Abstract

In the search for solutions to the "drug epidemic" of the 1980s, the country embraced the idea of drug testing as a way to stop people from using drugs. Drug testing spread most rapidly in the workplace, where by 1989, 20 percent of the workforce faced the prospect of drug testing, and it seemed likely the numbers would continue to grow.

Drug testing could not have moved so firmly into American society without the technical capacity to drug test, and a drug testing industry ready to deliver the necessary products and services. This article maps out that industry and the evolution in technologies that allowed drug testing to go from a drug use monitoring tool, used mostly in drug treatment settings, to a major innovation in social control, used to deter drug use among more general populations. The drug testing industry has grown to meet a growing demand for its products, but it has also worked to accelerate demand by offering a technological solution to a problem it also helps to define. The presence of this now well established industry makes it less likely that drug testing will fade in popularity or that the forces in opposition to drug testing will make headway in reversing the trend of increasing use.

Introduction

Social control strategies and technologies are not just part of a society's ideas, habits and practices. They are also economic and social institutions which create a world for the people they employ and process and which have consequences for the larger society. Thus, for example, jails and prisons directly employ hundreds of thousands of personnel, command huge budgets, and shape the nature, images, and extent of incarceration in American society. They also employ tens of thousands of others in construction, food and medical services, and vocational and academic education. Over time, jails and prisons have become a vested interest in American society as well as a cultural institution and mindset.

The war on drugs, which encompasses a number of social control strategies and technologies, has supported, reinforced, and sometimes changed existing economic and social institutions devoted to social control and has given rise to a number of new ones. This paper focuses on the industry and economy of drug testing, one facet of the war on drugs. Drug testing has "taken off' since 1986 when, as part of an emerging "demand side" drug reduction strategy, the President's Commission on Organized Crime recommended extensive public and private sector drug testing and President Reagan's executive order mandated drug testing for federal employees in "sensitive" positions. Drug testing is now being implemented, in one form or another, by the armed forces, many large companies, local governments, professional sports leagues, criminal justice agencies, insurance companies, and even by parents, using home testing kits. Perhaps as many as ten million Americans per year are subjected to urinalysis to detect the use of illegal drugs. The purpose of this paper is to describe and take stock of the drug testing industry and economy that make modern-day mass drug testing possible.

Drug Testing Technologies

The implementation of societal—wide mass drug screening, for purposes of social control, would not have been possible without inexpensive technologies, sensitive enough to detect low concentrations of drugs, and to do so with a negligible rate of false-positives. By the early 1980s, such technologies had emerged, the consequence of two decades of research and experimentation. Their evolution is intertwined with advances in bio-medical technology, since many scientific methods of urinalysis for illicit substances are also used to diagnose illness and chemical imbalances and to monitor the concentration of patients' medication.

Modern drug testing technologies were built upon these medical techniques, but their development also required a purposeful effort to create techniques especially suited to social control. The research that accomplished this was conducted by private pharmaceutical companies that stood to profit from their use, and by independent researchers who had received grants from the National Institute on Drug Abuse (NIDA),1 a federal agency charged with drug prevention, and committed to the widespread implementation of drug testing, especially in employment.

As drug testing technologies improved, drug testing programs expanded, to reach new target groups. Prior to the 1980s, the most extensive testing programs covered criminal defendants, probationers, and patients in methadone maintenance clinics, all groups where precision was not necessary because several failed tests were required before negative consequences occurred.2 The kind of drug tests used were not sensitive enough to detect the casual users who were the target of employment testing programs, nor reliable enough to be used as a basis for punishment, especially in the workplace. Thus, if employers had been uninterested in testing, there would have been little incentive for pharmaceutical companies to invest in research on advanced technologies.

The most common technology used in early drug testing programs was thin-layer chromatography (TLC), a urinalysis technique that applies solvents to a highly concentrated urine specimen that has been placed on a gelcoated glass plate. When the plate is tipped, substances in the urine separate and migrate; after approximately 30 minutes, with the application of colored sprays, "drug spots" emerge that trained technicians can compare to known standards.3

TLC could not have been used for the mass drug screenings of the 1980s for a number of reasons. Most importantly, TLC is not sensitive enough to detect drugs in low concentrations, and since marijuana metabolites always appear in low concentrations, it could not detect marijuana at all. It is not, therefore, an adequate technology for identifying or deterring the casual drug user who is the prime target of today's mass testing programs.4 In addition, TLC results are not easy to defend in court because they require a good deal of subjective interpretation, and the glass plates cannot easily be preserved for later use as evidence.

A more sensitive and reliable form of liquid or gas chromatography urinalysis was available in the 1960s, but the high cost of the machinery and the need for skilled technicians to operate it, made it too expensive for widespread use.5 Gas chromatography was used, however, in some small-scale testing programs where precision was particularly important, for example, in the testing of Olympic medal winners.6

These more sophisticated chromatography techniques operate on principles similar to those of TLC, but rely on finely calibrated machines to "read" the migration times of substances in the urine. The urine sample is first concentrated, then vaporized, and sent through a long glass column, filled with gas or a liquid medium. Drug molecules in the urine separate out and move through the column at a rate that is specific for each drug, and as they exit the column, the machine prints out their "retention times," by which they can be identified.

Although both liquid and gas chromatography techniques continue to be used, gas chromatography is more popular, and it is now often combined with a mass spectrometry machine, creating what is referred to as the GC/MS. The mass spectrometry works by "bombarding' drug particles with electrons as they leave the GC ma-chine, causing the particles to fragment and scatter, and creating a pattern that is unique for each drug. The machine then matches the pattern to the known "mass spectrum" of the drug initially identified by the gas chro-matography. The results from GC/MS are highly reliable.

While GC/MS itself is too expensive to be used as a first screen in mass testing programs, it may be affordable when used to confirm positives by less expensive technolo-gies, especially if the rate of positives is fairly low. This is the use to which GC/MS is now largely put, to confirm positives obtained by an immunoassay screen, the newest technological breakthrough in drug testing, and the one that made mass drug testing feasible.

The immunoassay test was developed by medical researchers who were dissatisfied with TLC's inability to identify some substances of interest to doctors, such as insulin. It earned Rosalyn Yalow the Nobel Prize for medicine and physiology in 1977.7 Even before the Prize was awarded, some criminal justice agencies, drug treatment programs, and the military in Vietnam were using immunoassays to investigate illegal drug use.

Immunoassay technology works very differently from both TLC and GC/MS. A sample of the drug being tested for is put into the urine, but with a "tag" attached to it. In one popular version of the test, EMIT (enzyme multiplied immunoassay test), the drug is tagged with an enzyme; in another, RIA (radioimmunoassay), a radioactive iodine is used to tag the drug.8 Protein—based antigens of the drug, to which drug molecules or drug metabolites normally attach, are also added to the urine.9 If there are no drugs already in the urine, the tagged drug and the antigens "match up;" if the urine contains drugs, those molecules bond some of the antigens, leaving some of the tagged drug unattached. A third substance added to the urine acts as a receptor for these unattached drug molecules, making them easy to "read" when a light beam is shone through the urine solution.

For drug control purposes, immunoassay technol-ogy has some important advantages. The process can (and has been) highly automated, making it cheaper and faster than TLC or GC/MS. In 1973, a single technician could complete between 200 and 225 immunoassay screens per day, compared to less than 100 tests using TLC, and 35 using GC/MS.10 Currently, the most sophisticated immunoassay machines can analyze up to 7000 urine samples per hour," and further technological advances are likely. Some labs now offer immunoassay screens for slightly over $1 per drug tested for.

Another advantage of immunoassay is that the results are less subjective than TLC, and because the amount of tagged drug left unattached can be measured, they can be expressed semi-quantitatively, making them more acceptable as evidence in court.12 Immunoassays can also detect marijuana; while this is not critical to criminal justice and drug treatment testing programs, where the primary purpose is to detect heroin and cocaine, it is critical to workplace testing which aims to identify a wide range of drugs.
Immunoassays are considerably more sensitive than other technologies, making them particularly useful for the detection of casual users, but the increased sensitivity also creates problems. Because the tagged drug and the antigens seldom interact perfectly, some drug molecules may be left unattached, even when no other drug is present. This could possibly result in a positive reading. The test might also wrongly identify as positive those who "passively inhale" drugs smoked by others in their presence. To prevent these "false-positives," testers must set "cut-off points" below which traces of the drug will be ignored. However, this also means that real drug users with only traces left in their urine will not be identified; that is, they will be false-negatives. Consequently, the full sensitivity of immunoassay cannot be capitalized on, but even so, it can identify many users that TLC misses.

Another problem with the immunoassay test is "cross-reactivity" — drugs that are closely related chemi-cally may interact similarly with the same antigen, gener-ating another type of false-positive reading. For example, a legal drug like codeine may bond to the opiate antigen, leaving some of the tagged opiates unbonded, and creating the false impression that the individual had been taking opiates. Non-drug substances like poppy seeds, taken in sufficient amounts, can produce a similar effect, as can some enzymes produced by the human body.13 This is why it is important that immunoassay results be confirmed by GC/MS, especially if they are to be used to apply sanctions to drug-positives; virtually all current testing programs, except those in criminal justice, automatically confirm all immunoassay positives with GC/MS.

Thus, the early 1980s researchers had developed an affordable and reliable drug testing technology capable of deployment on a truly massive scale. When, in 1986, President Reagan ordered the testing of federal workers, and urged business leaders to begin testing their employees, the technological capacity to do so was well in place. After some important improvements were made in the way in which samples were handled, the results could be confidently relied upon and defended in court if necessary.

Mapping the Drug Testing Industry

The key institutions of the drug testing industry are the laboratories that perform the urinalyses and the pharmaceutical companies that supply the equipment and chemicals. However, a number of other less visible businesses play a role in facilitating the design and implemen-tation of drug testing programs; still other companies, although not primarily a part of the urine testing industry, provide goods and services to the industry. Thus, it is appropriate to speak both of a drug testing "industry" and a drug testing "economy." Drug testing is often called a $300 million industry, but this figure refers only to the equipment and chemicals necessary for the urinalysis itself." It does not include all revenues within the drug testing industry, much less the drug testing economy. The total amount of money spent by all consumers for drug testing services has not, to our knowledge, even been estimated.

The Pharmaceutical Companies

The pharmaceutical industry is a major multi-national industry, involved in developing products for urinalysis long before the current surge in testing for illicit drugs. Today, dozens of pharmaceutical firms compete for a share of the drug testing market.15 The three largest are Syva, Roche Diagnostic Systems (a division of Hoffman LaRoche), and Abbott Laboratories, together accounting for over 75 percent of the business.16 Each company specializes in a slightly different version of the standard urinalysis technologies, and offers all of the chemicals and hardware necessary to run their tests. Some firms, like Abbott, manufacture the immunoassay machines as well, but more commonly, pharmaceutical companies contract with manufacturers of medical equipment to provide machines geared specifically to their chemical products.

Pharmaceutical companies make most of their sales directly to the testing laboratories, but occasionally they sell products to organizations and firms that want to do on-site testing. As the immunoassay process has become increasingly simplified and automated, it is possible for organizations to do their own testing; on-site testing is particularly useful when immediate results are desired. The District of Columbia Pre-Trial Services Agency (PSA), for example, does all its own testing, and is able to provide drug use information on criminal defen-dants to the court within an hour, if necessary. They contract with Roche, which provides all of the necessary chemicals, periodic training for PSA staff, and free use of an immunoassay machine. GC/MS testing cannot be done on site, and organizations that do on-site immunoassay screens are urged by the pharmaceutical companies to confirm positive test results with GC/MS.

Recently, pharmaceutical companies have begun to market on-site drug testing kits that do not require any machinery at all. They operate much like a home-preg-nancy test. These include Keystone Diagnostics' KDI Quick Test, Abbotes TDx and ADx Systems, and Environ-mental Diagnostic's EZ-SCREEN, all using a modified TLC technology. They generally sell for under $10, and have been marketed primarily to businesses doing a small volume of testing. These kits may eventually be marketed vigorously to the public, giving parents the capacity to determine if their children are using drugs, but the FDA has been slow to give its approval.17

The Testing Laboratories

Historically, both clinical and forensic laboratories have carried out urinalyses to detect the use of illicit drugs. However, prior to the 1980s, neither was well suited to large volume drug testing. Clinical laboratories are mainly medical testing facilities, not accustomed to maintaining tight chain of custody procedures or to de-fending their results in court. Forensic labs, which conduct tests largely for the police, are used to maintaining tight chain of custody procedures, but they traditionally oper-ated on a very small scale and could not handle the volume of urine tests associated with the demand for mass screen-ing drug testing programs in the 1980s.

In the 1980s, as the demand for drug testing programs mushroomed, there emerged a new type of laboratory, capable of processing the volume typical of clinical labs, but with the controls typical of forensic labs. Indeed, many of these new labs were formerly small-volume forensic labs or low-control clinical labs, each adapting in order to capitalize on the growing demand for drug testing services. New labs opened as well, specializ-ing solely in drug testing, or even a special form of drug testing, such as for insurance companies.18 Other labs continue to do medical testing or environmental testing, along with testing for illicit drugs.

The drug testing laboratory business remains geographically decentralized, with hundreds of labs around the country. Increasingly, however, the small laboratories are being bought up by a few large laboratory companies; SmithKline Beckman, for example, now owns about 30 laboratories that do drug testing. By 1989, five laboratory companies — SmithKline Beckman, Hoffman-LaRoche, Metpath, CompuChem, and Damon — domi-nated the drug testing business and together processed over a million urine samples per month.13 The large laboratories are highly automated and computerized, and frequently run 24 hours per day.

There is some diffeientiation among drug testing laboratories. Each lab offers a somewhat different combi-nation of tests, depending on which pharmaceutical companies they are linked to, and a different range of support services, at different prices. There are "up-scale" laboratories, catering largely to big business, and offering the most sophisticated confirmation process, highly trained staffs, urine collection and transport services, rigorous chain of custody procedures, and expert testimony in any legal actions that arise. "Down-scale" laboratories, which early out testing for criminal justice agencies and drug treatment programs, may only carry out immunoas-say screens, without confirmation, and offer few, if any, support services.

Improving Laboratory Performance

A study of laboratories used for drug testing from 1972 t6 1981 found an extraordinarily high failure rate,n leading many courts to find against drug testing programs and causing many employers to be wary of starting pro-grams. It also led some states to pass restrictive drug testing legislation. Just as basic drug testing technologies had to improve to support the mass drug testing screening programs that have emerged since the mid 1980s, so too, laboratory reliability had to improve to the point where it could stand up in court and pass the scrutiny of skeptical federal and state legislatures. Without this high level of reliability, private and public employers could not afford to discipline or dismiss employees identified as drug users.

The large laboratories have been in an especially good position to improve chain of custody procedures, adopting the same kind of procedures that the forensic laboratories had long used in receiving, analyzing, and storing drug samples in criminal cases. The procedures even reach back to the collection site, where urine must be placed in bottles supplied by the laboratory, and under procedures that prevent tampering by the person providing the urine, or anyone else. Each urine sample is accompanied by a chain of custody form, with a number to match the number on the bottle of urine, and everyone who then releases or receives the sample is required to sign and date the form. The urine is then transported by courier or mail to the lab, where the chain of custody form is checked, and the sample is entered into the lab's internal security system, with computerized bar code labels. The chances of human error are very slim.

Many private sector drug testing plans include chain of custody procedures and they are required in the federal governmenes policies for testing federal workers.n Federal law also requires that all laboratories contracting with federal agencies be certified by NIDA. As of March 1990, 44 labs had completed the process.n Even laboratories that are not obligated to become certified under federal guidelines may choose to do so, as clients demand near 100 percent reliability; in addition to NIDA, many state or city health departments offer certification as does the American Board of Clinical Chemistry and the American Board of Forensic Toxicology.

Part of every certification process is the laboratory's participation in an ongoing quality control program, and this has created a market niche for companies which monitor laboratory proficiency. A number of professional organizations, such as the College of American Pathologists and the American Association for Clinical Chemistry, contract with labs to provide proficiency testing. They regularly send in samples of "clean" and "dirty" urine, and monitor the results. Some clients, desiring assurance as to a lab's reliability, may contract with a private company to conduct site visits of the laboratory and do additional blind proficiency tests.

By the end of the 1980s, the drug testing laboratories had vastly improved their reliability, and were regularly producing results with only a small portion of false negatives, and virtually no false positives.23 The courts began to accept drug testing results as highly reliable, something like breathalyzer tests in drunk driving cases, or radar in speeding cases. Drug testing was graduating from an experimental to an established technology, part of the nation's social control landscape.

Drug Testing Information and Consultation

During the 1980s, the nation's employers have been deluged with information about drug abuse in the workplace. Claims about the dangers and costs of drug users to business are widely quoted in business journals, at conferences, in the popular media, and in materials produced by the Partnership for a Drug Free America. Even if President Reagan had not exhorted business leaders to join the war on drugs, they might have done so for perceived business reasons. Of several possible approaches employers might take to produce a drug free workplace, drug testing makes the largest promise, and seems to have had the largest appeal.24

Many businesses lack in-house expertise to design and implement an employee drug testing program, and there are a number of places to which they might turn for assistance. NIDA, through its Office of Workplace Initiatives, has made a commitment to provide information and consultation to businesses wanting to develop drug abuse programs. In January 1987 it established the toll-free number, HELPLINE, at which operators answer about 600 calls per month, mostly from small businesses.25 It sends out materials to callers, including a number of NIDA publications, lists of certified laboratories, and videos on workplace drug abuse and drug testing.

Drug testing has also proved to be fertile ground for the information mill, with a wide variety of products that cater to businesses' desire to drug test. The Conference Board, the American Management Association, Professional Management Seminars and other such organizations run conferences featuring "drug abuse experts," many of whom work in the drug testing industry. Many videos on drug testing are now on the market, for example the ABA's "Drug Testing in the Workplace." There has also been a proliferation of manuals, like the Conference Board's Corporate Strategies for Controlling Substance Abuse, Business Law Inc.'s The Legal Aspects of Substance Abuse in the Workplace, and Clark Boardman's Drug Testing Legal Manual.

For more individualized attention, businesses can hire consultants who will help them assess their drug problem design and implement an appropriate response. Some of the best known consultants in the field are Robert DuPont, Peter Bensinger, and Paul Mulloy, all former officials in the federal drug war establishment. Some consulting firms, like Diagnostic Dimensions, are subsidiaries of larger companies that include laboratories and pharmaceuticals.29 There is, then, some degree of vertical integration in the drug testing business, and when consultants encourage businesses to drug test, laboratories and pharmaceutical companies linked to them benefit as well.

Consultants may actively recruit clients by approaching CEOs or personnel directors, but there seems to be a ready demand for their services by companies that have already decided to drug test, and need help with implementations. These consultants know how to design and implement a drug testing program that will cause the least amount of disruption, and generate the least employee resistance. Whether they help create the demand for drug testing, or simply respond to it, they benefit from the steady increase in utilization of this new form of social control.

The Economy of Drug Testing

The drug testing economy is larger than the pharmaceutical companies, laboratories, and consultants which comprise the core of the drug testing industry. Each of these has to contract with other industries for numerous goods and services. For example, most pharmaceutical companies purchase sophisticated drug testing machinery from manufacturers of medical equipment like Hitachi, Instrumentation Laboratories, or Olympus. Labs contract with courier and mail services to transport urine speci-mens from collection sites to labs; with plastics' manufac-turers to supply millions of plastic bottles and plastic thermometer strips; and with office supply companies to provide printed chain of custody forms and the machine-readable identification strips that are attached to the millions of urine samples collected each year.

By far the largest "outside" service which drug testing has made necessary is drug treatment. Criminal Justice testing programs refer drug abusing defendants to government-run treatment programs that employ large numbers of doctors and other medical personnel, social workers, and drug counselors. The vast majority of public and private employers who drug test offer employees who test positive for the first time an opportunity for rehabilitation, and refer employees first to a company Employee Assistance Program (EAP), and perhaps to a private substance abuse treatment clinic, paid for through the employer's health insurance plan. Private treatment centers range from the Spartan to the plush, with the latter costing as much as $1,000 per day. The number of privately-owned substance-abuse treatment clinics -in the country almost doubled from 3018 in 1982 to 5360 in 1987, and during the same period, drug-related hospital admittances went from approximately 173,000 to 260,000.27 Drug testing has certainly helped to fuel t'his growth.

Employment drug testing has also been a boon to EAPs. First begun in the 1940s, as counseling services for employed alcoholics,28 EAPs branched out into other problem areas (drugs, divorce, family problems) and continued to grow in number throughout the 1970s and 1980s. By 1987, as many as 10,000 programs were in place, especially in Fortune 500 companies.29 Drug testing is not responsible for this growth trend, but it has contributed to it. A number of EAP providers originally opposed drug testing because they feared companies would substitute testing for EAPs.30 However, drug testing is often used as a tool for channeling employees into EAPs, an approach advocated by EAP consultants like Dale Masi, who encourages employers to rehabilitate rather than punish drug-using employees.31 Many workplace drug testing policies, especially those established by collective bargaining agreements require the presence of an EAP as a prerequisite to drug testing.

The Political and Social Impacts of the Drug Testing Industry and Economy

It would be simple-minded to argue that the rapid growth of drug testing in the United States can be attributed to a drug testing industry and economy. More likely, drug testing has been driven by the same factors that are responsible for fueling and sustaining the war on drugs. Nevertheless, it is possible to see an independent significance for the drug testing industry.

It was necessary for there to be a drug testing industry for mass drug testing to take place. Drug testing could not have developed the way it has just because, as an idea, it had specialized or popular appeal. There had to be a technology in place that met the requirements of reliability and cost, and a complex of institutions and services that could provide drug testing programs accept-able to a wide range of organizations in American society. How this industry evolved, through private efforts and government encouragement, is an important example of how social control is created in a modern society.

It is already possible to see how the drug industry reinforces the very forces that gave rise to its existence. The natural search for growth and profit leads the different firms that make up the industry to make their products and services as widely desirable as possible, and to that end, they often participate actively in promoting a particular view of "the drug problem" — one that includes casual and middle class users, and one that threatens the safety and productivity of the American workplace.

The large drug testing laboratories promote themselves to businesses with advertisements and direct mailings that stress the costs of drug abuse to employers. An ad-by the PDLA lab in New Jersey, for example, warns that "there is a crisis in the business community," and "no one is immune — no company large or small is unaffected." It identifies workplace drug abuse as "a $50 billion a year problem," and offers the best tool available to eradicate it — "a sound, responsible diagnostic testing program." The large pharmaceutical companies advertise directly to businesses as well, and help regional business organizations sponsor conferences and workshops on drug abuse and drug testing.32

By helping to define the nature of the drug problem, the drug testing industry automatically increases demand for its products. And in the repertoire of social control strategies in the United States, it is in the unusual position of having an unlimited capacity for deploying its services. Compare this to jails and prisons which are terribly constrained by resource limitations, and cannot be expanded beyond a certain point. Drug testing, however, can be expanded practically infinitely, with much of its cost born by the private employment sector. A comparatively small number of laboratories running around the clock could theoretically test on a regular basis much of the American population for evidence of illegal drug use. The logic of "zero-tolerance" provides a justification for such extensive use, and the drug testing industry makes it possible.

Lynn Zimmer and James B. Jacobs are with the Center for Research in Crime and Justice, New York University School of Law, 40 Washington Square South, New York, N.Y. 10012. (212) 998-6127.

Footnotes

1 See "Foreword" by NIDA Director Charles Schuster in Urine Testing for Drugs of Abuse, NIDA Research Monograph 73 (1986).

2 In fact, the tests were used more as aid to treatment than as a stimulus for discipline. For example, probationers who failed their drug tests could have their probation revoked, but it was more likely that probation officers would use the positive results to force them into treatment.

3 For this section on TLC and the ones that follow on the other drug testing technologies, we have relied upon the following literature: Lawrence Mikke and Maria Hewitt, "Accuracy and Reliability of Urine Drug Tests," University of Kansas Law Review. 36, 4: 641-81 (1988); Richard Hawks and C. Nora Chiang (eds.), Urine Testing for Drugs of Abuse, Rockville, Md: National Institute on Drug Abuse (1986); 'Drug Abuse in the Workplace: Prevention and Control," Clinical Chemistry 33, 11 (B) (1987); Don Catlin, A Guide to Urine Testing for Drugs of Abuse, Washington, D.C.: Special Action Office for Drug Abuse Prevention (1973).

4 Drug testing is hardly necessary for identifying heavy or addicted drug users. What differentiates the "drug war" of the 1980s from earlier drug control policies and initiatives is that the casual or recreational drug user is defined as an important target of drug control strategies. See National Drug Control Policy Strategy, Office of National Drug Control Policy, 1989.

5 Depending on the number of drugs tested for, processing a single urine specimen could cost anywhere from $25 to $100. See Bryan Finkle, "Drug—Analysis Technology: Overview and State of the Art," Clinical Chemistry 33, 11 (B), 1987, p. 14B.

6 See Rudy Baum, "Drug Testing is Integral Part of Winter Olympic Games," Chemical & Engineering News, 66 (1988): 23- 25.

7 The Who's Who of Nobel Prize Winners, B. Schlessinger and J. Schlessinger (eds.), Phoenix, Ariz.: Chyx Press (1986), p. 118.

8 There are also two less commonly used immunoassays: the free radical assay technology (PRAT) and the hemagglutinatim (HI).

9 These antigens are produced by removing antibodies from the blood of laboratory animals who had been injected with the drug, and then "bled."

10 See Catlin, note 3.

11 The cost of these machines varies with their speed. Hitachi's model 736-50 runs 7,200 test per hour and costs over $500,000 as compared with the Olympus AU--5000 that can run 4,000 test per hour and costs $350,000.

12 Results are usually presented as positive within a certain range rather than as a specific amount because the latter would be much harder to defend. It would be possible, using this technology, to specify the concentration of each drug in the urine, but this information would not reveal the amount of drug t,aken, when it was taken, or the degree of impairment at the time of the test.

13 For example, a liver enzyme that can be created in response to a number of kidney, bladder, or liver disorders will sometimes give a false-positive reading for marijuana. See Kevin Zeese, "Marijuana Urinalysis Tests,"Drug Law Report 1, 3: 25-31 (1983).

14 The latest figure available, calculated by the Robert Frost Report, a quarterly newsletter for the clinical lab industry published in Plymouth Meeting, PA, is $343 million.

15 For a complete guide, see the 1989 Medical Healthcare . Marketplace Guide, published by the International Bio-Medical Information Service.

16 See Patricia Winters, "Drug Test Prevents False Negatives," Advertising Age, May 4, 1987, p. 32.

17 See Joe Agnew, a$220 Million Market Seen by 1991 for Drug-Abuse Testing," Marketing News 20, 24:1.

18 The Prudential owns GIB Labs in Newark, NJ., which performs drug tests for policy applicants. Osborne Labs and Home Office Reference Laboratories, both located in Shawnee Mission, Kan., are exclusively devoted to insurance company clients.

19 See Jacob Clark, "Drug Testing: The Weeding Out Process," Law Enforcement News 16, 311: 1-2 (1990); Sandra Ackerman, "Science Observer; Drug Testing: The State of the Art," American Scientist 77: 19-23 (1989).

20 See Hugh Hansen et al., "Crisis in Drug Testing," Journal of the American Medical Association 253, 16: 2382-87.

21 See Federal Register, Vol. 53 No. 69 (11970-11989), especially Subpart B, 2.2 (f).

22 See Federal Register, Vol. 55 No. 46 (8896), March 8, 1990.

23 See, Clinical Chemistry 33: 1683-86 (1987); and 35: 891-94 (1980).

24 59.8 percent of firms that employ 5000 or more employees now have some kind of drug testing program, most commonly at the pre-employment stage. See Survey of Employer Anti-Drug Programs, U.S. Department of Labor, Bureau of Labor Statistics, 1989.
25 See NIDA Capsules, "NIDA's Drug-free Workplace Helpline," 1988, p. 2.
26 Diagnostic Dimensions (the consulting arm), Roche Diagnos-tic Systems (the pharmaceutical arm), and Roche Biomedical Labs are all subsidiaries of Roche Diagnostic Division. See Agnew, note 17.

27 This is NIDA data, reported in The New York Times, Oct. 2, 1989, p. B10.

28 See Wayne Lindstrom, "The Promise of the Employees Assistance Program: Well-Being in Our Institutions," Employee Assistance Quarterly 1, 1: 89-92.

29 See BNA Special Report, Employee Assistance Programs: Benefits, Problems, and Prospects, 1987.

30 See BNA Special Report, Employee Assistance Programs: Benefits, Problems, and Prospects, 1987, chap. X, "Drug Testing and EAPs."

31 See Dale Masi and Laura Burns, "Urinalysis Testing and EAPs,"EAP Digest, September/October 1986: 37-43.

32 The New York Times, Nov. 13, 1989, p. D4.