Home / Drug Testing in the Workplace

Drug Testing in the Workplace


The use of tests in the workforce gives rise to a serious conflict of rights. On the one hand, employers have the right to determine the competency of their employees for the tasks relevant to the job; on the other, the employees have a right to privacy. Neither of these sets of rights are indefeasible; a compromise must be reached in cases where these rights conflict. The BCCLA has taken positions on various types of tests used in the workforce: polygraphs(1), psychological tests such as intelligence or aptitude tests(2), and graphology(3). In each of these cases, not only had the above-mentioned conflict between privacy and the legitimate interest to gain information to be resolved, but two further questions had to be addressed: whether the test really measures what it purports to measure, and whether the test is the least onerous way of determining the information about the employee.

Drug testing raises these considerations again. However, the empirical facts involved in testing for drugs are different than in the other tests we have considered. In the cases of graphology and the polygraph in the workplace, the testing instruments are clearly so unreliable that there is no justification for an employee to give up his or her right to privacy. However, as we shall demonstrate in Part II below, the tests generally used to determine the presence of drug metabolites have been demonstrated, by proper scientific testing, to measure what they purport to measure. Thus, they cannot be universally dismissed as worthless. The problems with them are much more subtle: first of all, what they purport to measure is the presence of metabolites of psychoactive drugs, rather than the presence of drugs which are actually causing impairment at the time the test is taken. Therefore, they are not reliable as a measure of what the employer has a right to know about.

Second, we shall examine the tests which are in common use at present, and we shall see that despite the fact that some of them have a high (up to 95%) accuracy when administered by highly competent and motivated researchers, in the resolve the conflict between the employees’ right to privacy and the employers’ right (and even in some cases the duty) to obtain relevant information about the employee.

In Part 1 we consider the issues surrounding privacy. In Part II we consider whether the tests available have a high enough degree of validity to make it reasonable for an employee to give up her right to privacy. It will emerge from that section that the types of tests presently available are not accurate enough for this purpose. However, some might argue that the threats to safety and profits that result from what many perceive to be a ” drug epidemic” may justify employees’ sacrifice of their right to privacy in this case, despite the fact that such a sacrifice would not be reasonable in other areas. Thus, in Section Ill we examine this argument, and conclude that it fails. In Part IV we consider the alternatives to drug testing: monitoring of employees’ performance, and, if drugs are a problem in the particular workplace, the institution of an Employee Assistance Program (EAP). It will be argued there that these alternatives are not only more likely to succeed in real world of mass screening, where labs will be guaranteeing safety and productivity, which are the routinely doing hundreds of tests, accuracy levels legitimate concerns of the employer, they are far fall far short of this, due to such factors as less onerous and constitute much less of an boredom of the researchers. So, the issue of drug invasion of privacy for the employee. In Part V testing is far more complicated than the earlier we examine the legal questions surrounding drug issues we have dealt with; and the reader must be testing, and in Part VI draw our conclusions. We prepared to wade through a fair bit of empirical shall be maintaining that mandatory drug testing data before getting to the part where we apply the is entirely unjustified in the overwhelming majority line of reasoning developed in the earlier briefs to of workplaces for which it has been proposed.


I. Privacy

It is important to keep in mind what is involved in being submitted to urinalysis. A person is required to urinate into a bottle, or some other container, and in order for there to be certainty with respect to the source, observation of the act of urinating is usually required. There can be little question but that requiring a person to participate in such an act invades his privacy and is a denial of his personal freedom. The personal freedom of the individual and his privacy should not be invaded except for the most compelling reasons.

Requiring a person to provide a urine sample for analysis is basically a “search” of an individual that, without consent, would be considered an assault or a trespass. Requiring an employee to either submit to an urinalysis or be disciplined or dismissed infringes on what would otherwise be considered the ordinary civil liberties of an employee.

The right of a person to freedom from search is well-entrenched in our legal system and permitted solely as an incident of arrest, and even then, only when specific requirements have been met and there is a clear statutory authority for the search.

Where a search is done in response to reasonable grounds for believing that a criminal offence has been committed, then in the interests of detection of crime that can take priority over personal liberties. Where, however, this is not the case and there is not clear evidence of consent or necessity, then it cannot be the law that an individual, be he employee or not, can have his personal liberties and his personal freedoms violated by an employer.

In the absence of any concrete evidence of wrongdoing, it is unwarranted to discipline an employee for insisting on his right to privacy. To hold otherwise would rob that right of all meaning.

The preservation of the right of privacy ought to be jealously preserved. In the next section, we look at the conditions which would provide a prima facie justification for asking an employee to give up his or her right.

“The need to know”

First of all, this right of privacy should not be invaded except where there is the clearest provision in the contract of employment or the collective agreement to the contrary. Second, the taking of the drugs tested for must present a clear and present danger to the safety of others, or their property because of the nature of the employer’s business, or result in a serious loss of income to the employer. There can be no right in the employer to know whether or not the employee is engaged in drug use, unless the use of drugs is in some way connected to the work of the employee.

However, the satisfaction of these two conditions provides only a prima facie justification for a urinalysis test. A further condition necessary for justification of this test is that the test be accurate enough to provide the required information. In Part II we shall demonstrate that drug testing does not satisfy this requirement. Part of the reason that urinalysis fails this accuracy condition is that even the most accurate tests fail to discriminate between a person who is presently under the influence of the drug in question, and a person who has used the drug in the recent past but who is not now under its influence. Therefore, the argument that an employer has the duty to test for drugs in order to ensure the safety of others collapses. We shall address this point in Part III.


II. The Validity of the Tests

What is the current state of the art regarding the sensitivity, accuracy, specificity and reliability of the bio-assays under discussion?

Although Zeese (1987) acknowledges steady overall improvement over the last 15 years in 99 accuracy, precision, sensitivity and specificity of toxicological testing,” he identifies “numerous problems with testing bodily fluids for indications of drug use.” He is far from alone in holding this opinion.

Reviewing the state of the art in urine screening procedures, George B. Lundberg of the U. of Southern California Medical school writes in a lead editorial in the Journal of the American Medical Association.:

While the theory of drug testing has improved greatly, unfortunately, the practice has not Improved nearly as much. Survey data from the Centers for Disease Control published in J.A.M.A. demonstrated woefully poor performance by many laboratories on blind testing. I have reviewed urine toxicology survey data from the College of American Pathologists from 1984, 1985, and 1986. Hundreds of laboratories (284 to 600) participated in these proficiency programs. The results of these surveys are better, with laboratory accuracy of 75% to 95% for many drugs, but for some analytes and some common methods, accuracies in the 30% to 60% range were still frequent. Most disturbing of all, despite the fact that any positive result must be confirmed by a technically different method has been standard operating procedure for many years for lab personnel, false-positive reports of drugs like morphine, codeine, cannabinoids, phencyclidine, benzylecgonine, secobarbital, and amphetamines continue to appear in distressingly large numbers.

It should be noted that laboratories participating in this program are among the very best because they undergo such proficiency testing for the sake of laboratory improvement. However, these surveys by the College of American Pathologists were not blind so one could anticipate that the laboratory performance would be at its very best for these specimens.

No matter what the test – be it mammography or urine cocaine – a false-positive is always a problem. But there is a tremendous difference between reporting a false positive urine acetone to a patient’s physician and a false-positive urine morphine to a person’s employer or probation officer.(4)

Error rates reflect both inherent deficiencies in the technologies and mistakes in handling and interpreting the results. Published results of 99 blind” tests of lab accuracy by relevant governmental agencies and professional associations (using known urine samples) include false positive rates as high as 66% and false negatives as high as 100%.(5)

In the U.S., the “Einsel Commission” was struck to review allegations that a massive number of false accusations had resulted from errors by the military urinalysis program – in 1982 and 1983, for instance, the sources of 90,000 urine samples could not be verified and 46,000 soldiers were told they could appeal disciplinary action levelled on the basis of these urinalyses. E. J. Imwinkelreid (consultant to the Einsel Commission and Professor of Law at the University of California, Davis) writes(6):

There are many ways to test urine samples for evidence of drug use, but most laboratories employ at least one of three basic procedures: imrnunoassay, thin-layer chromatography, and gas chromatography combined with mass spectrometry. The tests vary widely in cost and reliability, the least reliable [i.e., immunoassay] being the most widely used.

While tests such as the Syva Co.’s EMIT (Enzyme Multiplied Immunoassay technique) are extremely sensitive, they have been found in many independent quality assessments to produce an unacceptably high level of false positives under real-life conditions (i.e., indicating presence of drugs that could not be confirmed by more specific methods). There have been problems of false negatives as well and while these are of less concern to those tested, they should disturb employers who order the tests. ‘ne magnitude of the disturbing failure rates in several methodologically rigorous quality-control field tests and reasons for them are discussed in several places, e.g., Zeese (1987), Lundberg (1986, see note 1), Morgan (1984)(7) Dubowski (1987)(8), Hansen et al. (1985)(9) and a recent Research Monograph of the U.S. National Institute on Drug Abuse.(10)

It should be emphasized that tests such as the EMIT detect metabolises (usually-inert breakdown products of formerly psychoactive chemical agents) rather than actual blood levels of the agents themselves. For this reason, they will react positively to a number of related but innocent substances that might be in the urine.

The reasons immunoassay alone cannot be relied upon to support the grave decisions entailed are:

(a) Cross Reactivity

Numerous over-the-counter medications, a variety of legally prescribed medicines, and even some foods, can produce false positives (e.g., nonprescription decongestants such as Contac and Sudafed, certain diet pills, and asthma medications can register as amphetamine on these tests and consuming poppy seeds in baked goods has resulted in accusations of opiate abuse based on immunoassay techniques). Similarly, codeine in over-the-counter pain preparations, and dextromethorphan in non-prescription cough elixirs will test positively for opiates with these tests. Users of nothing more sinister than certain herbal teas and prescribed antibiotics have been accused of imbibing cocaine (O’Keefe, op. cit., p. 35).

Zeese (1987) cites letters from the Syva company to its customers advising that anti-nflammatory drugs such as Ibuprofen, Fenoprofen, and Naproxin will trigger false positives onsome of its EMIT tests for marijuana metabolises. These are the active ingredients in some of the most widely-used non-prescription and prescription medications in North America: e.g., Advil, Motrin, Anaprox, etc. Only the EMIT screen for cocaine metabolises has the desired specificity.

(b) Interfering Substances

According to U.S. Food and Drug Administration internal documents summarized by Zeese (1987), it is possible to produce false negatives on the EMIT test by the addition of toilet soap or NACI (table salt) to specimens, by voiding the specimen at certain hours of the day, or by simply diluting the specimen in a number of possible ways.

(c) Reactions Due to Endogenous Human Biochemicals

Morgan (1984, op. cit., p. 312) cites evidence that some of the enzymes employed in the EMIT can appear naturally in human urine and trigger false positives in versions of the test employing these enzymes. One study cited by Morgan found levels of the enzyme lysozyme high enough to interfere with the test in 10% of the subjects tested, though Morgan cautions that because this was a hospitalized sample it may be an overestimate for the population at large. However, people with borderline, undiagnosed renal disease might be at risk for false positives.

There is currently a hot debate over claims of the Atlanta chemist, James Woodford, that the human skin pigment melanin might cross-react with tests for certain cannabis metabolises, putting non-white racial minorities at greater risk for false positives. More research is needed to settle this controversy.

(d) Human Error

If human beings are involved in a process, occasional performance lapses can be expected, no matter how qualified and dedicated the personnel and how stringent the quality control measures. This applies to all methods of urinalysis and to the collection, transportation, and identification of specimens, and to all record keeping and notification processes. Human involvement in the maintenance and calibration of equipment is also of concern here. As the tests themselves get more reliable, human factors will account for a greater percentage of the total errors of the enterprise and while there are ways of reducing there are no known ways of eliminating errors due to human factors such as boredom, fatigue, attentional lapses, flagging motivation, forced haste, or intentional sabotage.

The Need for Independent Confirmation of lmmunoessay Screens With Another Method Based on Different Physico-chemical Principles of ldenitification

Knowledgeable experts, including those employed by the companies marketing the immunoassay screening tests, admit that positive tests with techniques like the EMIT must be confirmed by more specific tests before a positive result is reported.

There is nearly unanimous consensus that if one is willing to spend the money to acquire the appropriate technology, train and motivate the operators, and to ensure meticulous record keeping, specimen handling and chain of custody and reporting, accurate and specific identification of drug metabolises can be achieved. From reviewing findings and recommendations of expert consultants to the military, U.S.ED.A, N.I.D.,k, Dept. of Health and Human Services, and the Toxicology Section of the American Academy of Forensic Sciences, it is obvious that there is an increasing consensus that the only confirmatory test with sufficient sensitivity and specificity to balance the risk of falsely depriving people of their livelihood or freedom is gas chromatography with mass spectrometry (GC/MS). The U.S. military has switched from confirming its immunoassays with chromatography to the more expensive GC/MS after encountering serious problems with the former (Zeese, 1987, p. 42). This course of action is in line with recommendations of noted toxicologists McBay, Dubowski, and Finkle writing in the Feb. 18, 1987 issue of the Journal of the American Medical Association (p. 881) and one of the armed services’ own expert consultants, Edward Imwinkelreid (1987, op. cit., p. 26).

Though the potential for virtually perfect accuracy is admitted (using GC/MS and given flawless conditions, adequate time and funds, and strictest adherence to all procedures), one U.S. Court has held that even confirmation by GC/MS is insufficient because of the possibility of human error.(11) Sceptical of the likelihood of greatly reducing the chance for human error when such a complex, multi-stage process is applied on a mass scale and cognizant of the agony caused by false positives, the states of Connecticut, Iowa, Minnesota, Montana, and Vermont have passed laws banning urinalysis for employees without probable cause.(12) The City of San Francisco has done likewise and seven other state legislatures are considering similar bills.

If the tests are sufficiently accurate under ideal condnons to reflect past contact with illicit substances, how has past performance of reputable testing labs stood up under rigorous evaluation in real-life field tests?

Do inevitable human errors in the long train from specimen collection, labelling and transportation, through carrying out and interpreting the tests, to keeping records and reporting the results present unacceptable risks of false accusations against innocent parties? Where will the tests be done? Will all testing personnel be adequately trained? Will they be sufficiently motivated to keep their attention constantly rivetted on the task at hand which involves much repetition and boredom? Will there be frequent, independent, and “blind” evaluations of the labs’ levels of accuracy?

Writing in the Nova Law Review (1987, v. 11, no. 2) urinalysis expert and toxicology professor Kurt Dubowski states:

Even some of the most highly motivated and most conscientious laboratories and laboratorians can and do produce erroneous results on occasion. Consequently, any given routine drug-use test cannot be regarded as infallible at today’s state of-the-art and practice. Only a full program of effective and comprehensive quality assurance practices can be expected to reduce the number, frequency and magnitude of laboratory errors.

Dubowski describes the minimum program of independent external quality control tolerable in a lab whose decisions will carry such weight and have such potentially drastic consequences for people. Such quality assurance, involving 99 blind” unannounced specimens sent by a central standards-control and licensing body, is expensive and time consuming but absolutely necessary in light of performance of such labs in blind tests by agencies such as the Center for Disease Control (see Hansen et al., above) and the Armed Forces Institute of Pathology. Differences in error rates when labs knew they were being evaluated (average < 10%) and when samples were sent 99 unannounced” for analysis (average 31% with a high of 100%) indicate that procedures are not always uniform or scrupulously followed (O’Keefe, op cit, p. 35).

Dubowski asserts that, given the gravity of the consequences of false positives for people’s lives, “20% of a laboratory’s total efforts should be devoted to quality assurance activities, and 10% is probably the lowest acceptable level of [such] activities.(13)

Noting a depressing lack of overall improvement in the industry since he reviewed the status of mass urine screening 14 years earlier (despite acknowledged technological improvements), pathology professor George Lundberg notes that “[t]hough in theory none of these things [i.e., errors] need occur, in fact they all occur, and often.”

Although the U.S. Armed Forces Study (Einsel Commission) has been accused of understating the actual rate of false positives in its evaluations of urine screening, it was devastating in its condemnation of the laxity of the chain of custody for test specimens and the quality of record keeping, making it “in many cases impossible to provide scientifically and legally supportable documentation.”(14)

Several commentators have stated that profit motives coupled with inevitable cost cutting measures mandated by competitive bidding for screening contracts virtually assure that corners will be cut, though, one hopes, not resulting in anything near the laxity exposed in the Armed Forces program.

Dull, repetitive work that nonetheless requires highly skilled technicians is a fertile breeding ground for human error – most tests will be negative, punctuated by the occasional, more interesting positives. The livelihoods of those being tested rest upon extreme diligence in routine tasks such as cleaning glassware, affixing and recording labels, reading meters, transcribing numbers, key punching and filing. Testing labs vigorously claim to have solved this problem, but nothing in the published error rates to date justifies these claims. Research on similar work conditions elsewhere would lead one to expect that the error rates will continue to be unacceptably high. Seemingly innocent acts unrelated to the tests themselves can also introduce errors. For instance, one reported series of false positive urine tests resulted from minute residues left when lab personnel warmed their lunchtime tacos in an oven used to dry test glassware. Highly trained and dedicated military personnel were nearly dishonourably separated from the service with resulting losses of lifelong benefits before the mistake was discovered.

Attempts to cover up such deficiencies have been exposed as well. A defence lawyer quoted by Zeese (1987) even goes so far as to recommend checking individual signatures on every chain of custody document because of the forgeries he has encountered in attempts to hide human errors of this sort. Expert testimony on the likelihood of such errors under these conditions has been recognized in judicial decisions of U.S. Courts (see note 8). Regular proficiency testing by independent accredited professional organizations (e.g., American Association for Clinical Chemistry) is considered essential.

Another cause for concern arises from the Bayesian Probability Theorem that underlies the Theory of Signal Delectability. The predictive value of a positive result (i.e., the chance that a positive result is accurate) decreases as the prevalence of the to-be-detected activity decreases in the tested population. As psychologist and lawyer Anne Marie O’Keefe writes:

Ironically, since the error rate inherent in the drug tests, are higher than the actual rate of illegal drug use in the general woridng population, as reported by NIDA, the tests are more likely to label innocent people as illegal drug users than to identity real drug users.(15)

Put another way, if a test is 95% accurate it will have a 95% predictive value when applied to a population containing 50% drug users. However, the predictive value of the same test applied to a population containing only 10% drug users will drop to 68% – the false positive rate will be 32%. With 5% of the population actually illicit drug users, the predictive value is only 50%. Half the accusations will be false positives.(16) On this basis, Harvard physician William Anderson argued in The Wall Street Journal that

[p]rovidence and fairness would suggest that tests be applied only when there is at least a 50% chance that the suspect is guilty. Furthermore, the test should have an accuracy of at least 95%, as determined by objective measurement. If these standards were applied, then only five out of every hundred of those accused would be innocent.

This, of course, is faint comfort to the 5% falsely accused, and 5% of a large population is a huge number of potentially ruined careers and lives. While, in theory, GC/MS meets the demands for sensitivity and specificity, the likelihood of human error at every step of screening process suggests that an intolerable number of innocent people will be severely sanctioned if enough tests are run.

If the results of urine screening could reliably suggest prior exposure to illicit substances, would this allow inferences concerning the magnitude or duration of that exposure and the time elapsed between exposure and testing?

Urine screens detect in various ways the breakdown products (metaboliyes) of drugs taken into the body. Thus they indicate that the person was exposed to a certain drug at some earlier time. Questions of “How much?” “How often?”, and “How long ago?” are critical if the presence of the metabolites is to have any practical significance.

The metabolites of cannabis may be detectable in urine up to one week after its use. The CDC Advisory cited above asserts that because metabolites of THC (the active ingredient in cannabis) can accumulate in body fat and be slowly released 99 the urine test for cannabinoids should be able to detect prior use for up to two weeks in the casual user and possibly longer in the chronic user.” Zeese (1987) cites a study in which the longest period of forced abstinence after which urinary metabolites were still detectable was 77 days.

According to NIDA scientists Hawkes and Chaing (cited in Zeese, 1987), the major metabolite of cocaine, benzoylecgonine, is detectable with immunoassay for 24 to 48 hours after an intranasal dose. The published figures for amphetamine are roughly comparable. The same authors indicate that for the shorter acting barbiturates urinary metabolites have been detected up to several weeks after cessation of chronic use. After a single dose, metabolites have been found as long as 76 hours later.

Concerning the questions posed at the beginning of this section, our review of the literature revealed general agreement with the following statement published in the Journal of the American Medical Association:

The results of urine drug testing should be thought of as qualitative, or, at the very most, seniiquantitative. Even if a drug level is given, it should not be interpreted strictly, as so many physiologic as well as analytic factors can distort the result…. Neither can time since intake or amount of dose be deduced from urine testing. (Lundberg, journal of the American Medical Association, Dec. 5, 1986, p. 3004).

[with respect to marijuana] Only blood-sample measurements are likely to correlate with a person’s degree of exposure. (“Urine testing for detection of marijuana: an advisory – from the Centers For Disease Control – Journal of the American Medical Association, Oct. 7, 1983, pp. 1665-6).

Proceedings of a recent N.I.D.A. symposium, quoted extensively by Zeese (1987), particularly the work of N.I.D.A. scientists Hawks and Chaing, reinforce these conclusions.

If urnialysis reflects prior exposure to illicit substances, does this allow accurate inferences about the extent of impairment at the time of that exposure or carry-over effects at a later date?

There is virtual unanimity in the literature that such inferences would not be reliable. To quote U.S.C. pathology professor, George Lundberg, “Specifically, under no circumstances can impairment be diagnosed or even presumed from a urine test result.” The U. S. Centers for Disease Control advise: ” . . . attempts to correlate urine concentration with impairment or time of dose are complicated by variations in individual metabolism, metabolise accumulation in the chronic user, and urine volume changes due to diet, exercise and age.” N.I.D.A. has acknowledged that urine tests are unable to prove intoxication or impaired performance, as has a commission struck by the state of Maine to review the use of urine tests in employment settings. The Maine Commission recommended a total ban on the use of urine tests in the workplace, in large part because

[t]here has been no scientific study that has established a urine concentration level for any drug, or its metabolites, that has received general acceptance as a presumptive level of impairment. All of the urine testing experts who testified before the Commission agreed that the possible variations involved in urine testing render such a task nearly impossible; there are so many factors that can influence the concentration of substances of abuse in an individual’s urine, that a standard level of presumptive impairment becomes meaningless in practical application. . . . The only standard of impairment generally accepted at present is the 0.10% blood alcohol concentration level; note that this standard is set upon blood concentration levels. Blood tests provide a much more accurate picture of current impairment because they measure the substance [not its inert breakdown products] as it is still circulating throughout the body, presumably having the desired effect upon the individual as it does so. However, presumptive concentration levels for substances other than alcohol have remained stubbornly elusive. At the present time, there is no conuiionly accepted standard of impairment for any substance of abuse other than the 0.10% blood alcohol standard.(17)

The Syva Company has acknowledged this in their test product information:

A positive EDMT Cannabinoid Assay result does not indicate intoxication. The psychoactive effects of marijuana and hashish do not correlate with urinary metabolise levels obtained by any method (quoted in Zeese, 1987).

Note how the situation with drugs differs from that of alcohol. Although with alcohol too, there are wide ranging individual differences in the degree of intoxication observed in a given individual with a certain blood-alcohol reading, having to do with such factors as body weight, at least we have an relatively accurate and simple to administer test of the actual agent – alcohol – that causes the impairment. However, with drug testing, only behavioural testing following ingestion of a drug can establish impairment. That drug use by itself does not necessarily cause impairment was dramatically demonstrated by Wisotsky, writing in the 1983 Nova Law Review (p. 1410). Wisotsky lists several professional athletes involved in “drug scandals” who were among the best performers on their teams. Deterioration in on-Cield performance was not what directed investigators to these players. Tessting for stimulants is routine in high level competitions because of the possibility that they might improve performance.

The research experience of the senior author as well as reading of the psychopharmacology literature provides ample evidence that individual differences in response to any psychoactive substance can be huge. Not only is there great variability in how different people react to an identical dose, but the same person at different times can show wide variation in response to a standard dose. Predictability is hampered by the ability of numerous physiological and psychological conditions to interact with pharmacological effects of a drug.

Variations in response to a drug can be due to four main classes of variables: organismic variables; drug variables; environmental variables; and task variables.

1. Organismic variables. Studies have shown that people differ in how they absorb, distribute, metabolize, and excrete drugs. These factors are affected by the individual’s age, weight, state of health, and position in the diurnal cycle. The state of arousal or fatigue may exert important influences on various measures of drug response. Many personality variables are known to interact in determining the effects of drugs as well. Mood at time of use and longer-standing emotional conditions are important. Different background physiological states of several systems can strongly affect the nature of the drug response. This background state is a joint function of current environment and genetic factors. Work in the laboratory of the senior author at SFU has shown how the interaction of heredity and environment can produce strikingly different outcomes in response to the same drug. There are often sex differences as well.

2. Drug Variables. The magnitude and sometimes even the direction of response to a drug can vary depending on dose range. Certain drugs act as stimulants in small doses but are depressants at higher levels, for instance. Naive and experienced users will react differently due to acquired physiological and/or psychological tolerance. Bioavailability, method of administration, and dosage regimen also contribute to response variability, as do interactions with other drugs that might be present. The same total dose of a drug may produce similar amounts of urinary metabolites if it is taken in one large dose or several small ones over a longer period, but the subjective effects and performance effects, which correlate with the level of available active molecules in the blood at any given time, could differ considerably.

3. Environmental variables. The “set and setting,” i.e., the meaning ascribed to the usage, the user’s expectations, and the social, psychological, and physical situation of the usage, can produce greatly different subjective and objective effects when drugs are imbibed. Much of the work in the SFU lab has focussed on these sorts of factors which have been known to swamp pharmacological effects on occasion.

4. Task variables. The ongoing behaviour of the recipient affects drug response. The degree to which a drug affects one task for an individual is not necessarily a good predictor of how other tasks will be affected, or how the same dose will affect the same task under changed conditions as described above. Motivation, attention, past experience with the task, concurrent tasks, are all important in determining the overall drug effect. Although averages adequately depict a group’s behaviour, large inter-subject variability is the rule, making predictions about individuals hazardous at best.

Would even accurate pro-employment screening of urinary metabolites along the proposed lines accomplish its stated objectives?

Given the factors discussed above, it seems highly unlikely that the results of a urine test could tell a prospective employer much of predictive value about a job candidate that the employer is entitled to know. The test might indicate prior exposure to a drug, but it could not distinguish chronic from casual use. Neither can assays of urine metabolites determine impairment at the time of use, or subsequently. They cannot distinguish responsible use on the employees’ own time from use on the job or use sufficiently close to assuming daily duties that might be dangerous or counterproductive – something all reasonable persons would condemn.

Research by the senior author’s group at SFU over the years has identified individuals who use considerable amounts of various drugs but maintain productive, responsible lifestyles. Much past research has centred on those who do have serious difficulties from drug abuse because they are easy to identify, often self-referred or detected by declining performance on the job. Research such as that of Patricia Erickson and her colleagues at the Addiction Research Foundation, by Peter Cohen in the Netherlands, and Melanie Dreher in Jamaica indicates that there are substantial numbers of people who are able to indulge in moderate cocaine or marijuana use, for instance, and still maintain highly productive careers. Much other research cited by Arnold Trebach in his recent book The Great Drug War indicates that mere use of a drug does not automatically imply an undermotivated, unsafe, or low productivity worker, though such drug users obviously do exist. Of course there are irresponsible, lazy abstainers as well. Performance over an extended period is the only reliable index.

Unintended Exposure

It is also worth noting that in the case of inhaled substances such as marijuana there is published evidence that being in the vicinity of users can result in passive uptake in non-users which can trigger a positive urine test, even if the passive inhaler experienced no subjective “high.” Thus, an employee’s livelihood could be threatened by attending the wrong concert or party or accepting a ride in the wrong automobile. If urine screening is instituted, it is of utmost importance that cut-off points for the tests be agreed upon in advance to protect such persons whose tests reveal trace amounts that could come from passive inhalation.

Consequences of Urine Screening

Pre-employment screening is not likely to have the predictive value employers seek, but it may have other unintended consequences. As the ubiquitous “grapevine” disseminates possible ways to “beat” preliminary screening tests such as the EMIT (in which case the more accurate and expensive confirmatory tests would not be done), will potential abusers not simply “stay clean” for a safe length of time before the test, switch to legal and/or non-screened intoxicants such as alcohol, benzodiazapines, or Darvon, for instance, or engage in one of several practices that are known to enhance the likelihood of “false negatives”? In addition, what is to stop potential employees who pass the screening from altering their drug use habits once inside the organization?

Ingenious schemes for substituting “clean” samples (now available on the Black Market or readily obtainable from friends) are becoming widely known. To quote Lundberg (Journal of the American Medical Association, Dec. 5, 1986, p. 3004), “Without large numbers of diligent and devoted micturition observers, the entire mandatory urine drug screening becomes a travesty and is certain to fail.”

Substitution and adulteration are preventable by sufficiently close scrutiny of tested subjects, but one must what ask what the long-term consequences of this intrusion into private biological functions, considered odious, embarrassing and demeaning by many, would be for the morale and climate of trust of the organization concerned.

To try to counter the numerous subtle ways of beating the initial screening, draconian procedures will be required for all applicants, innocent and guilty. This underscores the concerns about privacy raised in Part I. Also, how will the presumption of guilt inherent in this introduction to the workplace colour employee perception of the organization? E.g., how will it affect employee morale and the mutual trust that must underlie successful labour-management relations? How will this be reflected in productivity, employee turnover, the future negotiation climate, loyalty to the organization, etc.? An increasing number of lawsuits for invasion or privacy, defamation of character, and infringement of constitutional protections are being filed against employers in the U.S. who use urine testing in the absence of evidence of employee impairment.

Some employers such as the Drexelbrook Engineering Co. in Pennsylvania and Apple Computers and the Fairchild Camera and Instrument Co. in California have publicly rejected mass urine screening because of the negative effects they feel it would have on their relationships with their employees and loyalty of their workforce to the company.


III. The Problem of Drug Abuse

Are the illicitly used substances under discussion (viz., cannabis, cocaine, amphetamine, and barbiturates) used by a sufficient proportion of potential employees to justify the need to prove one’s innocence in advance of proving one’s competence for employment? While drug abuse, as opposed to use, is obviously a cause for concern, is the present danger so great as to force abandonment of the right to be judged on one’s performance rather than on personal habits or associations that cannot be demonstrated to have an impact upon work achievement? Are the rates of usage changing in ways that justify introduction of screening at this time? Most studies agree that, with the exception of cocaine, usage of illicit drugs has levelled off and declined slightly since the watershed year of 1979 (O’Keefe, 1987(18) – citing NIDA estimates; or Smart, 1983(19)). Similarly, Hollander and Davis(20)found in surveying Vancouver adolescents that there had been a decrease in the 1982 reporting period from previous periods in the use of alcohol, tobacco, hallucinogens, CNS [central nervous system] depressants and cannabis. Percentages reporting ever having tried heroin or CNS stimulants (other than cocaine) remained stable at about 2% and 13%, respectively. Those reporting ever having used cocaine rose from 8.5% to 10.1% in the interval from 1978 to 1982.

After reviewing their own and others’ surveys of cocaine use, in Canada and worldwide, Erickson et al.(21) summarize as follows:

No Canadian surveys conducted on adult populations have established trend data with respect to cocaine use. In the United States the extent of cocaine use appears to have peaked about 1979, much of the increase occurring between 1977 and 1979. There is also an indication that cocaine users in the population are ageing; in particular, while use seems to be declining among young adults from 18 to 25 years of age, it is increasing among adults 26 years of age and older. (p. 59)

In the same monograph the authors report on their study of cocaine use among Ontario adults. Those in this scientifically drawn representative sample who reported ever having tried cocaine amounted to 3.3%, the “majority of those . . . do[ing] so infrequently, once or twice during the previous year. (p. 59) Concerning negative consequences of cocaine use in the same sample, the authors conclude:

The precise nature of negative consequences stemming from use of cocaine remains ambiguous. Although negative reactions are reported among infrequent users, the nature, extent and seriousness of consequences appear to be strongly related to level of use. (p. 59)

The assessment of Erickson et al. is a refreshing change from alarmist scenarios painted by politicians and therapists with an interest in treating the alleged “cocaine epidemic.” The latter have generally relied on self-referred patients who have run into serious problems with the drug – those gathering data about marriage are likely to get a distorted picture if they draw their sample from appellants at the divorce courts.

The majority of users, who do not get into such difficulties, generally remain hidden unless a random sample of the population at large (such as that of Erickson and her colleagues) is studied. The Erickson study, which allows a distinction between use and abuse, concurs with the summary by Siegal(22) in his recent article on treatment of cocaine abuse:

Abuse of cocaine can lead to serious psychiatric and psychosocial problems. However, with the exception of acute and chronic toxic reactions, there have been remarkably few serious physical consequences of abuse.

The data of Erickson and her colleagues suggest that few users of cocaine in their sample do so with sufficient regularity to be in danger of significant problems of this sort.(23)Abusers who do reach this stage of difficulty should be detectable from previous job performance and other traits without examining their urine. Perceptions of former supervisors and co-workers are far more likely to predict satisfactory performance than urine tests (e.g., work performance, attendance, reliability, attitudes appearance and habits, safety record, financial situation, general health).

Jobs that involve special risks

There is, however, a special concern that is not addressed by the argument of the previous section. What about jobs that involve special risks to the employee or others, such as driving trucks loaded with dynamite or hazardous chemicals through densely populated and narrow streets? It might be argued that such risks are so serious that an employer owes the public whatever assurance she can provide that employees are operating at the peak of their efficiency. Even if drug testing provides a very slight increase in the ability to predict dangerous behaviour, so this argument runs, the slight increase justifies the violation of the employee’s privacy, given the seriousness of the risk.

This objection fails for two reasons. First of all, as we have already pointed out, there is no available test that discriminates between those employees who are presently incapacitated by the drug and those who are presently safe but who ingested the drug earlier. But wouldn’t drug testing provide some reassurance? In fact, we argue that this is not the case. Since it is the performance of the employee that is paramount, we submit that any test that deflects the employer’s attention away from monitoring this will increase the chance that an employer will miss the employee whose behaviour constitutes the risk. And remember, the results of urinalysis is not known immediately; the sample must be sent to a lab for analysis. So, an employer who is concentrating on collecting the urine sample is deflecting his attention from observing the employee’s behaviour. As well, a urine test taken before a trucker sets out on her run will not ensure that a ‘drug addict’ will not simply drive out of range of observation and ingest the drug. Thus, the false confidence of urine tests may in some cases actually increase the chance that an irresponsible employee may be engaging in dangerous tasks while impaired. Generally, if it is the behaviour the employer needs to monitor, he should monitor this, rather than relying on a ‘short cut’. This point is even more important when the employees’ tasks constitute a high risk to others.

Second, this argument presupposes that there is a reliable correlation between the amount of the drug taken and the degree of impairment. As we have already pointed out, this is not the case. As for performance with stimulant drugs(24) among functional people, Spiegel(25) summarizes the world’s literature as follows:

Mostly positive effects of amphetan-dne were reported In various tasks requiring sustained attention (letter sorting, card sorting, tracldng, pursuit rotor, signal detection, arithmetics). Reaction time to visual stimuli was improved in several studies, the improvement amounting to about 10% in one well-designed investigation. . . . There is some additional evidence that amphetarrdne can improve learning and memory.

Spiegal also summarizes a long series of “sophisticated studies involving large samples” by Hurst and his colleagues. These investigations, which included effects on driving behaviour, found:

there was no indication that the incentive value connected with the tasks was altered by the drug. Subjects judged their own performance more optimistically after amphetarmine. There was no indication, however, that subjects overestimated their performance or were willing to take undue risks, although they felt more vigorous and self-confident after amphetamine.

The ability of central nervous system stimulants to enhance performance under these conditions is not followed by a detectable behavioural “rebound,” i.e., by more pronounced fatigue, one day later.(26)

Thus the empirical evidence suggests that the risks to others’ welfare posed by those who use – in moderation – central nervous system stirnulants is greatly exaggerated in the public’s mind. Nevertheless, it must be recognised that there are jobs that are so dangerous that we have a right – perhaps even a duty – to monitor those doing them in ways that would not be justified for normal work situations. In such risky jobs, we may, for example, be justified in requiring an employee to take a breathalyser test (he might be drinking vodka and using mouthwash, so we cannot tell if he has been drinking) before starting his shift. If there were tests for drugs which similarly measured the active ingredient causing impairment, we would also he justified in using them in these circumstances, were they to become available. No such test is currently available; but if one were to be developed, the BCCLA would have no a priori objection to its use.

Also, for such risky operations where strict behavioural monitoring could be justified, perhaps a case could be made for urine testing for drugs as a part of that wider programme of intrusive observation. As things stand, we can think of no case where this would be justified; but we are willing to entertain an argument for these rare cases. In the last section we shall return to this point.

Routine testing in the workplace

The concession just made for extremely dangerous jobs should not be read as ‘the thin end of the wedge’ to justify routine testing for drugs in normal job conditions.

Working under the influence of depressants is to be avoided, but many workers are legitimately prescribed drugs such as benzodiazapines as anxiolytics or muscle relaxants. Since these and alcohol show cross-tolerance with barbiturates and are not the drugs generally screened for, they could easily be substituted by an employee abusing barbiturates for a period long enough to clear the urine of screened metabolites. This would stave off withdrawal effects in the interim.

Given the foregoing estimates of usage and the indications that the vast majority of users do so on their own time without escalating their usage and without detectable harm, it is questionable whether the privacy and dignity of all job applicants should be curtailed in a program of mandatory urine screening.

In his Dec. 5, 1986 Journal of the American Medical Association article, Lundberg asserts that “recent data indicate that deaths and serious impairment resulting from the use of psychoactive drugs in this country have been decreasing of late.” By far the most serious and prevalent threat in this regard continues to be alcohol, a fact reiterated in the 1987 Report of the Task Force on Alcohol and Drug Abuse in the Workplace, published jointly by the B.C. Ministry of Health, the Business Council of B.C., and the Workers’ Compensation Board of B.C. Lundberg in his 1986 Journal of the American Medical Association article decries the hypocrisy in that the drugs that most often kill are legal, and far less toxic drugs are illegal. The former have not been the objects of attempts to screen the urine of overtly functional workers. They have in recent years become the targets of other more humane programs to curb abuse.


IV. The Alternatives

Abjuring urinalysis, however, does not reduce the moral obligation of employers and employees alike to seek more humane and effective ways of reducing drug abuse. Many observers, including the senior author, have concluded that the use of urine screening is rising in popularity because it seems like a “quick fix” that is cheaper than serious sustained efforts to reduce drug abuse such as the mass educational efforts that have begun to reduce cigarette smoking.

The senior author’s research and that of others consistently shows that peer attitudes and influences are the most important determinants of drug use. Official attitudes and sanctions bear little relationship to the recurring cycles of rise or fall in prevalence of various substances. It is the fraction of recreational drug users who develop physical or psychological problems that affects job performance whose activities are the legitimate concern of the employer. Urine screening does not identify these people. Ale recent Report to the B.C. Minister of Health is a hopeful and humane document that outlines many successful Employee Assistance Programs (EAPs) that stand a chance of salvaging lives and turning impaired employees back into safe and productive workers rather than simply passing the problem on to someone else as pre-employment screening would do. Governments are becoming increasingly aware that because all of society is affected, public assistance with such programs is called for.

The money spent on urine screening and defending against challenges to it could be much better spent on positive measures to educate and to identify and assist employees who do have substance abuse problems. As the Employee Assistance Program instituted by Canada Post indicates, this can often be achieved most effectively by assisting the volunteer efforts of trained fellow workers. This also reduces employers’ direct costs. But for such a program to work, employee trust and active union participation is essential.

Workers who pass pre-employment urine screening may well develop alcohol or drug problems down the road. Thus, there are two alternatives to solve a problem that cannot be simply fixed by pre-employment screening: Continued urine monitoring, which further exacerbates the violation of privacy, or E.A.P. How will employees’ initial testing experience poison the atmosphere of trust and cooperation essential for E.A.P to be effective? The Report to the B.C. Minister of Health emphasizes that in a punitive atmosphere the most likely response to a substance abuse problem is to try to cover it up. When there is broad support for a co-operatively managed E.A.R it becomes increasingly likely that fellow workers will recommend troubled colleagues seek available help or that they will ask on their own. It is clear that continuing urine testing is not compatible with an effective E.A.P.

That an E.A.P. is the most effective means of identifying impaired workers and getting them to assistance was eloquently argued by the Association of Labor/Management Consultants on Alcoholism in their submission to the Taskforce:

How can the employer identify such an individual? By having an awareness of the signs of alcohol or other drug impairment and by using that awareness in performance monitoring…. The supervisor’s awareness, coupled with active monitoring and documentation allow for early identification.

This method of identifying alcoholldrug troubled individuals is known as the performance model. Its focus is limited to productivity and safety in the workplace; it does not deal with the issue of use away from work unless that use affects the job. The value of the model is that it allows management to intervene on the basis of legitimate performance expectations and to maintain union support in doing so.

In a milieu that lacks that mutual trust and respect, disputes can be expected increasingly to spill over into various courts and tribunals.


V. Legal issues

Are Canadian employers and testing labs prepared to bear the great expenses that the U.S. experience suggests are likely to ensue from legal challenges arising from attempts to screen people’s urine without probable reasons to suspect impairment?

In the U.S., legal defence against accusations or sanctions based on urinalysis is a complex, growing specially of the law. We predict that as Canadian courts are called upon to define the limits of our Charter of Rights and Freedoms, an increasing number of similar actions against employers will be launched here. Although the Charter arguably does not apply to drug testing by private companies, other statutes and acts do. U.S. courts have already upheld rights of accused employees and their counsel to examine in detail all steps in the accusatory chain. According to Zeese (1987) this has included right of access to

personnel records [and] time-off and leave data to establish who was responsible for test results. There should also be documentation of maintenance and repair logs [for all equipment], preparation and expiration dates for all reagents and up-to-date manuals for every method used in the lab. There should be retained on rile all raw data work sheets and printouts from the machine used. There should also be a “control chart” which is a graphical record of the laboratory’s results over a period of time. All of this material should be available in discovery.

Where tests rely on subjective interpretation such as reading the output of gas chromatographs, thin layer chromatographs, or mass spectrographs, all personnel involved have been required to be present for cross examination in court. Where computer data enter into the determination process, U.S. courts have required (e.g., U.S. vs. Rosso) labs to produce all personnel as well as “the underlying data used in the analyses, programs and programming method and all relevant computer inputs and outputs be available to the opposing party far in advance of trial.”

In the U.S. vs. Dioguardi the U.S. Court of Appeals for the Second Circuit held (cited in Zeese, 1987):

We fully agree that the defendants were entitled to know what operations the computer had been instructed to perform and to have the precise instruction that had been given. It is quite incomprehensible that the prosecution should tender a witness to state the results of a computer’s operation without having the program available for defence scrutiny and use on cross examination if desired. We place the Government on the clearest possible notice of its obligation to do this and also of the great desirability of maidng the program and other materials for cross-examination of computer witnesses . . . available to the defendant a reasonable time before the trial.

Zeese goes on to cite several other U.S. court decisions to the effect that if the foregoing items are not supplied prior to cross-examination of expert witnesses who rely on such data, the direct testimony of the expert must be stricken.

Legal redress in Canadian courts

1. Relevant Legislation

A. Privacy Act (R.S.B.C., c. 336) 

Violation of privacy actionable

1.(1) It is a tort, actionable without proof of damage, for a person, wilfully and without a claim of right, to violate the privacy of another.

(2) The nature and degree of privacy to which a person is entitled in a situation or in relation to a matter is that which is reasonable in the circumstances, due regard being given to the lawful interests of others.

(3) In determining whether the act or conduct of a person is a violation of another’s privacy, regard shall be given to the nature, incidence and occasion of the act or conduct and to any domestic or other relationship between the parties.

(4) Privacy may be violated by eavesdropping or surveillance, whether or not accomplished by trespass; but this subsection shall not be construed as restricting the generality of subsections (1) to (3).

It is certainly arguable that a demand by an employer for a urinalysis test is a violation of section 1. Arguments of reasonableness, the relationship of the parties, and the lawful interests of others could be advanced by an employer however. What would happen in the end result, were such an action launched, is difficult to predict, but even simply the existence of the Actprovides some basis for a strong anti-urinalysis policy.

B. Canadian Charter Of Rights And Freedoms

“7. Life, Liberty and Security of Person

Everyone has the right to life, liberty and security of the person and the right not to be deprived thereof except in accordance with the principles of fundamental justice.

8. Search or Seizure

Everyone has the right to be secure against unreasonable search or seizure.”

While the Charter may well not apply to the private contract employer-employee relationship, the general right of privacy contained in sections 7 and 8 is, nevertheless, worthy of recognition.

C. Canadian Criminal Code

The breathalyser is a test somewhat analogous to a urinalysis although far removed in terms of personal privacy. It is interesting to note that the right of a police officer to demand that an individual submit to a breathalyser arises from a statutory provision and can only be demanded if:

“235. (1) Where a peace officer on reasonable and probable grounds believes that a person is committing, or at any time within the preceding two hours has committed, an offence under section 234 or 236, he may, “

2. Analogous Situations in time Organized Sector

Arbitrators have looked at the question of searches of the persons of employees covered by collective agreements. In Re United Automobile Workers Local 44 and Chrysler Corporation of Canada Ltd. (1961) 11 LAC 152, the principles adopted in that case are set out conveniently in the headnote:

Apart from the restricted common law right of a private citizen or peace officer to search the person of an individual who has been lawfully arrested, a company can only justify the random spot checking of employees who are in no way suspected of theft, and who do not consent to a search of their person, by either an express or an implied term of the employment. For such a right to be implied in the contract of employment, it must have been established by past practice; the mere relationship of master and servant does not carry with it such a right, nor does the fact that the company is exceptionally large entitle it to carry out such a procedure for reasons of security. Thus, where there has been no arrest and the right to search is neither expressed in the agreement nor implied by past practice satisfactorily established in evidence, a discharge for refusing to submit to a search in such circumstances is without just cause and the employee must be reinstated.

In Canada Safeway Limited and Retail Store Employees’ Union, Local 832 [19801 3 WL.A.C. 333, the store manager had searched the grievor’s jacket because he was suspicious of all of the employees after one had been charged with possession of marijuana. On searching the jacket, marijuana was discovered. That employee was subsequently criminally charged and terminated by the employer.

The Arbitration Board held that it was an invasion of the grievor’s personal freedom to search his jacket. The grievor could only lose his fundamental right to refuse to be searched by the clearest provision in his contract of employment or collective agreement. The Board went on to say that, in any event, even where marijuana is found in an employee’s possession that is not a dischargeable offence where the employee clearly establishes that he did not intend to use the substance during working hours or on company property.

Arbitrators have also considered the right of an employer to require an employee to be medically examined by the employer’s doctor.

At common law, a medical practitioner who examines someone without first obtaining that person’s consent, express or implied, is guilty of a trespass or an assault. The existence of an employment relationship does not imply consent ‘ Although there are authorities to the contrary, the majority of the cases indicate that unless there is statutory or contractual authority permitting an employer to require an employee to see the company physician, the employee cannot be forced to do so. The underlying rationale of these cases is that an examination by a doctor is basically a trespass unless the individual patient consents.

Perhaps the best explanation of the basis of this position is set out in Re Thompson and Town of Oakville (1963) 41 DLR (2d) 294, wherein the Court stated:

We start with this general principle of law : a medical practitioner, who examines a person against his will and without authority to do so, commits a trespass

One has only to remind oneself what a medical examination means. A medical examination involves the confidence of the doctor if he is your own physician, but it is otherwise if he is maidng an examination on behalf of another. The right of employers to order employees to submit to an examination by a doctor of choice of the employer, must depend upon either contractual obligation or statutory authority. None exists in this case.

In Canadian Pacific Airlines and Brotherhood of Railway and Airline Clerks (Ilarding Grievance)(October 28, 1982, D. Munroe, unreported), when dealing with the question of requiring an employee returning from absence due to illness to see a physician, Arbitrator Munroe observed at page 25:

In the absence of contractual provisions signifying a different intent, employers are not

autocratically entitled to demand that the returning employee go through a process of “lie detection”, dignified only by the fact that it is administered by the medical profession. The loss of an individual’s privacy is not an incident of the employment relationship, at least not to that automatic extent.

In another arbitration award, Re Monarch Fine Foods Co. Ltd. and Milk and Bread Drivers, Dairy Employees, Caterers and Allied Workers, Local 647,(1978) 20 LAC (2d) 419, (Picher), the Arbitration Board stated:

The issue in the instant case is whether the company can require that the grievor submit himself to a medical examination by a clinic of the company’s choosing, strictly to test the truth of his asserted illness or injury. In other words, can the company subject the grievor to a medical examination that is, in effect, a lie-detector test? As the board has noted, an authority as extraordinary as that must be clearly expressed in the collective agreement or necessarily implied either from that document or from a longstanding practice of the employer. In this regard, useful reference may be had to the “lunch-pail” cases – generally speaking, boards of arbitration have held that there is no absolute right of an employer to search personal effects and the right to do so must be either an expressed term of the employment or a term to be implied on the basis of well-established past practice. (at page 422)

Similarly, with respect to demanding that an employee submit to urinalysis, the employer is basically requiring an employee to take a lie detector test since obviously, if the employee so approached admits drug use, the urinalysis would be unnecessary.

3. Protection, if any, for the employee who refused to submit to a test

If the employee who refuses a urinalysis test is in a union and protected by a collective agreement, then he cannot be disciplined or dismissed except for just cause. Unless there was some other evidence that would justify disciplining an employee, we do not believe an arbitrator would find that it was just cause to discipline or fire an employee simply because he refused to submit to a urinalysis.

Unorganized employees under federal jurisdiction would have a right to utilize Section 61.5 of the Canada Labor Code and go to arbitration. Under that Code, the arbitrator has the power to reinstate an employee who has worked for twelve months.

All other employees would have the right to sue for wrongful dismissal, that is, dismissal without just cause, but the remedy lies in damages and not reinstatement.

It is possible that an employee in such a situation might find some protection in the Human Rights Act (S.B.C. c. 22). Section 8 states:

Discrimination in employment

8.(1) No person or anyone acting on his behalf shall

(a) refuse to employ or refuse to continue to employ a person, or

(b) discriminate against a person with respect to employment or any term or condition of employment because of the race, colour, ancestry, place of origin, political belief, religion, marital status, physical or mental disability, sex or age of that person or because of his conviction for a criminal or summary conviction charge that is unrelated to the employment or to the intended employment of that person.


We have reason to believe that reinstatement can be ordered under this Act. However, the issue would only arise if an employer demanded a urinalysis from one employee because of a drug conviction that arose away from the workplace.


VI. Conclusion

History teaches us that the use of psychoactive substances has been with us throughout recorded time. The motives are varied and complex and where it leads to problems there are no simple solutions. The rates of usage for particular drugs wax and wane unpredictably and our own research has convinced me that these variations are largely independent of official attitudes and sanctions. No degree of law enforcement or retribution is likely to have much effect on fashions of drug use in a free society. In a perfect world no-one would feel the need to indulge, but in this imperfect one the intelligent course of action, is to concentrate our resources on minimizing the costs to individuals and society from those whose usage lands them in trouble. We must concentrate on identifying and rehabilitating the minority of users of all legal and illegal substances whose use causes significant harm to themselves or others.

Alcohol is by far our number one problem drug and the proposed pre-employment urine screening would do nothing to alleviate its abuse. In fact, it might exacerbate it.

Expert opinion cited above shows that preemployment and ongoing urine screening is unlikely to tell prospective employers what they wish to know about applicants, but it is likely to harm a significant number of innocent people and have unintended negative effects for the employer. An ongoing screening programme for employees already in the company would have to be so draconian as to seriously violate the employees’ privacy rights if it were to have other effects which were positive.

In many historical instances, drastic means have succeeded in reducing use of one substance only to see those desiring the subjective effects it provides substitute another, more dangerous substance. When a more dangerous substance such as alcohol is legal, available and not screened for, history suggests this is where people will turn.

Given the obvious bad effects and demonstrable uselessness for legitimate purposes of urine screening in the workforce, why are so many firms using it?

Reviewing the status of mass urine screening in industry, the military, and the prison system, the physician and professor of pharmacology, John P. Morgan (1984; p. 306) construes its underlying aims in the following way:

The real justification for the screening of urine in industry or elsewhere is the identification of deviant behaviour. That identification may be followed by segregation, punishment, firing, surveillance, treatment or any of a variety of interventions, humane or inhumane. The conventional excuses for the humane treatment should not obscure the fact that urine screening is a probe to identify deviance, not dysfunction – a technique to investigate humans, not accidents. (emphasis in original)

The BCCLA holds that it is not the job of employers to control deviance outside the workplace. And, as we have argued above, there are surer, less onerous means of investigating and preventing accidents. Therefore, we can see no good reason for an employee to submit to this kind of search. We unequivocally oppose the use of urine screening in the workplace, and offer whatever assistance we can provide to an employee who is coerced by her employer into taking such a test.

Since many employees are in a position where they are subject to various kinds of coercion to take a urine test, and not all employees are covered by a collective agreement which prohibits such a test, the BCCLA advocates legislation which would prohibit such testing in the workplace.(27) A model bill setting out such a prohibition is beyond the scope of this brief.

However, as we mentioned in the earlier section on extremely hazardous jobs, a blanket restriction on drug testing which prohibited such tests in extremely hazardous occupations may not be justified. Therefore, legislation which prevents use of such tests in normal circumstances should contain a provision for appeal for an exemption from these regulations for extremely hazardous jobs. The legislation should outline the creation of a government body to approve and oversee such testing. We foresee that such a body would, if knowledgeable and conscientious, approve very few exemptions.

Finally, we note that the objections to urinalysis testing in this brief are predicated on the current ‘state of the art’, which involves the test for metabolites, rather than the active ingredients which cause impairment at the time the test is administered. If a test were to be developed that gave reliable information about actual impairment, our objections against a test would be less severe. Such a test would give the employer information she has a right to be aware of, unlike the present tests. That is not to say that such a test would meet with our unqualified approval. For the overwhelming majority of jobs, behavioural evidence of impairment is sufficient for the employer; and the violation of privacy involved in being forced to urinate into a bottle while someone else watches simply is not justified by the added evidence about impairment this procedure would provide.



1. The Polygraph Procedure as a Truth Detector, BCCLA 1984.

2. Psychological Testing in the Workplace, BCCLA 1987.

3. The Use of Graphology as a Tool for Employee Hiring and Evaluation, BCCLA 1988.

4. G.D. Lundberg, “Mandatory unindicated urine drug screening: Still chemical McCarthyism,” Journal of the American Medical Association, Dec. 5, 1986, vol. 256(21), pp. 3003-3005.

5. A.M. O’Keefe, “The case against drug testing,” Psychology Today, June 1987, pp. 34-38.

6. E.J. Imwinkelreid, “False Positive: Shoddy Drug Testing is Jeopardizing the Jobs of Millions.” The Sciences. Sept/Oct. 1987, pp. 23-28.

7. Morgan, J.P. “Problems of Mass Urine Screening for Misused Drugs”. Journal Of Psychoactive Drugs, v. 16(4); Oct-Dec. 1984, pp. 305-317.

8. Dubowski, K, “Drug-Use Testing: Scientific Perspectives,” in Symposium: Testing for Drug Use in the American Workplace. Nova Law Review, v. 11, no. 2 (1987).

9. Hansen, H.J., Caudill, S.R & Boone, D.J. “Crisis in Drug Testing: Results of CDC Blind Study.” Journal of the American Medical Association, Apr. 26, 1985 – v. 253(16), pp. 2382-7.

10. Urine Testing for Drugs of Abuse. N.I.D.A. Research Monograph No. 73. Bethesda, Md., 1986.

11. N.T.E U. vs. Von Raab. No. 86-3522, Slip. Op. at 29 (E.D. La., Nov. 14, 1986 – cited in Zeese (1987 op. cit., p. 54).

12. E.J. Imwinkelreid, op. cit., p. 28.

13. Nova Law Review, v. 11(2), 1987.

14. Report of the Panel of Army and Civilian Experts in Toxicology and Drug Testing Legal Issues for the Surgeon General of the Army (Dec. 12, 1983).

15. “The case against drug testing.” Psychology Today. June, 1987, pp. 34-38.

16. George Lundberg, M.D.; U.S. C. School of Medicine writing in Journal of the American Medical Association, Dec. 5, 1986, p. 3004. Wm. Anderson, M.D.; Harvard Med. School writing in The Wall Street Journal, April 14, 1986.

17. Report of the Maine Commission to Examine Chemical Testing of Employees. Dec. 31, 1986. pp. 20-21.

18. “The case against drug testing.” Psychology Today. June 1987, pp. 34-38.

19. Reginald Smart, Forbidden Highs. Toronto Addiction Research Foundation. (1983) – See Ch. 2 for trends in cannabis use in Canada. Smart reports (p. 32) “some evidence of stabilisation after 1978” with 8.6% of Canadian adults reporting any usage in 1982.

20. Trends in adolescent alcohol and drug use m Vancouver: summary of findings. January, 1983. Province of B.C., Ministry of Health, Alcohol and Drug Programs.

21. P G. Erickson, E.M. Adlf, G.E Murrya, R.G. Smart, The Steel Drug: Cocaine in Perspective. (1987) Lexington, Mass., Lexington Books.

22. R.K. Siegel, “Treatment of cocaine abuse: Historical and contemporary perspectives.” Journal of Psychoactive Drugs, v. 17(1), Jan.-Mar. 1985, pp. 1-9.

23. Erickson’s findings were confirmed in a study in the Netherlands by Peter Cohen using a different sampling methods. Cohen presented his results in a seminar which the author chaired at the 1987 Institute on Drugs, Crime and Justice in the U.K.

24. Byck and his colleagues report that in “blind” subjects the effects of equated doses of amphetamine and cocaine cannot be distinguished. This is to be expected because the drugs affect levels of the same neurotransmitters by different mechanisms.

25. R. Speigel, “Effects of amphetamines on performance and on polygraphic sleep performance in man,” in P. Passouant and 1. Oswald (eds.), Pharmacology of the States of Awareness, Advances in the Biosciences, Vol. 21, Oxford: Pergammon Press, 1979, pp. 189-201.

26. P.M. Hurst et al. (1971), “Rebound from damphetamine’,” Psychological Reports, v. 29, pp. 1023-1033. P. M. Hurst (1976), “Amphetamines and driving behaviour,” Accident Analysis and Prevention, v. 8, pp. 912.

27. Again, subject to the qualification in the previous note, we foresee no reason to exempt any particular job category or type of employment. However, this legislation ought not to prohibit drug testing for athletic meets where the majority of athletes are agreed that the competitions is to be run without the use of drugs.