Drug Testing

DRUG TESTING ADVISORY BOARD
Scientific Meeting on:
Drug Testing of Alternative Specimens and Technologies (Part II)
Transcript - Day 1
September 9, 1997

TABLE OF CONTENTS

Opening Remarks - Joseph Autry, MD

Overview of Information Submitted in Response to

April DTAB Meeting - Carl Selavka, PhD

Forensic Workplace Drug Testing Program Requirements

US Probation/PharmChek Drugs of Abuse Patch Pilot

Program - Robert Fogerson

Administrative Office of the US Courts:

An Evaluation of Non-Instrumented Drugs Tests

- Robert Willette, PhD

Public Comments

Group Discussion

Public Comments

DRUG TESTING ADVISORY BOARD

JOSEPH AUTRY, MD (Chair), CSAP, Rockville, MD

DONNA BUSH, PhD, (Exec. Secretary), CSAP, Rockville, MD

ROSEMARY BAKES-MARTIN, CDC, Chamblee, Georgia

YALE CAPLAN, PhD, Quest Diagnostics, Inc.Baltimore, Maryland

CHRISTOPHER HOLLAND, MD, HRSA, Bethesda, MD

AARON JACOBS, PhD, USAMC, Fort Sam Houston, Texas

ALAN JONES, PhD, University of Mississippi, University, MS

TAI KWONG, PhD, University of Rochester, Rochester, NY

MELANIE MALLORY, Consultant, Scottsdale, Arizona

RICHARD PINDER, PhD, Public Health and Addiction Services, Hartford, CT

DIANA WILKINS, PhD, University of Utah, Salt Lake City, Utah

PARTICIPANTS

ROBERT FOGERSON, VP and Laboratory Director, PharmChem Labs, Inc

BRUCE GOLDBERGER, Ph.D., Diagnostic Reference Laboratory, University of Florida College of Medicine

RAY KELLY, Ph.D., Director of Toxicology, Associated Pathology Labs

CHRISTINE MOORE, Ph.D., Associate Scientific Director, US Drug Testing Laboratories

CARL SELAVKA, Ph.D., Director, Office of Forensic Services, New York State Division of Criminal Justice Services

ROBERT WILLETTE,Ph.D., President, Duo Research

PROCEEDINGS (8:54 a.m.)

DR. BUSH: Good morning. As the Executive Secretary of the Drug Testing Advisory Board, I would like to welcome you to this meeting of the Board. We are going to continue with what we have done with the April meeting in taking a look at alternative specimens and testing technologies for workplace applications.

Dr. Autry received his bachelor's degree from Rhodes University with majors in chemistry and psychology. He received his doctorate of medicine from the University of Tennessee. He completed an internship in internal medicine at Baptist Memorial Hospital, where he ran a coronary care unit for several months. He completed his residency training in psychiatry at the National Institute of Mental Health/St. Elizabeth's Hospital Model Residency Training Program. He was board certified by the American Board of Psychiatry and Neurology in 1976. Dr. Autry had an active career as a practicing psychiatrist prior to coming to the Washington area, joining the National Institute of Mental Health and becoming the director of extramural research, taking on responsibilities for basic research, clinical research, treatment research, and applied research. He took on the challenges of directing the Office of Policy Analysis and Coordination for the National Institute of Mental Health and then for the Alcohol, Drug Abuse and Mental Health Services Administration, the parent organization for NIMH. In 1990, Dr. Autry decided to move back into research and became the director, Division of Applied Research, at the National Institute on Drug Abuse, with responsibility for research on drug-free workplace programs, workplace policies related to substance abuse and AIDS in the workplace, health services research, research on HIV infection/AIDS in the community, and oversight for the National Laboratory Certification Program which certifies laboratories to conduct drug testing for federal drug-free workplace programs and the federally regulated industries. Following the reorganization of the Alcohol, Drug Abuse and Mental Health Services Administration in 1992, Dr. Autry became director of workplace programs for the Substance Abuse and Mental Health Services Administration, Center for Substance Abuse Prevention, and with oversight responsibility for all the federal drug-free workplace programs and the National Laboratory Certification Program for forensic urine drug testing. As the old saying goes, no good deed shall go unpunished. The reputation of his insightful leadership and his abilities to tame the wild bunch has resulted in his being assigned, at least temporarily, as the acting deputy director of the Center for Substance Abuse Prevention. Additionally, Dr. Autry continues his part-time private practice of psychiatry which he frequently describes as his link to the non-bureaucratic world and an opportunity to constantly infuse his federal research and services responsibilities with a continuing need for new knowledge to deal with the problems of mental illness and substance abuse in clinical practice. Dr. Autry.

Agenda Item: Opening Remarks.

DR. AUTRY: Good morning. It is good to see so many friends and colleagues here again. I am not quite sure what Donna meant by taming the wild bunch. I am afraid to ask. Let me apologize to you for being late this morning. I have a long and checkered history with northern Virginia. The first year I lived here I was convinced that you could not go from southeast Washington to northwest Washington without crossing the Potomac and doing a U turn in Pentagon South parking. I am now convinced that I cannot come to this side of the river without going to Pentagon City first and then going wherever it is that I am supposed to be going.

We have a very busy couple of days ahead of us. This is the second meeting of the Drug Testing Advisory Board that is focused exclusively on looking at emerging technologies and alternative specimens. It is an undertaking that we have been very pleased to do on behalf of the federal drug free workplace program. We have appreciated all your participation up to this point.

I also want to bring you greetings from Nelba Chavez and Paul Schwab, who are the Administrator and Deputy Administrator of the Agency.Unfortunately Nelba is out in Arizona and Paul is doing something down at the Department level in her stead and neither one of them could be here. They do both, however, send to you their greetings and let you know of their commitment to continue this very important undertaking.

There are a couple of housekeeping chores that we need to do. We are going to have a public comment period toward the end of this day and we would ask that you sign up in the back. We will take them on a first come, first served basis. Depending on how many people sign up, we will divide the available time by the number of people. You will probably have a period of about five minutes or so. We are also taking written questions throughout the day today and also tomorrow. Please, if there are any questions, get them up to us. We will weave them into the public comment period. We will also change the agenda slightly tomorrow afternoon to allow additional public comment and, again, ask that you sign up in advance.

In April we undertook, or started the process, I guess is a better way to say it, of looking at alternative technologies for specimens, to see how they might fit into the federal drug free workplace program. I think we were a bit optimistic at that point in time and thought that we could probably go through this pretty quickly in a couple of meetings because of the emerging science. However, I think we were quickly dissuaded of that disbelief.

Following that meeting we asked that the industry representatives coordinate getting additional information in to the board, which they did. We were scheduled to meet again in August. However, the Board needed additional time in order to go through what had been presented in April, to go through the extensive material that had been sent in following that meeting. So, the board members themselves had a meeting in August as a working group, not as a formal board. My staff and I were not involved in that.

So, we are hearing a lot of what you are hearing today, not exactly for the first time, but certainly in a more formalized way than we have had informal discussions with the board in the past. What we intend to do for this meeting is to try to take advantage of the board's work in August, the material sent in after the April meeting and the deliberations of the April meeting to come up to sort of a state of the science of where we are today with alternative specimens and technologies. Perhaps more importantly, where are we today in re-examining the underlying principles on which drug testing has been based for the past decade. I am pleased to see my predecessor, Mike Walsh, is here with us and we certainly look forward to having comments from him as we go through this meeting, too. I am not going to say anything more at this point because I think the deliberations of the board are what this meeting is all about. I will tell you that what we will try to do is at least once a year have a session in which we do look at the principles that underlie drug testing to see if they are still the gold standard. We will continue to look at all of the testing technologies and all the specimens that are available for testing on at least an annual basis. With that, I am going to ask that Skip Jones come up and talk a bit about where we are with the board's deliberations.

DR. JONES: I am going to use the floor mike. I am going to use the overhead in a little bit, so I am just going to use this mike to start with. Thank you, Dr. Autry. We are going to be passing out some efforts of the board and I am going to go over those in a few minute. I will let Donna and Laura pass them out. I have been asked to serve as spokesperson of the board, I guess, is the appropriate title this morning, to give you a little bit of an idea of what we have been doing, where we have come from, and where we are at present in our efforts. As Dr. Autry said, we started this program, this process, this trip back in April when we first had the first large open hearing and had a variety of submissions and presentations.

As a board member -- and I think I can speak for the board -- I certainly want to thank all of you who made presentations at that meeting, supplied additional information, supplied additional reading information. I think we had plenty of information to read. We had plenty of information to use in the gym to work out with as we lifted those boxes and packages of papers that you had submitted. It is that type of information that we certainly needed and we certainly appreciate your efforts in supplying all of that information. We certainly appreciate your efforts in making this meeting today. Our mission, our challenge, our charge, has probably changed.

When we started this process back in April, several of us thought -- and I think the board agreed in general -- that what we were going to do was probably have some presentations by a variety of people looking at some of the alternative technologies that are available, looking at some of the procedures and techniques that are available. Somebody was going to present a program for us and the board was going to come back and say yea or nay on that program that was presented. As we finished up the April meeting and started looking at all the material that was supplied, as we talked back and forth amongst ourselves via telephone, via fax, via conference calls, et cetera, over the following few months. Then we met in August and we have had a couple of conference calls again since then as a board, we came fairly rapidly to the position that we could not do that. We could not say yea or nay on a single technology or on a single process because we came to the realization that if we start changing and admitting or allowing one thing, then that one thing is going to impact upon the whole existing system, the system that we already have in place. So, to address that, what we did as a board, we decided that we needed to step back. Actually, we went all the way back to the April 1988 guidelines and started looking at, what are the required factors for an acceptable workplace forensic drug testing program.

We have established a series of ideas and comments and factors and that is what has been distributed to you, a table, a checklist, a measuring stick by which we are going to try to do these evaluations. I want to quickly say that this is a work in progress. This document is an evolving document and we would certainly appreciate your comments and your input as to whether or not it is complete, as to whether or not things need to be added, or whether or not things need to be deleted from it. It is a work in progress that we are attempting to establish a yardstick, a measuring device, by which we can evaluate these programs. So, with that, we came to the -- let me just spend a few minutes talking about some of these. You have all these in front of you.

What we attempted to do was identify various general characteristics, general categories of those factors that are included in a drug testing program that would be acceptable for federal workplace drug testing.

Initially we looked at this whole process. Of course, this whole process starts with the collection site. We heard issues about the collection site at the April meeting. So, collection site issues are certainly relevant to all of the methodologies. Many of my comments here today, I think, also reflect some of the deliberations of the board and reflect not only the science but the administrative aspects of a quality program. So, collection site starts the program, starts the issue. What we have tried to do is generate a matrix by which we can do evaluations of the five categories that we have indicated there.

Initially the first category is saliva, then the on-site urine testing, the sweat testing, the hair testing, and the current urine testing program that is laboratory based. We have also looked at that, and we hope to look at that. There may be issues and recommendations that need to be made for that issue as we currently have that formulated. So, we started with the collection site, looked at some characteristics of the collection site, moved on to the donor, some characteristics of the donor, and the specimen. We have included some descriptions of these characteristics. We hope that these descriptions are understandable. If there are questions, if there are issues that you see, that the clarity of the descriptor is not there, please let us know.

As I said, this is a document, a work in progress. It is certainly not etched in stone. It is the measuring stick by which we are trying to start this process. There may be some terminology that may be more appropriate in some of these. We have included in some of these descriptors, in some of these items, some parenthetical phrases that additionally elaborate on those.

We proceed on down, following the specimen into the collection device, and some characteristics of that device, and some characteristics of that device, and further then into the collector, the individual that is responsible for the collector.

Some of the things that we have included in here, and I think we have not come to closure on this by any means as a board, are such things as I have included here in the collection device.

FDA approval; that is an issue that has not been resolved. Do all the devices need to be FDA approved. Part of that is a program issue and the like.

A collector. Another issue that we have identified as an element of a quality program is collector training. How do we document that. What kind of requirements are required for documentation of the collector's training.

Certification; does that collector need to be certified, especially as we have said there in the parenthetical phrase, in the on-site arena. What kind of documentation and certification is necessary. If we are indeed to have a quality program, we want that program to be one that will stand up in litigation and arbitration hearings and court proceedings. These are types of challenges that we anticipate a program would have. That is what propagated some of these types of comments.

As we go on to the second page there, we have got some issues about the transportation. We identified some issues about the transportation of the specimen and whether or not there would be issues in this.

Then we had some issues associated with the on-site testing. What you might want to do -- and I don't have a marker -- section G only applies to on-site testing. If you want to put an X in the columns under saliva, sweat, hair and urine lab, and X those columns out, section G only applies to on-site urine testing. We recognized also that at present we are dealing with urine testing programs in the on-site mode. There may be in the future on-site protocols that are dealing with some other matrices.

There may be in the future some on-site protocols that deal with sweat, with saliva, with hair or with some other matrix that might come up. At present we are only evaluating this in the on-site mode in the urine testing. So, those other columns of section G do not apply. These are some criteria that we are discussing in terms of the on site for the on-site urine test protocol.

Some issues that we are discussing, as you will see here, are the specificity, the FDA approval again comes up on these, the documentation of the on-site test, the identity, the verification by a second certified technician; these are discussion points that are being deliberated by the board.

As you move out of the on-site mode into the laboratory mode, we are dealing with laboratory-based testing that is analogous to what we currently have, analogous to where we are there.

We felt that at least in an initial mode, that many of the protocols, many of the criteria that are currently applied to laboratory based testing should apply to all matrices.

It is that measuring stick, that evaluation, that scheme by which we are establishing this yardstick. We talk about the characteristics of the lab, the receiving, the documentation, the initial testing and the criteria of that initial testing as are delineated in the present guidelines.

Furthermore, we talk about the confirmation testing, that second protocol that is by an alternative methodology, something that is different from the initial testing methodology.

That confirmation test, again, currently the guidelines we are dealing with are GC/MS. We as a board, I think, recognize that there is a distinct possibility that the guidelines may need to be rewritten, if we start moving away from the position of these criteria that are currently established. These are the deliberations and the questions that are in the board's minds, I believe, as we go through these. The laboratory based criteria are not that distinctly different from where we currently reside as we move through there.

The quality control issues and quality assurance issues are, again, very similar to where we currently reside with the guidelines as they presently exist, internal and external QC, the documentation associated with that, and the ability to introduce blinds into the system.

Reporting, again, tries to address some of the issues. This is, I think, where we -- one of the places we first started seeing the fact that if we start moving into some of these other methodologies, that the things that are being proposed or the things that are being discussed will have an impact on the existing methodologies.

I am not attacking it in any way, but I am using as an example the on-site methodology right here, in reporting. If we allow, or if one allows, reporting of on-site test results directly to the employer or to the MRO, then what is to preclude the development of an on-site testing lab, what is to preclude the immediate results automatically coming out of an autoanalyzer in the resident lab, all of these types of things.

To have a program that is consistent and uniformly applied to all the individuals whom are affected by this program, we have to consider that.

As I said, this is one of those examples where we recognized fairly early on when we started looking at this, that if you start changing one thing, you are going to start impacting on the system as it currently exists. I think, at least in my own mind, the system that currently exists is a system in which the administrative component affords the protection to the individual that we desire, that I desire, as well as the scientific credibility of that result that I desire as an individual. So, that is one of those examples that, as I said, led us to where we are in terms of reporting.

Then finally, in terms of interpretation of those results, again, some of the challenges we face are illustrated in these elements. Certainly the issue of bias, we have talked about bias and we have heard some presentations on bias up to this point in time, the issues of cutoffs, what are the relevant cutoffs of some of these alternate methodologies, how do they relate to one another. How does an individual who is tested under a hair testing protocol or a sweat patch protocol or a saliva protocol compare to a urine protocol. We recognize that the windows of opportunity of detection with those other matrices are certainly different from urine and how does one correlate between those two. What are the pharmacokinetic and pharmacodynamic relevances of the cutoffs that are established and how are those cutoffs going to be established.

Some of the issues associated with the interpretation, looking at parent drug and/or metabolite, again, as we move into those, the guidelines that currently exist are very specific as to what the analyte would be. As we move into those other technologies, then the guidelines revisions are going to have to allow for those other potential analytes. This item four, this interpretation, this detection window we have termed there, the opportunity to detect a potential user, the contamination issue, alternate method explanation issues, all these impact upon these alternate technologies.

Then we get item seven, the source of the specimen. Are there variations in there from the source of the specimen, as to hair, as to where the patch might be selected, what the nature of the saliva might be.

Are there variations -- we know that there are variations in the source of the specimen in a urine test, what the hydration rate is on that individual, whether or not he or she might be hydrated or dehydrated. That certainly impacts upon the concentration of that, but that is variable across the population. Certainly the data suggest that it is variable across the population.Are those same types of variations going to be seen and expected across populations as we move into the other matrices.

So, we put together this chart, this matrix, this measuring stick, by which we are going to try to evaluate these specimens. As I said, as we go through the comments today, please look at this, see if we have omitted anything. If there are other items that need to be included, please let us know, please communicate with any of the board. There may be opportunities even to communicate that in the public forum, in the open hearing, in the open public questions. We want to make this as complete as we possibly can. What we did as a board to start this evaluation process is we started looking at these. This particular document, actually a version two generations prior to this because, as I said, this is a work in progress, went to all the board members.

The board members were asked to respond in each square with one of four responses. They were asked to put a Y in there, a yes, meaning that the protocols that had been described, that had been submitted to us, that we had reviewed, did meet the criteria that were established. For example, if we looked at the very first one -- flip back to the first page -- the board members were asked to place a Y into the box, that it might have been appropriate to put a Y into the box under security and privacy of the collection site for saliva testing.

Is that possible? Is that probable? Yes, so we saw Ys in there. If they didn't think it was yes, they put an N into it. A no, it can never be obtained.

So, a Y and an N were two possible responses. A third response was a P that it was probable but we didn't have the details on it, or an I, that it was incomplete, that we needed additional information to make a determination.

So, we went through this whole chart. Everybody, all participants on the board responded, and then we prepared a compilation of that. That is kind of where we are today. We had some issues and we identified some issues that we felt were still issues that we needed some additional information on. Then we also have had come out in the recent months, certainly since the last meeting, some additional review studies and multi-faceted studies that have come forward. So, we went back to staff and said, we need some more help. We need some more information. We would like some more information as we continue this deliberation process. Some individuals were identified as potential presenters, presenters of some of the summary data. We also went to staff and said, some of the summary data that have been presented in some of the summary studies that are presented here, and that is what is going to hear today and tomorrow. We also went to staff and said, we would like to have a summary of all the data that have been presented to date, from all the submissions, particularly in the April meeting, as well as submissions that came in subsequent to the April meeting. Then that was tasked out to an individual and Dr. Selavka, Carl Selavka, is going to make a presentation, a summary of that also, a little later today to the board to assist us. Even though most of us, I think all of us have had a chance to review all the documents that many of you have submitted, Carl is going to try to assemble those into some sort of a summary form for us to further work with. That is kind of where we started and where we are today. We are still looking and trying to get as much information as possible on all these various aspects. We have established this yardstick by which we are going to try to do these measurements. We certainly would like additional input from any or all of you as we go through these next two days of the session.

So, in summary, I guess that is where we are. If any of the board members that are present would like to make additional comments at this time, I would certainly be happy to hand the microphone over to them.

If anyone wants to expand or elaborate on anything I have said, please wave your hand and say so, or no.

Dr. Autry?

DR. AUTRY: Skip, are you saying that these are the criteria that any technology or specimen must satisfy in order to be acceptable for workplace testing?

DR. JONES: Yes, I would think that is the interpretation that we have put on there. Notice the title on this, factors required for reliable workplace drug testing. I think that we are at the position today that even if a new technology were to come on board, such that I could look at the number of eye blinks that might be coming out of the left eye as a result of being under the influence of a particular substance, that criteria would have to meet these factors. That is what we have tried to identify as the elements or the factors in any testing protocol that might be present.

DR. WALSH: Skip, you indicated that under some of these categories that some of these factors were still under consideration. Your last answer to Joe indicated that you had come to pretty much conclusion that these were.

DR. JONES: I guess I would respond to that, Dr. Walsh, in the following manner. As I said, these are the factors that we have identified to date.

We are at this present time in our deliberations, these are what the factors would be that we feel would be the elements of that program, of any program. As I said, I think it is a document that is a working document, that is a document in evolution. We have not made a recommendation to Dr. Autry yet as a board.That recommendation will be forthcoming sometime in the future.

DR. AUTRY: Part of what this meeting is about is to look at this document and see if it needs to be modified, scratched, whatever.

DR. JONES: So, that is why I urge you to give us your comments, to give us your feelings, to give us your input on these. I know that some of you will have some fairly strong feelings on some of the items that are here.

DR. WALSH: One last comment. In going back to the April 1988 regulations to determine what was originally intended to be the basic tenets of the program, you need to recognize that that exercise began in the fall of 1986, and at a time when my instructions from the secretary were to put together the best system of technology that was available at the time. My instructions from the Justice Department were that we needed to bend over backwards and be conservative because we knew the regulations were going to have to pass Supreme Court muster. I guess I would urge the board to remember that 10 years have passed and technology has changed significantly. Also, employee drug testing has really become a standard business practice in the United States. Some of those issues that we were dealing with back in 1986 and 1987 are no longer tenable, as being the primary focus of what we were trying to achieve at the time.

DR. JONES: And I would concur with that. I think this document here reflects not only a lot of the science, but a lot of the administrative -- if I can use that term -- issues associated with a testing program. The majority of the elements on this really are administrative and are not scientific in that sense. They are programmatic. They are organizational in nature. So, with those opening comments, then, we will proceed. We have a lot of time here. We have a lot of flexibility. When we laid this agenda out, we didn't know how long it would take to cover some of these items. But we can proceed on from where we are, Dr. Autry, Dr. Bush. Dr. Autry asked me if the board would be willing to take a few questions. I said, sure. I would ask, if you do have questions, if I can't get to you with this, use the mike in the center so that we do have a recording of this for transcript purposes. Questions and comments at this point in time? I guess I did an outstanding job. Thank you.

They were looking at the agenda. We were scheduled for a break at 9:45. Why don't we go ahead and have the break at the present time. Then we can come back and have Carl have his presentation right after the break. Is that agenda okay, Donna?

Okay, so, let's go ahead and have our break now and we will reconvene in about 15 minutes.

(Brief recess.)

DR. BUSH: Ladies and gentlemen, if you can take your seats please, the reason that we have this preliminary agenda is so that we can move presentations around as we need to, to fit the flow of the meeting. Many times what you plan ahead is not what you really need at the time. So, we are going to have Dr. Carl Selavka make his presentation at this time.

Ladies and gentlemen, I would like to introduce to you Dr. Carl Selavka. Carl Selavka is the director of forensic services for the New York State Division of Criminal Justice Services. He is responsible for coordinating the quality assurance activities of all public forensic toxicology and crime laboratories in the state, including the implementation of laboratory accreditation standards, training programs and provision of technical assistance to improve service delivery. In addition, the office of forensic services is the unit within DCJS which is responsible for insuring implementation and maintenance of the New York State DNA Identification Index. This data bank, which will be housed at the New York State Police Forensic Investigation Center, will contain the DNA records of certain convicted offenders for use in assisting criminal investigations, as well as identifying recidivistic activities. Prior to joining New York State in August 1996, De. Selavka served as the director of criminalistics at National Medical Services, a private forensic laboratory in Willow Grove, Pennsylvania from 1991 to 1996, and as the operations officer at the US Army Drug Testing Laboratory in Hawaii from 1987 through 1991. Other experience includes work with the Bureau of Alcohol, Tobacco and Firearms National Lab in Rockville, Maryland. Carl received his bachelors degree from Indiana University in 1982, and a master of science and a doctor of philosophy degrees in forensic chemistry from Northeastern University in Boston. He is a diplomate of, and serves on the board of directors for, the American Board of Criminalistics. In addition, Dr. Selavka is a fellow of the American Academy of Forensic Sciences. Carl serves on the editorial boards of several forensic science publications, and is active in organizations involved with the quality issues and standardization in forensic science. In addition, he has testified in numerous military and civilian criminal and civil trials on cases involving drug testing in urine and hair, drugs in street samples and paraphernalia, and several other areas of criminalistics. He has published and presented work on a wide variety of topics, including the forensic analysis of evidence for explosives, drugs in traditional and esoteric biological matrices, determinations of pesticides in biological samples, and leadership philosophy in the forensic laboratory. Carl lives in Chocopee, Massachusetts with his wife Carolyn, who is a veterinarian, and their assorted animals. Golf, basketball and sleeping for pleasure fill his odd hours. Dr. Selavka.

Agenda Item: Overview of Information Submitted in Response to April DTAB Meeting.

DR. SELAVKA: Thank you. I will have my mom stop writing these introductions for these talks. I can't start a talk without giving you a joke. It is not a true story, so you don't feel bad at the end. Since I have moved to New York state I have learned that the Hamptons is where the old money lives, and that funny things happen there, including Scarsdale diet things and other things.

A couple was having a dinner party and at the last minute the woman said, we need snails. We have no snails. How can we have escargot without snails.

So, she sent her husband down on the beach. He rolled up his pant cuffs and he went down to the beach and he is picking up snails and putting them into the snail pail. He looks up the beach and there is a beautiful woman off in the distance. He thinks, this snail picking up is pretty boring. I wish she would come over and talk to me. Oh, it would never happen. So, he is picking up the snails and all of a sudden a shadow crosses his path. He looks up and she is standing next to him. Not only is she a great conversationalist, but they spend the night together. The next morning he wakes up and says, geez, the dinner party, the snails, I am in deep trouble. So, he throws his clothes on and runs home, gets the snails in a pail, runs up to the doorway of their condominium on the beach.

Just at the top step, right at the landing, he trips and falls. The snails go everywhere. His wife hears the commotion and opens and door and she is just madder than heck and he can tell. He looks at her and he looks at the snails and he says, come on, fellows, we are almost home!

I did that because this is a really short talk and there is nothing else very exciting in it. A couple of weeks ago I did get a call. We have had quite a bit of information submitted to the Drug Testing Advisory Board. Quite frankly, in the years leading up to that April meeting, there is a good deal of information in the literature available for studying these alternative matrices and on-site testing methodologies in workplace testing environments or in applications thereof.

It is a pleasure to be here today. I appreciate the opportunity to do so. The usual disclaimers apply. Any attorneys in the audience will be happy I do this. Anything I am saying is from my own mouth and brain and doesn't have anything to do with anybody I have ever worked for or any of the labs I am talking about. In general, what I plan to do is very quickly give you a summary of the questions that were posed by the DTAB after the April meeting, for which information was elicited from those presenters and those that might go through those presenters to the DTAB to answer those questions.

I will go over what type of information was received, some of the responses to the questions, what got answered and what in essence didn't have answers provided in those additional materials.

The last time we talked, I was here summarizing hair testing reporting. Really we were talking as a group about thinking outside the box.

This discussion today is an ongoing part of the process of thinking outside the box of workplace drug testing. There is a flood of information available. Just in the meeting we just had -- in April you got a notebook. For those of you who surf the web, this is a double-sided down load of the verbatim transcript of the April meeting. This is half the information sent to the DTAB in response to the questions they posed. This is the other half. Today we are in a really skinny piece of documentation. So far, so good, really thin. But this is just since April. So, that is a lot of information. It is easy to drown in that information. So, what I hope to do is to take the material submitted and try to put it into a summary form that might be a little more digestible.

What I did for the board is, I presented to them about a six-page spread sheet that goes through the questions that were asked and then references a specific area of the documentation that were provided by the respondents to help them assess whether or not those questions were answered. It was not my task to decide what the information said. It would be very presumptuous for me to tell them what anything said. I did tell them where to look, in my opinion, within the package of information. Basically, there were nine formal questions and I added a 10th category or question at the end. They asked about information related to the future role of the matrix for on-site testing in workplace drug testing programs.

They looked for:

1. Correlations with urine drug testing;

2. Publications and supporting data for the alternative matrix technologies and on-site testing;

3. Proficiency testing results if they were available;

4. How the structure of the programs worked;

5. Any data that could be submitted to address lab intravariability related to that alternative matrix;

6. The preparation of calibrators and controls;

7. The issue and how it would be dealt with of sample unavailability, when a person from whom they desire a sample is unavailable to give a sample for whatever reasons are appropriate, that alternative matrix;

8. Definitions of unsuitable samples and how the technology goes about addressing those problems;

9. A summary of the best applications in the opinion of the respondent for that matrix or on-site test;

10. Then a lump at the end that I created called other significant information.

On-site testing and sweat testing packets were received from a number of organizations. The National On-site Testing Association, David Evans, submitted materials for them.

Robert Aramondo submitted materials from Roche Diagnostic Systems. Dr. Niebala sent materials from STC and sweat testing information was submitted by Neil Fortner at PharmChem Laboratories in Menlo Park, California. I didn't have a good picture for spit, so I just had to choose this one. Oral fluid testing and hair testing submissions were provided for the DTAB review. Oral fluid testing information was submitted by Dr. Niebala and Dr. Kippenberger submitted a packet from Psychomedics Corporation in California.

With respect to the first of the questions, the future role of testing and on-site testing, alternative matrix testing and on-site testing, there was material submitted by all of the alternative matrices, and on-site testing respondents. Correlations with urine tests were demonstrated in literature or information was provided in literature from all the respondents.

Additional publications and data to support the use of the testing in the workplace environment was submitted for all of them. Proficiency test information was submitted by two of the categories. Laboratory intravariability was addressed and information submitted by three of the respondents. Calibrator and control preparation information was submitted by the oral fluid and hair testing bodies. The issue of sample unavailability was addressed in information provided by all the respondents, as was unsuitable sample information. The best applications were addressed conventionally in summary format by all the respondents, as well as with some additional information provided by some. Significant information was provided by all of these respondents. There was a lot of information provided. To summarize them, basically what got answered, about seven of the 10 overall categories that I was asked to assess in the information provided to the DTAB were addressed by all the alternative matrix respondents and on-site testing respondents.

There were a few zebras outstanding, things that are not answered in the information provided to date. There are some examples herein, in these two slides. Basically, although information was not provided in the additional materials to the DTAB questions from April, there is additional information available, either in the notebook from the meeting in April or frankly, in some cases, the notebooks from the TF Soft conference that DHHS held at that time on hair testing. Of course, there is literature available that may be submitted hereafter on these questions for which information has not been provided in answer to the April questions. Proficiency testing data for sweat and oral fluid was not submitted as part of the packets that I reviewed, nor was laboratory intervariability for sweat testing. That may make sense because there really is not a large body of laboratories performing sweat testing. It may just be premature at this point.

Calibration and control information was not submitted as part of the formal follow up documents by on-site urinalysis testing organizations or sweat testers. Looking back at the information from April, the sweat testing calibrator and control preparation was described pretty thoroughly in that information. Maybe it was not needed or there was no additional information to be provided for DTAB review. Again, we are thinking of taking the urine programs and potentially augmenting them. As we do that, of course, it is easy to think that this new technology is a blast.

You don't want to get stung, and you don't want to get tied up trying to do it. So, what we really need is to take all the information, discern its critical elements and decide whether it does or does not answer the questions required for the workplace drug testing programs. In the end, as we go down the road with our future tools, we want to make sure we have the right tools available and use them appropriately.

I appreciate the opportunity to provide the service that I have to the DTAB and to be here today to present to you a very quick summary of what I did for them. Thanks very much for your attention and I will take any questions if there are any. In the absence of questions, I will give up the floor to Skip and move on. Thank you very much.

(Applause.)

DR. JONES: Thank you, Carl. That was a yoeman's task, to do that review, to summarize the elements of where many of the items are included in the packet or the packets, the volume of information that we have been supplied. That gives us a great cross reference to that.

Agenda Item: Forensic Workplace Drug Testing Program Requirements.

DR. JONES: Using that type of information, as I said, what we did as a Board is to go through this matrix and start trying to fill in line by line what we did. So, I would like to share with you for the next few minutes some of the deliberations of the board and where we are as we stand right now, in that evaluation. I will be getting copies of this. Some of you asked me during the break, are you going to be able to supply us with copies of the form as you have done these deliberations. The answer to that is yes. We will give you copies of where we are at present a little bit later, and I will distribute those with all the elements filled in on the form.

What I want to do right now is just start going through that and let you know, and explain in some detail, or attempt to explain in some detail, where we are and where we feel at present there are additional information items needed and the like. To do that, I have got some more overheads here. Let me just look at them an element at a time. As we looked at the various elements, what we did, we looked at each one of these items and looked across.

We said, we asked the question, in the collection site copy, is the security, privacy, custody and control form appropriate. Are they available. Are they possible. Do we need additional information. Can we never be satisfied.

As you look across the matrices in terms of saliva, on site, sweat, hair and urine drug testing as it currently exists in the laboratory environment, all the elements of all four of these have yeses in them, have Ys in them.

Everybody agreed that the security and privacy issue was not a problem, standardization of custody and control forms of all methods would not be a problem. Furthermore, the capability for observed collection would not be a problem and the preparation of the collection site would not be a problem.

So, when you looked at the responses of everybody, these all has Ys in them. They didn't have Ns or Is or Ps. They had Ys in them. If any of you feel that is not appropriate, I am going to take the liberty of doing the following. As I go through these, I would appreciate any comments from the floor, from any of you who have any experience with these various elements and these various matrices and these various environments.

If you think that certainly there are problems, please bring them to our attention, of our evaluation status on these. Hopefully that doesn't fall into a big vicious circle, but I would certainly appreciate any comments from here.

I would ask that you need to use the microphones as we go forward with that, because it is being recorded.

MR. FRANCE: This isn't really a comment, but a question. When you say you get a yes in these boxes, there is no comparison, then, as to say whether the capability for observed collection, you know, one is easier than another. It is just yes or no?

DR. JONES: Yes. One thing, state your name, please.

MR. FRANCE: I am sorry, this is Steve France with Workplace Substance Abuse Advisory.

DR. JONES: When we put a yes in an element here, that means that that particular protocol or methodology -- that criteria can be satisfied for that particular protocol or methodology. There is no comparison between the two at the present time. That simply means that this particular matrix, saliva, can satisfy the factor that we have identified as a required element for an acceptable drug testing program. Okay, so as I said, we felt that all of these were satisfied and we had no Is or Ps or incompletes or additional information needed. As we look at the donor, those items on the donor, the only one that popped up was, positive identification can be satisfied in all issues. The only one that popped up with an I was under saliva in preparation of the donor for collection. The I means that we would like to have additional information, that we felt there was incomplete information supplied to say whether or not the procedures that would be proposed or reported would be adequate for meeting these criteria. So, an I pops up in that element, and only that element, in the donor. Again, the board members are here. As I make comments, if anyone wants to make an expanded comment, please let me know and I will hand the mike to you.

Under specimen collection, specimen collection, again we felt that we would like to have some additional information at the present time on some of these. If you look at the size of the specimen in terms of saliva, is it consistent with multiple testing. Again, I think we did not know. We need to look at that. That is a question that we had. Is the size of the specimen in terms of the sweat patch consistent with the availability of multiple specimens. As long as we are at the position that we currently find ourselves of having a requirement for multiple specimens, and it is certainly the feeling of the board at the present time that that should be an element of such a program, then sweat patch has that issue of how do you, or what is proposed, what kind of things are proposed to supply a multiple specimen to the individual, a specimen for multiple testing, I am sorry.

Again, the hair issue, how do you supply specimen of sufficient magnitude for multiple testing and what are the criteria, what are the guidelines that are there. So, those were Is that we indicated there. The site of collection we had no problem with. Split specimen capabilities, if we stay with the split specimen capabilities, what are the proposals for split specimens for these particular alternative matrices in particular. How are split specimens going to be collected. What kinds of containers, what kinds of capabilities are there for split specimens in saliva, in sweat, in hair, even in on site.

We know that on site testing is there, but it hasn't been proposed or recommendations have not been submitted to us as to how split specimens are going to be collected with on-site testing.

If we stay with split specimen technology, that is the challenge that we felt there.

Item 4, evaluation of stability and storage, again, we did not have a lot of information and we felt that we would like to have some additional information on several of these. In terms of stability and storage of the specimen, again, this is analyte specific. I am not criticizing the saliva when I say this, but basically the saliva testing was limited to one drug, the data we have.

If we move to saliva testing for multiple analytes, we are going to have to have additional data there in terms of stability, in terms of storage of that specimen. The same is true with sweat, the same is true with hair, as we identified and reviewed the information that we had submitted to us. Specimen integrity evaluation, what we were looking at there was, again, the suitability of the specimen. We had issues with hair. We talked about that at the last meeting. We have had issues of the sweat patch. We have had the question there of the impact patch, whether or not it is intact, how its integrity is validated, et cetera. So, we come up with two I elements under that, under sweat and hair.

Under the tampering issue of the specimen, we did not have any inclusions in there. So, that is where we currently are with the specimen and those specimen items. Comments? Questions?

DR. AUTRY: A point of clarification. The I is not to say that these cannot meet these criteria, but that the data you had to review was not sufficient to let you know whether they did or not.

DR. JONES: That is correct. Under the collection device, again, we have some I elements under the collection device, particularly under saliva and sweat. What kind of containers are these, the appropriateness of the container. Many of these elements, obviously, have crossed over between other elements of this matrix. For example, the container in saliva, does it contribute to the stability of the specimen or does it allow for stability testing of the specimen to be conducted. There is an obvious cross over there. The same is true with the container that would be for the sweat patch. How do you supply that inappropriate container that maintains the stability and the integrity of the item, of the patch. So, those were the elements that we felt we did not have sufficient information on in the submitted documentation. The item II, capability to secure containers, obviously we felt that that would be okay. There were no challenges, no problems there. Item III, FDA approval if required. This is in part a policy issue. Is FDA approval going to be required for these containers or not.

We are going to hear some additional information on this issue from an FDA representative about the FDA approval process during this meeting, but at the present time we include Is in the four elements there, under the FDA approval if required of the saliva, on site, hair and sweat testing protocols and the containers used therein.

MR. SHULTS: This is Ted Shults with the American Association of Medical Review Officers. My thought process has slowed me up here. I wanted to skip back to section C item number 6 under specimens. We said the criteria here is, shall deter tampering. The first thing that crossed my mind is, are we satisfied that with current practices with urine testing that we are going to have adequate control over water dilution? At best, the best case scenario in my mind is that the answer would be I. If that is the case, how can we possibly say that the question of tampering with on site urine or test or saliva or hair have been addressed? It seems to me that the best case scenario for that would be I.

DR. JONES: I appreciate your comments. I think as we looked at that as a board or as I looked at that element -- and maybe the board can comment on that -- I looked on tampering to mean external tampering, not internal dilution. I am making a distinction personally between water dilution, as you have properly pointed out. I think many of us will agree that the biggest problem we face in urine drug testing in hydration and internal dilution. So, we have looked at it right now in this sense in terms of the integrity of the specimen from external tampering. Part of that interpretation of the impact of hydration, I think, comes up in some of the later elements in the matrix.

DR. SHULTS: I think that is a worthwhile distinction between external and internal tampering. However, we are still left with the situation in legal constraints by having an unobserved collection process with urine. We also don't have, in my mind, enough experience to know whether or not, in theory or practice, you can wash the drugs off your skin or shampoo them out of your hair or chew on the next internet product to tamper with your saliva testing.

Again, I am pretty well left with a personal view that we don't have enough experience or knowledge on doing this. Of course, if you are going to say we are okay with urine, we are probably okay with the rest. Do you see what I mean?

DR. JONES: I appreciate your comments. These are the types of comments that we as a board would appreciate hearing. Originally, from our evaluation, that is where we were. Many of these items right now are collection site issues. You are right. Who is going to check the internet tomorrow and there are 14 new products out there. If you download the internet, there is a plethora of material there. Again, I appreciate your comments on that.

MS. CHILDS: I am Paula Childs. I work at Lab Corp in North Carolina. One of the challenges that labs are facing in the urine program -- and Ted brought this up and I want to just focus on this a little bit -- is the adding of things to urine specimens. Probably one of the challenges that we are all dealing with right now is a product called KLEAR. I guess what I want to say is, as quickly as the laboratories are able to identify and try to overcome the challenges that these different adulterants present, there is something new that is coming along. Although we spend lots of time sifting the internet looking for those, I would estimate there are probably 50 to 100 problems that one could use to adulterate their specimens. I am sure that as soon as there are technologies that allow one to do on-site testing, for example, those same products may work very effectively with on-site testing and with hair testing and so on. We are looking at not only removing drug but doing something to sort of interfere that you don't really know is there and how to identify that. That is really a challenge that laboratories currently face and I am sure will provide adequate financial opportunity for people who want to go out there looking for ways to advise people about ways to beat their testing, whether it is urine based or some other kind of substance we are testing, hair or saliva or others.

DR. JONES: I appreciate your comment. I guess I would say that I have always looked at this program as a deterrent based program and not a detection based program. When we go with a deterrent based program philosophy, we are going to be encountering many of those. I think that is my own evaluation of the inevitability of that. Those are challenges that are going to impact on all these matrices. We don't even know what those challenges are going to be in the future. The labs, as you have rightly pointed out, have been operating in a reactive mode, in many, many cases, rather than having had the opportunity to be proactive. They have to be reactive. I appreciate that.

Okay, going on then, if I may, in the collection devices, as I said, the elements that we have entered into the collection device elements are the I in terms of the FDA approval and the questions we have there. An issue that we have identified in this whole thing, then, in the collector as we move on down to item E, the collector on this particular page, notice that we have included in the collector documentation, the training of the collector and certification of the collector as a desirable criteria for a factor required for reliable workplace drug testing.

All of these have Ps included in all elements of documentation across all matrices because drug testing does not require certification of the drug tester at the current time. As many of us know, that is often reported as being one of the weaker links of the program, is the collection system. So, ideally it would be very desirable to have that individual to be certified, that collector. So, that is where we came from in including these two elements, the documentation and the certification of the collector on that particular one.

Similarly, we included Ps in the requirements for special handling across all elements as we moved across the matrix. There may well be conditions out there, and it is certainly possible. Much of the data that was submitted to us did not include details of these. Certainly in our evaluation of the board -- and the representation on the board is fairly diverse, coming from a variety of backgrounds and disciplines -- many of us felt that those types of elements could be included. So, we included a P on all elements in the transportation of the specimen, requirements for special handling. Not knowing what those requirements are going to be in some of these instances, as I said before, these interface with other elements.

What are the stability issues, what are the handling issues for some of these other matrices, the saliva, the sweat. As we accumulate more and more data on these particular matrices, these are going to come into play and we felt those could be fairly easily addressed.

In terms of the on-site, let me just do it this way. As I indicated in my earlier comments, the on-site testing is strictly item G, on-site initial testing for drug or adulterant is strictly limited to urine, and we did not fill in any of the matrices under saliva, sweat or hair or urine lab. As I said earlier, there may be somebody out there who wants to propose an on-site sweat test or an on-site hair test or an on-site saliva test. We are not looking at that at the present time. We are looking only an on-site urine testing program. So, we had some elements in there that we had some questions for and many of these will probably be addressed by some of the review issues that Dr. Willette is going to share with us again later at this meeting.

As we started going through there, we felt we needed additional information on the specificity of the analyte class, some questions that we did not have total information on in the submission. Again, shall it have FDA approval if required. We don't know whether it is going to be required or not. I think that is where we are and so that is why we had an I in that element. Is it going to be required to have FDA approval. We will hear, again, more about that as we look at some of our comments from our FDA presentation. The provide documentation of the collection. Certainly that is possible. We felt that was certainly an element of P. We did not have the data submitted there.

This would be very similar to, in all probability, a typical urine collection. As long as the collection of on-site is consistent with the collection of any normal laboratory-based urine program, those two are certainly very feasible and very probable. That included an elemental P in there.

Documentation of the test results. I know we are going to hear some issues, some of those discussed by Dr. Willette today.

How are these results documented. I appreciate the information that has been supplied to us, again, by various submissions from Roche and from Mr. Evans and those things that came to the board allowing us to understand some of the details of some of these devices that are there, and some of those that Carl referred to in his comments.

We would like to have some additional information on the documentation of the result. Item 5, shall provide for specimen handling issues, specimen aliquoting for on-site testing, specimen retained in the test device.Some of the devices, as you know, are all in one devices and the urine goes in there. So, how is the result retained. We feel that to have at least under our initial look to have a valid protocol here, you need a mechanism by which that record can be maintained. As many of you are well aware, those are the challenges that you often face, is the documentation of that result. So, how is that result documented. What kind of sample retention capabilities are present. So, again, we entered an I in there. Shall use the HHS approved cutoff levels for that specimen. Again, many of the on-site devices have capabilities of a variety of cutoffs and we would like additional information on those.

Certainly the on-site testing protocols that are our thoughts at the present time, if those protocols are to be used and it is a urine-based protocol, then the same cutoffs that are based in the resident lab are going to be applicable in that screening test as those that would be used in the on-site testing protocol. Shall be directed toward DHHS approved target analytes. Again, we felt that this was a P. However, there is probably some additional information that we will receive from some of the presentations today and some of the more recent data that has been submitted to us on that.

The issue item 8, shall have objective differentiation of positive and negative with or without a recording device. The challenge that many of us looked at in reviewing some of these is the ability of that device to discriminate -- if I could use terminology that comes out of the guidelines -- discriminate reliability between a positive and a negative at or around the cutoff. Those are some of the challenges that we felt were opportunities for having additional data in there.

The issue of controls, we have already talked a little bit about controls in some of our previous discussions, and suitable controls tested and documented.

Item 10, a certification program for the analyst. If we are going to -- again, this is one of those -- item 10, 11 and 12, if these are going to be programs that are consistent with the existing program, we require certification of the laboratory based program. If we are going to allow results to be coming out of an on-site testing protocol, then initially it was felt that it would be appropriate to require certification of those individuals that are doing those analyses.

Rather than looking at the certification of a lab, it appeared to us, at least initially, that it would be better to look at the certification of the analyst. In an analogous manner, I guess I could say, we looked at the certification of the collector as an opportunity to enhance the integrity of the program, the certification of the analyst. Certainly the issue of the donor being blind to the analyst, shall provide for analyst to be blind to the donor identity, that is an issue and concern. That has been a major component of the existing program. As many of you know, in regulated testing, the identity of the donor shall not be known to the lab. So, translating that over to on-site testing, then the identity of the donor should be -- there should be a blind issue between the donor and the analyst in the on-site testing as an element of the requirement. If we do not require that, then the logic of it would be to not require it in the other side of the house. At the present time, that is where we are there. Thirdly, there, shall provide for verification of the results by a certified technician, by a second individual. We require that in resident lab environments. Even if we look at urine drug testing we allow for a negative certifying scientist to exist in that environment. That individual certifies a negative result and that result comes back to the MRO. If we accept that as a minimum standard of operation then it appears logical to require a certification of that result by a second individual in the testing protocol. Again, if you do not maintain the second tier of review here, then there is little logic in the argument, in my mind, of requiring that second tier of review in the resident lab testing protocol.

So, these are again examples of those issues, as we started looking at these, of why we went all the way back and started looking at the elements of a quality program as we defined it, rather than saying that this program or that program would be acceptable. Questions on the on-site issue?

MR. PINDER: I am Richard Pinder. I would just like to reiterate something that was said earlier; that these requirements that the board had listed here that are required for reliable workplace drug testing, positioned where we are now -- and we are open for additional requirements -- maybe we should consider for a minimum requirements for reliable workplace testing, also we are open to arguments about the appropriateness of the requirements that are listed here.

MR. KUNSMAN: Ken Kunsman, STC Technologies. What data do you have to suggest that in a presumptive positive or a screening test for on-site, that having the analyst and the tester and the collector creates a problem, certainly referring to the on-site alcohol testing, where that screen is then confirmed by another individual?

DR. JONES: Speaking for myself, I don't think we have any evidence there. Where I think we find ourselves is that we have a criteria that has been placed upon the in-lab urine drug testing system. That criteria has stated that there shall be that blind component. So, if that remains acceptable, then we felt it would be appropriate to apply it to the on-site arena. If that is not acceptable, then that probably should apply across the board. That is where we find ourselves as a board, I believe. Comments from the board?

MR. KUNSMAN: I certainly wouldn't speak for Dr. Walsh, but he did mention the idea that technology has changed to the time when that was the requirement. It was just a matter of function. Was there any debate on that issue during the board deliberations?

DR. CAPLAN: Yale Caplan. I will try to comment on that somewhat and maybe also reiterate some of the philosophical things that are going on, why we really are here and why we need your input. What we are trying to do is pose the questions and ultimately come to a decision or at least a recommendation of what a policy change might be and whether there is adequate scientific information to support it. So, as Dick has said earlier, the mind of the board is open. But what we did do at the result of the last deliberations that we had and our ability to look at the information and deal with these things is really try to pose the questions. We don't necessarily have the full answer. But we did realize, I think very significantly, that whatever we look at has to work across the board. It either has to be supported before the board is going to make a recommendation to HHS by some scientific basis that we are trying to determine. We may or may not be in a position to recommend a policy change. In other words, we can say we think that is what it is and that this requires a policy change. Then that policy would need to be looked at by parties above and beyond the board. I am glad to hear the discussion we are having now. It is hard to say that we, in the short time on telephone conferences, et cetera, came to consensus on every one of these issues. I feel that we are trying to look at these things in the most comprehensive manner. There is a big distinction, I guess, in what is going on now -- I guess I will take this opportunity to reiterate it since I have got the floor right now very quickly -- some of the things that dictated where we are going. I think the audience needs to understand that and give us the input back so that this can be altered or modified. That is to say, if we are going to change, what do we need to do to affect change. Our change has to be what does the public interest require. The public interest may require things differently today, as you and Mike have said, than it did before. But that may not be our decision. We could perhaps make a recommendation, but that is going to be what the public interest desires. The second major element of consideration was, what is the impact on current programs. We have to have, no matter what we do, something that will work across the board.

What became abundantly obvious in the deliberations and the process is that we cannot take each piece of this, each idea, each technology, each altered entity and say whether that is acceptable or not. We tried that. I think we made an honest approach by looking at that the first go-round and we could not come to an effective conclusion unless we went back to basics. So, philosophically we did, quite frankly, do a bit of a reversal. We are back to basics. It has become very clear that we need to develop the guidelines or redevelop the issues that are in the guidelines to be more appropriately stated, so that they could include multiple specimens. The guidelines deal only with urine, which was favored at that time, and the guidelines need to deal with this in an effective manner. I don't think we have an answer. In fact, as you can see now, we have got many more questions than answers. I know many of you here hoped that we would have answers and it would be nice if we did. But we actually have more questions than answers today. The important thing to the people here in the next hour or so and in the course of the day is to bring forth these things. We will try to make note of it and write it down. Yes, the on-site alcohol you might consider an anomaly. We have always considered alcohol an anomaly in drug testing for a variety of reasons. It is one of the substances that we can do an initial screening and confirmation on site by what have been recognized as evidentiary devices. We don't have that in this arena. Therefore, we can't make the jump to say, if you could do it for alcohol, you could do it for drugs. Please, I am happy -- particularly the other thing Ted mentioned, we might have skipped over these things in the vast issues that were there. It doesn't mean that we skipped over something because it was unimportant, but in the realm of things, that seemed to be more of a policy decision, whether or not in a deterrent based program there should be concern, fundamental concern, about the alteration of a specimen or not. We know that we can look at that later, but since it has already been established one way for urine, the others didn't seem to be any worse off for that.

Some of you may have comments that maybe they are better. If we are able to come up with a policy that is stronger than the one we say today, that there may be other specimens that can better meet that. I am going to reiterate what Skip said. The way these are worded, these are ideas. They are concepts. Our goal at the end of this is to write each one of those into a recommendation. In other words, write a one line, two line or three line or whatever statement, to say that it is our feeling that this is important, that this should be done or perhaps this shouldn't be done. That is the goal and we do need the input of everybody here. I would say with the 10 or 12 of us that are involved in doing this, there is a much vaster experience out there.

The other thing that we are always mindful of in this process is that what this -- and Mike again may attest for this over the years, even though things have changed, but what is this program based upon. It is based upon a process of integrity, insurability, reliability, met legal challenges, and we have to be very cautious that in making changes we don't compromise that. We have had a program that had two major elements; very tight control over the processes, which we are trying to review for the other substances and procedures which might come into play, and also the independent review of these processes by third party MROs. These are the things that have been the stalwart of the success to date, and it is certainly the responsibility of the board to look at these things in such a way that we can continue to ensure them from. I guess, a philosophical point of view that is where we are, and I want to reiterate the things that Skip said. We appointed Skip spokesman. He has done a very good job, but there is a deeper thought process here and we need your assistance in that as well.

DR. JONES: Thank you, Dr. Caplan.

MR. CROUCH: I am Dennis Crouch with the Center for Human Toxicology. I apologize because I missed the first meeting. I would like the board to consider blood testing for a couple of reasons. Maybe you have discussed it, but I think it is worth looking at. The negatives about blood testing have always been that it is invasive. I think it is arguable whether it is more invasive than plucking hair or wearing a sweat patch for a week or two. Secondly, a negative about it has been the sensitivity requirement to test blood. However, the sensitivity is at least in the ballpark of that needed for sweat testing, hair testing and saliva testing. There are a number of positives about blood testing. First of all, it would be difficult to adulter a specimen. We have a wealth of information about the pharmacokinetics in most drugs and metabolites in blood. In blood, often we can detect the parent drug or metabolite that would provide a lot more qualitative information about the identification of the drug.

Some of the new immunoassays and existing immunoassays target the parent drug so that they can be used on whole blood, plasma or serum samples.

Lastly, we have a lot of information about interpretation of results. There is a wealth of information in the literature about blood and the pharmacokinetics of drugs of abuse in blood, plasma and serum that is lacking with some of these other matrices. This can give us a lot of information, at least more information than we have about interpreting impairment, time of dose or possibly the dose.

I would like to just put that out. I don't know if they have been discussed before. Some of these matrices are so new that we don't have a lot of literature on them, while blood testing has been around for a long, long time.

DR. CAPLAN: Yale Caplan again. Just to respond to that point, part of what the deliberations that we had and are doing, while they may not have been specific on blood because I guess there are no proponent groups supporting that from an industry based point of view at this time, the goal of redeveloping guidelines along the lines that we are suggesting are such that they would allow the possible inclusion of anything as it came along.

That is a major difference in the thinking, I think. So that, if we say that there are certain elements -- and you can see probably in some of the information we have for saliva it is going to parallel blood in concentrations and maybe analyte detectability -- that should there be other specimens -- and blood is the primary one -- it could be included effectively along the same lines and wouldn't have to come, again, as another afterthought. So, part of the goal of the board is to say, what do we need. Then one could propose then whether or not current things out there or things which we yet have not considered may meet those criteria.

MR. THISTLE: Bill Thistle, Psychomedics. Just to clarify the reference to hair testing in the work place, I am not aware of anyone who plucks hair for testing.

DR. JONES: Other comments on the on site issue?

MR. FORTUNA: I am Joseph Fortuna with Chemical Detection Services. I am not sure I agree with the across the board factors being applicable to all different candidate systems. What we need to do, we have to look at how they are used in the workplace. On-site testing has a lot of problems. I can see it for my clients, some of whom don't want to pay the extra money to go to a laboratory. That is where the problem really becomes for me. How do I explain to my clients the risk they are taking by not having me take the samples to the laboratory. I will do both. I can do either one. I make a little more money, actually, doing the on-site testing than the laboratory, but I prefer the laboratory. The question that really comes up to this is we have got to do it in the context of what the workplace manager desires. For small businesses -- and that is who I serve -- they are very conscious about cost. Most of these factors are very good and most of the kits have these factors. Some of them aren't there, but most of them are. The question is, do you really want to have a uniform set of factors for all systems, independent of how they are used. That is what I would like to ask as a question.

DR. CAPLAN: I guess I will play the role in answering the questions, at least maybe commenting additionally, Skip, on the position of the board. What you are proposing is certainly secondary and was considered, in that we can't do everything at once, but right now we are talking about looking at equivalents for one particular task which is a workplace based deterrent program. It certainly doesn't preclude -- I think as Dr. Autry indicated, we are in a new time now. We spent the first 10 years doing one thing and I think we are going to spend the next 10 doing a lot of different, more diverse, things. Certainly the use of these specimens for pre-designated special purposes is something to follow. But it is too big a chore to bite off all at once. The first major question that we are dealing with is, what things can we do to support the current program. We use this term, alternative specimens. I am going to coin a phrase for you today. We are talking about not necessarily alternative specimens, but specimens which are complementary. We are looking, I think, at what are the various complements or complementary processes or complementary specimens or complementary other things as go on, which can allow us to produce an effective workplace result. If it is deterrence as we have now, that is one thing. If it turns out to be for cause and other things, then there may be other recommendations to come, as various specimens and technologies and various approaches complement one another in a comprehensive program. That is longer term, though, I think.

DR. JONES: I would also add, we discussed in a variety of times the use of these complementary matrices for specified purposes. There is nothing to preclude the idea at present of some of these matrices being appropriate for post-accident, for example, or uniquely for pre-employment or uniquely applied to rehabilitation type testing. I guess one thing that we have tried to look at it as, if they are going to be looked at it as, if they are going to be used in this general context of workplace drug testing, then that is where we are with the present matrix here, in the general application of workplace drug testing programs.

DR. CONE: I just had a couple of comments. We seem to be stuck right now on the on-site testing and my comments apply equally to the on-site situation as well as some of the other matrices. One of the things I think I lack in terms of understanding is I would like to focus whenever we can. I just want to make sure we are talking about the same things. The heading for this forum says, factors required for reliable workplace drug testing. I am assuming that means reliable federal workplace drug testing and not private sector.

DR. JONES: Correct.

DR. CONE: Several comments and questions. This may be getting ahead of the game, but I would like to think that what we are hopefully moving toward with each of these matrices and test modalities is the development of a certification program similar or dissimilar, but at least in effect some kind of an oversight of these test modalities. I am wondering if that is an assumed goal behind this forum or whether it is an implicit goal of a certification program, and all the sundry of programs that go along with that. The core of the NLCP program, I think, has been the strength of the certification, the PT program and the laboratory inspections. That triumvirate of strengths has gone over the decade toward producing an incredibly reliable technology in many respects; that is, urine testing. Are we moving toward these other matrices? If we are, I think I would like to see more in terms of how do we adapt these matrices to these programs. Along those lines, it would seem to me that maybe as part of your efforts to bring some of the people in the certification program the actual nuts and bolts of making that program work and say, if we were going to set up a certification program today, how would we do it or what are we missing. Where do we stand in terms of adapting these new test modalities to an actual certification program. So, I would like to see us, at some stage of this process, address the issues of how do we implement or design a certification program. I think in doing so we might then be able to open up some of these critical questions that haven't been answered but need to be answered.

DR. JONES: Thank you, Dr. Cone. I think some of the items in item H will address at least the board's current position on where we have looked at some of those elements that you referred to, because that addresses the laboratory process. So, some of that, I think, will come clear, at least, where we are at present.

MR. MC CANN: Thank you. Bernie McCann, Labor Health and Safety Fund, North America. The last gentleman made some astute remarks, I thought. What to me is clearly obvious is when I look at section G and section H, the major difference that I see is item number one, shall employ DHHS certified laboratory. The major concern I have with on-site testing is the lack of any certification process that an on-site facility -- or in fact an on-site capacity because it isn't really a facility in my mind -- would have to go through. Given the current political climate and the capacity of the certifying authorities of going into such on-site facilities and actually doing any type of certification, you know, it is just not going to happen. Therefore, what we really have is a Volkswagen and a Mercedes Benz, to use an analogy there. That is really the crux of the matter, to me. When we get to on-site testing we are really not going to be able to compare apples to apples. There are two very, very different, even conceptually, methodologies. I just don't really see how we can even put them up against each other with any expectation that they will be able to match up. I guess that is more of a comment than it is a question. I guess what I would like to hear from the board is how they expect to be -- or maybe hear from some of the folks who are here proponents of the on-site testing. How do they actually expect to be able to have a credible product that is going to stand up next to these other methodologies. I just don't see it. Basically the certification process I think is the way that we separate the two out.

DR. JONES: Thank you for your comment. Mr. Evans.

MR. EVANS: I am Dave Evans with the National On-Site Testing Association. My first car was a Volkswagen and it was a great car. The On-Site Testing Association is moving toward developing a certification program. We support certification. We have supported legislation. For example, in Alaska we supported legislation. A bill was passed that allowed on-site testing and that bill had certification requirements in it. We welcome anybody's input on the issue of certification. Certification has to do with not only collection sites, but test operators. We are moving in that direction. Similar to the National Association of Collection Sites is moving toward certification, we now have some very good programs for certification of MROs. Ted Shults is here. Those are all private efforts that were done without the government. Not that the government shouldn't be involved; we welcome governmental involvement. We agree that certification is the way to go, training is the way to go, meeting FDA standards. These standards, I think on-site testing can meet these. We welcome anybody's input on this, including labor, on how they think the certification should go.

DR. WALSH: I am Mike Walsh. I was very fortunate to kind of be with the framers of this program, and most fortunate in that the terms or concepts, efficiency and cost effectiveness, were never a consideration in the development of this program. We were breaking new ground. Our directives were to put together a system of checks and balances and multiple redundancies that would be legally unassailable. We did that. I think the very difficult balancing act that the board is dealing with now is the reality of the day, and that is that efficiency and cost effectiveness are primary concerns. They are very much so in the federally regulated, transportation, nuclear regulatory field, and so on; also, all private sector testing that generally has leaned on the federal regulations as the gold standard.

I think the board is -- I know at the last meeting I got the feeling that in many instances much of the discussion was focusing on really administrative aspects of the program. I felt that maybe these were more within the purview of the human resources section of various organizations and really are not issues that the board should be focusing on. The Bubba issue, which kind of dominated the last meeting, it seemed to me, was one that sort of fell into that category, within the DOT alcohol program that dealt with that. I suspect that it could be dealt with equally as well in a drug testing program. The difficulty is that some of the administrative aspects of this program are critical to the overall integrity and reliability and so on. I think what the board is being asked to do in terms of alternate technologies is, can we integrate this within this system.

How many factors that were key to the original program are still necessary today. Can we, in trying to balance the cost issue, the efficiency issue, the effectiveness issue where you are going to give on the one side, you are probably going to diminish the accuracy and reliability.

How far can you go and how can we come up with a system of federally regulated system for federally regulated testing that is still going to be able to meet today's legal challenges, but be more efficient and cost effective.

It is not an easy balance and I think intermixed in all of this are some very highly charged political issues. That is, if you believe that drug testing is an extremely viable tool, as I think everybody in this room does, that on the one hand, from a political perspective, you want to increase the use of testing as much as you can. Therefore, you want to increase the efficiency and cost effectiveness of these programs.

On the other hand, the only reason that these programs have evolved from the early 1980s to where they are today -- this is an issue that I come up against in Europe all the time where people say, where did you get to where you are now from where we are. The answer is, it is not easy and it didn't happen overnight. But I truly believe that because we built this iron clad system of protections and multiple redundancies and double checks and MROs and a sense of fairness in the overall policy end of the program but also an absolute confidence in the accuracy and reliability, as Dr. Cone mentioned, of the laboratory program, which for those of you who have been around for a long time was in serious question 10 years ago, that is the reason, built on this very strong cornerstone of science and technology, that the programs have been able to grow to where they are today.

I think the charge is where -- I know Dr. Autry struggles with this -- where to focus on the science and technology and how much of the policy and the cost effectiveness and cost and efficiency of the system -- how do you accurately mix that balance to come out where we ought to be.

DR. JONES: Thank you, Mike. As you well know, we met in this hotel some number of years ago. It was in the early 1980s, in the first meeting to get all the army, navy and air force together and have a discussion of these types of things before even the executive order came out. So, we go back a long way and we have seen the genesis of this. Thank you for your comments. You recognize the challenge that we as a board face in making recommendations to Dr. Autry and that group, too. In balancing those issues, if you back off on one point, you pick up on something else, but where does that balance lie. At least in our initial efforts we are trying to share with you where we are in that balance today.

Any additional comments before I proceed on with the laboratory side? As I said, then, the laboratory analysis, the laboratory process, it is still visualized that even the on-site will send, in some manner, those specimens to a laboratory. Under the current guidance of the laboratory being a DHHS certified laboratory, then all these complementary specimens, if I can continue to use the words that Dr. Caplan used, if all of these complementary specimens are still going to be tested in a laboratory, then it is certainly feasible, it is certainly possible that the DHHS certified laboratories could handle those.

Incorporated into the entry of a P into these elements is the fact that then a particular protocol would have to be developed to challenge these laboratories in a proficiency testing, et cetera, to handle these particular matrices that are there.

That becomes more apparent as you look at the other issues associated with the laboratory, of the utilization of current DHHS criteria involving receiving of the specimen, that provide for short term storage of the specimen consistent with stability issues, and provide long-term storage of the specimen, again consistent with stability.

Those have all been included in the top three elements, two, three, and four. It is certainly possible for all of those specimens, as they come into a DHHS certified laboratory.

Where we have found that we would like to have additional information is on the initial test, as it would be imposed on any or all of these.

Again, currently the regulations say that the initial test shall be an FDA approved test and an FDA approved protocol. So, additional information is being requested for not only a saliva but for the on-site and the sweat and the urine.

We have included the I even on the on-site there. That is probably the closest one that we probably have a P in, but at present we have left it in as an I. Again, shall be directed toward the target analytes. This becomes, in part, a regulatory issue, a regulation issue. The question is, what are the analytes that are particularly present in these complementary specimens. What is present in saliva? What is present in hair? What is present with sweat?

Are those consistent with the current regulations? Of course, in some of these, we don't even know because we don't have the data on all drugs or drug classes that are currently under the guidelines. Shall use DHHS approved cutoff for that specimen. Again, we don't have the information. We don't know what those cutoffs would be. Since we don't know for sure what the analytes would be in all cases, we don't know what the cutoffs would be. So, we have included a request for additional information there.

The issue of precision around the cutoff, again, we have some data that were submitted on some semi-quantitative and quantitative basis that Carl referred to. Certainly the sensitivity of these methodologies are methodologies as they apply to these particular complementary specimens and it allows for additional information to be submitted to demonstrate that precision in and around the cutoff on the screening assay or the initial test.

In the present arena, again, the second initial test is allowed, and translating that over to the other matrices, we don't know whether a second initial test, a protocol, exists, even for a second initial test. Are the alternative immunoassays applicable to these other matrices that are currently being used in a urine drug testing program. We don't know that, and that is why we have included Is in that. Again, this goes back in part to the issue of whether or not we retain all of these elements in this checklist. As the elements currently exist, that is where this list has been generated.

So, we have a lot of Is in the initial test and we also see some Is in the confirmatory test in the laboratory based testing protocol, as well as some Ps.

The confirmation test, at present we are again at GC/MS, under the regulations. I think as many of us are well aware, there may be some alternative methodologies there that are very applicable to these drugs as we go into some other matrices. It appears that the data that have been presented to date have all utilized the GC/MS methodology as that confirmatory test.

Certainly it appears possible and probable that GC/MS methodology can be applied to all of these matrices as we expand the analytes that are included in these matrices. As you move away, of course, there may be other methodologies and there may be something else that would go on there. But that is where we are with the methodology. The target analyte in the confirmation test appears as is the target analyte in the initial test. There is some questions about what those target analytes would be.

For example, in the saliva, as we move to some of the other drugs of abuse that are included in an allowed workplace -- federally regulated workplace drug testing program, we don't know what the analytes would be.

So, we need the additional data there. That would be the same with sweat and that would be the same with hair, as we move into all those particular drugs that would be used. Again, the cutoff issue is illustrated and the linearity around the cutoff, addressing the reliability, the discrimination ability of this particular assay, this particular procedure to discriminate reliably between those specimens that are positive and those specimens that are negative at or around the cutoff.

The verbiage there is very similar to the existing guidelines.

Finally, the challenge that we are faced with in part is relating one particular methodology to another. That is certainly not our goal, to have commonality between the methodologies. We all recognize, I think, that the different methodologies give different information.

The cutoffs to be employed, we use the terminology here that they are pharmacokinetically relevant. In other words, we have a feel for example, I believe -- most of us have some sort of a feeling of what it means to have a particular cutoff in a urine drug testing program, what that means in terms of recent use, what that means in terms of past history. However, as we move into these other matrices, we don't have that feeling yet. We don't have that data. Some of the data that we have heard reference to in blood, we don't have that data yet in sweat, in saliva. Those are some of the questions we have in saliva and hair. What are the cutoffs going to be in those particular matrices.

It is certainly a concern of mine, that if one recommends that a particular protocol be acceptable for federally regulated workplace drug testing programs, then that is one of the details about which we need to know something; what is the cutoff. So, from the laboratory side of the house, that is where we stand at least at the present time. Certainly we would be adding some additional information. Certainly from the laboratory side of the house, that is where we feel that the data that have been presently supplied to the board put us at present. Any comments about the laboratory sides or issues?

DR. SELAVKA: I am curious, after the April meeting. The language in number 7 is different. I am hearing it differently at this meeting than we heard it before. That is, cutoffs shall be pharmacokinetically relevant with respect to the analytes. The way you described it, or in April it was described as driving the process of setting cutoffs. That is where pharmacologic relevance at that time, that is how it was being used. Today I hear you saying number 7 relates to the ability to understand a time window around positive findings in any matrix, based on pharmacokinetic properties. That, I think, is a very different way of interpreting pharmacokinetic relevance. I guess the ultimate question is, are 5-B and C and 6-B and C, which is the setting of cutoffs and targeting of specific analytes going to be driven by pharmacokinetic relevance to given doses of abused drugs, or will pharmacokinetic relevance merely be used to understand and interpret appropriately positive findings once cutoffs are set. I wonder where the board stands on that question.

DR. JONES: I personally would say the latter. I don't know how other board members would respond, but for my own purpose I would say the latter, for interpretation. Historically, if you look back at where we are today with the urine program, where did those cutoffs come from. They were driven by a variety of parameters, among which were pharmacokinetics. There were policy, there were politics, and the availability of the methodology. That was one of the big driving forces, the availability of the methodology to produce a result at that particular level. That was a big driving force there. I think what we are looking at here, even though those cutoffs were established with all those other parameters input to them, from my perspective, we are looking at a similar type of thing, but there should be at least some correlation, if I can use that term, between the methodologies and presentation of that result and understanding of what that result means. That is the background behind 7 in my mind. The board may want to make other comments, too.

DR. AUTRY: It also strikes me that there is another element in pharmacokinetically relevant. That is, where do you draw the cutoff to strike the balance between true positives, false positives, true negatives and false negatives.

DR. JONES: Anybody else?

DR. SELAVKA: If I could just finalize a comment on your comment. I think that is a fairer way of approaching this problem than was addressed in April. Even urinalysis, because of the historical setting of cutoffs, it couldn't meet that test and it is impossible to say that it ever really could. Thank you.

DR. JONES: Thank you. Other comments, questions?

MR. VELASCO: Javier Velasco from Poisonlab. I have a question. When you are establishing a cutoff, how are you going to account for the differences from one individual to another, like female to male in the case of sweat, or hair growth or those kinds of changes from one individual to another.

If you establish a cutoff that would be applicable to the male population, let's say, how would it be applicable to the female population or between races for some of those analytes?

DR. JONES: I think that is why we have Is in some of these elements. These are types of information that we don't know the answer to, that we want more information supplied to us as board members before making those recommendations. I am getting head shakings up here from everybody in the front row. I guess I responded correctly. Other questions or comments? Then as we move forward into quality assurance programs, again we would like to have additional information across the board, QC specimens. I realize that there are systems currently in place and part of that is what we are going to see tomorrow and today, about the introduction of quality control specimens and a quality control sufficiency testing program for some of these matrices. We heard some about that at the April meeting. We are going to hear more tomorrow. But we certainly have Is in saliva, on-site, sweat and hair at the present time. The program should contain documentation of the validation of the test results. Again, that would be the next step.

Once the QC specimens are included, the documentation of the results should be obvious, and shall provide the introduction of blinds into these systems. Again, we are going to hear more about how blinds are introduced into these systems in today's and tomorrow's program. So, we did include Is in those four elements across the board there.

If I can just go on to the reporting, since this is on the same slide, again, the final review of all the results by qualified medical personnel, as we apply the same criteria to all of these that currently exist, that means a medical review officer in the loop. That is our initial interpretation.

So, all results would go to an MRO. It is certainly possible for all these matrices, whatever the results that are generated, to go to an MRO for a final review. That is why we have the piece in there, even though we haven't heard the presentations of these. We didn't see a need to make a challenges or hurdles to be overcome as you enter into the reporting arena with any of these other matrices. Similarly, reporting in writing in a confidential manner, can be just as easily achieved there, even though we haven't had the presentation of how to do that. A standard custody and control form across a given matrix technology on which the results are reported, it is certainly conceivable that a new standard testing control form could be created. However, how long did it take us to get the one that currently exists only for urine? To go across matrices, that is going to grow exponential, not just multiple factors. It is, again, certainly feasible to do that. Then, finally, the testing control form reports. It says, the results of the test shall be certified by the appropriate trained individual. That is different from the analyst. The test results are reported are reported by analyst class, and the positive results only if the initial and confirmatory test are positive, the application of the same criteria of the two-tiered testing program that currently exists. So, all in the reporting arena, all are possibles, even though we have not seen the actual proposals. We did not foresee hurdles that would offer any real challenges there, other than the establishment of a uniform testing result form for all matrices. Comments on those two?

Finally, interpretation of results, and some of this gets into issues that Dr. Selavka was addressing there. As we go into interpretation of results, is the result independent of bias toward specific populations. Those are questions we would like additional information on. We have heard some of those discussions about bias, particularly bias in terms of the hair. We don't know about bias in terms of sweat or saliva or bias in terms of on-site.

Are we willing to allow bias in urine testing? This gets back to some of the things we heard about earlier. Water loading. That is a bias. That is a bias of a population. I recognize that as a bias of a population and is that acceptable in workplace drug testing, as Ted Shults referred to a little earlier in his comments. Are cutoffs in their relation to passive exposure relevant? We know things about some urine. We know things about there, and I don't know whether that should be an I in urine there. I think that might be a typo. I don't think we had an I under urine under passive exposure. As I said, this is a document that is a document in evolution. It is a working document. So, that is certainly one that I think probably should not be there. What is the impact of passive exposure. We have heard some about some of hair. What is sweat? What is saliva? What are those. Part of that goes back into your comment, Carl, about what is that cutoff and how is it going to be established, and those are data that are relevant there. Are the results resulted for parent drug or metabolite. Again, we get into the issue of what is the analyte there. If it is parent drug, is that consistent with regulations? Are we going to have to change the guidelines or not. We know that we are talking about cocaine in some cases and not benzoylecgonine in some cases. In some cases it is benzoylecgonine that is in the confirmation tests. Those are minor technicalities but those are issues that we are facing in terms of a regulation and a regulatory kind of overview on this.

Again, item 4 addresses some of the things that Dr. Selavka was mentioning and some of my comments earlier about that window of detection.

What is that particular matrix going to be used for? I think it is certainly consistent, as I said earlier. It may be that some of these matrices are applicable to a particular type of workplace testing. As we look at their applicability in general, we have included these types of questions and these types of elements into this large shape or this matrix.

The interpretation in terms of contamination, internal and external, again raising the issue that Ted Shults raised a little bit. You can call it water loading, excessive hydration and internal contamination if you want to. That is certainly some of the thoughts that put the I in that element over there under the urine lab issue. Alternative medical explanations for the presence of the analyte. We are always going to see those. Again, the internet is replete with a variety of products and the herbal medicines is in its growth, in its particular issue there. So, that is a challenge and that is part of the challenge that the interpretation of the results has to face. Then the source of the specimen certainly could be overcome. It is possible, even though we did not have a lot of data on this. So, we put three Ps in the elements in saliva, sweat and hair on that.

So, that is where the board, that is where we were a week ago, I guess, when we had our final conference call, when we had our final submission of these things back and forth by fax, and where we kind of started this meeting after the review of all the literature that Carl summarized for us, and all the literature that has been supplied by many of you in reviewing the comments and the transcripts of the April meeting, and the deliberation in August that we had in a working meeting and subsequent deliberations on the phone and across e mail and fax. Questions or comments or anything are welcome at this time.

MR. FORTNER: Neal Fortner with PharmChem Laboratories. I might not have understood one of the sections in number 4, pursuant to on-site testing and the blank in that section relative to urine based testing.

DR. JONES: On that last element there, that last overhead?

MR. FORTNER: Yes, sir.

DR. JONES: Let's see if I can get my colleagues on the panel and the board to explain it.

MR. FORTNER: Pursuant to a question of inference of time of use, differentiating on-site versus the laboratory based?

DR. JONES: I think that is another typo. Yes, I think that should not be there.

MR. FORTNER: I just thought I missed it. Second, as a summary comment, we provided a lot of information as did all the other alternate matrices for the April hearings and as subsequent follow up questions are requested of us. I certainly commend the board for going through and absorbing all that information, which is quite a lot. However, at this point in time you are asking for even more. Do I understand that to be correct? If so, how does the board want this facilitated? Do you want this run through the original coordinators? Certainly there is evidence there are many other people involved in the different matrices other than just those individuals who were originally approached by HHS to coordinate the presentations for the alternate matrices.

So, where do we go from here? If you had a lot of information initially, now you are bound to get volumes and volumes of subsequent information because now your questions are more specific. As the original presentations were simply overviews of, this is what an alternate matrix could or could not offer to a workplace program.

DR. JONES: I think where we are and where I stand today is, that some of the presentations today and tomorrow will address some of these specific issues that we had identified as requesting additional information. I think where we are as a board is, we are going to hopefully say to Dr. Autry that here is where we feel that additional data is warranted and it would probably be out of his office or that organization, that such a call for additional data to address those particular issues would come. It would certainly be my hope -- and I think I speak for the board -- that within a matter of a couple of months, if not maybe even sooner, there will be a finalization of this document as it exists today submitted to his office. I think that is consistent with the board's feelings. Richard?

MR. PINDER: A question about a much more specific. If the information you provide to us can be addressed to the specific issues listed, it would be very, very helpful. If you submit a publication, could you outline that portion of the publication that you feel addresses the specific issue. The material that was given to us before was voluminous and very difficult to go through and identify what in that material was helpful and what was not. It would be very helpful if you, in your responses, would address these issues in a very specific manner.

MR. FORTNER: I certainly appreciate that. My comment was, if you think the original material that was submitted to you was voluminous, you haven't seen the rest of it. I would also suggest that as the questions become more and more specific and very technically oriented, if the board would consider, after receiving this second plethora of data, to actually sit down and have question and answer discussions. While it might be possible for the entire board to independently read, digest and understand all the different questions you have asked of all the different matrices, that is a very overwhelming task.

Certainly I believe that you could get support from the alternate matrices to help facilitate your endeavor in understanding the different questions that you are asking.

DR. JONES: Thank you for recognizing the task that we have, and I want to acknowledge the fact that all the board have other full time jobs and are not exclusively dedicated to this particular task. I thank them for going through all the data that have been submitted to this point in time.

Other comments or questions? Thanks. Dr. Autry.

DR. AUTRY: Thanks, Skip. I want to thank the board for work that they have done up to this point in time. Quite literally, there is no way that HHS could pay them for the amount of time that they have put into this endeavor. I also want to thank the people who have been presenters and have sent in information. There is no way the board could have gotten that information without your help. I think what you see here is that as the board has moved along in its deliberations, we have moved into the nitty gritty scientific issues and administrative issues that are still outstanding in order to make determinations about where we stand with these complementary technologies and specimens.

What we need at this point is just what Dick was saying, and that is very specific information summarized to address the specific questions that were put out to the group at the end of the April meeting. A lot of what we have has helped in that deliberation, but it has also left a number of questions unanswered. Some of these are probably fairly easily answered, but we haven't seen proposals. Similarly, there are some that are probably not answerable because we have not had a good critique of the scientific data that is still out there and had that summarized for us. We encourage all of you to please do that, to help us in our deliberations. Some of this we will have answered by presentations that the board has requested over the next day and a half here. They have very specific questions. They knew that there were some definitive studies out there and they have asked people to come and share that information with us.

In other cases, we know that people have tried to set up QC programs and they have run into problems. We have asked them to come and share that information with us, to try to fill in some of the gaps. When I met with the board after the first meeting to give them my instructions, or what my fantasy was at that time, for their working group, I said that ideally what we would like to come out from this, or what I would like to come out from this is a grid that lays out all the principles and criteria along one axis, all the testing technologies along the other axis, and then to simply fill it in.

It says meet or doesn't meet or can't tell. Where it doesn't meet or can't tell, give me a statement about what the industry needs to do to fill in the gaps for the can't tell or doesn't meet. That has not been possible to this point. But I think what you see and what the board has done is to lay out a grid to make that possible. The charge to you is to fill in what needs to be done for the doesn't meet or can't tell. So, that is going to be part of the charge of what we do over the next day and a half and part of the charge to all of you after the end of this meeting.

Again, I want to commend the board and thank them. I am painfully aware that they have other full time jobs and that this is only a very small part of what they have to do to make a living. I can't tell you how small it is in terms of what they have to do to make a living. But I think they deserve kudos for getting us to this point and I think they have given you at least an outline or a map about where we need to go to fill in the rest of the questions.

With that, why don't we break for lunch.

(Whereupon, at 11:45 a.m. the meeting was recessed, to reconvene at 1:00 p.m., that same day.)

AFTERNOON SESSION (1:00 p.m.)

DR. BUSH: Good afternoon. Let's convene our afternoon session. Our first presenter this afternoon is Bob Fogerson.

Mr. Fogerson joined PharmChem in 1975 as a laboratory analyst and has served as laboratory supervisor, laboratory manager, director of quality assurance and director of laboratory operations. Mr. Fogerson has more than 20 years experience in forensic toxicology. He is the designated alternate responsible person for PharmChem's forensic drug testing laboratory certifications and has acted as a certified Department of Defense laboratory inspector. He is a member of the American Association of Clinical Chemistry and a member and past president of the California Association of Toxicologists, the Society of Forensic Toxicologists and the International Association of Forensic Toxicologists. Mr. Fogerson is responsible for all technical aspects of laboratory operations. He directs laboratory methods development, including methods for the PharmChek sweat patch. He also provides in-service training on drugs of abuse for PharmChem customers and testifies as an expert witness in legal proceedings regarding laboratory test results.

Agenda Item: U.S. Probation/PharmChek Drugs of Abuse Patch Pilot Program.

MR. FOGERSON: Thanks, Donna. I also wanted to extend thanks to the board for giving us the opportunity to present a little bit more data on sweat testing. This is a little bit more data than we were able to present in April because it is actually relatively new data and most of it has been generated in the months following that last meeting. What I have been asked to do is present some information on our most recent rather large-scale field use evaluation of sweat testing and its comparison with urine testing in an existing program. That is the title of this particular presentation. We did a project with the Probation Division of the United States Courts to compare drug detection with their ongoing urine testing program, to see if drug detection could be achieved with a sweat testing alternative. Before I begin, I do need to extend some special thanks to an awful lot of people. Aaron Lucas with the Administrative Office of the U.S. Courts was instrumental in putting this entire project together. The other people listed on this slide were people in districts throughout the country who did an awful lot of work to get it organized. They began to recognize and appreciate some of the differences that I think the board members are having to think about from the other end of the process, when you do a process like this. Everybody has to figure out what they have to do differently in order to do their testing in this proposed new manner. I want to talk about this last set of data. I need to take a few steps back, I think, for you to really understand what the significance of the data are. I am going to back up a little bit, talk to you about the technology we are working with here some, and then talk to you about two previous sets of studies that form the background for this most previous one.

First of all, I am going to be discussing what we call the PharmChek monitoring and analysis system. We talk about it in those terms because the system or the PharmChek device is a device to collect sweat over a period of time. It is coupled with a particular analytical system, a particular set of testing methods and testing cutoffs. It is really the combination of those things that drive the performance that you can get with this particular system or with any alternative sweat testing system that you might come up with. This device and the system is an FDA-cleared system. That is one of the questions that was raised this morning. I will try, as I go through this, to try to answer some of the questions or the points that were posed this morning in the earlier discussion.

In the course of the clearance process, we had the opportunity or requirement, depending on how you look at it, to generate an awful lot of clinical trial data. We also did a lot of what we call field use evaluations. The distinction I am trying to draw here, in clinical trial data we are referring to controlled studies, typically dose administration studies. The field use data are simply comparative data. What does it look like. We take the same population and test them two different ways.

I am going to give you a few pictures here, because this thing is going to get horribly quantitative later in the presentation and there are a lot of numbers to look at.

This is a picture of the device itself and how it is typically worn. This gives you an idea of the scale, the size of the physical device, and that is the sweat patch collection device. Construction of the device is fairly straightforward. There is an absorption pad in the center of the device which is actually the material that will collect and retain sweat and the various components that are found in sweat. There is a polyurethane film which serves to hold this collection pad against the skin in a secure fashion. The device is numbered from a chain of custody identification with each individual device.

There is a release liner that isolates the absorption pad from the protective membrane. You can see that down here. It is designed to function in a couple of critical ways. If you think of the skin surface being on this side of the device and the environment being out here, this device is intended to permit water vapor passing through the skin to pass through the polyurethane film in a fairly unimpeded manner. It will to allow us to wear this for an extended period of time without affecting skin physiology too badly. It will allow drugs that may be carried in the sweat that is being produced to be absorbed here, but to prevent environmental contaminants from reaching the same absorption pad that we are using to collect drugs. This is accomplished through some of the physical chemical characteristics of this film. We talked about this in the past at length. So, I won't spend a lot of time on that. The polyurethane film, in addition to protecting from exposure out here, keeps this absorption pad in place by adhering very, very tightly to the skin surface.

You will see in this diagram the adhesive on this film actually kind of interpolates the exfoliated cells on the outer surface of the skin to provide a very tight bond and to keep contaminants from leaking in from the side. In the clearance processes that we have gone through, we have developed data and presented data to support the use of sweat testing with this particular device and technique for the classic NIDA-5 compounds. We have been involved in doing trials for all of these things over the period of time since we began the clearance process. A core part of that clearance process for us involved an awful lot of controlled dose administration studies. That is relatively unusual, at least in the outset of drug testing technology now. We were categorized as one of what the FDA refers to now as tier 3 510 k applications.

So, the foundation of our application is actually classic controlled dose administration studies using human volunteers in multiple research institute environments. That is something that has been done with urine testing subsequent to the outset. But all of us here have been working with urine drug testing for a long time, or had been, before that happened.

So, I think one of the comments I would make to the board to keep in mind is that we can go quite a ways down the road with less complete information than we have from having been on the road for 20 years or so. From these controlled dose studies we were able to establish a number of important things about sweat testing. One is, we know the metabolic profile of sweat, the analytes we expect to encounter when we test sweat.

Others include the time course of excretion, how long after a dose does one expect a drug to be detectable in sweat, minimum wear period of a device like the sweat patch. How long do I have to keep it on after administration of a dose to have a reasonable expectation of detection of the drug. There is what we call the signal storage and accumulation function. This is, after all, a collection device. It is intended to collect and hold the signal over the entire period it is warm. It is like the 24 hour urine taken to the extreme, the 160 or 170 hour urine or whatever it might turn out to be.

Very quickly, what we find for the drugs that we have been working with, these are the metabolites that we see and I have listed them in sort of descending order of concentration. We typically see parent drug, cocaine, heroin very significantly, methamphetamine, and THC. We see lesser amounts in the case of cocaine use, of ethylene methyl ester and benzoylecgonine, and with heroin lesser amounts of 6-acetyl morphine and morphine. So, this is a different profile than you see in urine. It does raise an issue when we have regulations to specify analytes. But this is what you would need to do if you want to effectively test sweat. Critical time frames that tended to emerge from the studies that we did, from the doses we were using -- they were modest doses, and these are doses you could give to volunteers in a research institute -- you could detect the drug coming out in sweat for two to three days after administration. So, the time frame isn't too terribly different from urine. We seemed to have to have the patch on for about 24 hours after dose to have a pretty good expectation of picking up a positive. These are some sort of basic numbers to keep in mind. The signal accumulation function that we referred to, we tried to sort of graphically show it here. If you administer cocaine at this time and begin to collect urine samples over time, we are graphing here the percentage of samples that are positive versus time. You see you get an increase up to virtually 100 percent of the urine samples at about day one are positive. Then as you collect urines later on, they become negative; no great surprise. The patch on the other hand is capturing drug. It has got a time course somewhat similar to urine in terms of percentage increase, but then it holds it. That is what it is designed to do. Along with these clinical trials we did some relatively small scale field use evaluations in the early days. This would have been about three or four years ago. What we wanted to do was work with some agencies conducting routine drug testing to look at how well this device worked in a real semi-controlled population. These typically are people in criminal justice agencies on probation or parole. There is some degree of control but it is not what you get in a hospital and it is not what you get in a clinical research institute. We wanted to look at clinical sensitivity relative to urine testing, was kind of the bottom line question for us to understand.

The scheme that we used for the early studies was this. We had an agency that was involved in routine urine collection. They simply continued that practice. Sweat patches were applied and were removed typically at the time a regular urine was going to be collected. All the urines, of course, were being analyzed for drug content. That continued to happen. Every time we had a urine that showed a positive drug result, we would analyze the patch associated with that urine. Ten percent of the patches associated with negative urines were also analyzed. Now this is an important thing to remember.

I will come back to it and you will understand more in a bit when we see some numbers. Our thinking at this time was to verify that we had a technique that was at least as clinically sensitive as urine testing. We wanted to make sure if we got a positive in the urine the patch would be positive, or at least find out if it would be. This was a bit of a control. This was the focus of the study and not the negatives.

To summarize some of these data, and this slide actually came with a presentation that we did for FDA some time ago, for the three drugs we were looking at at that time, we had the number of subjects, number of urine positives for cocaine, number of patch positives; similarly for opiates and amphetamines. For the most part, they looked about a wash. Now they are sort of equivalently sensitive. Maybe amphetamines, quite a bit better. We didn't have a lot of data. From our point of view at that time, from a sort of filing perspective, that wasn't unsatisfactory data. We looked at it in a little more detail and looked at the number of cases where we had a urine positive and a patch negative or a urine negative and a patch positive, and again, there were discordant cases, but they seemed to be about a wash, except for over here.

On the face of it, this data says, well, this technique is roughly equivalent to urine testing. You have got to remember that the number of urine negative and patch positive cases here came from only 10 percent of the urine negative cases.

So, it raised the question that we hadn't really focused on in the beginning, how many more drug use episodes were there in the 90 percent of the urine negatives that we hadn't checked because we hadn't done the analysis. That led us on to do a different design in our next really large scale evaluation of sweat testing compared to urine testing. Here we had a large agency. We did parallel testing of urine and sweat over a six-month period.

We were specifically trying to correlate our results to the best of our ability between all the urines and all the sweat samples that we collected over this period of time. It was quite sizeable. You see over a thousand subjects in that period of time, 2,900 or so sweat samples, and about 10,000 urine samples. In this case, we analyzed all the samples for all the compounds of interest. What we wanted to understand was whether or not urine and sweat tests gave us the same assessment of the drug use status of the people being measured over the period of time. Were they the same or were they different.

It raised a couple of methodological points. One of them was, how to decide what to compare. You will see some of the issues when we look at some of the numbers. One of the questions that comes up, if you think back to that signal accumulation slides, we are typically collecting a lot more urine samples, if you think back to the numbers. We had about four or five times as many urine samples as patch samples. So, we wanted to compare urines and patches one on one and how do we connect them. What we came up with was a scheme that looked like this. We have a time line where we put a patch on and remove the patch. That could be seven days. It was up to 14 days in some cases. And for the purposes of our analysis, we looked at each patch and looked at that time frame. Then we extended that time frame to two days prior to patch application and two days subsequent to patch application and said that any urine collected in here could be associated with the same drug use episode that might be reflected in the patch. So, when we did concordance analysis, we were matching individual patches with any urine sample collected over the time frame that patch was worn. This is where we get quantitative. But the numbers are important because there is a consistency here that I think is important.

In a real simple fashion, we have our 2,900 patches, our 10,000 urines. We see quite a few more cocaine patches than urines, very similar number of opiates. Again, there are more amphetamines in sweat than in urine, but not a lot of data, and quite a different pattern with marijuana, quite a few more urine positives than patch positives. This particular population didn't show us a lot of PCP use, so it wasn't terribly helpful for that particular analyte. Overall, there was an apparently significantly increased number of positive specimens.

We wanted to normalize that a little bit because if you are running a probation program or pre-employment program, what you really care about is if the individual you are working with is a drug user or not. Much less are you concerned about whether this sample on a given day was positive. Sometimes you are. So, we normalized to the subject as opposed to the test specimen; same data, but now we have the same number of subjects for both urine and patch. So, these subject numbers become a little more comparable. If you look at the data that way, you get about four times as many cocaine users identified with the sweat testing system as with urine, similar numbers with the opiates, amphetamines we saw before, and still a discrepancy in the other direction with marijuana. There was about a 50 percent increase in the number of individuals identified as drug users with the sweat system. So, from this data already we had our second rendition at quite a high level, showing us that we are going to have to develop a way of understanding and interpreting cases in which we have sweat results that are different from urine results. It was already pretty clear to us that it would be not unusual at all for sweat results to be different from urine results. We took the data down one more level of detail and then began to try to understand, on a sweat patch by sweat patch basis, how the results correlated.

I will go through this fairly quickly. You have all the numbers, I think, in your handouts. They can be looked at later on. This is pretty classic concordance analysis. This is based on that time line I described before.

These are urine positive results, negative, patch positive and negative. Overall, 28 cases where we had a positive patch that was matched by a urine, 128 cases of a positive patch where there was no urine collected over the time frame that was positive for the same drug. There were 59 cases in the other direction, where we had a urine positive not matched by a patch positive. Now this again, on this side, is by the test specimen, over here by the test subject. In some ways I think it is easier to think about it in terms of subjects. You can use percentages a little more cleanly. Here we have picked up almost twice as many users with patch testing as with urine testing.

PARTICIPANT: Do you have some kind of threshold on the urine tests?

MR. FOGERSON: Yes, there is a slide on that when we get to the next study which is the same as the threshold we are using here. If you give me just a moment, I will get to that. The slide will describe the federal probation study and this study as well. If we look at it in detail by drug, you will see that the discrepancies become very pronounced in the case of cocaine. Again, we see quite a lot more cocaine users through sweat testing than we are able to identify through sweat testing. It is interesting, because these patches were worn seven to 14 days. In many cases we are looking at four, five and six urines collected over that period of time. It still didn't show up.

Opiates, again we see more of a balance in terms of the detection capability. We see discrepant results, but it goes both ways.

One of the interesting things about this particular study is if you look at these 21 individuals here who were detected as opiate positive subjects, we could identify 14 of them as heroin users in sweat testing because of the fact that we do detect parent drug, and the acetyl morphine as predominant species.

The urine data on these guys would not have been suggestive of heroin users, certainly not conclusive. They tended to have fair to moderately low amounts of morphine.

THC, again, slightly more detail. The pattern is the same. Here we have it in the other direction.

The urine tests that we were using were more successful in identifying users and we had more positive specimens.

We concluded from this particular study that the testing system that we were using with sweat was significantly more effective in identifying cocaine users in this population, somewhat similar in identifying opiate users in a quantitative sense, but much more effective in differentiating heroin from other opiate use, and that the urine tests were more effective in detecting marijuana use.

Up to this point, the data we have got existed in April and we presented some of this to the board in April. One point I did want to bring up, because I am not sure we mentioned it before. Skip mentioned this morning that the federal program is conceived as a deterrence program. There was a little bit of information that came out of this last study that I actually found sort of interesting and we have not had the opportunity to follow it up in detail. I did the analysis because the population was big enough and the time frame was long enough that it made sense to do it. We wanted to test the theory that some people had that the visibility of a device like the sweat detection system, in and of itself, had a deterrent capability. So, we took the population being tested here. There was a general population in this probation system and then there was a subset of people who actually participated in the sweat and urine collection. Only a subset was involved in that. There was a period of time before the pilot, during the pilot and after the pilot. If you looked at what went on, these are the guys that were providing sweat samples and urine samples. They had a three percent positive rate by urine test. For about six to nine months prior to the pilot, it dropped to one percent while they were being tested by urine again, and all these numbers are urine positive rates to normalize the efficiency of the method. It went back to three once the program was over.

That wasn't a real shock. This is the funny one. These are the guys who weren't given any sweat samples and they went from three to two to three. This is an area of research that I think is still interesting and would bear some work in the future.

Now, on to the subject of this particular talk, at long last. We wanted to extend the work we had done with this large-scale pilot and pick up a couple of other dimensions to it. One of the things we wanted to do was to look at some different populations to try to get some other patterns of drug use. Up to this point we had virtually no use of PCP, not much amphetamine use to look at, cocaine use, marijuana use. So, we wanted to look at new, geographically distinct populations. The federal probation system is an ideal environment to do this sort of work. For those of you who are not familiar with it, I will give you an idea of who they are and what they are.

The Administrative Office of the U.S. courts is responsible for the probation and parole supervision of federal offenders. This could be people who are under supervision for drug use history, it could be a number of other reasons. But the people in their charge are being supervised for some sort of federal offense. This is a big testing program. These people do about 750,000 drug tests a year. For about the past year and a half they do their testing in a method that is familiar to all of us. From a urine testing methodology standpoint and from a volume standpoint and from a geographical distribution standpoint, this is a very good sort of test model for looking at what it might be like to do sweat versus urine testing in a workplace environment.

They also have people who are real, real skilled and real, real practiced at evading tests. You get to check all kinds of things in that regard.

These are the objectives of this pilot program as we set it out: to compare the results of drug testing using conventional urinalysis to a sweat testing technique;

To specifically determine the number of subjects we could identify as drug users with one technique versus the other; sort of catch all at the end, to gather whatever information came out of that, that might be useful to the agency in determining how and where and why they might want to use sweat testing as opposed to urine testing, in addition to urine testing, or vice versa.

The methodology on this particular study, patches were applied, again, for a seven to 14-day period and were left at the discretion of the participating district. Urines were collected at the time of patch application and removal. All the specimens, patches and urine, were analyzed for all compounds. We correlated the results of each sweat sample with the associated urine samples. Here we had 396 subjects in this particular study and 875 sweat samples, 782 urine specimens. You will notice already we have a much different relationship between the number of urine samples and sweat samples. This gives a slightly tougher target, in some respects, for the urine program to hit, based on the data we have prior. The geographical distribution of this study, you can see we had samples coming from all over the place, South Carolina, Texas, Michigan, Louisiana, California, North Dakota, Nebraska. We tried to have a significant number of the total samples come from all the sites. This is a bit hard to read here, but this goes back to the earlier question about the analytical methodology.

This methodology here was fundamentally the same as what we used in the previous study. We have got our sweat testing methods here. All the sweat samples are screened by an enzyme immunoassay, the microtiter plate technology. Cutoffs is at 10 nanograms per ml for opiates, cocaine, amphetamines, 1.5 for marijuana, 7.5 for PCP. Positives are confirmed by GC/MS at the cutoffs listed. Then the urine testing methodologies you can see over here, emit screens, GC/MS confirmations, with the cutoffs listed. These cutoffs are either similar to, or slightly lower, to the DHHS cutoffs. So, again, urine testing sensitivity is, if anything, slightly higher than what you would expect to get in a DHHS program. I will comment that, because of some discussion this morning, these cutoffs here, these immunoassay cutoffs and consequently the associated GC/MS cutoffs, were in fact, developed for pharmacologic relevance.

They were developed to be able to offer optimum clinical sensitivity and specificity in the samples produced by people administered known dose of drug. These weren't done just because they could be done. They were done to give us the best performance we could get, given the pool of samples in our clinical trials. Here is what we wanted to know again. Do urine and sweat test results give us the same or different information when tested in the same population over the same period of time.

Another set of numbers to kind of wade through but, again, I think it is important because you are going to see a very similar pattern. If we analyze the results by specimens, we have 109 allegedly cocaine-positive patches, 24 urine positives. Opiates, we have a different pattern here, far fewer patches than urine specimens. I will make another comment on that in another slide. More amphetamine data now, but again, a very much higher detection in sweat than in urine. The amphetamine urinalysis here is one that is somewhat more sensitive than the standard, using a 250 nanogram per ml cutoff in the confirmation assay. A similar pattern with marijuana, lower detection with the patch than with urine. Again, no PCP that we can really compare.

If we analyze by subject, about three times as many cocaine users detected in sweat, opiate patterns the same. Nothing much changes here. But that is not surprising because the discrepancy between the number of subjects and the number of samples collected is not as great as in the previous study.

The overall correlation in this study, we have 22 instances -- let's just call them subjects. It is a little bit easier to follow these numbers. There are 17 individuals in which we would detect them as positives by either urine or sweat, 66 individuals positive by sweat not detected by urine, and 19 cases in the other direction. If you go look at those in detail, you will see that an awful lot of the individual discrepancies come from cocaine users, 36 versus 5 in this particular case. As we cut down the frequency of urine testing, the gap that we had before between urine and sweat cocaine detection improved dramatically. If you remember back, the number was about three to four times as many individuals detected in the previous study. Now we are up to eight or nine times as many.

I think it is a function of the fact that we didn't have as many urine samples collected here over the same time frame.

Opiates, here we had a different result than we did in the previous study, or at least an apparently different result. You see we have quite a few more individuals detected in their urine test than in their patch test. When we look in detail at this data, one of the things that was quite different is that in these patch positives, we didn't have anybody that looked like a heroin user.

Everybody in both urine and patch data showed a profile you would expect of codeine use, large amounts of codeine, small amounts of morphine.

We knew from our clinical trial data that sweat is not as sensitive in detecting codeine use as it is in detecting heroin use. We just sort of came up with that inadvertently but the relative levels don't seem to be as favorable. That may have had something to do with this particular finding.

Amphetamines were quite a bit more dramatic. Here almost everybody we are detecting is with sweat. Isn't that an interesting finding for us now? There are some other agencies using sweat testing in a limited fashion in areas of considerable methamphetamine use. This is kind of what they are finding, lots and lots of instances of sweat amphetamine positives that are not being picked up in urine testing.

THC, again, the same pattern. Here the direction is not as favorable. So, detection of marijuana with the methods we are using is not as favorable as it is in urine.

Conclusions from this particular study, very similar. We have a significantly greater incidence of individuals detected as drug users with sweat analysis than we were able to get with urine testing. It is most dramatic with cocaine users and with amphetamine users, in this particular case. We began to get some significant data from amphetamine. The urine tests, again, were more effective in marijuana detection and in this group of opiate users. I think the opiate data may be affected by the fact that we were dealing with codeine use here more than heroin use. So, that is the outcome of that particular study. If we have any time, I would be happy to take any questions, if that is appropriate. Thank you.

(Applause.)

MR. TRACHTENBERG: Do you have any data as to whether the patch negative urine positive were codeine ingestors as opposed to poppy seed ingestors?

MR. FOGERSON: The profile that you saw would not suggest poppy seed. You would see codeine in an excess of morphine in a ratio of six or eight, ten to one. I wouldn't suspect poppy seed. It wasn't like low level morphine. These were very significant levels of codeine and much smaller levels of morphine. That is what led me to suspect that we were looking at codeine use. We actually have a case -- an individual case, now, that would be sort of interesting. Out of this same population with an individual who has been tested for urine for a long time and looks like a codeine user, that profile, very large amounts of codeine, much smaller amounts of morphine. Now, they happened to put a sweat patch on this guy and he shows lots of codeine, he shows some morphine and he also shows lots of heroin and 6 acetyl morphine. This may be an instance in which we have got somebody who has learned the trick of maintaining a codeine use level to mask heroin use. People do know how to do this.

DR. WALSH: I know I have asked you this question before, Bob. I know you have a lot more experience with the real world. How well are you doing with the adhesive on these patches. Can you give us some sense of the patches that come off for one reason or another during the 14-day periods?

MR. FOGERSON: Yes, we have got a lot more experience with that. One of the reasons we wanted the geographical distribution was to test that in a lot of conditions and environments. We tried to select districts that give us different weather. That turns out to be related to it. If we look at the history of this particular device and the technology behind it, it is affected by the weather and it is affected by skin type to some extent. I think the effect of skin type is probably much less than weather and that sort of thing. The adhesion rates would range from, in bad cases -- I am trying to remember numbers -- we might have had cases or 25 percent of them falling off in a relatively short period of time up to 95 percent adhering. When they typically fell off, they fell off quickly. In this particular study it is a difficult thing to assess. Although at least in the initial study everybody was told that no action would be taken on the grounds of the results, and that was in fact the case, no actions were taken, people were convinced that they were being injected with all sorts of compounds and there were a number of lawsuits filed against the agencies claiming they were being poisoned. It is difficult to tell how they are being adhered as opposed to being torn off. That is the only thing we struggle with. The end of your question is, there are adhesion issues that haven't been sorted out for very hot, humid environments in the summer time.

DR. WALSH: I wonder about people who are involved in heavy levels of exercise. Are there application procedures that you can --

MR. FOGERSON: Yes, we have developed a couple of techniques to deal with that. We have done a lot of experiments with various application procedures. Another thing that we do which seems to work pretty well is to actually extend the adhesive surface area. We developed this device we call window frame. We can actually put a patch on somebody and then you put a device that is like a bigger patch with a hole cut out in the middle. So, you sort of build a buffer of adhesion around it. So, if the adhesion is going to erode around the edges, which is what it tends to do, you build a longer barrier, if you will. That has been pretty effective, even in the hot, humid environment.

DR. WALSH: Did some of these people you have to drop from the study because it consistently just would not stick because of their particular skin type or environment?

MR. FOGERSON: Not very many. There was a study we did with a street population in New York, which never really got off the ground. Part of it was that it couldn't get off the ground because we couldn't get that to stick. The descriptions of the personal hygiene there made it easy to understand why.

MR. HOLLAND: One of the concerns with urine collections was privacy issues, intrusiveness. Of course the argument was, you have to provide unobserved urine donations throughout the day anyway, so it wasn't really that intrusive. A patch on your arm or elsewhere over seven to 14 days, it seems like it would be somewhat more intrusive. I am wondering about your thoughts on that.

MR. FOGERSON: It is clearly more obvious. People talk about the intrusiveness from lots of different directions. If you want to talk about it in the sense of people know you are being tested, I would have to agree. It is obvious you are being tested. One of the reasons we selected the upper arm, for most people for most of the time, if they choose not to have it known that they are being tested, they wear a shirt. Most people do anyway. So, in that particular case, unless you disrobe, it is not obvious. In fact, all the sites that we worked with, the arm, the torso and the back, would normally be covered. In Key West in the summertime, I will give you that. It is different.

MR. TOWT: I have a question regarding sensitivities. Obviously you are going for much more sensitivity with the sweat patch as opposed to urine. If you were to take the urine test down in sensitivity and increase its sensitivity down toward the levels that you are going with the sweat, I was wondering how it would do versus sweat. To address that, did you test any of the urine negatives, sweat positives by GC mass spec, the urines from those individuals? Did they have any detectable drug by GC mass spec?

MR. FOGERSON: In this federal study we didn't do that and we didn't analyze the data that way. In a previous study we did, to a fair extent. I actually have some slides that I thought about bringing and didn't. To answer your question, if you were to use lower cutoffs with urine testing, you can reduce the discrepancy that we saw, but it doesn't go away. If you look at the most dramatic one, the cocaine one, I could reduce the discrepancy by about 30 percent if I went to sort of limit of detection analysis of the urines. An awful lot of cases with these urine samples, we had no evidence of drug at all.

MR. TOWT: That was by the screen or GC/MS?

MR. FOGERSON: Most cases by the screen, some cases by GC/MS as well. In a lot of cases, I suspect that they knew, but in lab tests to this point. Beyond that, the amphetamine cases, I am not sure. We have a very, very consistent pattern with amphetamines from a half dozen or so different agencies. These cases are going to court. So, we are pulling these samples out and are doing retests. We cannot find the drug in the urine but it has been detected in sweat.

DR. SELAVKA: I am wondering about your interest in the deterrence slide. I am wondering if that is because you see an enhancement in price/performance ratio by doing appropriate sampling techniques on the recovered patches, meaning you don't have to test them all. You can get the same deterrent effect without it.

MR. FOGERSON: That question has been raised. I think it got discussed about a year ago at an NIJ meeting from about three different perspectives. I actually did this analysis because at the time we were doing this study, I was talking to a couple of people who were involved in drug testing from the rehabilitation and treatment side rather than the enforcement side. They were curious about the ability of the patch or something like that to support a rehabilitative deterrence. We thought, gee, let's go see if there was any evidence of that. That issue has been raised, but it wasn't the motivating factor for that particular analysis.

DR. SELAVKA: You said you were interested. If that is not your interest, what is?

MR. FOGERSON: If it in fact is an effective deterrent, one of the points that I make to people every once in a while is that in any drug testing system you have to be concerned about detecting positive results. The test has to be effective. Nobody in the business is really in the business of detecting positives. What they really want to do is preclude positives. They want to be sure they are really precluding positives and not having their tests beaten. So, from a treatment standpoint, from the probation supervision standpoint, they don't really want a lot of positives. They want them to be true negatives. If the device is able to contribute to that, in and of itself, that gives it a value, I think, as a testing technology that stands above the alternative technology.

DR. JONES: On one of your earlier slides you showed, I believe you called it an accumulation profile. I don't recall the terminology.

MR. FOGERSON: Right.

DR. JONES: Does that represent drug in the patch or does that represent continual drug being eliminated? The genesis of that question comes from, we know that many of these compounds are available for absorption via subcutaneous routes. So, my question to you is, in your patch design, is drug retained there or are we looking at a continuous process as we look at that protracted curve.

MR. FOGERSON: That particular slide was really designed to just graphically demonstrate this signal storage function that I mentioned. Basically the drug is being excreted in the sweat, being collected on the patch and then being retained. The real point there is that with urine testing, because of the window of detection, you have to time your collections fairly carefully to be effective. With the patch that eases a bit because it can be worn over a long period of time. As long as it retains the compound, you will be able to pull the patch off at any given time and detect use over the wear period. That is obviously limited by diffusion, if that occurs. We don't have a lot of really good data on that. I suspect that it might be a factor with marijuana. I doubt it is as much a factor with amphetamine and cocaine.

DR. JONES: Cocaine, of course, is one that really shows a significant percutaneous absorption. Have you done any experiments when patches have been so-called laced or spiked, and placed on there and looked at drug retention on the patch over time?

MR. FOGERSON: Yes, it is very good. The drug is stable on the patch in a number of different environments. From that standpoint, it is not so much a problem. There are some studies that I think Pascal Kentz did in Europe looking at a couple of other compounds, phenobarbital and some of the other benzodiazapines. He finds far lower levels on the patch than you would expect given the dosages used, if there were none lost back into the skin. His data suggest that might occur with some species.

Anything else? Thanks.

(Applause.)

DR. BUSH: Thank you, Bob, for that enlightening presentation. Now we have another Bob, Dr. Robert Willette.

Dr. Willette formed Duo Research in 1981 for the primary purpose of providing consulting services to employers in the public and private sectors on the design and implementation of workplace alcohol and drug programs. Dr. Willette has helped to establish policies, procedures, laboratory inspections, and blind quality control programs for the U.S. navy, administrative office of the U.S. Courts, the Federal Bureau of Prisons, the Drug Enforcement Administration, the Federal Bureau of Investigation, and numerous private companies. As Chief of the Research Technology Branch of the National Institute on Drug Abuse, Dr. Willette was responsible for overseeing the Federal Proficiency Testing Program for drug testing laboratories, conducted by the Centers for Disease Control, from 1973 to 1981. At NIDA, Dr. Willette was also responsible for multiple studies on the disposition of drugs in the body and their detection in biological specimens.

In addition to these specific efforts directed toward quality control programs, Dr. Willette is serving or has served as a consultant on drug testing programs to the White House Office of Drug Abuse Policy, U.S. navy, U.S. army, National Institute on Drug Abuse, Federal Aviation Administration, Federal Railroad Administration, National Highway Traffic Safety Administration, U.S. Postal Service, Drug Enforcement Administration, the FBI, U.S. Customs Service, Administrative Office of the U.S. courts, and others. This included the development and administration of the National Laboratory Certification Program for the Department of Health and Human Services, for which he currently serves as a certified inspector. Dr. Willette received a bachelor of science degree in pharmacy from Ferris Institute, now Ferris State University, in 1955 and a doctor of philosophy degree in medicinal chemistry from the University of Minnesota in 1960. He has held research appointments at the Upjohn Company, University of Michigan, Australian National University, and the Commonwealth and Scientific Research Organization of Australia. He has served on the faculties of the Schools of Pharmacy at Ferris State University and the University of Connecticut. Dr. Willette, if you will, please.

Agenda Item: Administrative Office of the U.S. Courts/An Evaluation of Non-Instrumented Drug Tests.

DR. WILLETTE: Thank you, Donna. I want to thank the board and the Division of Workplace Programs at SAMHSA for the invitation to present some of this data, particularly on a study that we conducted, and also to thank the administrative office of the U.S. Courts to present this today. Part of this presentation is to present some data and information on a study that we conducted for the Administrative Office of the U.S. Courts. I want to try to put it in a broader context, to expose the board and the audience to some of the issues surrounding on-site testing. I mentioned alcohol testing, as was mentioned earlier, as a contributing factor that we have federal regulations under the Department of Transportation that permit on-site testing for alcohol. Results are obtained immediately and can be used to take action within the workplace.

One of the aspects of these regulations is that it lays out a rather detailed procedure which we have suggested to people to use as a model on how to conduct on-site drug tests. There are differences. As Yale Caplan pointed out this morning, alcohol is an anomaly in the sense that you can do the screening confirmation in juxtaposition in a short period of time. At the present time, on-site drug tests are only screening tests, where confirmation is usually conducted elsewhere. There are a lot of parallels and a lot of things that can be learned from the alcohol model. For example, in the case of alcohol, the tester is also the collector, whereas in the on-site drug test one could more easily separate those functions, as is one of the issues raised in the board's list of factors. There are a number of devices available for doing on-site alcohol tests, and I don't mean to belabor this. This is one device that can be used not only for saliva but also for urine. So, what has happened in the arena of on-site drug testing? This is certainly not a new and novel topic because on-site drug testing has been around for at least 30 years. In the late 1960s, the military took advantage of doing on-site testing in the Korean War period, and on-site testing has been used throughout the 1970s and the 1980s and into the 1990s in the drug treatment community. This is not a novel concept. There is certainly a lot of information available. I haven't listed on this slide drug treatment. Drug treatment is still a main area where on-site drug testing has great utility.

If one looks at the literature that was generated during the 1970s in particular, there are a lot of publications, perhaps hundreds, that show the benefit of having an instant result to confront the individual with, and the therapeutic effect, the deterrent effect that that had. There are some distinct differences, however, between using on-site tests in the clinical arena, such as in treatment, and in the workplace and criminal justice arena. The major difference is that these two areas, listed on this slide, are exempt from the clinical laboratory improvement amendments of 1988, which impose a whole set of restrictions on on-site testing, particularly for drugs.

What has happened in the last 10 years or less is an expanded interest in doing on-site drug tests. The current thrust, and the thrust for the study that we conducted was really based in the criminal justice arena. As you heard earlier from Bob Fogerson, the Administrative Office of the U.S. Courts has an extensive drug testing program. The figure he gave of 750,000 urine specimens tested under the Federal Pretrial and Probation Contract, was in the central laboratory. It does not take into account the amount of on-site testing that is done in that same jurisdiction. We currently inspect and monitor 36 federal courthouses and a few probation sites where on-site testing is being conducted. So, that 750,000 is really an under-estimate of the magnitude within the federal system alone. There is also a great deal of on-site testing going on at the state level, in terms of arrestees, probation and parole.

What triggered the study that we conducted in October last year was an executive order issued by President Clinton approximately two years ago, directing the Department of Justice and the Administrative Office of the Courts, in collaboration with the people at the Division of Workplace Programs, to look at how the government may go about expanding the testing of federal arrestees to the selected number that are done today, to all federal arrestees.

A committee were formed and a number of issues were addressed and looked at. That led to a project, Operation Drug Test, that is going on at the present time, where the Administrative Office is conducting a study comparing non-instrumented drug tests to instrumented tests in some 28 or 29 locations.

However, beyond the criminal justice arena, we consult for a large number of companies that have had an increased interest in doing on-site drug testing.

As was pointed out this morning, the board in particular, and certainly the Division of Workplace Programs and others in the audience have to bear in mind that most drug testing done in this country is not covered by federal regulations. When you add together criminal justice testing, which is exempt from the HHS guidelines, and look at all the employment testing that is done outside the federally-covered positions, not just within the federal government, but certainly those covered by it, the Department of Transportation, Department of Energy, Nuclear Regulatory Commission, and so forth.

So, the rules, I think, that go on outside tend to track the federal regulations insofar as possible, because that is the standard to which they are going to be held in court and in any kind of disciplinary hearing. So, there is a tremendous shadow effect or halo effect around the federal guidelines. I think the consensus conference that was held in 1989 was an attempt to address the recognition that there really need to be better guidelines and rules for testing outside of the federal regulations. We now have a lot of clients that are interested in, or are doing, on-site testing. There are a number of scenarios where on-site instant results have a very applicable use. Remote locations. We have a company building a factory in South America that are doing pre-employment tests. They can't get specimens to the United States laboratories for testing. They want to screen out the negatives on site and send only the positives on. These two categories are the biggest demands that we have. In the chemical industry, petrochemical, as well as in the nuclear industry, contractors are brought on site to perform maintenance, break downs, construction jobs, in rather dangerous environments.

Typically they are brought into a security or safety center, they are given certain instructions, they are badged and permitted on the premises. Many of those companies collect specimens and send them off to a laboratory. They are on site two or three days, perhaps, before a positive result comes back, which causes all sorts of concerns. One of the things they want to do is to screen them out before they are badged. If they are positive, they are told to return in two days when the confirmation results come back. Not that many companies are hiring like crazy but we have a few clients that are doing massive hiring. They put every applicant through a one-day physical examination which is extremely expensive and consumes a lot of time and resources. What they would like to do, the first thing they do when they walk into the center, is to give them an on-site drug test. If it is positive, they then do not put them through the complete physical, and tell them to come back in two days when they have the confirmation result. So, there are a number of uses that are growing that are really putting the pressure on the development of these assays, and certainly the flood of non-instrumented drug tests that are currently available.

Some of you have a good understanding of how immunoassays work and I only want to just do a very simple graphic review to stress the point that the devices that we are going to be talking about and the kind of devices that people are using are all immunoassays. They are based on the same principle that is used in the central laboratory. For those of you who are not technically grounded, antibodies that have been raised to have specificity toward a particular type of drug molecule binds to those drugs.

A variety of mechanisms are then used to measure the degree of this binding to tell whether a sample has drug present or not, or present at a sufficient concentration to give a positive response. Most but not all of these non-instrumented devices use the same concept, but it is not in solution as it is in the laboratory. The drug that the antibody is designed for is bound to an absorbent strip. These are actually sprayed on in a manner that they stay in this position. The antibody is tagged with a dye. When the urine is applied to a well -- you see this isn't the only configuration, these come in different layouts -- as the urine passes up this absorbent strip, it carries the antibody and dye with it.

If the specimen is negative, the antibody then binds to the bound drug and a line is produced. I will show you some examples of this. If it is positive, and drug molecules are present, they take up the binding sites on the antibodies, so that as it passes over the strip, it doesn't bind and therefore no line appears.

Now, a few of the devices work in a slightly different format, but they are still all immunoassays and they all rely on some kind of end point, which is the appearance of or the lack of appearance of a color.

Back at the end of September when we were commissioned by the Administrative Office of the U.S. Courts to evaluate all non-instrumented drug tests that were currently available or known, this is the list that we were able to compile. There were three instrumented devices available and I have forgotten what the count is here, but several non-instrumented devices. Those that are in white are ones that were included in the study. This one here, the rapid drug screen, unfortunately was on the market but was not available in sufficient quantity, so it was not included. The Emit technology using an ETS instrument was selected as an instrumental reference method, because that was what was being used within the U.S. court system. This is obviously an ETS running EMIT reagents. To show you what some of these devices looked like -- at least all those that were included in the study and some newer ones -- these are single test devices. It happens to be upside down, but it really doesn't matter because in some cases you can't read what is on them anyway.

These are what single test devices look like. We had some choices. In some cases we compared what would be called a strip test as opposed to the dip stick. This is the same device, just in a different format. We had two such choices of those, and these were all the single ones. In other words, each of these has reagents for one single drug. Then we also included as many as possible of multiple drug strips, using the same technology. Instead of having one drug on a band, we had multiple drugs.

Most of these were for five drugs, this one six. I believe this is the five drug panel. Many of these come in different configurations. You can order three drugs, four drugs, five or six, depending on the purpose for which it is being used. Also, we included the four-drug version of the Roche test cup, which is a self-contained system, collection bottle and test strip. The other cup that is available, this Rapid Drug Screen, this is the one, unfortunately, we weren't able to obtain because there were not quantities available. Since we did the study there are at least six or seven or eight new devices, including one which has revived the old name of SYVA.

There is now a SYVA rapid drug test. There are probably another half a dozen that have come out since October.

So, what is the study we conducted? The idea was, in a short period of time -- you know, the way the government works, it is hurry up and wait and you wait and wait and all of a sudden they want something in a month. So, we were given a short time frame to conduct an evaluation of as many of these devices as we could obtain. With one exception, we actually purchased these.The government doesn't allow you to go out and ask for hand outs, but one manufacturer did supply devices gratis. So, we did a side-by-side comparison of all 16, where they were tested at the same time with the same samples.

The design was to take 100 samples, 90 of which were real specimens, that were arriving at the U.S. Court's central laboratory from federal probation sites, and then 10 controls. We looked every day at the initial test results. These were screening results in the central lab using EMIT reagents, with the intention of trying to select a concentrated group of samples around the cutoff. There are a lot of different ways of doing these kinds of studies. With the goal in mind of looking at how well the devices define their cutoffs, we tried to pick approximately 60 percent of the samples within plus or minus 25 percent of the screening cutoffs. I should mention that this is under the U.S. Court's protocol. You may have noticed on Bob Fogerson's slide, that they use the same cutoff in the screen that is used under the HHS workplace rules. However, on confirmations, opiates and amphetamines are one half the HHS cutoff. All the data I am going to show today are under the HHS criteria, but we did have to obviously evaluate the criteria against the court's criteria. Needless to say, more positives were confirmed under the court's criteria using those lower cutoffs. We were able to test all the devices except the test cup using aliquots that were selected each day. They require, at most, five or six drops of urine.

The test cup package insert requires or at least recommends 30 milliliters to have sufficient urine in the cup so that when it is tipped there is enough urine that can enter the test zone and be required. So, the Test Cup, because we didn't have 30 milliliters of urine each day, we had to do this with a separate set of samples with about a 30 percent overlap with the other set, where we could take frozen specimens selected under the same criteria that the original samples were collected under. Then we used the EIA reagents on the ATS as a reference method, because this is what is being used in the federal courts system. The study procedure, we hired from a laboratory temp agency three laboratory technicians, all licensed in California. Two were bachelor degree, one was a master level degree. They were trained in the manufacturer's protocol which we followed religiously. We never made any alterations if we had problems unless the manufacturer had recommended it.

The samples were coded to the test operators. The test operators did know that today there are cocaine samples. The problem is that they were testing multiple test panels along with single test panels on the same day. So, we couldn't test, let's say, 100 specimens on a cocaine strip if we knew that the sample selected were opiates. So, this was in order to try to conserve the number of devices, and the time and everything required. However, they were blinded to the make up of the sample, whether it was a negative, below positive, above positive or a strong positive. We were set up -- because the federal probation contract for central testing is with PharmChem laboratories, PharmChem graciously set us up in a separate facility from the main laboratory, removed from any of the laboratory personnel. Specimens were then retrieved from the main lab, brought over to our location and the three technicians then performed the work. Specimens came in bar coded, so the technicians had no further information as to the nature of the sample. In some cases, devices require timing, so all the protocols were followed as strictly as possible. Now, one of the major issues -- I shouldn't say issue, but a problem or an observation that we made in conducting these, as I pointed out, most devices exhibit a colored line for negatives, and the absents are positive. If you have had my experience, it is much more difficult to see the lack of something than its appearance.

Now, the Triage device is the opposite. It is the only one that really shows a line for a positive and of course the EMIT produces a numerical print-out. I think in this light it is difficult to see some of these, but I am going to show you some of the devices and how the results appear. One thing about an instrumented device is that you have got a number that you can directly compare to another number. There are no borderline results; it is either positive or negative. If the cutoff -- if the sample gave a result of 716, it is called negative.

Now, this is one of the devices. I think most of you can tell that the line is missing at the cocaine line. That would denote that this is a cocaine positive. You can see lines -- I can see them up here. Maybe you can't.

This is another device. Does anybody see a line here? So, what is this? This is a negative. It is counter-intuitive. It takes a while to get used to the idea. You think a positive is when you see something, but this is the positive.

One of the dip sticks, the same thing. Now, you will notice they all have a control line. This is an important feature. It is not so much a control in the sense of quality control. It is more of a validity line. It shows that the urine did pass up the strip. You need to see this control line or validity line to accept any kind of test result. Which one of these is positive and which is negative? See, this is one of the difficulties we found in the training period.

The operators who read the devices -- we only did single readings because that is what happens out at Federal Probation. They don't have a second reading. The person that collects the sample tests the sample and there is nobody to look over their shoulder. We tried to model that protocol in doing the study. The operators would say, well, I think I can see a line there, but I am not really sure if the line is there or not. Most of the package inserts -- and some more emphatically than others -- say that if there is any hint of a line, you call it negative. We were concerned about the phenomenon. We told the operators, look, you have got to score these positive or negative. There are no maybes.

If you look at it and say, well, I think it is positive but I am not sure, score that as a borderline positive. If you think the line isn't there but you aren't sure, score it a borderline negative. We did an extensive analysis on the borderline results. I tell you, there is a tremendous variation across the 15 products, how many produce borderline results or not. I don't have any slides included on it, but the bottom line is, it didn't make a lot of difference on the accuracy. In some cases, certain products would have been favored if we called borderline positives positives. In other cases, they wouldn't have been favored.

Either you are going to get more true positives or you are going to get more false positives. But this is one of the things. Now, all the data I am going to show you is scoring all borderlines as negatives. It is a very conservative approach and one that most package inserts advise.

The test cup, it is a very clever design because negatives show up as a little negative bar and if you have all white, that is a positive. This, I think, was from a control sample. So, you can see there are three positives on it.

Triage, now, is just the opposite. There is a band that appears at opiates. That means it is opiate positive. It is the opposite of the other devices.

The only other one that is different is the EZ Screen profile. You have a control spot and then the individual drug spots, and you are actually comparing the intensity of the color to that of the control, to determine the difference between positive and negative.

Now, we evaluated the data using standard approaches, such things as sensitivity, specificity, positive predictive values, negative predictive values. In layman's terms, you understand true positives and true negatives and the opposites are false positives and false negatives. In layman's terms, these positive predictive values and the negative predictive values are that a positive result will be positive by reference method. In this case, we used GC/MS. The same thing for a negative result, a result is negative by the reference method.

For those of you who like mathematics, these are the formulas, and they take into account the issue of prevalence. I want to point out that because we focused on specimens, approximately 60 percent of the specimens was our goal, around plus or minus 25 percent of the cutoff, this is our positive prevalence was something like 40 to 45 percent. You cannot extrapolate from these values to the population at large, which we did. We took data over a 12-month period from the same population from which these specimens were collected. You cannot predict what the sensitivity and specificity is from this data. It is just too concentrated. It is not a random representation of this test population. I am not going to spend a lot of time on these. They are in the handout material. I just want to illustrate that across devices the variations in the results that were obtained -- just to orient you, the dark bar is the true positives, the yellow bar the true negatives, and the greenish line is the false negatives and the reddish bar are the false positives.

So, these represent all of the negatives and these represent all the positives. Now, everybody wants to know what device A is, because device A for amphetamines had no false positives, no false negatives. That is the device we call the perfect device. It is what everybody wants, every manufacturer would like to make. So, just for reference point, you will see this A on here. This is the goal that we are heading for. Just on amphetamines, look at the variation between devices. Here are some with false positives and some with false negatives and some that may have maybe a little more balance between them.

This is important in governing the selection of the devices, as to what the goals of the program are. We will talk about that a little bit later. I will just run through these quickly.

Cocaine, a fair number of false positives, certainly fewer false negatives.

The same is true of the opiates, very few false negatives, but a large number of false positives. This population uses hydromorphone. There were a few, but not many. There were a few specimens in here that were positive for hydromorphone. That obviously would be false positives under the HHS criteria. So, if you use the U.S. courts test protocol, which includes hydromorphone, the data look better because there are fewer false positives.

Cannabinoids, you see the same kind of variation between the devices.

The phencyclidine, we only had seven of the devices that at the time of the study were available for PCP. So, only seven were included. The other difference between the other analytes is that even in this federal probation population there weren't a lot of PCPs. What we had to do to complete this step is to take a number of high positive PCPs and dilute them. So, we don't see matrix population effects in this group, because a lot of them were diluted with negative urine.

If we take all four of those drugs -- I am mentioning the four because all the devices were available but one for these four drugs, and then we had just a few for PCP.

If you average all four drugs together, looking at false positives and false negatives, this is the overall profile.

If you go back and look at the individual drugs, you see that there are variations. Something might look good here because the false positives and false negatives cancel each other out. If you are doing all four drugs -- and remember, a number of these are multiple tests, the four drugs are all in the same test as opposed to a number of these are single, I guess I should have made the distinction here which were the multiples and which were the singles.

At the present time we are only presenting this information in an anonymous fashion. The named products have not been released publicly yet.

We thought this was another way of expressing the data, just a quick summary. If you read computer magazines, when they evaluate computers, they evaluate performance versus costs. This is the most bang for the buck. These are the positive predictive values and negatives. Unfortunately, I noticed in the hand out that the white letters didn't print as contrast. So, if you want to mark on your handout, negative predictive values on the Y axis and positive on the X. Now, that perfect score, that product A is up here. This is where you would be if you had no false negatives and no false positives. You can see the scatter of these devices throughout. Obviously those that get closer and closer to this have the best balance between positives and negatives. The further down they are, it means they have higher false negative rates. The farther back they are, the higher false positive rates.

Now, part of certainly the analytical performance is a major criteria in selecting, but there are other issues that one has to take into account, such as ease of use, reading the end points and, as I mentioned, the operator differences. We compared our three operators every way that we could. Across the three operators, one operator had a lot more borderline scores than the other two. In terms of overall accuracy, there was not a significant difference. Certainly the prevalence is a factor.

Now as far as ease of use, there is quite a variation across these devices. In this particular device, which is one of the earlier ones on the market, the urine has to be placed on each individual spot, then the reagent A is added, then a B wash, then substrate. So, this is a more tedious procedure than you will see that followed.

Triage is also more complicated. This is a well that contains reagent. The urine is added and incubated for a period of time and then the pipette tip is used to spread it over the device, which is blotted and then a buffer wash. By going through these steps, they are able to reverse the process so you get a positive as a line. Again, it is somewhat more time consuming. This device, again, the drugs are individual pads where the urine has to be added and then it wicks up through the test strips. A device like this, which is the more typical strip device or card device, is the urine is taken up in a small pipette and then added to a well, which then wicks up through the multiple or single drugs that are available on the strip. This takes three to five minutes. The dip sticks, again, is a little convenient. You don't have to have a dropper. You don't have to put urine in. Depending on who is conducting the test, they may have more or greater aversion to handling urine. So, this offers an alternative where you just dip it into the urine.

The test cup, this is the best designed device. This serves as a collection container which then can be sealed. At the time of the test, the lid is turned to the test position, tipped on its side for a very short period of time. That allows the urine to wick into the test strip which is covered at the time. That cover is removed and you can see the test results. If it is positive, it can then be sent into the laboratory. Again, this testing is done in front of the individual. That is not the case with alcohol. The sample is then extracted by using the dropper in some cases, or tipping the bottle or dipping the stick in, whichever device you are using. The urine is added to the test device, the test then should be done strictly in accordance with the package insert. Then we recommend -- this is bottle A. We recommend resealing that bottle with evidence tape, signing the tape. Now the test then is conducted. If it is positive, the custody form then can be annotated positive for the drug name. In some cases they may want to record that result in some other fashion.

There has been some success in photocopying the devices. Most of them don't retain the color permanently. Some say read it in 10 minutes and don't read it after that, because some of them will continue to turn color and you can't tell whether they are positive or negative. So, there are some variations there, but certainly that result has to be recorded. Then all positives in the workplace situation would be sent on to a laboratory, directing the laboratory to confirm only for that particular drug. If they are negative, we recommend at least 10 percent of the negative samples to be sent in as well to be tested under the usual fashion. You have to put this note on, tested on site. The laboratory will say, the original seal is broken, and it has been resealed by somebody else.

Recall, bottle B is still there sealed by the donor and if there is any challenge on the result you still have a secured sample that can be sent on to another laboratory ala the federal regulations.

The interpretation and use of results, an on-site test is obviously a screening with presumptive results. We certainly recommend confirmation of presumptive positives. Certainly that is true in the workplace. In criminal justice, it varies. It depends a little bit on the district, it depends on the judge and whether the judge will permit screening results to come in now, as they do in many jurisdictions. So, there are some variations here. But certainly in the workplace, any employer who takes action on a presumptive result has to have some good support behind it. For example, in the federal regulations issued by the Nuclear Regulatory Commission, they permit somebody to be removed from service if they screen positive on site. The NRC has regulations governing on-site testing, currently limited to instrumented tests, under the same rules that the laboratories are under for screening. They can remove them from service if they are positive for marijuana and cocaine, but not if they are positive for opiates or amphetamines because of the phenomenon that we know about, cross reacting substance, codeine and legitimate prescription, and if that test operator has a proficiency of better than 90 percent confirmation rate.

So, they take into account the proficiency of the operator as well as the nature of the drug. Certainly whether immediate action is taken, like in NRC, or delayed is going to depend on the situation. If you have got a pilot that, for example, you have just tested on site and he is positive for cocaine and marijuana, are you going to let him fly off the carrier, or are you going to let him fly a commercial flight, if in that circumstance the pilot is not subject to DOT regulations. Then, one also has to look at local and state regulations. As David Evans pointed out, there are more states that are addressing this issue.

Is there only one that has banned it? I think there are three or four now that have --

PARTICIPANT: (Comment off microphone.)

DR. WILLETTE: For the record, what David said is that there are at least up to maybe five states that have laid out requirements for on site, but no state has yet banned on-site testing. Now, certainly in the workplace we recommend strongly to our clients that they use an MRO process. Within the nuclear regulatory regulations the MRO must review the initial results. Certainly they are a critical role in interpreting the positives. In some cases the MRO then serves to track proficiency. We do an open proficiency testing program for the federal courts. There are 36 federal sites that do on-site testing. In almost all of those, the collector is the tester. We have not been able to get blinds into them. In a few sites they have a separate tester than the collector. They do submit the PT samples in a blind. But in sort of a corporate setting, the MRO should be keeping track of what the initial test results are and what the lab results are. If they see a significant discrepancy in the number of unconfirmed positive, I think that is one of the functions they serve. They are the only one getting all the results.

Another issue that I want to raise is that the board has laid out some critical factors in this long section G in particular, but in general that apply across all of them. There are other issues surrounding how devices are actually tested, the data that is available and how the data is presented. We see in many cases -- there are many publications where other types of studies have been done. Certainly it is our intention to publish the U.S. courts study. It will be presented at the SOFT meeting next month, the first public release of the identity of the products. So, often we see that the accuracy claims are based on quite a disparate group of samples where there are frank negatives and strong positives and hopefully they all do quite well. In the study protocols, we did one kind of a study that was focused on concentrating on the cutoffs. There are certainly many other types of study protocols that can be done. I think it is only fair to point out the differences between them. Data can be taken from studies selectively. We get a lot of calls from companies saying, I have received this literature. Somebody is promoting this particular non-instrumented drug test product, and you know, what does this all mean. Certainly it can be distorted. Other products certainly using selective data can also misrepresent what is actually publish. I want to illustrate this in a recent example. Our study has not been released publicly. It was distributed as a confidential document to federal district courts. However, one manufacturer happened to get a copy of it and pre-empted our opportunity to publish it in a public forum. They distributed a small piece of the study in this manner. I just want to show you how you can sort of favor your product without giving a balanced view so that the buyer can really see what they are getting.

They took product number one and showed, this product had very low false positives. These are obviously less than 10 percent. They compared it to two other products showing that they had much higher false positives results. We wondered why cocaine and opiates were selected out. The main reason is that if you look at amphetamines and cannabinoids, the other products did much better where the worst cases were with opiates and cocaines. However, they did not show the false negative results. These are the same products, and if you look at the false negative results for these products, this is what you see. If you recall back on those bar graphs, you can see there was generally a trade off. Devices with high false positive results have low false negative results and vice versa. Also, if you then look at the false negative results for all five drugs, you see that product one, which released this excerpt from this study, didn't do too well with these two drugs. Certainly these devices did much better.

Now is this good or bad? How do you present this to a corporate HR person or a medical director? I think what you really want to do is say, well, depending on the objectives of your program, like in criminal justice you may want to detect all five. Maybe you can afford to have a high false positive rate or a higher rate than you might. The amphetamine case is even more interesting. Those of you who are familiar with these terms know that this is just false positives and false negatives from the same sample population, you subtract this from this and you don't get anything new.

I don't know if you followed me. In other words, if you have got 100 specimens and you have got 22 of them that are false positives, you are going to have 73 that are false negatives. Not very information. What I would have done is take these and compare them to the false negatives. You see this throughout all the devices, not just the four that are shown here. This is logical. The more false positives you have, the less false negatives, and vice versa. Now this may be fine for your program objectives, but if you really want to detect all positives, you may want to strike a balance or you may not want to miss any positives. In the criminal justice arena, our experience with federal probation is that a vast majority of screen positives confess. It is the same things that folks maybe like Bob Dupont and Avram Goldstein and a number of people published in the 1970s in the drug treatment community, is that you don't need to send it to a lab if the person confesses. In many of the on-site positive results they get a lot of confessions. It doesn't cost any money. You don't have to get a confirmation.

The other kind of thing is we were quoted as authors of this confidential report, certain conclusions, and this publication that was released makes a statement like this. I will show you the context. What they left out was this, with the down side of missing more drug users. Whenever you see ellipses, you really want to look at what is missing. So, the manufacturer had the highest PP values of the four drugs, which it did. Of all the positive results, a high percentage were confirmed. It had the lowest negative predictive value of all the devices, which means it missed more true positives than any other device. That may be good or bad; I am not saying. I am saying that in terms of informing the buyer of what the nature of the product is.

This is just another excerpt which has one connotation. We said, for example, in criminal justice settings in which on-site tests will be utilized, if the purpose is to identify as many drug users as possible, devices with low false negative rates would be favored. That is certainly true within the criminal justice arena that we work in. It may not be true in the workplace, if you are going to take immediate results. I don't know what the board or Division of Workplace Programs can do in terms of laying out guidelines. But laying out criteria for how studies could be conducted and how they could be interpreted, and to urge a full disclosure of the pros and cons, the weaknesses and strengths of the devices would certainly be welcome.

I do want to acknowledge certainly the charge by the administrative court and the funding of this study, and also thank PharmChem Laboratories who we kept in the dark. We had space away from the main laboratories. Nobody was allowed in. It was done under cloaked conditions. We would select the specimens and did all the testing and wiped the computer disk clean before we left, and certainly the three technicians who did all the work. Thank you.

(Applause.)

Do we have time for questions now, Donna? Do you want to take the break and then questions? Questions now, Dr. Autry says.

DR. CHILDS: You didn't do any screen of the positives that were on-site positives. Your instructions were that when they came into the laboratory you just did the confirmation test; is that correct?

DR. WILLETTE: No, just the opposite, Paula. We would go to the screening records every day to select -- well, we could only manage 20 or 30 specimens a day with three technicians. We were doing 15 devices in the first part of the study and two in the last part. So, we used only the screening records. We didn't have all the confirmation data until three months later. In some cases the specimens that may have tested positive for other drugs had to go into 90-day frozen storage under the federal probation protocol. So, we didn't have access to them to do GC/MS for other drugs. So, we didn't have GC/MS data until after all the screening was done.

DR. CHILDS: I understand that. Actually, I probably phrased it wrong. I was referring to your recommendations for how to do the testing, was to submit the specimen to the laboratory?

DR. WILLETTE: If they screened positive, yes.

DR. CHILDS: And that they would not repeat the screening part of the test. The laboratory would not repeat the immunoassay test.

DR. WILLETTE: We have had mixed experience on it, Paula. One of the problems is you are testing a fresh specimen. In most cases the on-site is done so you get a result right away. In some situations they may put them in a refrigerator and test them at the end of the day. By the time that takes place and they are packaged up and sent to the laboratory, specimens that are around the borderline will then screen negative in the laboratory. We have looked at it both ways. In the early day when the navy had portable drug test kits on every ship and they were sending them in, CompuChem Laboratories, who had the contract, would run the screen as a quality control check but still go on to confirmation.

The laboratory's responsibility for the screening test is then really sort of usurped. It is the responsibility of the on-site operator to have the right documentation and training to be able to defend that result.

Now, if you look at the literature -- David is probably more familiar with this than I am -- in legal cases, there are many states and many courts that uphold screening results only, in the criminal justice arena. I don't think they are going to do that much in the workplace arena. I don't see a problem with going directly to the confirmation. It is like a retest, in a sense, but still using the original cutoffs.

DR. CHILDS: The reason I asked that is because we have had quite a bit of experience in the last two years with on-site tests being performed with different types of devices and then sending either portions of the samples or the original sampling to the laboratory. Of course, most of them do confirm in the positive. We do get some with absolutely no drug there at all by GC/MS, even with the limited detection types of levels. Others that come in that the request comes in to please rescreen and confirm these, because that is the way the program is set up, we do get some that do not screen positive at all. There is nothing there by screening and nothing by GC/MS. Then there are some others that are borderlines that you could see some deterioration of sample. I am just curious to know. I see in your recommendation for how to do that, that you basically recommend not to do a rescreen of that.

DR. WILLETTE: It may be more costly, in terms of that sample that is sent in, if you are doing the screening and confirmations. If the protocol is to screen it at the laboratory and not confirm, then there may be some money saved. As somebody pointed out -- I think Dr. Walsh -- is that one of the reasons, but certainly not the only reason, that companies are looking at on-site is to reduce costs.

I didn't mention costs in this, but cost is a big factor in the choice of which way you go. Most of the clients we deal with aren't looking at it so much from the cost saving but from a safety standpoint. In the application where they are testing applicants to eliminate a long medical exam, that is a cost issue, but it is not the cost of the test that is the factor. They will pay more for that to save a few hundred dollars in the medical exam. However, I am not saying that rescreening at the laboratory and following the standard protocol may not be well advised in some cases. Without the confirmation, you have got no quality control over what the on-site operator does. I would like to see confirmations in every case.

DR. CHILDS: The other question I had was, you had made a comment that if they were negative on site, then a certain percentage of those should be sent to the laboratory for a screen and confirmation. That certainly was one of the recommendations of the on-site testing panel from 1989. One of the questions I had, though, is how does that get tracked and monitored? Is that like the collector interacts with the reporting format and says, I am sending a negative report back to the employer at that point? Then what happens if the lab screens it as positive and tests it and, in fact, it is an incorrect identification. We have run across that. The way that tests are being performed on site is that there are incorrect positives and negatives.

DR. WILLETTE: We have had more experience with that in the nuclear regulatory field where they do on-site testing. The MRO is really the person who reviews that information. In the one relationship we have with a nuclear plant, that MRO, if they get a discrepancy they will usually contact us and ask us to do an analytical investigation to follow up on that, when they are doing an on-site test blind, in the sense that they don't know whether it is a true specimen or a blind specimen and a certain percentage has to go on to the laboratory.

DR. CHILDS: The last question I had had to do with the relative cutoffs. I am presuming that the cutoffs for all the devices were at the HHS cutoffs; is that correct?

DR. WILLETTE: Almost. The only difference is with the amphetamines. You know the problem with amphetamines. You have got amphetamine, methamphetamine, you have got DML. One of the devices actually had a 500 nanogram D meth amphetamine cutoff and one device had a 1,000 nanogram DL amphetamine cutoff. All the rest were 1,000 D meth. Many of these products are available as an amphetamine specific and some as meth amphetamine specific. We ordered meth amphetamine specific wherever we could. You will see the disparities in the amphetamine results because some of these devices were amphetamine specific and had very high false negative rates because they weren't picking up the meth amphetamine. So, the amphetamines were the biggest problem. All the rest were at the HHS cutoff; well, supposedly.

DR. WALSH: I have got two or three questions. To a certain extent, all the results depend on the skill of your technicians to follow the procedures properly and so forth.

DR. WILLETTE: Right.

DR. WALSH: I wonder if you could elaborate a little bit on what the background experience of these people were, whether they had had experience with hand-held kits, and whether the manufacturers were involved at all in the training of these individuals, or how much practice time and training time did they really have before you got into the analysis?

DR. WILLETTE: Their backgrounds vary. It turns out that in the San Francisco Bay Area there are temp agencies that provide laboratory personnel. After going through a number of resumes, we picked those who had the most clinical experience. They were all licensed clinical laboratory technicians. One of the operators, in fact, had some experience and did some temporary work for one of the manufacturers of these devices. Of the three, he had the more experience with it. We did not use the manufacturers for training. There were 12 manufacturers and distributors involved. So, we followed the package inserts, protocols were worked out. Then they set out to test some initial samples. They were overseen by my associate, Dr. Kadejian, virtually every day of the test. I would say it was limited but certainly not extensive training. They were all skilled laboratory technicians.

DR. WALSH: Most technicians that begin to use these various kits get better with a little experience, once they have been on the job. It is going back to the old probation and parole program, a lot of these guys got to the point of being very good, in terms of almost having a calibrated eyeball, in terms of being able to differentiate a borderline positive.

DR. WILLETTE: Depending on the color on these strips, too. That goes back to the old TLC requirement days where you had to do color blindness tests.

DR. WALSH: Did you drug test these guys before you hired them?

DR. WILLETTE: No, unfortunately we didn't have that kind of contract. The point I was going to make is what you are saying, Mike, is the same thing that is in the board's factors that they are considering. Certainly training is an important issue. Even with training, if you have had 30 years of experience with on-site testing, that training is only as good as how often they test, how frequently they test even with instrumented devices. There are a lot of variables in that. It is not as simple as up front training.

DR. WALSH: Having just recently done a comparable study evaluating kits for police officers, one thing that isn't covered, I don't think, in any of the manufacturers package inserts has to do with using frozen samples. All these kits were really designed to be used with fresh, warm, 90 degree specimens. What I found, since were collecting specimens from DUI suspects and then freezing them and taking them back to the laboratory and analyzing them later, is that when you allowed them to warm up to room temperature, 72 or 73 degrees, they run much more slowly, four or five or six times the amount of time that is recommended. If you normally are running these things with a fresh, warm specimen according to the package insert and it says read the results in three to five minutes, as soon as you see that control line, typically we had to wait 14 or 15 minutes before the tests would be fully completed.

DR. WILLETTE: That is certainly a factor and we look at the real world situation and you see the whole spectrum. Rarely are they frozen. We didn't have a choice in terms of how the study went because of the time frame. Each device, tested on a set of samples, stand on their own. It may be hard to compare devices tested on frozen specimens with those that are tested on fresh specimens.

DR. WALSH: One just real quick specific technical questions. Since 60 percent of your specimens were plus or minus 25 percent of the immunoassay screening level, is that the HHS level?

DR. WILLETTE: Yes, that was the goal based on the initial screening. As you well know when you look at screening results and you look at GC/MS results, we didn't end up with plus or minus 25 percent of the samples within the GC/MS cutoffs.

DR. WALSH: For marijuana, for example, you were shooting for 60 percent of the specimens between 37.5 nanograms and 62.5 nanograms, something like that?

DR. WILLETTE: Yes.

DR. WALSH: One last question. In the grand scheme of things, I know you retested -- you selected the specimens on the basis of initial EMIT screen, retested them with these non-instrumented kits but you also retested them with EMIT and EMIT-II.

DR. WILLETTE: That is right. That was the instrumental replicant.

DR. WALSH: My impression looking at some of the data was that overall a number of the non-instrumented devices performed very comparably to the instrument-based assays; is that right?

DR. WILLETTE: Yes, in the report we issued to the courts and in the public paper we point out that the technology in doing non-instrumented drug tests has really advanced significantly. Many of the devices gave comparable results to the instrumented device. We are certainly a cautious advocate of doing on-site testing with non-instrumented devices, and there are many settings where it is very necessary, perhaps applicable, if you are trying to have a drug free work place. Many of them did very well. These have come a long way over the last 10 or 20 years.

DR. WALSH: I think this study represents a really good start and I hope there will be a lot of others in the next year to look at a number of the different aspects of how these things can be integrated into the system.

DR. WILLETTE: We are hoping to capture data from the operation drug test that is going on within the courts right now.

DR. AUTRY: I am going to ask the group's indulgence and we will take all the questions in line. We have one person who is here who is signed up to make a public comment and has a 4:00 o'clock flight. Will you grant him five minutes at this point in time?

I think it says Benjamin Hoffman.

Also, we want to take a break after his comments and then come back to the comments. What is the group's feeling? Okay, thank you very much.

Agenda Item: Public Comment.

DR. HOFFMAN: Good afternoon. Thank you for letting me speak. My name is Dr. Ben Hoffman. I am the president of a consulting firm called Business Health. I specialize, and the firm specializes, in occupational environmental medicine. I am a physician with graduate training in internal medicine, in toxicology and epidemiology. I did my training at Yale. I act as an MRO for hundreds of companies. My offices are located outside Boston. I have developed and managed substance abuse programs and have been in this field for over 10 years. While I have many different companies that I work for, my clients tend to be the large, blue chip, Fortune 100 firms.

Over the past five years I have had a significant involvement with the use of hair in testing for illicit drugs, primarily with Psychomedics. I have come here today to share with you my positive experiences on behalf of these clients, and to provide you with their position on several relevant aspects of hair testing. I have been responsible for designing and running numerous drug testing programs over the years. I have been involved in over 50,000 hair tests. I obviously have a lot of real world experience with hair testing.

Many of my clients have selected hair testing to augment their comprehensive substance abuse programs. These programs include urine testing, employee and supervisor awareness programs, employee assistance programs and rehabilitation when needed. Hair testing was selected as an alternative or to augment urine testing when it became obvious to these corporations that urine testing was in itself quite inadequate, due to the problems with short detection windows, dilution, adulteration and all the things that you are aware of. All my clients quite carefully scrutinized hair testing from many perspectives -- legal, scientific and ethical -- and many consultants' opinions have been reviewed. The various scientific controversies were extensively reviewed, and were not felt to be show stoppers or could be handled through various policy considerations.

These scientific issues included really three-fold, passive inhalation, hair color or so-called racial biases, and thirdly, the lack of standardized lab procedures or lab certification. I am briefly going to review these issues along the lines of my clients and review their positions.

The first one is passive inhalation. Passive inhalation is no more an issue with hair than it is with urine. In fact, in my recent review, as an epidemiologist and prior bench researcher, the data used to support the DHHS urine drug thresholds, particularly for marijuana, are of questionable accuracy, since it is based on old data and there have been recent substantial increases in the concentration of THC in marijuana. In addition, studies that have raised the concern of passive exposure to coke have very serious epidemiologic and methodologic flaws. Real world data fails to substantiate these concerns. My adjudication of thousands of positive hair tests has failed to justify empirically that passive inhalation is an issue. I realize that people can't see this, but I recently reviewed the last several hundred positive hair and positive urine tests.

Interestingly, the denial of use with hair was only eight percent, as opposed to 49 percent with urine. Passive exposure was used as an explanation in four percent of the hair and three percent of the urine. Real world data, on the last 400 or so tests, doesn't substantiate at all that passive inhalation is an issue.

The second scientific controversy that has come up is the issue of hair color or racial bias. I believe as a toxicologist that it is inappropriate to believe that biases based upon human factors such as gender, race, et cetera, don't enter into all types of biologic testing, including urine testing. The data from the last 5,000 tests, positive tests, when we reviewed them along racial lines, we failed to see any evidence of racial bias whatsoever. Again, it is empirical data.

The third, and the last controversial area, has to do with the lack of federal certification and standardized methods to measure drug content in hair. This is of concern to the business community. My clients would like the federal government to step in and provide lab standards in this area, particularly since there seems to be so much political infighting among the scientists involved.

As an outsider, the politics and the vendettas appear to impact the scientific issues and much of the energy of quality researchers, perhaps even using federal money. There seems to be more money spent trying to discredit this technology than improving our understanding of the technology and widening its application of federal drug testing programs.

My summary, my take home message, coming from clients -- and these are multiple Fortune 50 and 100 clients -- is that hair testing is being used and widely supported. The business community will continue to use the technology, despite the perceived limitations among the vocal minority in the scientific community. Thirdly, there is a request among these clients that the federal government focus their energy, their talent, their research dollars into research that will eventually validate the use of hair as a methodology, and to regulate and certify labs that are performing hair analysis. Thank you.

DR. AUTRY: Thank you, Dr. Hoffman. If you will leave copies of that, we will make sure all the board members have copies.

DR. HOFFMAN: I will provide copies. Thank you very much for your indulgence.

DR. AUTRY: I thank the group also. Why don't we take a 15-minute break. When we come back, Bob will take some more questions. While you are gone we will hand out the filled in grid from this morning. I will ask Skip to point out the couple of corrections on the last page that are different from what was on the overhead. So, we will see you in 15 minutes. After the next Qs and A, we will go into public comment and we have some written questions to read in. Thank you.

(Brief recess.)

DR. BUSH: We will resume where we left off with some questions to Dr. Willette.

Agenda Item: Committee Discussion and Questions.

DR. WILLETTE: Do you want to go ahead?

DR. KWONG: Bob, I would like to ask for clarification on your study protocol. Was each of the specimens in a study analyzed independently by all three technologists or analyzed only once by one technology?

DR. WILLETTE: They were each tested once. What we did, each day we grouped the devices into panel A, B and C. Each day they rotated to try to have each technician test approximately the same number of devices as the other technicians.

MR. EVANS: Bob, I wanted to ask you, the test devices that you selected, was the selection made based on any criteria? Were they all FDA cleared, for example, or some were, some weren't?

DR. WILLETTE: Some were, some weren't.

MR. EVANS: Did you do an analysis of how the FDA cleared tests compared to the non-FDA cleared tests?

DR. WILLETTE: No, we haven't looked at that. Most of them were either FDA cleared or in the process, insofar as the manufacturers, the information they provided. The selection was to try to get everything that was on the market at the time. There were one or two products that were known to us that there was no forwarding telephone number, they had dropped out of sight. The one, they were between manufacturing locations and they didn't have enough to sell. We tried to get every one we could.

MR. EVANS: I just wondered, because as far as we were concerned, the only ones we would recommend would be ones that were FDA cleared and could meet the NIDA cutoff levels. The same thing with an alcohol test. It would have to meet the DOT standards before it would be recommended by us. We also, in an employment context, always recommend confirmation.

DR. WILLETTE: I would agree with all that.

MR. EVANS: I just wanted to make sure everybody was clear about that. Thanks.

DR. WILLETTE: We had a little bit of latitude. We expected all these to be FDA cleared at some point. The federal criminal justice system doesn't need FDA clearance. The concern was that some of these products may still find their way into criminal justice even though they hadn't been FDA cleared. So, we couldn't use that as a criteria.

DR. CONE: Bob, I want to compliment you on a very nice study. Before I ask you a question I was going to respond to Tai's question about reader concordance. We have done some studies over the years. In different ways, and usually quite different from the way you did it, where we have evaluated reader concordance under different circumstances, where we have had readers reading the same test result within a few minutes of each other but still under blind conditions. You can get actually reasonably trained operators in excess of 90 percent concordance, but you always get a little discrepancy and that is usually with samples around the cutoff, like you were looking at. You were looking at some of the most difficult ones and that would be the biggest challenge. Still, you get very good concordance with trained operators. Of course, training has a lot to do with it. We have never tried to vary that variable. Our operators were trained with starting them out each day with running a negative and a positive and observing the result as a refreshment for their initial training.

A question that I have for you, Bob, and it has to do with documentation. We looked at different ways to document results. We found some ways that were successful and some ways that didn't work at all. Did you think about that? We looked at things like photocopying. I guess you could take pictures and so forth. What did you try to do or did you think about that?

DR. WILLETTE: Yes, we considered it. We didn't attempt to do any permanent documentation during this study but we considered all the possibilities. Some seemed to photocopy all right, but that was probably a little clearer with the ones that really were positive where there was no line. Some of these borderlines would be difficult. Again, if all borderlines are called negatives, it doesn't make a difference. You are really looking for an absence of the line.

DR. CONE: When you do get a positive, did photocopying bring out the line on most devices? I guess the negatives, I suppose.

DR. WILLETTE: I think in the few we tried to do. We didn't make it a point of the study but we wanted to see what was possible.

DR. CONE: One thing we tried was photocopying, and then having another reader read the photocopy at some later date, just to see the accuracy. We actually got about 85 percent concordance with those. Photocopying is usually about the best way to document. People usually have photocopiers around.

DR. WILLETTE: That is a very good suggestion.

DR. AUTRY: I want to ask you a question that you may not be able to answer, but I will ask it anyway. Given your true and false positive and negative rates and your negative positive or false positive, false negative rates, what is likely to happen to the positive and negative predictive values as the prevalence goes down?

DR. WILLETTE: We have some other data that is available from federal probation locations. We took -- we just wanted to test that and found that there was no way we could extrapolate from this set of samples. We do know that in the real world situation, that the results are much better, because the drug concentrations are more randomly distributed. They aren't all concentrated around the cutoff. So, from experience with perhaps maybe 10 or 12 of these devices, they all performed better under, say, the routine day-in, day-out, with fairly high confirmation rates across the drugs. This is probably the worst case scenario, is to concentrate the specimens around the cutoffs. As was pointed out, you are obviously going to get a lot more borderline readings.

DR. JONES: Bob, these are authentic samples. Were there any spiked controls included in your battery of specimens that were analyzed by these protocols?

DR. WILLETTE: Yes, out of the 100 samples for each drug, we included 10 controls and 90 real specimens. The data I showed you earlier were real specimens.

DR. JONES: Could you summarize how the controls performed?

DR. WILLETTE: The controls included also plus or minus 25 percent borderline. They tend to parallel. The ones we saw higher false positive rates were picking up the minus 25 percent controls. They seemed to parallel the controls as well as they did with real specimens.

DR. AUTRY: Thank you. I want to thank Bob Willette and Bob Fogerson. I also want to thank our colleagues at the Administrative Office of the U.S. Courts, both for their collaborative activities with us and also for their willingness to have this information shared with you and also shared with the board. I think it does help us fill in some of the grid that the board had some question about. That is why we asked for this type of presentation across the various specimens and technologies.

DR. BUSH: At this time, we would like to entertain the questions from the members of the public who had signed up to do so, at this time. We have already heard from Dr. Benjamin Hoffman, a little bit out of order, but well received. Dr. Bob Dupont; is he in the room?

Agenda Item: Public Comments

DR. DUPONT: I didn't expect to be the first one on. I thank you very much for that. I don't know how to be so fortunate, but I am very grateful.

I am Bob Dupont. I know many of you folks and I feel very proud to be here and feel very grateful for this opportunity. I am speaking on behalf of a sympathetic view toward the use of hair testing in workplace drug testing. I want to go back a little bit with you in my experience. I am proud to be here with one of the other gang of four that were singled out by Abby Hoffman in his book, Steal That Urine Test in the 1980s, along with Bob Willette. He is also on that list, because he and I and Mike Walsh and several other people here were very important in development of workplace drug testing. I was the first director of the National Institute on Drug Abuse, and have been vice president of Bentsinger Dupont and Associates, focusing on workplace issues, including drug testing, since 1982.

My interest in hair testing began when I approached the Psychemedics Company about the use of hair testing in the late 1980s, because I was very interested in its potential for having a longer surveillance window, particularly in a pre-employment context. I use some of the tests in my own practice with my own patients. I had a pretty good idea of what was going on and was frankly surprised and pleased by the concordance between the results that I got when I had a pretty good idea what was going on with those patients, and have been active with the company since, including serving as the chairman of their scientific advisory board.

I also, like Ben Hoffman, am a medical review officer and do a good bit of MRO work for large companies that do hair testing in pre-employment contexts. I have found that it works very well in that context. I would call the board's attention to the fact that the marketplace is speaking rather loudly about the use of hair testing, particularly in pre-employment testing. As Dr. Hoffman said, it is the larger companies that seem to be particularly attracted to this and quite a large volume of testing is going on and it is growing quite rapidly.

With respect to the results of hair testing, I would like to remind you how things happened with urine testing. That is, the testing went on a long time in the workplace prior to the first guidelines from the government. There was a lot of activity in the marketplace to define what the testing was and how it was going to go on. The guidelines came really late into the process and not early. Many of the issues we see in hair testing we also saw in urine testing, particularly the issue of racial bias and passive exposure were very big issues in urine testing as well.

What I would ask the board respectfully to do is to recognize and, to the extent possible, benefit from this very large naturalistic experiment that is going on in any way that is possible to make use of that. I think a lot is happening. It is very attractive in terms of a pre-employment test. I think it is a fair statement to say that a scheduled test using a urine test is much more of a test of intelligence than drug use. All a person has to do to pass the test is not use for three days and they are negative. The fact that a large number of companies are using a pre-employment screen of a urine test suggests that they don't have a very good understanding of that. It also shows the very high prevalence of both ignorance and drug use in the population because you get such high levels of positive results of it. But this is an example of an application that is extremely important for the hair testing.

With respect to the passive exposure, I think it is very important to find ways to use naturalistic, real life situations and not laboratory cooking procedures to establish this, which is exactly what happened with urine testing. I think that would take care of that issue.

With respect to racial bias, I think it is very important to recognize that variations do occur in all our testing. For example, in alcohol testing, we have one BAC for people, no matter what their body weight is and no matter whether they are male or female, although we are well aware of the fact that body weight and sex is a very important determination of the relationship between use of the substance and BAC level. This is not surprising.

If we want to look at the issue of racial bias, then I would think that we would want to do that with urine testing and other things as well. All I am saying is that the role of the government and the board should be to encourage the development of alternate technologies in the public interest. The identification of drug use, including in the workplace, is an extremely important part of this nation's drug abuse prevention activities.

I think hair testing plays a very important role in that. I would encourage you all to encourage that process. I think when you do get involved in it, a lot of concerns that you have will diminish and this public interest will be well served by adding this, as I have been a very strong supporter of other alternate technologies such as the sweat patch and on-site testing. So, this goes into the armamentarium and it should be judged on the same basis as the other. Thank you.

DR. BUSH: Thank you, Doctor. The next individual is named Pat.

PAT: Good afternoon. My name is Pat. I am not a scientist, a doctor or a lawyer. I am not associated with any company, agency or organization. I represent myself and I would assume hundreds if not thousands of people in the past and the future who were, or will be, unjustly accused of illegal drug use through hair analysis, as I was. For those of you who were at the three-day conference in April, or for those of you who had the opportunity to read the transcript of that conference, you might be familiar with the comment made by Dr. Blank, who is a deterrence branch head for the Department of the Navy.

He had stated that years ago the navy had opened its doors to a Dr. Baumgartner for the purposes of exploring hair analysis. Dr. Blank said that, although the results were promising, they were far from conclusive, and that contamination of hair by passive exposure is a scientific fact. Then he proceeded to read a letter that was written by a police officer. In that letter, the police officer stated that he was recently suspended due to a positive reading for cocaine through hair analysis. He went on to say that he did not use cocaine, and that during the past two years as a police officer he had never missed one day of work, had given numerous negative urine samples, and for the immediate 17 months prior to his suspension, he had worked in a drug prone location on a daily basis. His results were 24 nanograms per 10 milligrams of cocaine, no cocaethylene and a trace of BZE. Dr. Blank concluded by saying, what do you say to this guy? Well, that was my letter. I am that police officer and still searching for the answer to that question.

Dr. Baumgartner's response to Dr. Blank's question was, "of course we know that drug users do deny." My reply to that is, of course we know that people who do not use drugs will always deny.

A Dr. Sachs also had a response to Dr. Blank. He said, "It is of no use to take one single letter to discriminate against one alternate methodology. We make mistakes in urine labs and hair labs."

I really do not know what point of view he was coming from when he said this. Just the thought of him saying we make mistakes is frightening enough, but I feel that the main issue is getting lost in the shuffle.

I am not part of a study of 40 undercover narcotic officers. I am not part of a study of 29 children living in a contaminated environment. I am not part of a controlled experiment of drug free hair in a 7x8 unventilated room, and I am definitely not a single letter. I am a real, live human being who lives and works in the real world. I am a person with morals and reputation and a family. I am a person who did not use cocaine whose friends and neighbors are now talking about him behind closed doors. I am a person who can never again get a city job in the city that I live in. I can never again get a law enforcement job in our country. I am a person whose life and family have been turned inside out. I am a person who was unjustly accused because of a problematic test. Since my suspension eight months ago, I have been officially terminated. During that time, I have been in contact with numerous politicians, agencies and experts. I have literally been in touch with well over 100 different people regarding my situation. Out of that number, I can count on one hand the number that have truly shown concern and rendered helpful assistance. To my disappointment and amazement, many of the other contacts who are aware of, or have been made aware of the controversy with hair testing aren't able to render me any assistance. I have received replies such as, we would like to help, but, or although we understand, there is nothing we can do. It was even suggested to me that, although I would be perfectly welcome to make a public comment today, that this is not the proper forum. Obviously, I disagree.

As of today, the only forum for myself and people like me, who have been unjustly accused, has been after the fact, after we have been terminated, after our reputations are advertised and our future has been put in jeopardy. That forum is in the court of law, where after spending thousands of dollars, we have only a hope of the slightest chance of winning. Well, whether it takes me one year, five years or 20 years, I plan to change that. This is the only place where I can get the leading scientists around the country and around the world in one room. This is the forum, right here and right now.

It is up to those of you in this room that understand as a fact that hair analysis is still problematic, that it shouldn't be used in the workplace, to speak out. If you have been speaking, I implore you to speak louder. I understand that you do your research and publish your research in scientific journals, but that is not enough. Just as the scientist will go out and sell his product to companies emphasizing its advantages, it is up to you to contact state and local governments and advertise its disadvantages. People such as politicians, police commissioners, CEOs of companies, will listen and give credence to what it has to say. There is a big difference when it comes from your mouth compared to mine. When it comes from mine, they see a person accused of drug use. They see a statistic. No matter what, there are two facts that assure that I will prevail. The first is that I know that I did not use cocaine.

The second is I will not stop until my name is cleared and I prove that this type of testing cannot distinguish between drug use and external environmental contamination in all cases, especially at the low cutoff levels that are being used. I stand before you as living proof that this is an unreliable test. The pride I had in being a police officer has been unjustly stripped from me. This is the mere beginning of my crusade.

I now have two simple questions. The first is for Dr. Baumgartner. I hold in my hand a copy of a detailed letter dated March 25, 1997, a letter that I sent to you. In that letter, besides certain particulars that I mention here today, I also informed you of discrepancies between collection procedures, the protective measures that you state in all of your manuscripts must be taken and what the police department's medical division actually did. Is there any reason why you never responded to my letter?

The second question is for anyone, and I mean anyone in this room. It does not matter if you are a scientist or a secretary. If you feel that at this time hair analysis is still problematic and should not be used in the workplace and are willing to help me in any way, I will be in the lobby during the break today.

My intentions were to be present throughout tomorrow, but due to a tragedy yesterday, the unexpected passing of my beloved brother-in-law, I will return home immediately following today's agenda to be at my wife's side. So, again, I repeat myself. Please feel free to approach me. I would like to extend my gratitude to the Substance Abuse and Mental Health Services Administration for affording me this time to speak today. Thank you.

DR. BUSH: Next on the list of public commenters is Amy Taylor, Director of Congressional and Regulatory Affairs, National Association of Collection Sites.

MS. TAYLOR: My name is Amy Taylor. I am the director of congressional and regulatory affairs for the National Association of Collection Sites. We are a relatively young association. We were developed about two years ago. On behalf of the National Association of Collection Sites, I am offering these comments to the Drug Testing Advisory Board, on the drug testing of alternative specimens and on-site devices with a special emphasis on collection. NACS will submit written comments in the next few days. The association believes that collection problems need to be more closely addressed and collection training must be emphasized before DTAB makes its recommendations to DHHS. A comprehensive training and certification program would provide universal collection standards for each collection site. This will increase the professionalism, the validity and the safety of each collection and will mean that less mistakes will be made that jeopardize the accuracy of the testing process.

This training program could be modeled after the voluntary accreditation program that NACS has just finalized for urine collection, which is aimed at training the trainer at each site. Even though urine collection has been federally mandated for almost 10 years, the collection industry is just now addressing the quality issues associated with urine collection. This is why collection standards for all specimens must be defined early in the process, so that they become a standardized component of any drug testing program, to ensure that collectors have sufficient guidance on how to properly conduct the collection.

NACS is also concerned that with several different types of alternative specimen testing being offered, there may be confusion about how the collection procedures differ for each alternative specimen. With an increase in the specimen types that may be tested, there is certainly more room for mistakes during the collection process. This is why industry collection standards and training are so important. Each technology is only as good as the collection and DTAB should closely consider the collection process in each of the technologies being examined.

DTAB should also provide adequate information on the collection techniques for each specimen and implement collection training requirements.

If DTAB decides to recommend the use of any of these specimens in a federal program, NACS would like to be an integral partner in the creation of collection standards to ensure the validity of each collection. Thank you.

DR. BUSH: Thank you. Now, Skip, do you have those three sheets of written questions? I don't have those sheets. We have someone who desperately wants to ask some questions and has laryngitis.

DR. JONES: A couple of questions and comments that she did submit to us. I will read these. Some of these are, in part, addressed by some previous comments, but I think they are points that we certainly need to include.

The statement goes as follows: Why do requirements not require documentation of a balance of the true positive, false positive, true negatives and false negatives as a driving consideration for the choice of cutoffs; for example, in ROC analyses.

I interpret that as the guidance regulations should include -- your recommendation is that the guidance recommendations should include what is an allowable false positive rate, what is an allowable false positive rate, et cetera.

Is that what you are getting to? I have got to make her speak, because I am not sure.

DR. SPIEHLER: I am just impressed by the level of analysis that is being carried out by the people who are presenting to the board, and using appropriate gold standards and appropriate analysis, not for example, but id est, that is, ROC analysis.

This has certainly been done on a scientific level. So, I was a little disappointed that the board didn't require that scientific criteria, even taking it to the extent of setting some standards for clinical sensitivity or clinical specificity. I know that the FDA certainly has them. So, it was just a mild remonstrance. But I feel sad to present that because we know as scientists -- after that last speaker, I was really touched. We know as scientists that our tests are not 100 percent. There are going to be people like Pat. I know of a case in the military drug testing which I testified has false positives at the one in 10,000 level. Of all the testing they do and the positives they have done, I do know one person who was falsely accused and left the military career that he loved and wanted above all else, and there is that human side of it. As scientists, it is a statistical thing. That is all I am saying, that the proper statistical analysis has been done and the presentations made to you. So, you should take up the courage of putting that in your recommendations.

DR. JONES: Thank you. One other question that you had, what pharmacokinetic studies support the choice of current cutoffs. Her education did not include any. Would anybody like to address that question?

We cannot ask Dr. Walsh to answer that question. But I think it is a very valid question. I think the current cutoffs, as they were established back in 1986 when they originally came out -- really, we were looking at similar cutoffs to these even prior to that in the military. We certainly didn't have the kinetic studies that supported those even back in those days.

DR. WALSH: I will follow up to what Vina was saying. There was a lot of lobbying by a number of the different manufacturers back in mid-1986. Many of them had a great deal of access to the secretary, who took turns at beating the stuffings out of me, in terms of trying to set up special cutoff levels if you used a particular manufacturer's assay. We spent a lot of time looking at the reliability from one lot to the next in terms of, you know, getting ROC curves for each lot of antibody being generated by a manufacturer. In over a period of year we may have five or six or 10 or 12 lots, depending on how many of the rats or mice died or the sheep or goats died. You can have a significant amount of variations. In setting the original cutoff levels, a lot of the consideration was, what can we reasonably expect the manufacturers to produce and what can we reasonably expect the labs to do on a day to day basis and running calibration curves and so on.

Obviously we relied very heavily on the military experience at that time. I remember being in one meeting where somebody was even suggesting that almost monthly we would send out a catalogue of a matrix saying, if you use this particular manufacturer's assay with lot numbers X, Y and Z, then you could screen at 47.9 nanograms. But if you are using this guys, you have to screen at 65 nanograms or so on.

So, realizing that it just wasn't practical in doing it that way, we just took a step back and just tried to set a standard that we could live with at the time. As many people have reminded us, that was 11 years ago and it is time to look to the future.

DR. BUSH: Donna Bush, Division of Workplace Programs. I am personally grateful that the cutoffs that were established back during the early development of these guidelines were chosen as they were. In fact, they were quite conservative. In the federal system, when you are blazing new trails, indeed it is a positive thing to be conservative, so you can make a case for the concept and get past some initial criticism of why you picked certain numbers.

Yes, we know they are conservative, and yes, we know they work well in the workplace. It is not necessarily based on a pharmacokinetic study, a typical, classic pharmacokinetic study on which the cutoffs are evaluated at this point in time, not exclusively pharmacokinetic. We know a lot about pharmacokinetics and excretion of drugs in urine now, compared to what we did when the cutoffs were first established. We also know an awful lot about passive inhalation issues, based on research work that has been published in peer reviewed scientific journals, that have helped us in taking a giant step forward in changing those cutoffs.

The THC cutoff is reduced from 100 nanograms per milliliter to 50 nanograms per milliliter based on the evolution of immunoassay technologies as well as immunoassay studies and GC mass spec studies on the passive inhalation issue. Similar studies have been done for cocaine, on which we may base future recommendations concerning cutoff changes. Another action which we are proceeding with, which is not pharmacokinetic but actually based on assessment of reality at a given time, is our proceeding forward with a desire to change the opiate cutoffs, to actually increase morphine and codeine cutoffs, but additionally add 6 acetyl morphine, in response to the overwhelming number not of poppy seed positive cases, but rather, urines positive because of the legitimate use of medical products containing codeine. However, the cutoffs initially came about historically, they were sound, and we are proceeding ahead with what we know on pharmacokinetics, as well as reality checks on a regular basis.

DR. JONES: I guess I would add, and it may be that that wording needs to be changed in the list, too, to talk about pharmacokinetic. That is certainly available and subject to recommendation and change. Pharmacokinetic studies, truly as we all know, are very difficult to do with any urine specimen. Urine specimens, I always say that the only thing you know from a positive urine specimen is that the drug was on board at some point in time prior to the collection of the urine specimen. That is some terminology I use. So, maybe pharmacokinetic studies is the wrong term to use because excretion studies require urine volumes and repeated urine volumes and total urine collections and the like.

Those types -- I think I am speaking for the board, for the choice of that, or the idea in that item is the relevance of the cutoff to a subject that has used the drug. We know, at least in urine testing now, and we are learning more and more about the other complementary technologies, about what that window of detection is. That is coming from, maybe not critically designed pharmacokinetic studies, but at least from anecdotal data in many cases, and user population data in many cases, and more and more of the data that are coming from both the regulated and non-regulated arena as to where we are there. I would certainly entertain some alternative wording and I think the board would, too, to those types of things such as the choice of the word, pharmacokinetics in that sense.

MR. SHULTS: Let me take that one step further. One of the things that has struck me this morning in listening to the presentations is, do we have the wagon attached to the right end of the horse. We have basically, in terms of the discussion of the cutoff levels, have admitted that the program has basically been a technology driven process. We have these immunoassays. They are available. They are established. We can support them. So, we will build a program around that. We now have some additional methodologies. We can do hair testing. We can do sweat testing. We can do a variety of other methodologies. We have the technology, so we have some basic principles. Do they meet this set of principles.

Well, my set of principles are different. Mine are not solely a technological set of principles. The principles that we have for urine testing are technical principles. The real principles that I think are visceral principles that we all stand for, and what support drug testing, are number one, this issue of accuracy and reliability. We all understand that. Are the methods accurate and reliable; no false positives; no Pats. We don't have Pats. Our society, despite the fact that we pay a high price in civil liberties surrounding fighting substance abuse issues in this country, we will not tolerate Pats, in terms of being able to make credible allegations that they have been falsely accused.

The second principle is a principle that is integrated into this, but it is the principle of confidentiality, both from a physical perspective in terms of how the information is collected, and informational. What type of information are we getting from that. That is both social and legal norms.

Another principle is the principle of reasonableness, and here is where I think the rubber meets the road.

The principle of reasonableness is, from a legal perspective, is to what degree are these processes, technical processes, minimally intrusive and germane to the goal that we are wishing to accomplish.

That is where we are missing. That is where I think the essential elements are. What are the goals. What are the questions we are wishing to solve here.

There are a number of goals of drug testing. We have talked a lot about deterrence being a fundamental goal. We want the program to be a deterrent process. That makes it sort of a punative process. We also want to identify drug users. That is primarily a safety goal. We also want to be able to make diagnostic decisions based upon drug testing. That is a different goal. That is a clinical medical type of goal with treatment and rehabilitation.

Treatment and rehabilitation and follow up is also a goal of testing, as well as investigatory work. How did the accident -- what is the proximate cause of the accident in Paris.

There are those types of very disparate goals that I think need to be articulated. What I think is missing here is identifying what those goals are. What is the best technology for those particular principles.

It goes beyond the scope, beyond the mandate of what the board is given. I don't want to in any way slight the tremendous amount of work and the important part of the process going on here.

I actually want to commend the board and Dr. Autry for having us go through this process. But again, the question comes down to me, which end of the horse have we attached to the wagon.

I think what we are left with now is that if the mandate of the board is two-fold, as Yale articulated this morning, what is in the public interest. Well, part of that is what are the goals we intend to accomplish. If the other mandate or restriction is how do the current technologies impact, or do they adversely impact current urinalysis testing, then I think it is a fait accompli. I think the only specimen that does not adversely impact the type of urine testing -- I don't make this farcical -- but is a methodology nobody wants. There is no interest group. That would be fecal samples. I mean, in other words, we have got all these other specimens, but wouldn't that be the perfect specimen. But no one is interested in it because nobody wants to do that.

Dennis suggested blood testing which has its own limitations. But it fits the model pretty well and everything else is going to fall out of it. I think this part of the process, I think it is going to be fulfilling these requirements and then asking the basic paradigm question, what is the goal and then what technology is out there to answer it.

DR. JONES: Thank you, sir. Dr. Autry signalled me to do an errata sheet. You were distributed earlier this afternoon a copy of the sheets as we had them filled in. There are corrections on two pages that I would like to make, at least from the board's position at the present point in time when we started this morning.

The first correction appears on page two under item F, requirement number one. I have deleted the Ps under the probability of being satisfied under requirement for special handling for both the urine in the lab and the urine on site. I think it would be appropriate to delete those. Just scratch those two Ps off of there. That is edit number one. Edit number two is on page five of the document distributed to you. There are three deletions.

Item number two, the Is should be deleted from on site and urine. Item number four, the I should be deleted from urine. Ignore the circle that I have on there on item number five. I wrote the wrong one and I can't erase it. My marker is indelible.

Oh, there is an I on your four? I don't know how that got there. That is the way it should be. Where did that copy come from. It is amazing what technology will do. It is an administrative false positive.

That is the way I think it should look from the board's perspective at the present point in time.

DR. AUTRY: Thank you. We have a choice at this point. We can go ahead and have some additional public comments if you like. What we decided to do earlier this morning is, since the public comment and question period is going so well, is instead of having the written questions submitted tomorrow, to have all those sessions be public comment. So, you won't have to sign up for that and we will just take the comments as they come up or questions as they come up. So, we will have a lot more time for that tomorrow. So, your choice is, do you want to spend some more time on public comment this afternoon? Do you want to go ahead and break? It is now 4:35 and we will reconvene, I think, at 8:30 tomorrow morning. See you at 8:30 then. Thank you.

(Whereupon, at 4:35 p.m., the meeting was recessed, to reconvene the following day, September 10, 1997.)