaudid positives in this population of samples. Dilaudid is a commonly abused drug in federal probation. That attributes to some of the false positives, but a lot of these are codeines and morphines below the confirmation cutoffs. This is all GC/MS at the HHS cutoff.

Cannabinoids. As in the first study, the biggest variability between devices is in cannabinoids. There are big ranges in false positives to big ranges in false negatives.

I should have pointed here that U is the adulterated. This is the Adultacheck. Obviously, these aren't positives and negatives. What these are, are correct identification of dilute specimens based solely on creatinine. A false positive bar means that the Adultacheck identified a specimen as being dilute when it wasn't, in other words, over 20 nanograms. We had creatinines run on every sample in the study. A false negative means that it failed to identify a dilute specimen. The Adultacheck creatinine reads in increments of 10. So, you go 5, 10, 20, and then it goes to broader increments.

PCP. In the first study, only 7 devices were available with PCP. In this study, all but one had PCP available. When you look at the variability in PCP, about the only acceptable test is the ETS. It had very tight false negatives and false positives. We don't know the reasons for these. We just finished the study last week and this is the first data analysis. We have a number of other things to look at, but a lot of false positives on the devices, but not on the ETS. In most cases, people are using multiple test devices. If one is using a 5 drug panel, this is the overall performance when you average all the correct and incorrect results together. Interestingly, there are not a lot of differences when you average them out. As I showed, there are big differences drug to drug. Some of them, the selection that may be required in, some of the standards or guidelines that need to be promulgated, it has to really take into effect drug to drug variation as opposed to overall. We also did the plots that we did before. This is the typical best bang for your buck. Our perfect device is up here, using positive predictive values and negative predictive values. The Adultacheck, using the same approach, whether it is correctly identifying dilute specimens or not, appears here.

The instrumented system, as one might expect, shows the best overall performance in an approximately 80 percent negative predictive value. In other words, of all the negatives that are identified, eight percent were correctly negative, and just a little less than 80 percent of the positives were correctly positive; a very tight bunching when you look at an average of all 5 drugs. I haven't prepared these for the individual drugs, but it would be similar to the bar charts.

Next slide. These were the 15 devices in the original study, using the same analysis. There was a lot more spread and scatter. Anything below the 50 line means that of your positive results, you have less than a 50/50 chance of being right.

In the year and a half to 2 years between the evolution of these devices, it would appear that there certainly has been improvement either in the manufacturing process or other factors that we don't really know what they are at the moment. Julie Murdock went into this. She discussed their selection for the U.S. Postal Service. There are a lot of factors that go into selecting an appropriate device. How many of these really need to go into the guidelines, on ease of use and reading the end points. Some of these end points are not that clearly definable. Certainly, operator differences which argue strongly for some kind of operator training, I think David Evans made that clear. Certainly these are presumptive results and confirmation is required. We see big variations from device to device in confirmation rates, certainly from the positives, but also in terms of false negatives.

One of the critical factors that is not really taken -- in the guidelines currently for urine drug tests, this is a topic that is covered. It is not in the matrix. It is, who knows about that original test result? Is it only the test operator? What a lot of people are using these on-site devices for, if somebody screens positive on site, they are removed for service. If they are a job applicant, they are told to come back in 2 days. They are not processed. I don't know what they are doing at the Postal Service. This has always been one of the biggest problems with on-site testing outside criminal justice, is who knows about this. If the person is removed from service, do all the other employees find out through various means that the person tested positive? Is the person stigmatized? In company policies, this is a very critical issue on how that initial positive test result is communicated, who knows about it, and so forth. It is a major factor.

Do you take immediate action or delayed action? Certainly in many of the situations, people are removed from service based on that presumptive positive. Not to do so would open people up for liability.

I think these are two major policy issues that people have to deal with. Then of course, David Evans went into federal, state, and local regulations that either permitted or restricted on-site testing. Of course, my lettering is the old matrix. They have moved or changed some of the letters. Certainly, everything but the section on on-site testing are really the same.

I have been involved with on-site testing for 26 years, going back to starting with drug treatment, going through criminal justice. We have on-site testing going on within workplace settings. We also audit and inspect nuclear power plants for on-site, and so on. All of these can be the same for any form of urine testing, whether it goes to a central lab or not.

Factors unique to on-site initials tests, certainly the analyte classes is the same. FDA approval can be the same. Documentation and collection is the same, how the result is documented, there are a variety of things.

Many of these do photocopy, so you can make a photocopy of the device result. Certainly signed reports and so forth, there are ways of dealing with that.

Specimen handling, in the most conservative approach, we recommend also split specimens, similar to what can be done.

Four of the devices are actually a cup. They serve as a cup and the test is also done in it. There are still other reasons for having the split.

The use of HHS cutoffs, they claim to be the same, but how are these established. I think the guidelines don't really lay out criteria. They say what the cutoffs are and they dictate the kinds of controls that the laboratory uses to ensure that on that particular test day, that the screening tests are meeting the guideline requirements for cutoff. It is very difficult to do, even with all the different kinds of quality control things that you can do with on-site. You really don't come close to what laboratories can do in establishing. That is why I think the ETS or any of the instrumented systems show better performance. You really know, at the time of testing, how it is performing.

Shall be directed toward targeted analytes. Right now we haven't seen a device, out of the 30 different products we tested, that has both amphetamine and methamphetamine in a single test. It is separate. It really requires 2 tests, and 3 of the devices actually did come in both forms, so you were able to do that.

Shall have objective differentiation of positive and negative results. The end points aren't that clear in many of these cases and it varies from product to product. Some are much clearer than others.

In a subsequent report, we will deal with these borderline readings, as to what products seem to have much clearer end points than others.

Controls, Stewart Bogema talked about these. Certainly you have to evaluate each lot. You must have periodic controls above and below cutoff. Then, of course, blind QCs. We have been doing open and blind QCs on a number of programs for years. It is just like you can do with the laboratory. It is a little more difficult in situations where the collector and the analyst is the same. It turns out in criminal justice, typically the collector is also the one who is doing the testing. You really can't get blind QCs into it, unless you have the analyst separate from the collector. The analogy -- and I think Julie Murdock mentioned it earlier and David Evans did, too -- is that on-site testing, if it is more analogous to breath testing, all the nurses that we have talked to that are doing on-site testing, treat it just like they are doing a breath test or an on-site oral fluid test. I agree with Julie, that the best approach is to test in front of the individual. If you treat it like a DOT protocol for on-site oral fluid, the collector and the tester are the same.

Shall provide for verification of results by a second certified technician. This is not very common. We have found in the real world that it is not very practical in most cases because most of these have to be read within 10 minutes. It really takes away one of the advantages, both the cost and time advantage, to have two people reading these.

Now, all the data I showed you were single reads. We were simulating what happens out in the field and didn't compare it with second reads.

We do compare -- the 3 technicians rotated the devices between them, so we could see if there was any operator bias in the results. In the first study, that was not a significant factor.

Then detection of adulterants, as the Postal Service is doing, we looked at the Adultacheck to see if it could -- actually, the Adultacheck performed extremely well on the creatinine. As I said, we didn't have any samples with adulterants in them.

Certainly, I thank the Division of Workplace Programs for its support. Dr. Leo Cadejian who, unfortunately, is speaking in Ohio, so wasn't able to be here. He is the man on the scene giving day to day direction to the technicians. PharmChem, again, provided us with great assistance in a privacy away from the main lab. Thank you.

MR. STEPHENSON: Thank you very much. Again, very informative and a nice presentation of the material. We look forward to the actual development of the PD device. At this time, Bob Schoening would like to have a few minutes to talk about the SAPAA survey.

MR. SCHOENING: I am representing the Substance Abuse Program Administrators Association. For those of you who are not aware of who SAPAA is, it is an association of people and companies that manage drug testing programs -- TPAs, third party administrators, who manage programs for generally a large number of companies, most of them being under DOT regulations.

In the course of events with the DTAB meetings, several of us within SAPAA thought it might be appropriate for SAPAA, with its members, to develop a survey to determine the use of alternative technology in workplace drug testing. Dr. Donna Smith -- and she does send her apologies, because she was supposed to do this today and I am pinch hitting for her -- evolved a survey.

We emailed it to several members. We have been receiving responses. The data is in raw form and it is not available for release at this particular time.

We went into several different areas in the survey. First was the demographics of types of alternative testing being done; yes, we have clients; no, we don't have clients; we are currently investigating. A large number are saying, yes, we are looking at this; we haven't done anything yet; we are not sure what we are going to do.

Area of the country in which they may be operating, and also industry type. What type of industries are they dealing with for the different types of alternative testing -- construction, retail, manufacturing, service, and the always good answer, other, for those not listed.

Then we broke the survey into four different parts to include on-site, non-laboratory, urinalysis, drug screening, with a whole series of questions after that.

We then went to hair testing, with almost an identical set of questions.

We then did sweat testing with an identical set of questions.

Then, we tagged oral fluid on the survey here, but we will change it to oral fluids.

Basically, we started off, do you have any clients that use on-site screening technology; yes or no. We want yes or no, or very simple number-type answers. How many tests are conducted annually using on-sites. What brands are used. List all that are being used.

We also asked, what type of testing are you doing, and by that, pre-employment, post-accident, random, reasonable suspicion, et cetera, like that.

Then we asked what I would consider a very key question here: are all presumptive positives sent to an HHS lab for confirmation. That is a straight yes or no answer.

Then we go into some policy questions here. If yes, what actions does the employer take on screened presumptive positives prior to receiving a confirmed laboratory result. We broke it into pre-employment. As an example we put down, not hired, allowed to retest in the future, delay hiring action until confirmation results return from the laboratory -- this is to take care of those false positives on screening, or another accident might take place.

We did almost the same for all the other different areas of testing. We also have, how those folks who do not send the positives on to an HHS lab for confirmation, and what actions do they take as far as policy is concerned. There is, not hire under pre-employment or allow to retest in the future or they terminate, et cetera, things like that. We are trying to get some policy actions here.

Then the other question that came up in line with it was, do the action or actions taken by an employer depend on the drug or drug classes that are identified. If yes, please ask for an explanation. If no, the drug is there, fine. We have found that in some companies, if it is marijuana, there may be no action. If it is cocaine, it may be a very severe type of action. So, that may vary.

Another question we ask, are any negative screens sent to the laboratory for quality control. If yes, please indicate what percent.

If you have data of what percentage of screened positives are confirmed positive at the laboratory, please provide.

If you have data on what percentage of the test were not a valid test or were unable to confirm at the laboratory, if the reason is suspected adulteration, please provide invalid tests, unable to confirm. I think Julie Murdock has indicated this has happened. We go on and ask a lot of questions about that.

We do ask, what is the average cost of on-site screening, breaking it down to cost of the device per unit, cost of the specimen collection per test, cost of the laboratory confirmation. We also ask, does the medical review officer review the laboratory confirmed positive tests. Some of the responses -- I am not trying to divulge the actual data -- but we have gotten back yes and no for both here. If yes, what is the charge per review of these confirmed positive tests.

Then we have also asked if the specimen collector, and we have emphasized, does not do the screen analysis of the specimen, who does the on-site analysis. We have used four different categories here, the medical professional, which could include an RN or an LPN, an MT or MLT, a test technician or HR/safety-type personnel, or other company personnel who may be involved in the testing process here.

Those are some of the basic questions we have gone through. We have tried to break it down.

As far as hair analysis, all hair analysis does go to a laboratory for analysis here. So, we basically asked, does the laboratory do automatic confirmation by GC/MS/MS or by GC/MS.

What types of testing are done, and then what drugs are being tested for. We included more than the basic, what is commonly called in trade the HHS-5, to include barbiturates, benzodiazepines, propoxaphenes, things like that.

We have also asked a policy question here, and then we asked about adulteration of hair samples as suspected, yes or no. If yes, how determined. Was it an invalid test, unable to confirm, or was there another reason. Then we asked does the MRO review all the positive drug tests that come back from the hair testing laboratory. Then, can you compare the hair testing positive rate to the laboratory urinalysis positive rate. We are asking to see that particular type data.

Along that particular line, we asked the same questions on sweat analysis and also on the oral fluids analysis. There was one question that came up with oral fluids. This was, that there are some oral fluid devices that can do an actual on-site analysis. We had to make allowances for that within the survey here. The answer was yes, but they can't do it yet. We will see what happens in the future.

If you are interested in seeing the actual survey, it is available on the web. The web address is www.sapaa.com. If you wish to fill it out, please fill it out and email it back to me. My email address is attached on the survey.

I thank you for your time and we will make the results of this survey known to the DTAB. It will also be put on the SAPAA web site. Thank you.

MR. STEPHENSON: If you are going to have the responses coming in from the web, how are you going to credential to make sure you have a legitimate entry by a knowledgeable person?

MR. SCHOENING: We will have their email address and we will track it back to the person.

MR. STEPHENSON: We have a presentation on on-site oral fluid.

MR. FOLEY: I am not prepared to make a presentation, but I really appreciate the opportunity to say a few words. I am head of a small company in Southern California called Lifepoint. We are in the process of developing an on-site method, including instrumentation, for the simultaneous assay of drugs of abuse and alcohol. The device differs from those we heard some talk about this morning, in that it involves both the automatic acquisition of sample, processing sample. Without sending the sample to the laboratory, the sample is automatically tested for drugs of abuse at that point.

Our concerns are somewhat along the lines that were being spoken about this morning with regard to the guidelines we are attempting to put forward. They do differ in that we have a closed system here, and our concerns reach a little bit beyond those that were spoken of this morning. We use an immunochemistry method that is patent based, and I won't burden you with all the details. It is a method that has a sensitivity down to at least a nanogram per mL, so we are well in the region of doing the assays for drugs of abuse as they are found in oral fluids, at a cutoff of around 10 nanograms per mL or less.

We will be providing a cassette in a variety of formats, so you can actually choose you own panel of tests. The technology basically is applicable to other analytes beyond drugs of abuse.

I don't want to duplicate a lot of what was said this morning with regard to the guidelines and the issues that need to be addressed for oral fluids. What I would like to point out is a few things. Although, as I said, the current methods discussed about oral fluid collection relate to products of which the sample is collected and then something is done subsequently to the sample, i.e., it is taken to a laboratory for testing. We are developing an oral fluids based testing method, which can deliver lab quality results rapidly and on time, immediately on site, without touching the sample.

Since the collection and the testing of the specimen is done while observing the donor, the risk of adulteration is removed. Maybe removed is a bold word, but at least it is minimized. If the collector/tester is concerned the donor may have a contaminating substance in the mouth, then he or she may require the donor to rinse the mouth with water and then wait 5 minutes before running or re-running the test. That is perfectly appropriate.

All other requirements for on-site oral fluid testing should follow the same formula and procedures as currently followed by oral fluid or breath-based alcohol testing, which is typically done on site and the donor is in attendance.

The other areas that we have addressed very briefly were the laboratory requirements, the initial test. We would say that since with oral fluid the levels of drug detected are directly related to those found in a positive blood specimen, these levels which are generally found to be in the 10 to 25 nanogram per mL, we should regard this as an initial test, or a screening test, as the term is used, for urine specimens.

On the issue of confirmation testing, if that is required or deemed to be required, a second oral fluid specimen should be collected and sent to the laboratory for GC/MS. We are in the process of developing a procedure to collect and transport such a specimen to the laboratory, should it be needed.

In terms of long-term storage, oral fluid, similar to urine, can be frozen for considerable periods of time, and that is an acceptable method, so that positive specimens can be available for the laboratory retest if required.

Retesting of the specimen, we may experience some deterioration, of course, in storage. Since this is not a quantitative test that we are developing, but rather, a screening, cutoff based method, that should suffice.

Our instrument will automatically check all instrument functions to verify that they are functioning correctly. Additionally, it will ensure that the cassette or the reactor in which the reaction takes place is not outdated, but that an adequate volume of specimen has been collected and that the reagents are working properly. All of these controls would be built-in. With all of this done automatically, is there a requirement for further QC testing.

With regard to agency blind sample programs, or PT programs, this can be accomplished by using outside vendors, as we have done in clinical chemistry, for long periods of time, to provide blind proficiency testing samples. We are working on, and can provide, oral fluid matrices, artificial or otherwise, to outside vendors, that can be used to spike with drugs to test for laboratory proficiency.

In terms of MRO, the MRO can be used the same way that they are currently used for positive urine test results. The testing site can set up an arrangement of its own with an MRO to review the results before sending the results on to an employer, for example.

In the area of certification, our product is being designed to be exempt from CLIA regulations. Under the same requirement, it should be exempt from lab certification.

The product has been designed to be fool proof, if one can use that term, with an untrained user being able to run the system, using only directions written at the seventh grade reading level.

These are my comments. My initial intent was to come here and determine the status of where we are in alternative sample testing, and to participate in establishing, to the extent that I can, any procedures and guidelines for the testing of oral fluids. We at Lifepoint, make ourselves available to the Board or any of the working groups that are working to establish such guidelines. Thank you.

MR. STEPHENSON: I would suggest that you actively consider participating in the oral fluids working group. Is that acceptable? This is an area where you have a new technology. Did I misunderstand or is your product not yet available?

MR. FOLEY: Not yet available.

MR. STEPHENSON: You are really still on the drawing boards in some ways.

MR. FOLEY: Well, it is important for us to be here and understand which way the thinking is going, so we can build it into the final development.

MR. STEPHENSON: Yes, absolutely.

MR. FOLEY: I am more than happy to help with the working group.

MR. STEPHENSON: It is my understanding that we have some information submitted by Intoximeters. I don't think we have anyone here, but the material was made available for you at the sign-in table.

[Brief recess.]

MR. STEPHENSON: From the number of people leaving the room, I assume there are going to be very few public comments. Are there any individuals who wish to make public comments at this time. Are there any individuals who have basic questions?

DR. TAYLOR: My name is Dr. Howard Taylor. I am with National Safety Alliance. I would simply like to ask a procedural question of the board. You described the small working groups or working committees with the various groups. Certainly, there is one thing that comes to mind, that you are going to have competitors in the industry. Let's say, for example, in the hair testing arena, you have different laboratories that have different thoughts and different ideas of how things should be done. Can you describe the process of how this will work, how information will flow, and how decisions are to be made, and essentially the process for how this group is to work together.

MR. STEPHENSON: First, we are not going to pick on the hair people, per se, but we are going to use the same basic rules of engagement for all the technologies and alternative specimens. We would like this to be an inclusionary process, one in which individuals who feel that they would like to belong to a small working group, can. We would ask that they pledge to each other to take off their corporate hats, per se, and not fight the battles for market share or for particular bragging rights or the publicity aspects inside the working group. In fact, what they do is address the issues. They accept the input from each other.

In terms of the exchange of information, I would call for individuals who have an interest in working on any of the alternative technologies or specimens to contact the industry representative and coordinator at this time, to make sure that the information is even-handedly exchanged.

We would expect that the materials would be sent to each individual that would be working on that group and that we, in turn, would be copied on those exchanges.

We will be the holder of the copies, but in the context of being objective about material being provided or not, or the timing of it, or the contents of something that is provided, we will see the material.

We will not interact with the small working groups other than to assist individuals who have felt that their perspectives, views, inputs have not been fairly received within a group. I can't perceive that happening right now. I think what is going to have to happen is, in some of these areas -- you had referenced hair -- we are going to have to get people locked in a room. I don't want to even hand out short-handled baseball bats. But I want to lock them in a room and get folks to at least begin to address the science issues.

Two things you have got to deal with. What is missing from the grid. What research is needed to address those issues. What specifically would you like to see designed in terms of the research. Then, let us know so that we can help identify funding and support for that research as a priority.

This is what this process is going to be about. I would like to suggest that, in each one of your areas, you take a hard look at a time line, and that we move forward aggressively. We will assist the small working groups in the same way we assist in assembling these meetings. As there is a need for people to do specific things, let us know. We will provide that support to the best of our ability.

We have said this in our written testimony before Congress. All the work that needs to be done can't be paid for by our division, nor by the Department of Health and Human Services. This is going to be a sharing process in which the industries are going to have to assist in doing this. And in many cases, you guys have already done the lion's work.

As new issues arise, new technologies arise, I hope we are going to be able to look back on this process as the model by which we achieve a good understanding and set forward some changes in public policy and guideline uses that we are going to be proud of in the years to come.

I believe this process will work. I think that we are finally on a path of doing something that gets us beyond simply laboratory based urine sampling. It served us well in the past, but the technologies and the applications and the opportunities are overwhelming.

MS. DAUGHERTY: I am Elizabeth Daugherty. I am with the Department of Health and Mental Hygiene in Maryland, licensing and certification.

I listened to an evaluation of on-site drug testing kits. I saw on the statistics that there was an approximately 50 percent failure rate, where the results were either false positives or false negatives when confirmed by another method. I am trying to determine whether the committee or the Board will look into why there is this high rate of failure. Is it due to something inherent in the method? Was it due to poor personnel standards or lack of training for the people who were doing the test which, since they were very well qualified technicians, I doubt. Was there a problem with the specimens or the specimen handling that maybe we have not addressed yet? I would like to encourage you to make sure there are good personnel standards for the drug testing site, and some certification process for the sites.

MR. STEPHENSON: Thank you for your comments. Now, given that you have made that series of observations, I am going to call upon you to join and participate. You don't have to be a federal employee or just a member of an industry. You represent a state that is rather forward looking and currently involved. We have a representative from another state who is doing work in QC in Florida. I would call upon you to join with us. If you wish to work specifically in the on-site testing working group, David Evans is right there in the back of the room. I am sure he will be more than glad to talk with you. You can make sure that your thoughts are included by having that dialogue with him. At that point in time, when it comes forward, we will be looking to make sure that the on-site testing technology report addresses the issues that you have raised today. Is that fair enough? Anyone else?

DR. BOGEMA: I think part of the response to the last question is that the study protocol was meant to look very closely at specimens around the cutoff. The sample selection criteria chose primarily specimens that were tested by the DRI that were either very close below the cutoff or above the cutoff. That is going to exaggerate the false positive and false negative rates, as opposed to using a random selection of positive specimens or negative specimens.

DR. WILLETTE: Yes, that is right. It was in the slide and I didn't, for the sake of time, stress it. This was a protocol that is in the draft or the proposed FDA guidelines for these devices. It was really to see how well devices are able to establish their cutoffs, in contrast to what happens in the laboratory where you can run calibrators or controls and know where your cutoff is. That is one of the factors and differences in performance between these. Also, similar to the immunoassays that are available for laboratories, one of the biggest differences between the devices is the same issue which is the difference between the laboratory reagents. They have different cross reactivities based on what the antibodies are that are used. So, I think one of the biggest variables is the antibody choice. What I think we have shown is that you have those multiple factors showing. If you look at how well they perform around the cutoff, it is the way you really magnify some of those issues.

Dr. Bogema is absolutely right. If you take a random population of specimens, performance is going to improve, because you will have a smaller percentage around the cutoff. That is why we cannot use these negative and positive predictive values to project what the false positive, false negatives rates are for other populations, because this is not a random distribution. Those prevalence predictions only work in a random distribution of samples.

Two follow up questions, though. One of the points I made about who gets to know about the presumptive positive, I was going to ask Julie Murdock how that works within the U.S. Postal Service. On the quality control side, I didn't hear mentioned if negatives from on-site tests were sent to the laboratory to monitor false negative rates. I just wanted to clarify.

MS. MURDOCH: On the who knows, in the Postal Service program now, there is a form that they use called the personnel notification form. It is a bureaucracy and all forms must have a name. On that form the individual is identified as either qualified for employment if the test was negative, not qualified for employment if something happens that caused the testing process to be terminated -- for example, the person wouldn't sign the form, with a quantity insufficient provided, twice that was insufficient, or those types of things, or pending. That can be the specimen sent to the laboratory for a variety of different reasons, not just that it is positive. That goes to a single person in personnel. So, the collector, the analyst that is, and the personnel person handle those sensitive records. Those same records, by the way, go back now. There is sort of no difference. They had sensitive information on them previously. Those people handle them. Then the specimen is reported to the laboratory using normal procedures, right back to the district occupational health nurse administrator's office, since that person works for the MRO or, if it is a contract MRO, directly to the MRO's office.

Negative specimens, we actually haven't implemented our negative QC program, mostly because we are just in the finalizing roll out of the main program. As complicated as this is -- and despite manufacturers' representations, it is a complicated process, to institute an on-site program -- we didn't want to confuse them too much by saying, in addition to all these other procedures, save this aliquot, this reserve specimen to be sent to the laboratory for negative QC.

We do have all the program design. We will have negative QC identification numbers that the laboratory will get, so that they will know that is what is coming in to them, and a selection process for identifying who will be required to send in the negative specimens. We are using a stratified random selection program for that.

I did have a financial question. You said you are going to be providing support for these working groups. Does that mean like travel money if the working group decides to convene, that sort of thing, or is this just moral support?

MR. STEPHENSON: Not in Hawaii, and not in a warm spot if you are in a cold spot at the time you are asking for it. Other than that, to the degree that we can, to the amount that we can, yes, we will provide facilitation to make it happen. That is the purpose of this.

MS. MURDOCH: I thought that was against the government's policy.

MR. STEPHENSON: To make it happen, to facilitate? You are probably right. You have more experience with the government than I have.

MR. EVANS: The center of the on-site working group will be in New Jersey. I can imagine you will have to pay people to be there.

MR. STEPHENSON: Nobody will volunteer to come work with you. That is about the only thing I can think of.

MR. EVANS: Who is going to find out about the results of the on-site test I think presumes that the on-site test is going to be done at the workplace. One of the things that we have found is that -- and it still preserves the value of on-site testing -- is to have the test done at a local specimen collection site. In March of this year, the National Association of Collection Sites endorsed on-site testing in their organization of at least 1,000 or more collection sites, professional collection sites. You could send somebody to a collection site a block away from the workplace, have the on-site test done there. The results could be phoned in to management, so that only management, the same way that it is done with a laboratory test, would find out. No employee would have to find out. Also, if you had a large enough company, there could be a company medical office that could perform the on-site test, and then report that to management also. The company medical office already has all kinds of confidentiality procedures for preserving that type of information. They are also removed from daily contact with the employees. So, you could preserve the confidentiality in that way also. These are certainly things that we could encourage. We want to encourage on-site testing being performed by certified trained people, such as certified trained collection site specialists. I will point out that that is how a lot of the federal testing is being done now, not by employers, but by sending people to collection sites. They could do the same thing with on-site testing, get the results back immediately, which preserves the value of the on-site test, and everyone is protected.

DR. HUESTIS: I was very happy to hear Dr. Bogema talk about sending the new program to include in the postal study the negatives to be sent to a laboratory, a percentage of negatives. I think that is a really critical point for the quality control, and also for the issues of control that a single individual who might be taking specimens would be doing the analysis, that there is some check on this process by having the negatives sent in, not only the quality control aspects.

One thing that is going to be difficult is going to be for the industry to decide what percentage of performance that you would accept on these negative specimens.

I know in the military program, where they had on-site tests being done and being sent in, you have totally different antibodies, different cross reactivities, different immunoassays being done for the initial test. You can't expect that there is going to be 100 percent agreement. There won't be. It will be important to decide what level of performance and to evaluate those samples that disagree.

I think also it is very important to determine that when you send in the percentage of samples, there is some way to regulate how those samples are selected, and that it is not up to the same individual who is actually doing the testing.

MR. STEPHENSON: In other words, what you are suggesting is that it may be a random assignment of your negatives that would be predetermined and that specimen, if it is negative and comes up on that random generator, would be the one selected. That would make a lot of sense. Your comments will be captured for, I think, one of the issues that I didn't raise earlier. There are two aspects of what we will be doing in the small working groups. One will be to identify issues that are specific to that technology or specimen. The second will be to identify issues that, in fact, are cross cutting, that will have to be addressed across several of the alternative technologies that we haven't thought of yet, that don't exist, perhaps, now, as a grid matrix, but are going to have to be addressed as we actually move into writing the text for the guidelines, and going forward with what we have to do.

DR. BUSH: I have for the record a couple of emails we have received at the office. One has to do with on-site urine drug testing and concerns about failure of the on-site assay, and then how that particular specimen should be handled and sent to the lab for screening and adulteration panel. I am going to turn this question over to the working group for on-site drug testing, to address that question. The second comment comes to us concerning hair testing, and specifically concerning drug cutoff levels and specifically, even further down, marijuana, concerns about THC and/or metabolite detection. I will turn that over to the hair testing coordinator.

MR. STEPHENSON: Are there any other questions? If there are no more, we will conclude this meeting of the Drug Testing Advisory Board. I thank all of you who have been in attendance, those of you who have presented, and those that I know are going to continue to work together over the years.

[Whereupon, at 3:28 p.m., the meeting was adjourned.]