ABSTRACT
Introduction
I. Alcohol and drug testing defined
II. Nature and extent of alcohol and drug use
III. Workplace strategies for managing alcohol and other drug-related difficulties
IV. Workplace alcohol and drug testing
V. Trends and perspectives
VI. Summary and conclusion
Author: M. HANSON
Pages: 3 to 44
Creation Date: 1993/01/01
A flashpoint in the debate over workplace responses to alcohol and drug use by members of the workforce centres on the chemical testing of current employees and job applicants for alcohol and drug use. Drug testing may be the most contentious issue faced by enterprises struggling to develop fair and effective programmes. to deal with the consequences of substance use in the workplace. The present paper examines scientific evidence on the nature and extent of alcohol and drug use by members of the workforce, evidence linking alcohol and drug use -to workplace problems, workplace strategies for managing ,alcohol- and drug-related difficulties, and arguments for and against drug and alcohol testing. To date, the evidence supportive of alcohol and drug testing is, inconclusive. Testing programmes may be useful in identifying drug users in the workforce. Their deterrent value is uncertain, however, and they are not efficient -tools for linking drug users to assistance programmes. Enterprises that. are contemplating establishing testing programmes should consider:, (a) whether substance use is a problem in their setting; (b) whether testing will respond to the' problem; (c) the costs and benefits of testing; and (d) any ethical and legal questions raised by the programmes.
Alcohol and other drugs have had an extended and varied association with the workplace. At one time, for example, the British navy paid one third of each sailor's salary in rum. The first federal law on alcohol in the United States of America was passed when that nation was barely a year old; a 1790 statute authorized that every soldier be given a daily ration of a quarter pint of rum, whiskey or brandy ([116] , p. 129). In some countries, such as Germany, beer consumption by workers on the job has been a long and guarded custom ([20] , p. 92).
Despite this legacy, the ill effects of alcohol and other drugs have periodically alarmed many who are interested in the smooth and efficient functioning of enterprises. In the early 1800s, farmers in the United States and small business owners were distressed by the manner in which employees' excessive drinking affected their work performance. Such concerns were among the driving forces behind the emergence of the temperance movement in that country [93] . In the late nineteenth century, industrialists in the United States who were worried about their employees' health and such "undesirable" activities as heavy drinking, which endangered the operations of their factories, hired welfare secretaries to shape a "new, improved and contented" workforce that would not disrupt factory life.* A major motor vehicle manufacturer based in the United States hired investigators to probe employees' drinking habits during their off -work hours. Persons who would not, or could not, abide by corporate expectations often were summarily dismissed [25] , [31] , [50] , [55] , [111] . In the 1940s, management and workers' groups developed occupational alcoholism programmes (the forerunners of modern employee assistance programmes) to identify alcohol-related problems among workers and to persuade alcohol-impaired employees to seek assistance.
Recent responses to alcohol and other drug use by employees represent a continuation of earlier cycles of concern. A major difference in the current flurry of activity, however, is a growing sensitivity regarding the adverse consequences associated with illicit drug use by members of the workforce. In the 1960s, heroin addiction was deemed a threat to worker health and safety [1] . In the 1970s and 1980s, attention shifted to other illicit drugs, most notably marijuana and- cocaine. Regardless of the specific drug, the focus switched away from alcohol use (which continues to be the drug most widely used and abused by employees) to illicit drug use. Illegal drug use by employees was defined as a "serious" problem. The problem's severity was uncertain; ways to resolve it were unclear.
With the shift in attention to illegal drugs, enterprises were confronted with a new set of issues. In addition to the impact of drug use on work performance, productivity and business costs, new questions arose about such matters as workplace security, public confidence and unlawful activities by organization members. Traditional methods for dealing with substance use in the workplace were challenged; new, more comprehensive strategies were developed. In some countries (most notably the United States) the workplace was designated as a crucial battlefield in a "war on drugs".
A flashpoint in the current debate over workplace responses to alcohol and drug use centres on the testing of job applicants and current employees for drug (and to a lesser extent alcohol) use. Drug testing may be the most controversial and contentious issue faced by enterprises struggling to develop fair and effective programmes to deal with alcohol and other drug use in the workplace. It is the component of workplace substance abuse programming on which there is the least agreement.
*See Blair [22] for an early report on drug abuse in the workplace.
Chemical testing for alcohol, drugs or their metabolites in biological tissues and fluids is part of a broader, category of drug detection strategies designed to establish the presence of drugs, alcohol and/or related impairment among employees and other individuals.. Besides drug and alcohol testing, drug detection procedures include, but are not limited to, direct observation of drug use, observation of intoxication, direct observation of impaired behaviour, reports from informants, self-reports, neurological testing and behavioural (i.e. performance) testing.
In this paper, workplace drug testing (or drug screening)* is defined as "the process of obtaining samples of body fluids or tissues (e.g. urine, blood, hair, breath) from [job applicants and] employees and conducting laboratory analyses to detect the presence of certain drugs, including alcohol, and their metabolites" ([72] ,p. 1). The most widespread forms of drug testing are breath testing for alcohol and urine testing for other drugs. Although accepted standards for-impairment have been established for alcohol (e.g. through Breathalyser -readings), in most cases impairment is inferred from screening and testing results.
and related problems in the workplace
People have used and misused mood-altering substances since before recorded history, when the, first alcoholic beverages were produced, probably accidentally, by natural fermentation. Several factors converge, however, to frustrate e f forts to gather accurate information about alcohol and drug use and abuse patterns in any society [130] , [156] . For example, substance abuse is a stigmatizing activity in most cultures. Thus, biased reporting can occur in drug use surveys if individuals minimize the extent of their drug use and deny encountering problems. This tendency is magnified by the responses of others in society who may also downplay substance use and related difficulties in the community and the workplace to "protect" themselves, their community and their livelihood.* Because of these tendencies, many analyses of drug use rely on such indicators as arrest records, aggregate data on economic productivity, and hospital emergency room reports to develop inferences about the nature and extent of substance use and related problems. Although these measures can pro - vide gross estimates of substance use and problems linked with substance use, they underestimates many difficulties associated with alcohol and drug use, and they do not permit causal inferences about the relationships between alcohol and drug use and particular societal problems. As a result, there is considerable disagreement about how accurately national trends characterize and predict substance-related problems in the workplace [53] , [131] .
*Except where indicated, "drug testing" refers to both drug 'testing' and drug screening". According to Montagne, Pugh and Fink [95] , although the terms are used interchangeably, drug "screening" refers to a more non-specific qualitative analysis of a tissue or fluid sample for the presence of a particular drug. "Testing" involves a quantitative analysis of the same sample to confirm the presence of the specific drug identified in the screening analysis.
A second obstacle is created by the fact that the use of some illicit drugs (e.g. heroin) is relatively limited in many countries. Since small numbers of persons use drugs such as heroin it is difficult to obtain accurate usage estimates and to discover particular problems related to the use of those drugs. General population surveys tend to underestimate the presence of all illicit drug use in communities. Thus, extensive and intensive sampling of smaller "at-risk" subgroups is necessary to retrieve reliable and valid information. Because this type of sampling adds greatly to the cost of general population surveys, few countries conduct such epidemiological drug use surveys on a regular basis [156] .
A third barrier is related to the definitions of drug abuse and drug- related harm, which vary from country to country. Definitions of alcohol and drug abuse are shaped by cultural and social factors unique to different settings'. Therefore, cross-country comparisons of use patterns and difficulties associated with use are problematic. This dilemma is illustrated clearly by the experiences of researcher William R. Miller of the United States, who observed that, although his international audiences recognized the pragmatic value of classifying drinking problems based on adverse consequences correlated with alcohol consumption, perceived problem thresholds differed across countries:
*Managers may be slow to recognize drug abuse in their companies and may tend to se esubstance use as more problematic in firms other than their own [153] . Furthermore, cultures emerge in many enterprises that normalize and 'conceal' the presence of excessive substance use [45] . A recent survey of physicians [82] illustrates these tendencies. While the respondents viewed drug use as a problem for other health-care personnel,three fourths regarded physician substance abuse as non-existent or a minorproblem, an impression that is belied by reports that suggest that the level of drug use and abuse among physicians is comparable to that among the general population [21] , [114] .
"I, observed this in presenting the same data in various countries. The U.S. samples that I defined as 'problem drinkers' ... reported ... an average of 40-50 drinks per week (20-25 ounces or 0.6 to 0.75 of a litre of absolute alcohol). In Norway and Sweden, the audiences tended to be shocked by the amount of drinking and argued that my samples must consist of chronic addicted alcoholics. In Scotland and Germany, by contrast, audiences evidenced skepticism as to whether these clients had a real problem at all because such drinking, albeit heavy, seemed within normal limits" ([94] , p. 111).
Despite the above-mentioned difficulties, some general observations can be made about the nature and extent of alcohol and other drug use and related problems in the workplace.
Prior to the 1940s, alcohol and drug use and abuse were confined by drug type to relatively circumscribed regions of the world [15] As the world became smaller with the development of rapid transportation and communications, as well as the emergence of an international economy, all forms of drug abuse swept into wider global arenas. For example, heroin and opium use, once found mainly in Asia and North America, spread to Europe and developing countries; cannabis, which has a long history in south-east Asia, the Middle East and parts of Africa, is smoked now in all parts of the world; cocaine use, which at one time was concentrated primarily in Latin America and some sections of the United States, has extended to Europe and more regions of North America [42}.
Worldwide per capita alcohol consumption increased by, 15 per cent between 1965 and 1980 [43] , [145] . After doubling between 1950 and 1975, however, annual per. capita alcohol consumption in developed countries declined slightly from 8.4 litres to 8.1 litres between 1975 and 1980 [109] , [115] , [128] , [129] .
In commenting on these trends, Walsh and. Grant ([145] , p. 133) observed that, although Europe and North America remained the leading areas of alcohol production and per capita consumption, "the most notable feature of the 1960s and 1970s was the rapid growth of both production and consumption ... in areas where commercially produced alcohol was previously virtually unknown". Smart ([129] , pp. 102-103) asserted further that "alcohol is increasingly a global problem. Many developing countries will soon face major difficulties if present trends continue unchecked". While alcohol consumption has been low in Africa and Asia, consumption levels in some developing countries are approaching those of European countries.
Similarly, although the United States seems to be the major consumer of illicit drugs [49] , [61] and some countries are not yet noticing extensive illicit drug use by members of the workforce [19] , [69] , [77] , illegal drug use and misuse also have expanded into new areas. A 1989 World Health Organization (WHO) report [155] indicated, for example, that health problems associated with drug abuse were multiplying most rapidly in Africa, Latin America and the western Pacific. When international substance use trends are analysed against the background of global demographic trends, it appears that many developing countries are catching up with their industrialized counterparts. In many poor countries with exploding birth rates, more young adults entering the workforce are likely to bring with them problems associated with alcohol and illicit drug abuse.
Epidemiological data suggest that the substance use patterns of employed individuals resemble those for society in general [53] . Although some research suggests that unemployed persons have the highest rates of substance use [76] , virtually all experts agree that the majority of alcohol and drug users - and quite possibly a majority of those experiencing substance-related problems - are in the workforce [67] , [68] [112] , [122] .
A survey conducted in the United States, for example, found that 13.2 per cent, of employed men and 5.9 per cent of employed women could be diagnosed as alcohol-dependent [8] . These rates compared to 9.6 per cent and 3.2 per cent respectively for all men and women in the United States. Nearly 70 per cent of all United States citizens who disclosed in 1988 that they were current illicit drug users (i.e. had used illicit drugs in the past month) were in the workforce (over 10 million people); and over 8 per cent of the full-time United States workforce admitted to current illicit drug use, most often marijuana or cocaine [80] .*
Comparable findings have been uncovered in other countries. According to a survey of work sites in Ontario, Canada, up to 10 per cent of the workforce drank excessively and between 3 per cent and 7 per cent used drugs other than alcohol and tobacco occasionally or frequently [3] . Four per cent of all Canadians admit to troubles in their work lives due to drinking [6] , [27] ; heavy drug users represent around 1 per cent of the Canadian workforce [4] . Reports from Poland suggest that the majority of heavy drinkers in that country are in the national workforce. According to surveys carried out in Poland in 1980, 5.7 per cent of the respondents had had a "recent drink" in the workplace; over half had drunk at some time in the workplace; over the course of a year the average number of work-site drinking episodes was 17 for men and 9 for women [97] . In Germany, too, a majority of drinkers are in the workplace, with more than half (52:per cent) admitting to drinking on the job from time to time and approximately 10 per cent acknowledging workplace drinking every day or nearly every day [159] .
*Urine screenings of United States employees and job applicants corroborate the epidemiological findings. A United States postal service study [163] found that 7.8 per cent of newly hired employees tested positive for cannabinoids; about 2 per cent were positive for other illicit drugs. Results from the SmithKline Beecham Drug Testing Index showed that 8.8 per cent of those, screened tested positive for drugs in 1991 [132] . A 1992 American Management Association survey [12] found that in 1991 4.6 per cent of job applicants and 2.7 per cent of employees tested positive for drug use.
The implications of findings such as these are clear: although trends suggest that overall drug and alcohol use is on the decline in many countries, most people who use alcohol and other drugs are employed. Furthermore, while most individuals experience no adverse consequences associated with substance use, a substantial minority of workers encounter problems due to their substance use. Consequently, employee (and management) alcohol and drug use represent major challenges for the workplace.
Anecdotal and inferential evidence suggests that illicit drug use and alcohol abuse produce numerous problems in the workplace, including threats to work-site safety and security, higher accident rates, lower productivity, legal liabilities,-and dangers for the public [18] , [37] . Inter- nationally, business leaders, workers and others view substance use and abuse as critical issues confronting business and industry [10] , [61] ,[113] . Consequently, efforts to establish drug -free work environments are supported widely by most citizens [6] , [48] , [74] , [81] [86] , [90] , [105] , [147] .
It seems clear that drugs and alcohol can impair individual performance (e.g. perceptual ability, visual-motor coordination, interpersonal behaviour) [40] , [78] , [104] . Empirical data linking alcohol or drug use to workplace problems, especially when employees are not under the immediate influence of mood-altering substances, are equivocal, however [44] , [51] , [89] , [98] , [99] , [149] .* For example, respondents to a review of alcohol in employment settings, sponsored by the International Labour Organisation (ILO) and WHO, concluded that the '"overwhelming majority" of persons experiencing alcohol-related life problems are located in the active workforce [33] . An examination of the country reports, however, suggests that many countries (a majority in some regions) had no mechanism to collect systematically data about employee drinking practices. Similar findings were presented at a 1991 international symposium on alcohol and drug abuse prevention and assistance programmes in the workplace. For almost three fourths of the 53 workplace programmes surveyed as part of the symposium, systematic studies of employee alcohol and drug use either were not available or were not carried out prior to programme implementation [131]
*Most research connecting substance use to workplace accidents and other difficulties has examined alcohol consumption. Some experts have observed that findings from studies on the effects of alcohol on work performance can be misapplied if they are used to justify the need for particular workplace initiatives such as drug testing to counteract the effects of illicit drug use [64] , [98] .
Alcohol and other drug use has been cited as a factor in higher job turnovers, lower productivity, increased risk of, workplace accidents, greater absenteeism and other workplace problems. A YMO-sponsored international survey of drug abuse in 29 areas found that, although empirical studies are rarely conducted, 'in all countries drug abuse has important adverse economic effects. ... Drug use is recognized to affect employability in ... Indonesia, Japan and Peru. ... Careful estimates for Japan place the costs of drug use at 10 billion yen per year (nearly US$ 70 million). ... For the, Philippines, the total cost of drug abuse to the country is US$ 22,507,467' ([130] , p. 5). A 1984 study estimated, that the United States experienced a US$ 26 billion productivity loss due to illicit drug abuse [39] . Findings from the 1990 national household survey on drug abuse in the United States showed that respondents who reported that they had used marijuana or cocaine in the past year had household incomes that were lower than those of persons who denied any drug use [103] . Canadian productivity losses due to substance use have been estimated to be between US$ 1.7 billion and US$ 2.3 billion [2] , [5] .*
An investigation of Canadian railroad workers found that over one fifth (23 per cent) were "problem drinkers" and that 5 per cent admitted that they appeared for work "very drunk' or became 'very drunk" on the job one or more times during the study year [60] . Another Canadian survey found that 21 per cent of the respondents had seen co-workers drink on the job where it affected their performance and that 17 per cent had seen or heard of illicit drug use on the job [6]
*These figures should be used only to suggest trends and not to draw firm conclusions about the economic costs of illicit drug use. The lead author of the 1984 analysis stated that there was no significant difference between the household incomes of current illicit drug users and the incomes of comparable households having no drug using members (Henrick J. Harwood, referred to in [64] ; [98] . Furthermore, the authors of the report on the 1990 national household survey on drug abuse point out that 'few of the comparisons of income groups were statistically significant' ([101] , p. 133). .
An early study by Observer and Maxwell [134] discovered that alcohol abusers had twice as many accidents on the job as a comparison group of workers of the same age, sex, ethnicity, job tenure and job class. Two more recent studies showed that alcohol was the most common drug involved in occupational fatalities, being present in 10.7 per cent and 13.3 per cent of the cases [9] , [84] . Other evidence tentatively 'suggests that alcoholics/problem drinkers are approximately 2 to 3 times more likely to be in industrial accidents than non-problem drinkers' ([3] , p. 6; [63] ).
A study of job applicants for the United States postal service at Boston, Massachusetts, found that newly hired employees who tested positive for marijuana or cocaine had higher rates of job turnover, industrial accidents, absences, occupational injuries and disciplinary episodes during their first year of employment [163] . In a two-year follow-up study, it was discovered that the work-related difficulties experienced by persons testing positive for cocaine remained relatively unchanged during their second year of employment; however, the risk of work-related problems appeared to decline slightly for employees who had tested positive for marijuana use [121] .*
Another blind, multi-site study of the United States postal. service covering the first year of employment found that newly hired employees with drug-positive urine had higher absenteeism rates than other newly hired employees (6.6 per cent versus 4.2 per cent); however, it did not find statistically significant differences between the two groups, in accidents, injuries or involuntary turnover [106] .**
A study by a major motor vehicle manufacturer based in the United States found that drug -dependent employees who utilized employee assistance programmes experienced twice as many occupational injuries as non-drug dependent workers [150] . A case-control study involving hourly workers of a large manufacturing plant found that, over a five- year period, substance-abusing employees averaged significantly more absences than non-substance-abusing workers. They also were more likely to experience accidents and injuries [26] . Several studies of United States military personnel have concluded that illicit drug users are more likely than others to encounter work-related problems and to be discharged for unsuitability or performance problems [23] , [92] , [100]
*In commenting on their findings, the authors stated that the results at two years were consistent with those at one year; however, the findings "raise the possibility that a pre- employment screen may be decreasingly effective in predicting adverse employment outcomes associated with marijuana use after the first year of employment" ([121] , p. 1062).
**This study included over 4,000 newly hired employees in 21 sites throughout the United States..
Although some studies report a link between substance use and adverse work -related consequences, others suggest that an association may not be present. In a comparison of 72 diagnosed alcoholics and 204 randomly selected workers [137] no statistically significant intergroup differences were discovered in number of accidents, extent of injury and the number of lost-time accidents in the year preceding diagnosis.
According to autopsies conducted on 41 victims of workplace accidents in Allegheny County, Pennsylvania, there was one case in which alcohol might have been a factor and no cases in which drugs were pre- sent in the victims' systems ([136] , referred to in [44] ). In an earlier review of 1,800 employees over a 15 -year period in an unnamed industry, only one fatal accident was linked to drinking ([117] , cited in [44] ). An analysis of post- accident drug test results by the Federal Railroad Administration in the United States concluded that only 1 per cent of the train- men involved in accidents in 1986 and 1987 tested positive for abused drugs [142] .
A prospective study examining the association between pre- employment drug-testing results and employment status among a sample of 180 hospital workers found no statistically significant difference between the drug test-positive employees and drug test-negative employees in job turnover in the first year of employment [110] . Results from a study of utility plant workers revealed that promotion and demotion records did not clearly differentiate drug users from the total workforce. In fact, the rate of demotion for those testing negative for drugs was significantly higher than the rate for the total workforce [126] .
The reviewed literature suggests that a small but noticeable proportion of workers experience work problems associated with alcohol and/or drug use. It is difficult, however, to draw firm conclusions about either the extent of work-related problems associated with employee substance use or any causal role that substance use plays in the emergence of those problems. This does not mean that statements about the adverse impact of workplace substance use are inaccurate; it simply suggests that insufficient evidence exists to make any global assertions.
Since alcohol and drug users differ from non-users in many ways, it is difficult to conclude exactly what is causing observed work difficulties. Furthermore, given the influence of instructional and contextual factors on the effects of some drugs [127] , [144] , [160] , the effects of drugs in particular settings like the workplace may not be understood fully, unless evaluations are carried out "under conditions which approximate those in which people might be using [drugs], while at the same time controlling extraneous variables and protecting the participants from possible deleterious effects" ([46] , p. 126).
The goals of workplace strategies to reduce and prevent substance use-related difficulties are multiple and complex, and different constituent groups are apt to have different standards by which they judge the value of any alcohol and drug abuse management strategy. For example, although interests will overlap, government representatives may be most concerned about how workplace programming affects the overall drug use rate in a community. Management personnel may be more interested in reduced costs and increased productivity. Worker groups may place relatively greater emphasis on the protection of worker rights, guarantees of confidentiality and assurances of workplace safety.
Numerous factors are thought to motivate enterprises to establish substance abuse programmes and initiatives [91] , [122] , [133] . The following are among the motives frequently cited:
"Selfless altruism", a concern about the well-being of employees and the promotion of the common good as ends in their own right;
"Selfish altruism", a belief that in the long run firms will save money by assisting workers (Curran and Kiefhaber, cited in [133] ;
Pressures to comply with legal regulations and mandates;
Concern about legal liabilities;
"Public relations", a desire to promote the corporate image;
Safety and security concerns;
A wish to improve productivity and profitability.
It is apparent that programmes established with different purposes will have differing goals and objectives; and they will utilize different criteria for determining their success.
A recent ILO-sponsored review of workplace drug and alcohol abuse prevention and assistance programmes found that a majority of the surveyed programmes combined a focus on substance abuse problems with attention to other difficulties faced by workers., More European than North American programmes attended only to alcohol and drug problems, however. An important finding of the review was that employee welfare, health and safety were the primary programmatic concerns. Improving productivity and meeting legal mandates were given lower priority by the study respondents [131] .*
The ILO findings are consistent, with those of other reports. Since the mid-twentieth century, assistance and prevention efforts have evolved considerably from the days of organizationally based occupational alcoholism specialists through the rise of external consultants who offer expert services to enterprises of which they are not members [124] . With this evolution has come a shift in emphasis - programmes have placed increasing significance on worker lifestyles, health promotion and primary .prevention ([31] ; [34] cited in [124] ). 'Long-term employment has become an objective, ..., which necessitates actions toward maintaining the employees' well-being and health" ([124] , p. 36). Consequently, work- place prevention and assistance programmes are seen less and less as management tools. More often they are perceived as employee benefits [35] , [68] , [91] , [138] .**-
The reports cited earlier underscore the fact that the relationship between substance use and workplace problems is not constant. Associations vary depending on factors such as industry type, job location, job classification, worker age and worker experience [125] . Thus, experts assert that enterprises must conduct surveys and needs assessments to establish the nature, extent and consequences of substance use in their particular settings before programmes are designed [71] . It is critical that the specific needs of different organizations and localities are identified and understood. By using focus groups, self-report surveys, and indicators such as accidents, employee turnover, absenteeism rates, insurance claims, grievances and disciplinary incidents, and changes in productivity, enterprises can begin to identify their needs.*
*Although the study respondents did not we their programmes as extensions of legal control mechanisms, legal mandates may have an indirect effect on workplace programming. In the United States, for example, there has been a marked increase in workplace drug testing since the mid-1980s, when the federal Government became more concerned about drug use by workers and enacted several new laws and regulations.
**This position is reflected in the fact that the cost-effectiveness of assistance and prevention programmes is not a major issue for many corporate executives [56] , [65] .
In addition to establishing need, experts state that programme planners should identify any legal requirements that apply to their organizations and that might shape programme design and development; they should familiarize themselves with the policies and initiatives of other organizations; they should assess the organization's capability to deal with the problems; and they should identify the external and internal resources available to aid in programme development, implementation and operation (e.g. budget, staff, treatment facilities, information clearing- houses, health insurance) [73] , [139] . Once need is established and the initial assessments are completed, programmes can be created and tailored to respond to' the particular problems encountered by different organizations [118] .
It has been argued that strategies to manage workplace alcohol and drug problems should fit within an organization's overall structure for ensuring the quality of work life and maximizing organizational effectiveness and efficiency. Because the needs of an organization's many constituents will differ, strategies should be jointly developed to reflect workplace diversity, as well as the interests of organizational members, government, and worker groups [11] , [29] , [70] , [102] .
Several core components have been described as essential for any comprehensive substance abuse management strategy [102] :**
Explicit policies on alcohol and other drug use are the foundations of comprehensive alcohol and drug abuse management programmes [73] . Policies should include:
Clear, written statements of goals (e.g. creating work environments that are free of substance use and abuse and that are not disrupted by 'alcohol and other drugs);
Measurable objectives (e.g. establishing drug detection programmes, employee education, assistance programmes and supervisory training by specified dates);
Delineation of operating procedures (e.g. statements about whom the programme covers, who operates the programme, and employee rights and protections) [61] , [146] ;
Employee education should be present. Included in educational initiatives should be primary and secondary preventive efforts that inform employees about alcohol and other drugs and their adverse consequences; about the organization's drug policy; about their responsibilities 'and options under the policy; about the roles of company representatives and supervisors; about options for assistance; and about drug-testing policies and procedures, if present. The goals of employee education include pre- venting drug use by providing employees with information, by influencing organizational attitudes and values and by mobilizing individual action to deter substance use;
Other primary prevention efforts, such as health promotion campaigns, corporate wellness initiatives and family education, should also be established. It has been pointed out that primary prevention requires a shift from an individualistic focus to one that is systemic and contextual. The targets of workplace primary prevention efforts include the work- place (e.g. inculcating healthy values), the individual (e.g. developing problem-solving skills) and the community (e.g. promoting efforts to reduce drug-related harm and to create tolerant communities) [71] ;
Supervisory training includes activities that will help supervisors to identify employees with impaired work performance, to document poor performance, to take appropriate disciplinary steps, to mobilize troubled employees to seek assistance, and to reintegrate them into the work setting following treatment. Supervisory training should enable supervisors and managers to fulfil their responsibilities under an organization's alcohol and drug abuse prevention and assistance policy [61] , [73] , [139] ;
Employee assistance includes all organized efforts that exist in work-sites that are designed to identify troubled employees and problematic working conditions to assess the nature of the difficulties to make appropriate responses; and to monitor workplace and workers' conditions to prevent recurrence of the difficulties [73] . Assistance programmes may be in-house programmes, external (contractual) programmes, or consortia, which are owned by, and provide services to, several (usually smaller) businesses. intervention efforts may be directed at individual employees and their family members and/or they may attempt to bring about change at the organizational level [7] ;
Alcohol and drug detection efforts are designed to identify and deter employee substance use. Although there are several ways to spot employee substance use, the most controversial methods are those involving drug and alcohol testing. The goals of alcohol and drug detection efforts include discouraging and deterring substance use; identifying, treating and rehabilitating current substance abusers; and identifying and dismissing from employment persons who will not or cannot be rehabilitated [139] . Although there is considerable international debate about the need for alcohol- and drug- testing programmes, most authorities believe that, when they are established, testing programmes should be part of a total organizational package to reduce and prevent workplace alcohol and drug use [47] , [57] .
*The assessment by Yamatani and others [157] of workers' problems, associated work performance difficulties, and financial costs in a Pennsylvania manufacturing company with 1,800 employees is a useful example of the manner in which needs assessments can inform programme development. Anglin and Westland's paper [14] is an illustration of how urine analysis data can be used to establish baseline data that, in turn, can be used to monitor changes in drug use following the implementation of a workplace initiative.
** These components have been developed most extensively in the United States. Except for alcohol and drug testing, most international experts and practitioners are in agreement about their importance.
Methods for drug and alcohol detection have existed for centuries. Haber [52] believes, for example, that urine testing (through visual inspection) dates from the time of Hippocrates. Miller and his associates [36] establish the origin of workplace drug testing with the rise of the American occupational medicine movement in 1916. According to Montagne, Pugh and Fink [95] , several tests for detecting alcohol use were developed in the early twentieth century; they were based on physical assessment, crude biochemical measures and behavioural observation. By the 1920s, glass capillary kits were used by some police departments to spot drunken drivers. In the 1930s, when more portable devices were developed, breath testing replaced blood testing, and chemical testing for alcohol use became more widely accepted [24] , [95] .
Observational and clinical tests for narcotics use were employed by hospitals, drug treatment programmes and the criminal justice system before 1950. By the latter half of the 1950s the nalorphine pupil test emerged as the standard means to uncover narcotics use by parolees [95] . The technology that enabled relatively accurate mass urine screening in the workplace was not developed and refined until the early 1960s, however [11] , [57] , [151] . For the most part, these initial mass screening methods were moderately specific (i.e. they correctly identified drugs) but relatively insensitive (i.e. they did not detect low concentrations of drugs) [152] .
Drug (and to a lesser extent alcohol) testing emerged as major work- place issues in the 1980s. Several forces seemed to converge and become catalysts for the renewed interest in drug testing - especially mass, compulsory testing of current employees and job applicants:
Technology developed to the point where inexpensive and reliable means were available for spotting and identifying drugs and their metabolites;
Concern about the adverse consequences of drugs in the workplace reached a critical point [12] , [147]
Rising-health care costs and liability insurance became critical factors in some countries (e.g. the United States) [12]
In the United States, governmental policy and regulations, such as the Drug- Free Workplace Act of 1988 and Defense and Transportation Department mandates, encouraged drug testing as a means to reduce and prevent drug-related harm in the workplace.* Younger [158] , for example, discovered that employers who thought they were accountable under the Drug-Free Workplace Act were much more likely than other employers either to have, or to be planning to develop, a testing programme (81 per cent of those who believed they were accountable compared with 56 per cent, of all respondents).
A number of arguments have been put forth to justify workplace guments have been put forth to justify workplace drug testing [28] [72] [123]
The primary rationale for testing programmes is the belief that they will promote work-site safety and reduce accident risk. It is argued that, by deterring drug use, testing programmes will ensure the safety of both co- workers and members of the general public who might otherwise be, injured in accidents caused by drug-,impaired workers;
A second rationale is that drug testing will enable enterprises to identify employees who might develop drug problems and to encourage them to seek assistance before their conditions deteriorate. By using the threat of job loss as the "stick" and assurances of continued employment as the "carrot", it is believed that drug users will be mobilized to seek help rather than be dismissed;
A third rationale for the existence of drug-testing programmes is that they will lead to increased organizational productivity and lowered workplace costs. Drug use has been correlated with higher absenteeism and job turnover rates in some studies. Thus, it is asserted that reducing and eliminating drug use in the workplace will lead to better attendance and corresponding productivity increases, as well as cost savings that accompany the retention of experienced workers;*
A fourth rationale-for establishing drug-testing programmes is that they will improve the quality of work life. It is argued that, by promoting drug-free values and norms, testing programmes ultimately will lead to higher employee morale. Furthermore, it is believed that, by demonstrating a commitment to a drug- free work environment, organizations employing testing programmes will foster public trust and improve their corporate images;
Another rationale for the existence of drug-testing programmes is the need for organizations to meet legal obligations. Occupational safety laws in many countries dictate that companies must provide for the safety of the work environment. Drug- testing programmes are one way to do this. In addition, by establishing mechanisms for identifying and responding to drug -impaired workers, corporations may be able to reduce the extent of any legal liabilities they might incur as a consequence of employee drug use.
*Although policies and regulations supportive of drug testing were introduced most extensively in the United States, policies in that country had a major impact on other countries (e.g. through regulations that applied to international transport workers and by the responses of international corporations based in the United States [28] .
For every argument in favour of workplace drug testing there is an argument against it [28] , [31] , [41] , [54] , [59] , [64] , [75] , [99] , [107] , [108] ,
The most frequently voiced challenges to drug-testing programmes include the following.
(a) They are not justified based on the evidence. Opponents adopting this view point out that not only is the evidence linking drug use and workplace difficulties weak, but also testing programmes themselves do not measure impairment that is likely to lead to workplace problems.** It is argued, therefore, that random and not-for-cause testing represents unreasonable searches under the circumstances;
(b) A second argument against drug testing is that the testing of bodily fluids is an unwarranted invasion of privacy. Testing procedures can be intrusive, demeaning and embarrassing. In addition, test results can provide information unrelated to an employee's drug use that can be
*Related cost savings are said to be the lowered health-care costs incurred by a drug- free workforce.
**A related argument is that the evidence that does exist establishes alcohol as more problematic than other drugs in the workplace [4] . Yet, in most settings, testing programmes focus only on illegal drugs and not on alcohol use [154] . used in a discriminatory manner (e.g. information about medical conditions and pregnancies);
(c) A third argument asserts that mandatory drug testing is a form of social control that taps employee lifestyles and not their work performance. Drug tests most reliably indicate the presence of drugs (or their metabolites), not impairment. The only legitimate concerns of employers should be whether their employees can complete their work safely and effectively. Since employers are. not law enforcers, whether or not persons use drugs should be of no interest to them;
(d) A fourth argument is that testing procedures are flawed. Some tests yield high rates of false-positive results, and companies do not always conduct confirmatory tests [28] . Furthermore, drug tests do not measure impairment or recency of use;* they do not establish that a user is aAn abuser of drugs [95] , [123] . Thus, results which are misinterpreted and misapplied can have devastating effects on workers and in work- places;
(e) Another argument opposed to drug testing suggests that testing programmes are not effective and that better options than drug testing exist for identifying and assisting drug users in the workplace [41] , [108] . According to this argument, drug testing does not actually identify drug abusers and that abusers may be more likely than drug users to success- fully sabotage tests; testing does not deter drug use [41] ; it does not encourage users to get help [68] -, and it may not be cost-effective in all work settings ([85] cited in [96] ; [162] ). Because of these limitations, it is argued that alternatives such as performance testing, educational initiatives and assistance programmes may be more useful means of identifying drug users and of encouraging persons with problems to seek assistance [3] , [41] .
Positive drug and alcohol test results indicate that a test has uncovered the substance that was tested for or its metabolite. Positive results may mean that the worker is any of the following: an intermittent user of the substance; a chronic user; addicted to the drug; under the drug's immediate influence; or taking the drug according to a physician's prescription. It may also mean that the result is incorrect (e.g. a false-positive result of an initial screening test). Negative test results may mean that the worker either does not use the drug that was tested for or has used the substance but in insufficient quantities or too infrequently to be detected. A negative result also might indicate that a specimen has been tampered with [30] .
*Except for alcohol, drug dosage has not been linked reliably with behavioural impairment [128] . The National Institute on Drug Abuse of the United States acknowledges that 'positive results of a urine screen cannot be used to prove intoxication or impaired performance' ([122] , p. 131).
As has been asserted several times, most tests can establish only that a drug (or, more frequently, its metabolite) is present; most cannot prove impairment. Therefore, drug screening tests alone - particularly urine tests - lack the validity needed for either disciplinary responses or referrals for substance abuse treatment [66] . Because of limitations and risks inherent in urine tests, most experts urge that clear guidelines should be established and followed when tests are employed. First, screening tests that are qualitative and fairly non-specific in nature should always be followed up with confirmatory tests. It is critical that confirmatory tests are based on chemical principles that differ from those of the initial test and that they are not prone to the same type of errors to which the initial screening tests are. vulnerable. For example, if the initial screening test utilizes an immunoassay technique, the, confirmatory test should rely on chromatography, preferably gas chromatography/mass spectrometry [141] , [101] .
Secondly, the testing standards of the laboratories conducting drug tests should be monitored closely. Laboratories vary widely in their accuracy, diligence and reliability; it is essential that quality-control procedures are instituted to ensure, the accuracy and reliability of testing programmes [128] . It has been suggested, for example, that split portions of the same urine specimen should be compared to determine whether they yield, the same results, and that specimens with known quantities of a drug should . be sent for analysis and that the results should be examined [101] .
Thirdly, clear procedures (e.g. chain of custody guidelines) for ensuring the integrity of each urine specimen should be established for certain kinds of tests to prevent substitution or contamination of specimens. In addition, an independent physician (i.e. medical review officer) who is knowledgeable about substance abuse and the interpretation of laboratory results should be employed to clarify the meaning of all test results, taking into consideration the worker's medical history and other relevant information [61] , [87] , [135] .
Fourthly, fair and equitable procedures should be in place to allow workers to challenge test results. For example, workers should be able to present information that explains the results; they should be able to request re-analysis; they should be allowed to submit specimens for analysis by their own experts. Final disposition of cases involving positive test results should not occur until the worker has been assured of due process.
Once the safeguards have been employed and organizations are certain that the positive test results are valid, action appropriate to the work setting should be initiated. In the case of job applicants, the individual may or may not be hired. If the applicant is not hired, he or she should be informed that a job was not offered because of a positive test result. In the case of current employees, workers should be referred for assistance, should be reassigned to non-sensitive positions or should receive disciplinary action, including termination, depending on the circumstances involved [101] .
The vast majority of enterprises, including those in the United States' where testing programmes are most widespread, do not test for alcohol or drugs. Evidence suggests, however, that drug, testing may be the component of comprehensive workplace assistance and prevention programmes that has grown the fastest in recent years.
The American Management Association's sixth annual survey of its membership reported an increase of 250 per cent in the number of drug- testing programmes between 1987 and 1992 [12] . Whereas 21.5 per cent of its membership reported that they tested for drugs in 1987, by 1992 the percentage had risen to 74.5 per cent. The 1988 Conference Board survey of United States businesses revealed that just under half (49 per cent) of its 681 responding firms had drug-testing programmes [16] , [17] . Also in the United States Bureau of Labor Statistics surveys of private-sector, non -agricultural business establishments found that 3.2 per cent had testing programmes in 1988 and 4.4 per cent had testing programmes in 1990 [58] , [140] .* A survey of 44 small service -and manufacturing companies in the Chicago area reported that 56 per cent of the respondents had drug-testing programmes or planned to establish them [158] .
Pre-employment testing is the form of testing employed most frequently by United States businesses. Respondents to the Conference Board survey reported, for example, that 92 per cent of the firms with testing programmes conducted pre-employment tests [16] . Among the respondents of the Bureau of Labor Statistics survey, 85.2 per cent of the establishments with drug - testing programmes tested job applicants,, while 63.5 per cent tested current workers [140] . Testing programmes are more likely to be located in larger firms. They also are more common in industries such as mining, manufacturing, transportation, utilities and communications.
*Although the Bureau of Labor Statistics survey discovered no statistically significant change in the incidence of drug- testing programmes from 1988 to 1990, this outcome was explained by the lack of change in small businesses employing fewer than 50 people. Among medium-sized establishments (with 50-249 employees) and large-sized establishments (with 250 or more employees), the incidence of drug- testing programmes increased to 26.1 per cent and 45.9 per cent, respectively. Small firms also were more likely to discontinue testing programmes and were less likely to implement new testing programmes between 1988 and 1990 [58] .
A survey of personnel managers in the federally regulated Canadian transportation system found that nearly one fifth of the organizations (19.5 per cent) had drug-testing programmes and 14.5 per cent engaged in alcohol testing. As with United States businesses pre-employment testing was the most common type of programme (present in 80.6 per cent of the organizations with testing). Less than half of the testing organizations (44.4 per cent) conducted periodic testing with medical examinations; one third (36.1 per cent) conducted post-accident tests [88] .
An ILO review found that one third (18) of 53 surveyed organizations conducted some type of a drug-testing programme [131] . Testing was more common in North America than in Europe. For-16 of the programmes, testing occurred at pre-employment and following work- related incidents. Only four firms conducted random, tests. Urine tests were the most common testing methods used, occurring in 17 instances. Interviews (6), breath tests (5), blood tests (5) and other measures (5) were employed less often.
The effectiveness of drug- testing programmes can be assessed in a number of ways, including the following:
Their capacity to identify drug users in the workforce,;
Their deterrent impact,
Their effect on worker morale;
Improvements in productivity;
Decreases in accidents;
Their linkage and relationship with other components of a comprehensive drug-free workplace programme;
Their cost-effectiveness.
Few evaluative studies that respond systematically to these assessment issues have been published. By extrapolating from the current literature, however, some inferences can be drawn.
By comparing studies that report results from the drug testing of current employees and job applicants to the findings of epidemiological surveys, an estimate can be made of the capacity of drug-testing programmes to identify current drug users. In the United States, the 1988 National Household Survey on Drug Abuse estimated that about 8 per cent of full-time employed persons reported that they had used an illicit drug in the past month; 6.8 per cent reported current use of marijuana. Among unemployed individuals the rates were 18.2 per cent and 14.8 per cent, respectively [80] . Findings from the drug screening of employed persons and job applicants in the United States include the following:
The 1988 Bureau of Labor Statistics survey found that 8.8 per .cent of current employees who were tested had positive test results; and 11.9 per cent of all job applicants who were tested were positive for drug use [140] ;
Among applicants for the United States Postal Service at Boston, 7.4 per cent tested positive for marijuana; 2.0 per cent tested positive for cocaine; and 2.7 per cent tested positive for other drugs [163] ;
The SmithKline Beecham Drug Testing Index revealed that in 1991 8.8 per cent of American workers and job applicants tested positive for drugs [132] ;
The 1992 American Management Association survey reported that in 1991 2.7 per cent of current employees who were tested for drugs tested positive; and 4.6 per cent of the job applicants who were tested did so [12] .
These findings suggest that drug testing may effectively identify drug users in the workforce. The SmithKline Beecham, Bureau of Labor Statistics and United States Postal Service findings parallel closely those of the National Household Survey. The American Management Association results, however, suggest that drug testing may under-identify drug users in some businesses.*
A possible deterrent effect of drug testing can be assessed by observing decreases in the rate of positive test results over time. Two United States reports offer information on changes in test results over time. SmithKline Beecham plotted a steady decrease in the rate of positive test results between 1987 and 1991.'the overall drug-positive rate for current employees and job applicants was 18.1 per cent, in 1987, 13.6 per cent in 1988, 12.7 per cent in 1989, 11 per cent in 1990 and 8.8 per cent in 1991 [132] . The 1992 American Management Association survey reported
*When interpreting the results, the reader should be aware that the studies' samples were not comparable. Also, the Postal Service study included only job applicants, while the SmithKline report combined the results for applicants and current employees. Thus, their results might be higher because unemployed persons were included in their samples. similar trends. The test-positive ratio among current employees was 8.1 per cent in 1989, 4.2 per cent in 1990 and 2.7 per cent in 1991. Among job applicants, the test-positive ratio decreased from 11.4 percent in 1989 to 5.8 per cent in 1990 and to 4.6 per cent in 1991 [12] .
Although these results may indicate that there has been a decline in drug use in the workplace and that that decline may be due to workplace drug testing, taken alone they do not allow such a conclusion. Two alter- native explanations for the declining test-positive ratios illustrate the dilemma:
(a) Earlier testing results yielded test-positive ratios of 14-18 per cent. Later testing results showed significant declines in the test-positive ratios. During the same period in which there was a decline in test- positive ratios, national surveys in the United States and elsewhere documented an apparent decline in drug use among the general population. The decline in test-positive ratios in the workplace may be reflective of this broader trend and might be explained by age/gender changes that occurred during this time rather than the presence of testing and preventive activities alone;
(b) There was a marked increase in both the number of drug tests and random drug tests during the period that saw a decline in the rate of positive tests. As the authors of the American Management Association survey noted, "statistically, the most important factor in the decline of positive test ratios [among current employees] is the increase in the testing. pool, due to the growth in random or periodic testing. As more employees are tested for reasons other than suspicion of use, the test-positive rate falls" ([13] , p. 3 [emphasis in the original]). They add that the best statistical indicator of declining drug use in the workforce is the 'dramatic decline' in the test-positive ratios for job applicants. Even with job applicants, however, certain factors, such as company policies that explicitly inform applicants that they will be tested and an increase in older job applicants (due to the recession) who are less likely to be drug users, cannot be ruled out as explanations for the decline.
Taken as a whole, the results permit the conclusion that the proportion of positive drug test results has declined steadily. When combined with national epidemiological data, they suggest that drug use in the workforce is subsiding. They do not demonstrate that testing, prevention or any other particular workplace initiative is responsible for the observed changes. Other explanations, such as changes in demographic trends and statistical artifacts, must be ruled out as possible reasons for the apparent decline in employee drug use.
The impact of drug testing on employee morale is an important evaluation issue, especially since work performance can be affected by morale. Public opinion polls in the United States suggest that the public and the workforce are sympathetic towards drug testing. A 1989 survey of 919 active union members who belonged to 13 major United States unions found that 76.6 per cent of the respondents supported drug testing under controlled conditions where the individual's privacy is protected". Two per cent favoured unrestricted testing. Respondents who favoured drug testing tended to believe that unions should share in the responsibility for implementing testing policies and that drug testing should be negotiated [83] .
A study of 500 intercollegiate athletes required to participate in mandatory drug testing provides insight into the impact of drug testing on morale [32] . Most (71.4 per Cent) of the participants were not greatly disturbed by the testing experience, and some perceived beneficial out- comes, including a socially acceptable way to refuse drugs and a decline in their "partying'. Some, however, experienced adverse consequences, including a fear of being falsely accused, embarrassment and 'hassles'. The respondents suggested that the acceptability of drug testing could be improved by providing for participant dignity, having effective orientation and educational sessions, and more rigorous testing standards.
It is unclear how closely linked drug -testing programmes are to other components of comprehensive assistance, and prevention programmes. The 1992 American Management Association survey reported that 57.2 per cent of its respondents combined drug testing with some other anti-drug programme. In only 6.5 per cent of the firms were employees tested without other programmes being present [12] . [16] .
The testing of current. employees may not be a very efficient way of linking individuals with treatment. Surveys suggest that nearly three fourths of the positive test results are for employees who work in organizations with employee assistance programmes. Yet, one survey of employee assistance programmes (Backer and O'Hara, referred to in [68] ) found that practically none of the clients sought assistance because of a positive test result (the report does not indicate how many persons were referred for treatment but did not make contact with an employee assistance programme).* The potential gain from combining testing with other drug - free workplace initiatives is illustrated, however, by the 1992 American Management Association survey. The majority of respondents that combined drug testing with other programmes reported significantly lower test-positive ratios among current employees in 1.989, 1990 and 1991 than did companies that had testing programmes only. In 1991, for example, the test- positive ratio for businesses with combined programmes was 2.6 per cent; the test-positive ratio for firms with testing only was 4.1 per cent [12]
*The authors of the report of the Institute of Medicine point out that most Surveys of employers drawn on-representative samples and have low response rates. Thus, their results must be interpreted with caution. Although most companies indicate that they will refer workers with positive test results for treatment, there is no clear estimate on how often this happens ([68] , p. 124; [143] ).
When they were asked whether they thought drug testing was an effective way to deal with workplace drug abuse, 82.3 per cent of the companies that tested replied "yes" in the 1992 American Management Association survey. Respondents (12.8 per cent) who did not believe that testing was effective tended to be from smaller firms in service industries, who tested because of government mandates. Only 7.5 per cent of the firms had attempted cost-justification studies of, their drug-testing programmes, however, [12] .
Anglin and Westland, based on an examination of the completed urine analyses of commercial drug-testing laboratories in California, question the value of drug testing in the workplace: "When we consider that 1 or 2 out of each 100 tests in the employee population are positive ... we have to decide whether we really need to test for drugs in employee populations" ([14] , p. 95). Zwerling and. Ryan [161] point out that whether drug testing saves money depends on both the prevalence of drug use in the populations screened and the costs related to adverse outcomes such as accidents. In a cost-benefit analysis of the Boston Postal Service study, it was concluded that drug screening could have saved the Postal Service US$ 162 per applicant hired. If, however, the prevalence of drug use in the target population had been I per cent rather than 12 per cent, the programme would have lost money., Furthermore, if the cost per urine sample screened had been US$ 95 instead of US$ 49, the programme would have lost money even if the prevalence of drug positives were as high as 9 per cent [162] .
The number of workplace drug-testing programmes has grown noticeably, especially in the United States and primarily in larger business firms and companies responding to legal mandates and governmental regulations. Although there has been a growth in the absolute number of all categories of testing programmes, the most dramatic increase has been in the periodic and random testing of current employees. Drug-testing programmes have some utility for identifying current drug users in many businesses. Although the degree to which they may deter drug use is uncertain, their presence is associated with an apparent overall decline in the rate of drug use in the workforce. The acceptability of drug testing by employees in the workplace may be contingent on programme design and existing legislation. Drug-testing programmes have not been shown to be efficient means of linking employees with treatment services. Their effectiveness seems to improve, however, when they are combined with other workplace assistance and prevention initiatives. Although most employers feel that testing programmes are effective means to deal with drug use in the workplace, little has been done to evaluate systematically their effectiveness or to establish their cost-effectiveness.
Workplace alcohol- and drug-testing programmes are affected by, and raise, multiple and complex legal and ethical issues. As observed earlier, the presence of laws in many countries have spurred not only the development of testing programmes, but also the form they take. In the United States, for example, mass random testing increased in interstate transportation industries following the enactment of specific legislation. In Canada, however, current interpretations by human rights commissions seem to preclude mass random testing, especially pre -employment screening that occurs before employment has been offered.
Any drug- or alcohol-testing programme must weigh the rights of individual workers against the need to ensure the public's safety and health. Since any testing programme carries costs associated with its intrusion on individual human rights, these costs must be offset by the benefits to be gained by the programme - both for the worker whose rights may be compromised and for the general public.
Several ethical and legal questions emerge when drug-testing programmes are discussed [30] , [96] , [123] , for example:
Does random drug testing constitute an unreasonable and illegal search and seizure, since no obvious "cause" precedes a test?
Does drug testing violate a worker's right to privacy by revealing information unrelated to work and unrelated to any drug use?
Will drug testing intimidate workers and lead to unwarranted discrimination in the workplace?
Do testing programmes violate "due process', especially in the absence of proper confirmatory testing and vague chain-of-custody procedures?
Will testing programmes divert attention away from alcohol abuse, which continues to be the major drug of abuse in the workplace?
Many different responses have been recommended to avoid the legal and ethical pitfalls that may befall testing programmes. In Canada, for example, the Ontario Human Rights Commission has stated that pre- employment drug testing, before a written offer of employment, should be prohibited [3] . The laws of some countries preclude mass, random workplace testing. The following are among the many guidelines that have been offered to ensure that testing programmes operate legally and ethically [101] , [123] :
Before establishing testing programmes, organizations should familiarize themselves with the laws that apply to the workplace in question-,
Testing programmes should be jointly developed by labour and management; they should be subject to standard collective bargaining principles;
A firm's testing policy, including its goals, objectives and procedures, should be written and communicated to all affected parties before a programme is implemented;
Competent, certified liaboratories should be employed to conduct all analyses. Results should be reviewed by expert, independent medical specialists. Quality -control procedures should be instituted to monitor the analytic and testing procedures;
Testing programmes should be part of a comprehensive organization-wide effort to reduce and prevent alcohol and drug problems in the workplace. Like other work place prevention and assistance initiatives, testing programmes should focus on alcohol and drug use only in so far as they relate to the workplace and work performance;
With the exception of certain specified high-risk and safety- sensitive positions, the testing of employees should be conducted only when there are reasonable grounds to suspect health, safety or performance problems.
It is generally agreed that workplace substance abuse initiatives, including drug-testing policies, should be designed and developed 'con- textually" [70] , [71] . That is, policies and programmes should reflect the needs, realities and constraints of particular communities and work set- tings. National attitudes and experiences with workplace drug and alcohol problems vary widely. Thus, 'standardized' policy packages that are imposed from outside and that do not respond to the unique circumstances of specific enterprises in particular settings seem doomed to failure.
Unfortunately, in an interdependent world, the policies of one nation (or organization) often have unanticipated effects in other nations (and organizations). Initiatives with benefits for specific localities can pose major costs in other jurisdictions.
Drug- testing policies developed in the United States provide a clear illustration of the way a nation's policies create dilemmas for other countries. Workplace drug-testing policies have been developed more extensively in the United States than in other countries. Several experts have observed that United States drug-testing policy has had a major impact on organizations in other countries [19] , [28] , [70] , [118] . For example, American policy affects foreign companies that do business in the United States. It also penetrates other countries when companies based in the United States try to enact their testing programmes in their foreign subsidiaries.
The dilemmas posed by local policies seem to be particularly problematic in the transportation, maritime and petroleum industries. It has been observed that United States Department of Transportation regulations require drug testing for aviation, pipeline, maritime, rail and trucking industries operating in the United States. Thus, foreign transportation firms doing business in the United States "will be compelled to introduce testing programmes in order to conduct business across the border" ([28] , p. 3). While many countries have no laws prohibiting such programmes, the manner in which they are introduced can have major ramifications for constitutional guarantees in sovereign States, and they may force Governments and other interest groups to propose legislation and policy to protect their own self-interests. In addition, even if countries accept the United States policies, technical problems are created because' there are few testing facilities in some jurisdictions and there is no international standard governing testing procedures.
To date, there has been little concrete discussion aimed at developing uniform international standards for drug testing. And there has not been much consideration of how to better coordinate drug-testing policies at the international level. Constituent groups have called for greater coordination of effort, a clearing-house for national and international polices, and additional study on the global impact of local initiatives [70] .
Because most drug testing tends to be invasive, a threat to individual rights, and not clearly linked to impaired work performance, a number of alternatives to testing (in particular chemical testing for illegal drugs) have been proposed. Some experts have urged that greater attention should be directed at legal drug use, especially alcohol consumption. Alcohol abuse has been linked to workplace difficulties more often than has other drug use. In addition, impairment standards for bodily alcohol levels are more apt to be agreed upon. Therefore, it is argued that, if chemical testing programmes are to exist, they should include alcohol testing*.
Alternatives to the chemical analysis of bodily fluids that are more directly linked to performance and that do not raise the complicated ethical and legal dilemmas associated with chemical testing also have been suggested. For example, it has been, proposed that supervisors be better trained to improve their, ability to identify and respond to safety - related performance: problems ([120] , referred to in [72] ). In addition, behavioural and computer tests have been developed to measure performance directly. A decade ago, a review identified over 50 tests -measuring psychomotor functioning, [62] . Computer tests that measure eye-hand coordination are used by truck drivers at a California petrochemical company [38] , [61] .
According to Kleiman [79] , tests -of impairment may be performance-based and. detect generalized substandard performance (e.g. a flight simulator) or they may be neurological based to spot specific signs of drug7induced neurological impairment (e.g. an electrosynstogmograph, which reportedly can detect drug-specific muscular movements in the human eyeball). There are obvious advantages of performance-based alternatives to chemical testing:
They have immediate relevance.to work performance;
They can detect levels of impairment;
They shift focus away.from drug use per se and towards performance and safety issues.
There are also some disadvantages associated with performance- based testing:
The technology is still being refined and may not be feasible for mass use;
To be utilized properly, all workers must have baseline performance levels established against which test results can be compared, which adds to the cost of testing-,
Some drugs may interfere with higher mental processes that are not readily detected with performance tests.
*Recently, the United States Department of Transportation proposed random alcohol testing for transportation workers. Under the proposal, transportation workers would be subject to breath tests at pre-employment, when there is 'reasonable suspicion', and after accidents and to monitoring following alcoholism treatment [148] .
A final set of alternatives to drug and alcohol testing suggested by some authorities includes increased utilization of existing prevention and assistance technologies (e.g. health promotion, employee education, organizational development). Advocates of this position assert that effective means of preventing and reducing problems related to substance use already exist. It is believed that, by focusing on individual and organizational factors associated with substance abuse, organizations will develop more humane responses to substance use and abuse and that they will be more likely to create climates in which substance abuse cannot thrive. Performance evaluation in the form of supervisory observation and constructive intervention has long played a central role in workplace efforts directed at substance-impaired workers [41] , [55] . While the evidence supporting its value is contradictory [67] , proponents of performance-based alternatives to drug testing suggest that they direct attention towards work performance and away from worker lifestyle issues. Consequently, they are the more appropriate response to work- place performance problems, especially since the evidence linking drug- testing results to impaired work performance is weak.
This overview has illustrated that programmes for alcohol and drug testing in the workplace can be understood as a continuation of a long line of efforts by Governments, employers' organizations, workers' organizations and enterprises to prevent and reduce alcohol- and drug- related problems in the workplace. The issues dealt with in this paper underscore the complex and often competing ethical, humanitarian, economic, legal and scientific questions raised by the existence of testing programmes.
All workplace initiatives that respond to alcohol- and drug-related problems are based on at least four interrelated assumptions:
Substance use poses serious difficulties in the workplace;
The particular response that is implemented will resolve those difficulties;
The benefits gained by implementing an initiative outweigh the costs of the initiative;
The response is consistent with ethical and legal standards.
When drug- and alcohol-testing programmes are assessed against these assumptions, it becomes apparent that there is insufficient evidence either to justify their existence in all cases or to call for their elimination:
Empirical evidence linking substance use with work-related difficulties is equivocal and study results are inconclusive. Paradoxically, the evidence is stronger for the adverse effect of alcohol in the workplace. Yet most testing programmes test for drugs other than alcohol;
Studies demonstrating that testing programmes effectively resolve workplace substance abuse problems are, for the most part, lacking;
Given the weak evidence linking alcohol and drug problems to the workplace and the lack of proof for the effectiveness of testing programmes, it is difficult to assert that the benefits of the programmes either justify or do not support their existence;
Testing programmes obviously can threaten individual rights, and the testing laws of one, country may cause problems in another country. Courts have not established conclusively the legitimacy of all forms of alcohol and drug testing.
Further experience and study is needed to demonstrate both the need for, and the effectiveness of, drug- and alcohol-testing programmes. Such programmes should be examined in the context of a broad range of societal and organizational responses to substance abuse. Whether or not particular programmes are of value depends on a number of factors, including the nature and extent of workplace substance abuse, the type of testing programme that is designed, the programme's costs, and any unanticipated consequences accompanying the programme.
In conclusion, it is unlikely that the international community will arrive at one set of standards that apply equally well to all workplaces. Rather, the needs of each organization (and each sector) should be assessed individually. If a drug- or alcohol-testing programme is instituted, it should be appropriate to those needs, it should be part of a comprehensive effort to respond to substance abuse problems and it should contain safeguards to monitor its procedures and to evaluate its impact.
Ackerman, D. L. A history of drug testing. In Drug testing: issues and options. By R. H. Coombs and L. J. West, eds. New York, Oxford University Press, 1991. p. 3-21.
002Addict ion Research Foundation. ARF annual report 1988-1989. Toronto, 1989.
003Report of the interdivisional task group on employment related drug screening. Toronto, November 1990.
004Workplace drug and alcohol screening: where to draw the line. Toronto, 1991.
005Adrian, M., P. M. Jull and R. T. Williams. Statistics on alcohol and drug use in Canada and other countries. Toronto, Addiction Research Foundation, 1988.
006Ahlstrand, S. S. Alliance for a drug-free Canada: nationwide study. Toronto, Gallup Organization, July 1992.
007Akabas, S.,H, and M. Hanson. Organizational implications of drug abuse programming: making the organization work for you. In National Institute of Drug Abuse. Drug abuse curriculum for employee assistance professionals. Rockville, Maryland, 1989.P. III 1-111 57.
008Alcohol use and dependence among employed men and women in the United States in 1988. By T. C. Harford, and others. Alcoholism: clinical and experimental research (Baltimore, Maryland) 16:2:146-148,1991.
009Alleyne, B. C., P. Stuart and R. Copes. Alcohol and other drug use in occupational fatalities. Journal of occupational medicine (Baltimore) 33:4:496-500, 1991.
010Alvi, S. Corporate responses to substance abuse in the workplace. Ottawa, Conference Board of Canada, 1992. (Report 87-92)
011American College of Occupational Medicine. Committee report: Drug screening in the workplace; ethical guidelines .(Council on Social Issues). J ournal of occupational medicine. (Baltimore) 33:5:651- 652, 1991.
012American Management Association. 1992 AMA survey on workplace drug testing and drug abuse policies. New York, 1992
0131992 AMA survey: workplace drug testing and drug abuse policies, summary of key findings. New York, 1992.
014Anglin, M. D. and C. A. Westland. Drug monitoring in the workplace: results from the California Commercial Laboratory Drug Testing Project. In Drugs in the workplace: research and evaluation data. Rockville, Maryland, National Institute on Drug Abuse, 1989. p. 81-96. (National Institute on Drug Abuse Research Monograph Series, No. 91)
015Arif, A. and J. Westermeyer, eds. Manual of drug and alcohol abuse. New York, Plenum, 1988.
016Axel, H. Corporate experiences with drug- testing programmes. New York, Conference Board, 1990. (Report No. 941)
017Drug testing in private industry. In Drug testing: issues and options. R. H. Coombs and L. J. West, eds.New 'York, Oxford University Press, 1991. p. 140-154.
018Bensinger, P. B. Drugs in the workplace. Psychiatric letters (London), 4:8:39-44, 1986.
019Bijl, R. Views on alcohol and drug testing in the Netherlands. Zest, Netherlands, ALCON Foundation, 1992.
020Bilik, S. EAPS in Germany: options for transatlantic exchange. Employee assistance quarterly (Binghampton, New York) 3:2:83-97, 1987.
021Bissell, L. andP. Haberman. Alcoholism in the professions. New York, Oxford University Press, 1984.
022Blair, T. S. The relation of drug addiction to industry. Journal of industrial hygiene (New York) 1:6:284-296, 1919.
023Blank, D. L. andJ. W. Fenton. Farly employment testing for marijuana: demographic and employee retention patterns. InDrugs in the workplace: research and evaluation data. Rockville, Maryland, National Institute on Drug Abuse, 1989. p. 139-150. (National Institute on Drug Abuse Research Monograph Series, No. 91)
024Borkenstein, R. F. Historical perspective: North American traditional and experimental response. Journal of studies on alcohol (Piscataway, New Jersey) Supplement 10:3-12, 1985.
025Brandes, S. D. American welfare capitalism, 1880-1940. Chicago, University of Chicago Press, 1976.
026Bross, M. H., S. K. Pace and 1. H. Cronin. Chemical dependence: analysis of work absenteeism and associated medical illnesses. Journal of occupational medicine (Baltimore) 34:1:16-19, 1992.
027Canada. National alcohol and other drugs survey (1989): highlights report. Ottawa, Health and Welfare, 1990.
028Ontario Law Reform Commission. Report on drug and alcohol testing in the workplace. Ottawa, 1992.
029Ontario Ministry of Labour. Report of the advisory committee on EAPS. Ottawa, 1990.
030Privacy Commissioner of Canada. Minister of Supply and Services. Drug testing and privacy. Ottawa, 1990.
031Conrad, P. andD. C. Walsh. The new corporate health ethic: lifestyle and the social control of work. International journal of health services (New York) 22:1:89-111, 1992.
032Coombs, R. H. and C. J. Coombs. The impact of drug testing on the morale and well- being of mandatory participants. International journal of the addictions (New York) 26:9:981-992, 1991.
033Corneil, D. W. Alcohol in employment settings: the results of the WHO/ILO international review. Employee assistance quarterly (Binghampton, New York) 3:2:5-58, 1987.
034History, philosophy and objectives of an employee recovery program. In Employee recovery program: consultant training course participants' reading. Ottawa, Canadian Labour Congress. p. 20-51.
035De Bernardo, M. A. Drug abuse in the workplace: an employer's guide for prevention. 2. ed. Washington, D.C., Chamber of Commerce, 1988.
036Drug testing: medical, legal, and ethical issues. By N. S. Miller and others. Journal of substance abuse treatment (New York) 7:4:239-244, 1990.
037Du Pont, R. L. Never trust anyone under 40. Policy review (Washington, D.C.) 48:52-57, Spring 1989.
038Dusek, D. E. andD. A. Girdano. Drugs: a factual account. 5. ed. New York, McGraw-Hill, 1993.
039Economic costs to society of alcohol and drug abuse and mental illness: 1980. ByH. J. Harwood and others. Research Triangle Park, North Carolina, Research Triangle Institute, 1984.
040Effects of alcohol intoxication on risk taking, strategy, and error rate in visuomotor performance. By S. Steufert and others. Journal of applied psychology (Washington, D.C.) 77:4:515-524, 1992.
041Employee assistance and drug testing: fairness and injustice in the workplace. By W. J. Sonnenstuhl and others Nova law review (Fort Lauderdale, Florida) 11:3:709-731, 1987.
042Extent of drug abuse: an international review with implications for health planners. By P. H. Hughes and others. World health statistics quarterly (Geneva) 36:3/4:394-497,1983.
043Engelsman, E. L. Alcohol policies in the Netherlands: a three-pronged attack. World health forum (Geneva) 11:3:257-263, 1990.
044Feinauer, D. M. The relationship between workplace accident rates and drug and alcohol abuse: the unproven hypothesis. Labor studies journal (New Brunswick, New Jersey) 15:4:3-15, Winter 1990.
045Fine, M., S. H. Akabas and S. Bellinger. Cultures of drinking: a workplace perspective. Social work (Silver Spring, Maryland) 27:5:436-440, 1982.
046Fischman, M. W., T. H. Kelly andR. W. Foltin. Residential laboratory research: a multidimensional evaluation of the effects of drugs on behavior. In Drugs in the workplace: research and evaluation data. v. 11. Rockville, Maryland, National Institute on Drug Abuse, 1991. (National Institute on Drug Abuse Research Monograph Series, No. 100)
047Francek, J. Boarding the testing train. Employee assistance quarterly (Binghampton, New York) 2:3:26, 60, 1989
048Frieden, J. Corporate America's response to substance abuse. Business and health (Washington, D.C.) 32-42, July 1990.
049Ghodse, H. Drugs and addictive behaviour: a guide to treatment. London, Blackwell Scientific Publications, 1989.
050Googings, B. andJ. Godfrey. The evolution of occupational social work. Social 14,Vrk (Silver Spring, Maryland) 30:6:396-402, 1985.
051Gust, S. W. andJ. M. Walsh. Research on the prevalence, impact, and treatment of drug abuse in the workplace. In Drugs in the workplace: research and evaluation data. Rockville, Maryland, National Institute on Drug Abuse, 1989. p. 3-13. (National Institute on Drug Abuse Research Monograph Series, No. 91)
052Haber, M. H. Pisse prophecy: a brief history of urinalysis. Clinics in laboratory medicine (Philadelphia) 8:3:415-430, 1988.
053Hammer, T. Unemployment and use of drug and alcohol among young people: a longitudinal study in the general population. British journal of addiction (Abingdon, Oxfordshire, United Kingdom of Great Britain and Northern Ireland) 87:11:1571- 1581, 1992.
054Hanson, F. A. Some social implications of drug testing. Kansas law review (Lawrence, Kansas) 36:4:899-917, 1988.
055Hanson, M. Serving the substance abuser in the workplace. In Occupational social work: strategic response to workplace and workforce. ByP. A. Kurzman andS. H. Akabas, eds.Silver Spring, Maryland, NASW Press, 1993. p. 218-238.
056Haughic, 0. E. Evaluation. International information exchange on drugs in the workplace. Bergen/Os, Norway, International Labour Office, 1991.
057Hawks, R. L. andC. N. Chiang, eds.Urine testing for drugs of abuse. Rockville, Maryland, National Institute on Drug Abuse, 1986. (National Institute on Drug Abuse Research Monograph Series, No. 73)
058Hayghe, H. V. Anti-drug programmes in the workplace: are they here to stay? Monthly labor review (Washington, D.C.) 114:4:26-29, 1991.
059Hecker, S. andM. S. Kaplan. Workplace drug testing as social control. International journal of health services (New York) 19:4:693- 707, 1989.T
060Heffring Research Group. Substance use in transportation: airports aviation, surface (bus/trucking) and marine. Integrated report. Ottawa, Transport Canada, January 1990.
061Heller, D. andA. E. Robinson. Substance abuse in the workforce. Ottawa, Canadian Centre on Substance Abuse, 1992.
062Hindemarch, I. Psychomotor function and psychoactive drugs. British journal of clinical pharmacology (Oxford) 7:Supplement:77S-82S, 1980.
063Hingson, R. W., R. I. Lederman and D. C. Walsh. Employee drinking patterns and accidental injury: a study of four New England states. Journal of studies on alcohol (Piscataway, New Jersey) 46:4:298-303, 1985.
064Horgan, J. Test negative: a look at the "evidence" justifying illicit-drug tests. Scientific American (New York) 262:18, 22 March 1990.
065Houts, L. M. Survey of the current status of cost-saving evaluations in employee assistance programmes. Employee assistance quarterly (Binghampton, New York) 7:1:57-72, 1991.
066Imwinkelried, E. J. The debate over drug testing in the workplace: a novel opportunity for public scientific education. Behavioral sciences and the law (New York) 9:3-305-322, 1991.
067Institute of Medicine. Broadening the base of treatment for alcohol problems. Washington, D.C., National Academy Press, 1990.
068Treating drug problems. v. 1. Washington, D.C., National Academy Press, 1990.
069International Labour Office. Establishment of resource centres for rehabilitation, workplace initiatives and community action on drugs and alcohol. Report of the subregional seminar, Harare, Zimbabwe. Geneva, March 1991.
070International information exchange on drugs in the workplace. Proceedings. Geneva, August 1991.
071Washington tripartite symposium on drug and alcohol abuse prevention and assistance programmes in the workplace. Proceedings. Geneva, 1991.
072Working paper on drug and alcohol screening issues in the maritime industry. Geneva, 1992. p. 1.
073Workplace initiatives to prevent and reduce drug and alcohol problems. Geneva,1988.
074International Transport Workers' Federation. ITF briefing paper on drug and alcohol abuse. London.
075Jacobs, J. B. andL. Zimmer. Drug treatment and workplace drug testing: politics, symbolism and organizational dilemmas. Behavioral sciences and the law (New York) 9:3:345-360, 1991.
076Kandel, D. B. Drug and drinking behavior among youth. Annual review of sociology (Palo Alto, California) 6:235-285, 1980.
077Karlsson, C. Workplace drug and alcohol programmes in Sweden. Washington tripartite symposium on drug and alcohol abuse prevention and assistance programmes in the workplace. Geneva, May 1991.
078Kelly, T. H., R. W. Foltin andM. W. Fischman. Effects of alcohol on human behavior: implications for the workplace. InDrugs in the workplace: research and evaluation data. v. 11. Rockville, Maryland, National Institute on Drug Abuse, 1991. p. 129-146. (National lnstitute on Drug Abuse Research Monograph Series, No. 100)
079Kleiman, M.A.R. Beyond excess: drug policy for results. New York, Basic Books, 1992.
080Kopstein, A. andJ. Gfroerer. Drug use patterns and demographics of employed drug users: data from the 1988 National Household Survey on Drug Abuse. In Drugs in the workplace: research and evaluation data. v. 11. Rockville, Maryland, National Institute on Drug Abuse, 1991. p. 11-24. (National Institute on Drug Abuse Research Monograph Series, No. 100)
081Latessa, E. J. L. F. Travis and F. T. Cullen. Public support for mandatory drug-alcohol testing in the workplace. Crime and delinquency (Sacramento, California) 34:4:379-392, 1988.
082Lemon, S. J., D.G.Sienko,and P.C.Alguire. Physicians' attitudes toward mandatory workplace urine drug testing. Archives of internal medicine (Chicago) 152:11:2238-2242, 1992.
083LeRoy, M. H. Discriminating characteristics of union members' attitudes toward drug testing in the workplace. journal of labor research (Fairfax, Virginia) 12:4:453-466, 1991.
084Lewis, J. R. andS. P. Cooper. Alcohol, other drugs, and fatal work-related injuries. Journal of occupational medicine (Baltimore) 31:1:23-28, 1989.
085Lewy, R. Pre-employment qualitative urine toxicology screening. journal of occupational medicine (Baltimore) 25:5:579-580, 1983.
086London Underground. London Underground alcohol and drug policies: briefing note to all staff. London, 1992.
087MacDonald, D. I. The medical review officer. journal of psychoactive drugs, (San Francisco) 22:4:429-434, 1990.
088MacDonald, S. andS. Dooley. The nature and extent of EAPs and drug screening programmes in Canadian transportation companies. Employee assistance quarterly (Binghampton, New York) 6:4:23-40, 1991.
089Manley, S. A- and G. S. Gibson. Drug-induced impairment: implications for employers Psychology of addictive. behaviors (Seattle,, Washington) 4:97-99, 1990.
090Mass transit warms up to drug testing. New York newsday, 11, 5 October 1991.
091McClellan,, K. andR. E. Miller. EAPs. in transition: purpose and scope of services. Employee assistance quarterly (Binghampton, New York) 3:3/4:25-42, 1988.
092McDaniel, M. A. Does pre-employment drug use predict on-the-job suitability? In Drugs in the workplace: research and evaluation data. Rockville, Maryland, National Institute on Drug Abuse, 1989. p. 151-167. (National Institute on Drug Abuse Research Monograph Series, No. 91)
093Metzger, L. From denial to recovery. San Francisco, Jossey-Bass, 1988.
094Miller, W. R. Haunted by Zeitgeist: reflections on contrasting treatment goals and concepts of alcoholism in Europe and the United States. In Alcohol and culture: comparative perspectives from Europe and America. T. F. Babor, ed. New York, New York Academy of Sciences, 1986. p. 110- 129.
095Montagne, M., C. B. Pugh and 1. L. Fink. Testing for drug use. Part 1: Analytic methods. American journal of hospital pharmacy (Bethesda, Maryland) 45:6:1297- 1305, 1988.
096Testing for drug use. Part 2: Legal, social, and ethical concerns. American journal of hospital pharmacy (Bethesda, Maryland) 45:7:1509-1522, 1988.
097Morawski J. and G. Swietkiewicz. Alcohol in employment settings in Poland. Employee assistance quarterly (Binghampton, New York) 3:2:105-119, 1987.
098Morgan, J. P. The "scientific" justification for urine drug testing. Kansas law review, (Lawrence, Kansas) 36:4:683 - 697, 1988.
099Morgan, J. P. andP. S. Puder. Urinary testing for drugs of abuse in the military. Behavioral sciences and the law (New York) 7:3:379-386, 1989.
100Mulloy, P. J. Winning the war on drugs in the military. In Drug testing: issues and options. R. H. Coombs and L. J. West, eds.New York, Oxford University Press, 1991. p. 92-112.
101National Institute on Drug Abuse. Comprehensive procedures for drug testing in the workplace. Rockville, Maryland, 1991. p. 133.
102Model plan for a comprehensive drug-free workplace programme. Rockville, Maryland, 1989.
103National household survey on drug abuse: main findings, 1990. Rockville, Maryland, 1991.
104Nicholson, A. N. and J. Ward, eds.Psychotropic drugs and performance. British journal of psychopharmacology (Basingstoke, Hampshire,) lg:Suppl.1:1-139, 1984.
105Nordheimer, J. Drinking an job: getting help is a problem. New York times, C1, C4, 20 November 1991.
106Normand, J., S. V. Salyards and J. J. Mahoney. An evaluation of pre-employment drug testing. Journal of applied psychology (Washington, D.C.) 75:6:629-639, 1990.
107O'Keefe, A. M. The case against drug testing. Psychology today (New York) 34-35, 38, June 1987.
108O'Malley, P. andS. Mugford. Moral technology: the political agenda of random drug testing. Social justice (San Francisco) 18:4:122-146, 1991.
109Osterberg, E. Alcohol-related problems in cross-national perspective: results of the ISACE study. In Alcohol and culture: comparative perspectives from Europe and America. T. F. Babor, ed. New York, New York Academy of Sciences, 1986. p. 10-20.
110Parish, D.C. Relation of pre- employment drug testing result to empoyement status: a one- year follow- up. Journal of general internal medicine (Philadelphia) 4:1:44- 47
111Popple, P.R. Social work practice in business and industry, 1875-1930. Social service review (Chicago) 55:257-269, 1981.
112Potter, B. A. and J. S. Orfali. Drug testing at work: a guide for employers. Berkeley, California, Ronin, 1990.
113Prescott, J. BP tests officers for drink and drugs. Lloyd's 4v international (Colchester, Essex) 15 May 1991.
114Prevalence of substance use among U.S. physicians. By P. H. Hughes and others. .Journal of the American Medical Association (Chicago) 267:17:2333-2339, 1992.
115Pyorala, E. Trends in alcohol consumption in Spain, Portugal, France and Italy from the 1950s until the 1980s. British journal of addiction (Edinburgh) 85:4:469-477, 1990.
116Ray, 0. Drugs, society, and human behavior, 2. ed. St. Louis, Missouri, C. V. Mosby, 1978. p. 129.
117Roberts, J. I. and E. A. Russo. The alcoholic in industry and his rehabilitation. Industrial medicine and surgery, 24:270- 276, 1955.
118Robinson, A. E. Drug use testing: Canadian scene. Journal of forensic sciences (Philadelphia) 34:6:1422-1432, 1989.
119The role of needs assessment in the design of employee assistance programmes: a case study. By P. S. Berman and others. Employee assistance quarterly (Binghampton, New York) 6:3.21- 35, 1991.
120Rothman, M. Random drug testing in the workplace: implications for human resource management. Business horizons (Bloomington, Indiana) 23-27, March-April 1988.
121Ryan, J., C. Zwerling and M. Jones. The effectiveness of pre-employment drug screening in the prediction of employment outcome. J ournal of occupational medicine (Baltimore.) 34:11:1057-1063, 1992.
122Scanlon, W. F. Alcoholism and drug abuse in the workplace. 2. ed. New York, Praeger, 1991.
123Schottenfeld, R. S. Drug and alcohol testing in the workplace - objectives, pitfalls, and guidelines. American journal of drug and alcohol abuse (New York) 15:4:413-427, 1989.
124Shahandeh, B. Rehabilitation approaches to drug and alcohol dependence. Geneva, International Labour Office, 1985.
125Shain, M. A. A paper on the relationship between alcohol and drugs, and accidents in the workplace. In Towards safe production. Volume 2: Report of the Joint Federal- Provincial Inquiry Commission into Safety in Mines and Mining Plants in Ontario. Ottawa, Joint Federal-Provincial Inquiry Commission into Safety in Mines and Mining Plants in Ontario, 1980.
126Sheridan, J. and H. Winkler. An evaluation of drug testing in the workplace. In Drugs in the workplace- research and evaluation data. Rockville, Maryland, National Institute on Drug Abuse, 1989. p. 195-216. (National Institute on Drug Abuse Research Monograph Series, No. 91)
127Social facilitation of marijuana intoxication: impact of social set and pharmacological activity. By A. S. Carlin and others.. Journal of abnormal psychology (Washington, D.C.) 80:2:132-140, 1972.
128Smart, R. G. Is the post-war drinking binge ending? British journal of addiction (Edinburgh) 84:5-743-748, 1989.
129World trends in alcohol consumption. World health forum (Geneva) 12:1:99-103, 1991.
130Smart, R. G., G. F. Murray and A. Arif. Drug abuse and prevention programmes in 29 countries. International journal of the addictions (New York) 23:1:1-17, 1988.
131Smith, J. P. Project overview and findings. Washington tripartite symposium on drug and alcohol abuse prevention and assistance programmes in the workplace. Geneva, International Labour Office, 1991.
132SmithKline Beecham. Press release: SmithKline Beecham index shows drug use decline for the fifth straight year. Philadelphia, 10 February 1992.
133Spicer, J., P. Owen and D. Levine. Evaluating employee assistance programmes. Center City, Minnesota, Hazelden, 1983.
134A study of absenteeism, accidents and sickness payments in problem drinkers in one industry. By Observer and M. A. Maxwell. Quarterly journal of studies on alcohol 20:302-312, 1959.
135Swotinsky, R. B., ed. The medical review officer's guide to drug testing. New York, Van Nostrand Reinhold, 1992.
136Traumatic workplace death in Allegheny County, Pennsylvania, 1983-1984. By D. Parkinson and others. Journal of occupational medicine (Baltimore) 28:1:100-102, 1986.
137Trice, H. M. Alcoholic employees: A comparison of psychotic, neurotics, and "normal" personnel. J ournal of occupational medicine (Baltimore) 7:1:94-99, 1965.
138Trice, H. M. and J. M. Beyer Employee assistance programmes: Blending performance-oriented and humanitarian ideologies to assist emotionally disturbed employees. Research in community and mental health (Greenwich, Connecticut) 4:245-298, 1984.
139Trumble, J. Designing and sponsoring effective programmes. Washington tripartite symposium on drug and alcohol abuse prevention and assistance programmes in the workplace. Geneva, International Labour Office, May 1991.
140United States of America. Department of Labor. Bureau of Labor Statistics. Survey of employer anti-drug programmes. Washington, D. C., 1989. (Report 760)
141Department of Health and Human Services. Mandatory guidelines for federal workplace drug-testing programmes; final guidelines. Federal register (Washington, D.C.) 53:11970-11989, 11 April 1988.
142Federal Railroad Administration. Summaries of post-accident testing events, February 10, 1986-December 31, 1987. Washington, D.C., 1988.
143General Accounting Office. Employee drug testing: information on private sector programmes. Washington, D.C., 1988.
144Vogel-Sprott, M. Alcohol tolerance and social drinking: learning, the consequences. New York, Guilford, 1992.
145Walsh, B. and M. Grant. International trends in alcohol production and consumption: Implications for public health. World health statistics quarterly (Geneva) 38:130-141, 1985.
146Walsh, J. M. and J. G . Trumble. The politics of drug testing. In Drug testing: issues and options. R. H. Coombs and L. J. West, eds. New York, Oxford University Press, 1991. p. 22-49
147The politics of drug testing. In Drug testing: issues and options. R. H. Coombs and L. J. West, eds. New York, Oxford University Press, 1991. p. 22-49.
148Weintraub, R. M. Random alcohol tests on transport workers proposed. Washington post, A1 and A41, 11 December 1992.
149Weiss, R. M. Writing under the influence: science versus fiction in the analysis of corporate alcoholism programmes. Personnel psychology (Durham, North Carolina) 40:2:341-356, 1987.
150Wiencek, R. G. Drugs in the workplace: retrospective and prospective. In Drug abuse in the modern world. G. G. Nahas and H. C. Frick, eds. New York, Pergamon, 1981. p. 121-126.
151Willette, R. E. Techniques of re-liable drug testing. In Drug testing: issues and options. R. H. Coombs and L. J. West, eds. New York, Oxford University Press, 1991.
152Willette, R. and L. Kadehjian. Drug testing in the workplace. In The medical review officer's guide to drug testing. R. B. Swotinsky, ed. New York, Van Nostrand Reinhold, 1992. p. 23-52.
153William M. Mercer. A survey of CEO views on substance abuse - and its impact in the workplace. Toronto, 1990.
154Wisotsky, S. The ideology of drug testing. Nova law review (Fort Lauderdale, Florida) 11:3:763-778,1987.
155World Health Organization. Global strategy for health for all by the year 2000. World health statistics quarterly (Geneva) 42:4, 1989.
156WHO expert committee on drug dependence 27th report. Geneva, 1991. (WHO Technical Report Series, No. 808)
157Yamatani, H., M. P. Ballas and G. E. de Silva. The troubled employee: an assessment of problem areas and related financial liabilities. Journal of applied social sciences (Cleveland, Ohio) 12:2:170-185,1988.
158Younger, B. The Drug-Free Workplace Act of 1988: government intent and employer response. Employee assistance quarterly (Binghampton, New York) 7:2:15-40, 1991.
159Ziegler, H. Workplace drug and alcohol programmes in Germany. Washington tripartite symposium an drug and alcohol abuse prevention and assistance programmes in the workplace. Geneva, International Labour Office, May 1991.
160Zinberg, N. Drug, set, and setting: the basis for controlled intoxicant use. New Haven, Connecticut, Yale University Press, 1984.
161Zwerling, C. and J. Ryan. Pre-employment drug screening: the epidemiologic issues. Journal of occupational medicine (Baltimore) 34:6-595-599, 1992.
162Zwerling, C., J. Ryan and E. J. Orav. Costs and benefits of pre-employment drug screening. Journal of the American Medical Association (Chicago) 267:1:91-93, 1992.
163The efficacy of pre-employment drug screening for marijuana and cocaine in predicting employment outcome. Journal of the American Medical Association (Chicago) 264:20:2639-2643, 1990.