U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Research Council (US) and Institute of Medicine (US) Committee on Drug Use in the Workplace; Normand J, Lempert RO, O'Brien CP, editors. Under the Influence? Drugs and the American Work Force. Washington (DC): National Academies Press (US); 1994.

Cover of Under the Influence?

Under the Influence? Drugs and the American Work Force.

Show details

BThe Legal Environment of Drug Testing

This appendix is based on a paper commissioned by the Committee on Drug Use in the Workplace from David Wasserman of the Institute for Philosophy and Public Policy, University of Maryland, and James Jacobs of the School of Law, New York University.

The law today mandates and defines permissible drug-testing programs. All drug testing by public bodies and some drug testing by private entities is done pursuant to legal authority or command. This authority may not only mandate testing, but may also constrain the conditions under which drug testing proceeds and the consequences that can attach to positive results. There are also limits on what the law can authorize or mandate with respect to drug testing. These limits are established by courts in suits that challenge the legal authority or constitutionality of drug-testing programs. Finally, the law, either through statutes, regulations, or court decisions, may control what private entities can do on their own initiative by way of drug testing. This appendix addresses these matters as we describe the legal environment in which drug testing occurs today.

Sources Of Authority

Over the past decade, the President and the Congress have not only authorized drug testing by public and private employers, but have also required or encouraged it in some workplaces. Their actions provide legal authority for drug testing that overrides all contrary authority except the U.S Constitution.

Executive Order 12564 (1986) introduced the concept of a "drug free workplace" and set in place many of the features of workplace drug testing that have now become standard for public and regulated private employers. It required all federal agencies to adopt employee drug-testing programs, which included mandatory testing for employees in "sensitive" positions involving law enforcement, public health and safety, and national security, with the extent of testing and the criteria for testing being left to the discretion of each agency head. It also permitted the testing of employees in other positions: (1) when there was reasonable suspicion that the employee was using illegal drugs; (2) in an examination regarding an accident or unsafe practice; and (3) as part of, or as a follow-up to, counseling or rehabilitation for illegal drug use through an employee assistance program.

Under the executive order, visual monitoring of the giving of a urine specimen was not permitted unless there was reason to believe that the employee would alter or substitute the specimen. The order also mandated procedures to protect the confidentiality of test results. In addition, it required each agency to establish a comprehensive employee assistance program involving education, counseling, and rehabilitation.

In Section 503 of the Supplemental Appropriations Act of 1987, Congress required federal contractors as well as agencies to guarantee a drug-free workplace; although drug testing was not mandatory, it was recommended. Section 503 also required all federal agencies to draft detailed testing plans to be reviewed by the U.S. Department of Health and Human Services (DHHS), thereby promoting the uniformity of federal testing procedures. The DHHS "Mandatory Guidelines for Federal Workplace Drug-Testing Programs" contain stringent procedural safeguards that have now become standard: requirements for the collection of urine specimens, the avoidance of sample mix-ups, the certification of and quality control measures for testing laboratories, and the confirmation of positive test results.

In 1988 Congress passed the Drug Free Workplace Act (DFWA), which required all federal contractors with contracts of $25,000 or more to certify to the appropriate contracting agency that they are maintaining a drug-free workplace. If they failed to do so, the contract could be canceled. Contractors were not required to implement a drug-testing program, but could do so in order to demonstrate compliance.

In 1991 standards for industries regulated by the U.S. Department of Transportation (DOT) were unified and consolidated in the Omnibus Transportation Employee Testing Act (OTETA) of that year. OTETA mandates drug and alcohol tests for all individuals who hold or apply for "safety-sensitive" jobs in the transportation industry. Public-and private-sector employers must set up drug programs in conformity with the DOT regulations.

Legal Constraints

Any program of workplace drug testing must operate within legal constraints—not all testing programs are legally permissible. It is important, however, to remember that legal constraints are not necessarily binding for all time, even when they are constitutionally based. Changing technology, changing perceptions of the drug problem, and changes in the perceived relationship between drug use and workplace performance may all influence legal decision makers, be they judges or legislators, in determining what should be permitted. This committee's report, in contrast, is time bound. We can only recount what as of this writing seem to be the constraints that the law places on workplace drug-testing programs.

In surveying the law as it currently exists, three important distinctions emerge. First is the public-private distinction. Drug testing done by a government agency or by a private business in compliance with a government mandate will face legal constraints that will not bind a private firm that chooses to test for drugs at its own initiative. Second is the distinction between preemployment and postemployment drug testing. Generally speaking, both governmental agencies and private businesses are freer to test potential employees for drug use than they are to test those already employed. Third is the distinction between state and federal law. Drug-testing programs that are permissible under federal law may nonetheless be impermissible under the law of a particular state. However, state law will not prevent the federal government from testing its own employees in conformance with federal law, nor can it preclude employers within the state from complying with federally mandated drug-testing programs.

The Fourth Amendment

The public-private distinction is important because drug testing normally involves the taking and examination of material from a person's body that is not normally exposed to public scrutiny and so is considered a search under the Fourth Amendment to the U.S. Constitution. This amendment prohibits unreasonable searches and seizures by governmental agents and, subject to certain exceptions, limits governmental searches to those made pursuant to a warrant upon a showing of "probable cause." The Fourth Amendment does not, however, restrict the actions of private individuals or organizations so long as they are acting on their own initiative and not in response to some governmental mandate.

Universal or routine workplace drug testing obviously fails to satisfy the warrant and probable-cause requirements of the Fourth Amendment. If the amendment were applied to drug testing as it is applied in ordinary criminal investigations, routine drug testing when conducted by public officials or by private parties acting in response to a legal mandate, such as the Omnibus Transportation Employee Testing Act of 1991, would not be permitted. The "probable-cause" language of the Fourth Amendment has, however, been held inapplicable to administrative searches that serve primarily a regulatory rather than a law enforcement purpose, so long as the search does not contravene the Fourth Amendment's more general preclusion of "unreasonable searches." The reasonableness or unreasonableness of an administrative search depends on the balance between the governmental interests served by the search and its intrusiveness.

A special rule for administrative searches was first carved out by the Supreme Court in 1967 in Camera v. Municipal Court (387 U.S. 523, 1967), which held that city housing inspectors could secure warrants to conduct areawide searches for housing code violations without having probable-cause to believe that any building or buildings were in violation of the code. In abandoning the probable-cause requirement for warrants, the Court pointed to several factors clearly present in the Camera case: a neutral plan that limited discretion and possible bias in conducting searches, very limited intrusion in conducting the searches, and a showing of a compelling government need for the plan in the absence of less intrusive alternatives. However, since Camera, "[t]he 'tests' have been reformulated, qualifications have been stripped away, and the elements have been applied ever less strictly" (Schulhoffer, "On the Fourth Amendment Rights of the Law-Abiding Public," 189 Supreme Court Review 87, 94). In particular, in certain circumstances not only is there no need to justify searches with a showing of probable-cause, but also the warrant requirement itself has been abandoned. Among the contexts in which the Supreme Court since Camera has upheld searches without probable-cause and/or without a warrant are: an inspection program of federally licensed firearms dealers (United States v. Biswell, 406 U.S. 311, 1972); an inspection by the Occupational Safety and Health Administration to determine compliance with workplace safety regulations (Marshall v. Barlow's Inc., 436 U.S. 307, 1978); safety inspections of mines and stone quarries (Donovan v. Dewey, 452 U.S. 594, 1981); a search of the handbag of a student suspected of smoking (New Jersey v. T.L.O., 459 U.S. 325, 1985); an inspection of an automobile junkyard done with crime control goals in mind (New York v. Burger, 482 U.S. 691, 1987); and stops at roadblocks set up to control drunk driving (Michigan Dept. of State Police v. Sitz, 496 U.S. 444, 1990). Finally, in a pair of cases decided in 1989, Skinner v. Railway Labor Executive's Association (489 U.S. 602, 1989) and National Treasury Employees Union v. Von Raab (489 U.S. 656, 1989), the Court added drug testing to this list.

Skinner and Von Raab

In Skinner the Supreme Court was faced with a challenge to the constitutionality of the Federal Railroad Administration's (FRA) mandatory and permissive alcohol-and drug-testing procedures. Testing for drugs and alcohol was mandatory after a major train accident involving (1) a fatality, (2) the release of hazardous materials accompanied by evacuation or reportable injury, or (3) damage to railroad property of $500,000 or more. Furthermore, blood and urine samples had to be collected after an "impact accident," one that resulted in a reportable injury or in damage to railroad property of $50,000 or more. Finally, a railroad had to test after "any train incident that involves fatality to any on-duty railroad employee." If an employee declined to give a blood sample, the railroad could presume impairment, absent persuasive contrary evidence.

The Court, in a 7-2 decision, upheld the regulations. The majority began by holding that a search conducted by a private railroad under the auspices of a federal regulatory agency constituted state action under the Fourth Amendment. The Court next found that both the collection of blood, urine, and breath and the subsequent chemical analysis of these substances constitute searches for Fourth Amendment purposes. But, having found that the testing program was a search within the meaning of the Fourth Amendment, the Court analyzed the FRA's drug-testing requirements under the reasonableness test used for administrative searches rather than by reference to the law enforcement standard of probable-cause or a lesser standard of reasonable suspicion.

The administrative search standard required the reviewing court to balance the governmental interest served by the testing program against its intrusiveness. Just as the adoption of this balancing test was significant, so was the way the Court weighed the factors in the case before it. In assessing the infringement of privacy involved in blood, urine, and breath testing, the Court pointed out that railroad employees already consent to significant restrictions on their freedom of movement and that "any additional interference with a railroad employee's freedom of movement that occurs in the time it takes to procure a blood or breath sample for testing cannot, by itself, be said to significantly infringe on privacy interests." While the urine test raised greater privacy concerns, the fact that urine collection did not involve visual monitoring and that it took place in a medical setting led the Court to conclude that the infringement of privacy was not significantly greater. Finally, the Court argued that, since railroads have long been a heavily regulated industry, their employees must expect greater intrusion. "Though some of the privacy interests implicated by the toxicological testing at issue reasonably might be viewed as significant in other contexts, logic and history show that a diminished expectation of privacy attaches to information relating to the physical condition of covered employees and to this reasonable means of procuring such information" (489 U.S. 602, 628).

Perhaps the most significant indicator in Skinner of how the Court might rule in future cases was its assessment of the government's interests served by testing. The Court deferred substantially to the FRA's claim of alcohol or drug problems on railroads, although there was only weak factual support for that claim in the FRA's data on accidents over a 20-year period. Almost all drug-related railroad accidents were alcohol-related, but the Court, like the agency, spoke of "alcohol and drugs" in a single breath: "… the government's interest in testing without a showing of individualized suspicion is compelling."

Employees subject to the tests discharge duties fraught with such risks of injury to others that even a momentary lapse of attention can have disastrous consequences. Much like persons who have routine access to dangerous nuclear power facilities, employees who are subject to testing under the FRA regulations can cause great human loss before any signs of impairment become noticeable to supervisors or others (489 U.S. 602, 628).

The dissenters argued that the approach that led the Court to accept postaccident drug testing in this case would likely lead to a broad acceptance of drug testing in many contexts. Their point seems to be supported by the Court's holding in Von Raab, the companion case to Skinner.

In Von Raab, a union of federal employees and a union official challenged the U.S. Customs Service's drug-testing program, which made drug tests a condition of promotion or transfer to positions that met one or more of three criteria: (1) direct involvement in drug interdiction or the enforcement of related laws, (2) the carrying of firearms, or (3) the handling of classified materials. Employees who tested positive for drugs and offered no satisfactory explanation were subject to dismissal. However, test results could not be turned over to any other agency, including criminal prosecutors, without the employee's consent.

The Customs Service characterized drug interdiction as its primary enforcement mission and argued that "there is no room in the Customs Service for those who break the laws prohibiting the possession and use of drugs." While no serious drug problem had been uncovered within the Customs Service, the implications of such a problem, if it did arise, were offered as a justification for a wide-ranging drug-testing program.

The Supreme Court, this time by a 5-4 vote, once again held that neither probable-cause nor reasonable suspicion were needed to test government employees in "sensitive" positions. Justice Kennedy wrote: "Our precedents have settled that, in certain limited circumstances, the Government's need to discover … latent or hidden conditions, or to prevent their development, is sufficiently compelling to justify the intrusion on privacy entailed by conducting such searches without any measure of individualized suspicion" (489 U.S. 656, 668).

The Court's opinion emphasized the job demands in the Customs Service that would be specially threatened by employees who used drugs. Thus, the Court noted that "the physical safety of these [Customs] employees may be threatened, and many may be tempted not only by bribes from the traffickers with whom they deal, but also by their own access to vast sources of valuable contraband seized and controlled by the Service" (id. at 669).

The reasonableness test adopted in Skinner and Von Raab requires courts passing judgment on drug-testing programs to balance the intrusion on tested individuals against the government interests served by the test. While this standard is easy to articulate, it has no clear meaning. In Skinner the standard justified the testing of employees in highly safety-sensitive positions, predicated on suspicious circumstances. The Court's deference to the FRA's safety concerns suggest that drug testing throughout the transportation industry and in other areas in which safety concerns are easily and plausibly invoked, such as fire departments and public utilities, will be upheld against Fourth Amendment challenges, at least so long as the testing program is required of employees in safety-related positions and no more intrusive than the urine testing in Skinner. Von Raab extends Skinner's permission for warrantless searches to situations in which there is no reason to believe that drug use has been a problem and to employees whose performance will not necessarily be adversely affected by drug use.

Justice Scalia's dissent in Von Raab suggests the potential scope of this case:

[In] extending approval of drug testing to that category consisting of employees who carry firearms, the Court exposes vast numbers of public employees to this needless indignity. Logically, of course, if those who carry guns can be treated in this fashion, so can all others whose work, if performed under the influence of drugs, may endanger others—automobile drivers, operators of other potentially dangerous equipment, construction workers, school crossing guards. A similarly broad scope attaches to the Court's approval of drug testing for those with access to "sensitive information."

Since this category is not limited to Service employees with drug interdiction duties, nor to "sensitive information" specifically relating to drug traffic, today's holding apparently approves drug testing for all federal employees with security clearances—or, indeed, for all federal employees with valuable confidential information to impart. … Moreover, there is no reason why this super-protection against harms arising from drug use must be limited to public employees; a law requiring similar testing of private citizens who use dangerous instruments such as guns or cars, or who have access to classified information would also be constitutional" (489 U.S. 665, 686, 1989).

It seems likely that law enforcement agencies will see the Court as having approved drug testing for all employees who carry firearms. Many other agencies will take Von Raab as approving drug testing for all employees who have any access to "sensitive information." In addition, law enforcement agencies may attempt to justify drug-testing programs solely on the basis of a claimed need to assure the public that there are no drug users serving in their ranks.

Two other aspects of Skinner and Von Raab are worth noting. First, in neither case was the urine sample subject to visual observation. The need for reliable drug testing might have to be greater than it was in one or both of these cases to justify a drug-testing program that required an employee to urinate in the presence of an observer. Second, in each case only certain employees defined by the nature of their job responsibilities were required to provide urine samples. Agency-wide drug testing programs remain constitutionally suspect.

Applying the Reasonableness Standard

While Skinner and Von Raab opened the door to substantial drug testing, several cases that have since been decided in the Federal Courts of Appeals indicate that the courts will scrutinize the constitutionality of drug-testing programs and that lower court judges do not see Skinner and Von Raab as eliminating all constitutional barriers to governmentally mandated drug testing.

For example in Harmon v. Thornburgh (878 F.2d 484, D.C. Cir. 1989, Cert. denied 493 U.S. 1056, 1990), a case brought by the U.S. Department of Justice employees challenging that agency's drug-testing program, the D.C. Court of Appeals read Von Raab as "suggesting that the government may search its employees only when a clear, direct nexus exists between the nature of the employee's duty and the nature of the feared violation." Applying this standard, the court upheld random testing for employees with access to top secret classified information and suggested that the drug testing of drug prosecutors would be permissible. However, the court held that the government's interests in work force integrity, public safety, and protection of sensitive information did not, for Fourth Amendment purposes, make reasonable a plan for randomly testing employees in less sensitive positions, including Antitrust Division attorneys. The appellate court also rejected the department's plan to test all criminal prosecutors and all employees with access to grand jury materials. In a later case, a different panel of the D.C. Court of Appeals, in evaluating the U.S. Department of

Defense's drug-testing program for civilian employees, distinguished between laboratory personnel on one hand (random testing not allowed) and aviation workers, police guards, and drug counselors on the other (random testing allowed) (National Federation of Federal Employees v. Cheney, 884 F.2d 603, D.C. Cir. 1989, Cert. denied, 493 U.S. 1056, 1990). In a similar vein the Seventh Circuit Court required the Cook County Department of Corrections to limit its unannounced drug-testing program to employees with regular prisoner contact, with opportunities to smuggle drugs, or with access to firearms (Taylor v. O'Grady, 888 F.2d 1189, 7th Cir. 1989). The D.C. Circuit Court has, however, upheld the testing of mail van operators, not because of safety risks but on the rationale they are vulnerable to blackmail (American Federation of Government Employees v. Skinner, 885 F.2d 884, D.C. Cir. 1989, Cert. denied, 495 U.S. 923, 1990), and it has upheld the urine testing, without probable-cause, of a person who had been tentatively accepted for employment by the Department of Justice's Antitrust Division (Willner v. Thornburgh, 928 F.2d 1185, D.C. Cir. 1991).

Factors to Be Considered

In striking the balance mandated by Skinner and Von Raab, there appears to be an emerging consensus among federal courts reviewing drug testing by public employers about what makes a governmentally mandated program reasonable within the meaning of the Fourth Amendment.

First, the courts have been concerned with who gets tested and whether there is a reasonable basis for demanding that they do get tested. When drug-testing programs have been found unconstitutional, it has usually been because they targeted too wide a range of employees. In such cases, the remedy has typically been to narrow the program to employees whose drug use poses an actual or symbolic threat of harm to fellow employees, the public, or an agency's mission.

While few courts demand proof of significant employee drug use or drug-related impairment for postemployment testing, the employer must offer some reason why drug abuse by employees could impair the safety, integrity, or productivity of the operation; a threat to the employees' own health or safety or the moral desirability of a drug-free workplace is not enough. Employers have been found to lack the compelling safety interest that would justify random or routine testing of transportation maintenance custodians (Bolden v. SEPTA, 3rd Cir. April 1, 1991) or sanitation enforcement agents (Watson v. Sexton, 755 F. Supp. 583, S.D.N.Y. 1991).

An exception to this requirement of job-specific dangers may arise in heavily regulated industries like transportation, in which a reduced expectation of privacy on the part of all workers seems to obviate the need for an inquiry into the specific risks that drug abuse would pose to their work (IBEW v. Skinner, 913 F.2d 1454, 9th Cir. 1991). But those expectations of privacy are often reduced precisely because intensive screening, regulation, or supervision are thought to be necessary to protect the safety or integrity of the operation—which makes it difficult to assess the independent significance of privacy expectations in the judicial review of workplace drug-testing programs.

Second, the courts recognize three distinct but overlapping factors as justifying random, routine, or universal testing for whole job categories. The most important is the actual risk to safety or integrity posed by employee drug use, particularly to nonemployees, as, for example, in the case of workers at a nuclear power plant (Ensor v. Rust Engineering Co., 704 F. Supp 808, E.D. Tenn. 1989, Aff'd 935 F.2d 269, 6th Cir. 1991). A second factor that reinforces the first is the special responsibility of the tested employees for the safety of those who might be endangered by their drug use—bus drivers are an example (Holloman v. GCRTA, 741 F. Supp. 677, N.D. Ohio, 1990, Aff'd 930 F.2d 918, 6th Cir. 1991). The third factor is the need to maintain public trust and confidence. For example, the district court in O'Connor v. Police Commissioner of Boston wrote, in upholding a drug-testing program for police cadets, ''[P]ublic confidence in the police is a social necessity and is enhanced by procedures that deter drug use by police cadets" (557 N.E.2d 1146, 1150, Mass. 1990. See also: Gauthier v. Police Commissioner of Boston, 557 N.E. 2d 1374, Mass. 1990).

While there are cases that suggest that programs that mandate the testing of all employees within a particular category are more likely to survive constitutional scrutiny than those that randomly select individuals for testing, random testing has been allowed in situations in which it seems reasonable for deterrence or other reasons. It also appears that, although the Supreme Court in Skinner pointed to the absence of visual monitoring as a factor supporting the constitutionality of the program at issue in that case, monitoring seldom appears as a crucial factor, perhaps because visual monitoring of the urine sample collection process seldom seems to occur.

The reliability of testing techniques in detecting the presence of illegal drugs (or their metabolites) has been noted by some courts as a factor in their decisions, but little attention has been paid to whether the type of drugs tested for or the amount detected reflects a threat to safety, integrity, or productivity. Thus, by far the most common illicit drug revealed in workplace testing is marijuana, the adverse effects of which are comparatively slight and disputed. Even less attention has been paid to the preventive or deterrent value of the testing, that is, to whether a drug-testing program really reduces employee drug consumption. The courts seem surprisingly uninterested in the empirical research bearing on these questions.

Although some court decisions mention the fact that a positive drug test will not lead to the dismissal of an employee or will not be shared with law enforcement agencies as a factor supporting a testing program, by and large the courts have not expressed much concern about the consequences of positive test results. The judicial silence on these issues may result in part from the growing uniformity of public-and private-sector testing programs, promoted by statutes like the DFWA and OTETA, which impose stringent restrictions on the sanctions that can be imposed for positive test results and on the disclosure of those results. The private employers whose testing programs lack these restrictions are less likely to be subject to judicial scrutiny.

The Private Sector

Private businesses are not subject to the strictures of the Fourth Amendment unless their testing programs are mandated by law. Thus the factors that lead courts under the Fourth Amendment to approve or disapprove of drug-testing programs in the public sector will not in themselves validate or undermine the drug-testing programs that private businesses establish at their own initiative. The other two dichotomies that we have mentioned, the distinction between pre-and postemployment drug testing and the distinction between state and federal law, do play an important role here. Generally speaking, both public and private sectors have considerably more leeway in establishing preemployment drug-screening programs than they do in establishing postemployment drug-testing programs.

In the private sector, the potential limitations are greatest among the unionized segment of the work force. This is because the National Labor Relations Board has held that the institution of a drug-testing program for current employees constitutes a "material change" in the employees' working conditions that is subject to collective bargaining (Johnson-Bateman Co., 295 N.L.R.B. No. 26, June 15, 1989), but it has reached the opposite conclusions with respect to preemployment programs (Star Tribune, 295 N.L.R.B. No. 63, June 15, 1989). This means that preemployment drug-testing programs may take whatever form management prefers, but postemployment testing of unionized employees is subject to negotiations that usually produce compromise programs, often restricting the testing of employees to cases of reasonable suspicion and almost always mandating treatment rather than discipline as the initial response to positive test results. Also, it is possible for employees in unionized industries to file grievances regarding certain characteristics of drug-testing programs. Arbitrators have struck down drug-testing programs on such grounds as (1) the testing procedures were not accurate, (2) a testing program was not within the employer's prerogative, and (3) the testing program was not reasonable given the lack of evidence of a workplace drug problem. But as unions and management agree on the details of such programs and as drug testing in the workplace becomes more common, such decisions are increasingly rare (Hopson, 1986; Veglahn, 1988; Labor Relations Week, 1989).

In the nonunionized sector, existing employees, but not job applicants, may benefit from the erosion that has been occurring in some jurisdictions in the "employment at will" doctrine; that is, the rule that gives employers wide latitude in dismissing employees for any reason that is not prohibited by law. This body of law is beginning to change so that in certain situations courts will hold that dismissals for certain reasons are against public policy. Given the national policy against drug use, employees to date have had little success in claiming that discharges based on positive drug tests violate public policy, but one court did indicate that a firing in contravention of a state statute regulating workplace drug testing would be against public policy (Johnson v. Carpenter Technology Corp., 723 F. Supp. 180, D. Conn. 1989).

Private employers may be reluctant to fire employees who refuse to submit to drug testing if they will be held liable for that employee's unemployment compensation. A discharged employee is not eligible for unemployment compensation if she was fired for misconduct, but her refusal to obey an employer's rule is held to constitute misconduct if and only if that rule was reasonable as applied to her. Although no court has held that an employer's drug testing must conform to the Fourth Amendment to be reasonable, courts have found drug testing predicated solely on unsupported allegations to be unreasonable. Testing based on "suspicious behavior" has, however, been found reasonable (Annotation "Private Employee's Loss of Employment Because of a Refusal to Submit to a Drug Test as Affecting Right to Unemployment Compensation" 86 A.L.R. 4th 309, 1991).

In a related area, courts and unemployment boards generally uphold the denial of unemployment benefits to employees fired because of arrest or conviction for off-the-job drug possession, on the grounds that the employees have not become unemployed "through no fault of their own," as most applicable statutes require. Recent cases have also upheld denial of compensation to school teachers and job counselors on the grounds that drug possession is a form of misconduct incompatible with their responsibilities. Other cases have required compensation when the statutory exclusion covered only employees discharged for misconduct "in connection with … employment." A workplace drug test may reveal misconduct that warrants discharge without compensation. One recent decision found that a bus driver was not entitled to compensation after discharge for a positive drug-test result, since by coming into the office during his vacation to report an earlier accident, he violated a company ban on employees having drugs in their system while on the premises (Shaw v. Unemployment Compensation Board of Review, 539 A.2d 1383, Pa. Comwlth 1988. See also Annotation "Jobless Pay—Off-Duty Conduct" 35 A.L.R. 4th 691, Secs. 5,8, 1991 Supp.).

Employees and Applicants

The preemployment/postemployment distinction is important in the public sector as well, for it is a factor that may bear on the constitutionality of a drug-testing program. One reason relates to the property interest that an individual may be seen to have in the expectation of continued employment and the protection that due process gives such interests. Another reason relates to the reasonableness of a drug-testing program. One factor that might lead a court to find a preemployment drug-testing program reasonable whereas a postemployment program would not be is that an employer cannot observe an applicant at work and so has little basis apart from a drug test for determining whether his or her work might be impaired by drug use.

A pair of cases from the D.C. Circuit indicate the potential importance of the distinction. In Harmon v. Thornburgh (878 F.2d 484, D.C. Cir. 1989, Cert. denied 110 S. Ct. 865, 1990) the appellate court held that a drug-testing program for current Justice Department employees could not include, among others, the department's antitrust lawyers because there was an insufficient nexus between their work and drug use to justify the privacy invasion. (The same decision upheld the testing of prosecutors who handled drug cases.) In Willner v. Thornburgh (928 F.2d 1185, D.C. Cir. 1991) a different panel of the same court approved a program that required an applicant for a position with the department's Antitrust Division to submit to a drug test. An important rationale for this decision was that the Fourth Amendment precludes only unreasonable intrusions on privacy, and the fact that applicant drug testing is common in the private sector "is some indication of what expectations of privacy society is prepared to accept as 'reasonable' (id. at 1192-93)." If this rationale were generally accepted, it could eliminate the importance of the public-private distinction with respect to job applicants, as private-sector behavior unconstrained by the Fourth Amendment would ultimately determine what were reasonable expectations of privacy under the Fourth Amendment. At least some other courts, however, have relied on Harmon in job applicant cases and have not made this bootstrap argument (Georgia Ass'n of Educators v. Harris, 749 F. Supp. 1110, N.D. Ga. 1990).

State Law

The third distinction that we alluded to at the outset is between state and federal law. When federal law, including the valid regulations of federal administrative agencies, mandates drug testing, that testing is required, regardless of any state law to the contrary, so long as it is otherwise constitutional. A federal agency testing its own employees pursuant to federal law is similarly unaffected by state legal restrictions. However, private entities and state agencies may in particular states be barred by state law from engaging in drug testing in circumstances in which federal law and the U.S. Constitution pose no legal barriers.

State legal barriers may arise because identically worded state and federal constitutional provisions are interpreted differently by the state and federal supreme courts, because state constitutions may protect rights that are not explicitly protected in the U.S. Constitution, and because state legislatures may enact statutes that relate specifically to drug testing.

Ten states have explicit constitutional rights to privacy, and most other states have recognized an implicit right to privacy in their constitutions (Silverstein, 1989). California courts interpret their explicit constitutional provision to apply to private as well as public employers. California constitutional law requires a showing of "reasonableness" for preemployment drug testing and of "compelling interest" for postemployment testing. In 1987, a California appellate court affirmed a $485,000 jury award for wrongful discharge to a programmer for the Southern Pacific Railroad, who had been fired for refusing to provide a urine sample for drug testing. The court held that the railroad lacked a safety interest strong enough to justify testing. Not all states, however, would apply constitutional privacy provisions to private actions (The Alaska Supreme Court, for example, has held that its constitutional right to privacy only protects against state action and could not be violated by a private employer's drug-testing program. Luedtke v. Nabors Alaska Drilling, Inc., 78 P.2d 1123, 1989) and, in another California case, a state appellate court held that preemployment drug and alcohol testing through urinalysis was not so intrusive as to violate that state's constitutional privacy protection (Wilkinson v. Times Corp., 215 Cal. App. 3rd 1034, Cal. App. 1 Dist. 1989).

At least seven states have legislation that restricts the right of employers to order their employees to submit to mandatory urinalysis for purposes of identifying or deterring illegal drug use. Montana, Iowa, Vermont, and Rhode Island have banned all random or blanket testing of employees (without probable-cause or reasonable suspicion) and Minnesota, Maine, Connecticut, and Oregon permit random testing only of employees in safety-sensitive positions or otherwise limit the circumstances in which employees can be tested. These states also mandate confirmatory testing, certified laboratories, confidentiality of test results, and other procedural protections. Louisiana, Maryland, Nebraska, and Utah also regulate test procedures and protect the confidentiality of test results, but they do not limit the circumstances in which employees can be tested. In addition, several municipalities, including Boulder, San Francisco, and New York, have ordinances that restrict workplace drug testing for some employers. Municipal restrictions are valid to the extent that they do not conflict with applicable state or federal law.

Fearing that a patchwork of restrictive laws would deter employers, especially multistate employers, proponents of drug testing have sought to create national standards for drug testing through federal laws that preempt state and local legislation. With the enactment of OTETA, with a strong preemption provision, they have gained a major victory in the transportation field. Parts of the Vermont and Rhode Island statutes have already been held to be preempted by federal regulations; other state statutes explicitly defer to conflicting federal laws and regulations.


The scope of workplace drug testing is limited by certain state and federal constitutional restrictions, particularly in the public sector and in postemployment settings, but these limits are generous and allow a broad range of employees to be tested using a wide range of reliable methodologies. If there is a general limitation discernable, it is that the drug testing of public employees or of any employees pursuant to a governmentally mandated program violates the Constitution if the testing program appears directed at the general enforcement of the nation's drug laws and lacks a particularized occupational nexus. Even this principle, however, appears to be substantially relaxed in the case of job applicants, at least as a practical matter. Although there are a few states that legislatively protect certain workers from drug testing, most laws and regulations that regulate drug testing are important not for restricting who may be tested but for mandating testing in certain situations, for requiring protections against test error, and for restricting the consequences of positive findings. Indeed, the imposition of drug testing on a wide range of public and private workplaces by federal law and regulation has standardized testing programs and alleviated many of the early concerns about reliability and confidentiality. It is virtually impossible for an "innocent" employee to be fired or blackballed because of a false positive result in a program that complies with the DHHS guidelines.

Accuracy and confidentiality, however, remain concerns for nonunionized private employers not regulated by the DHHS guidelines. Except in states with restrictive legislation or constitutional provisions, private employers in industries unaffected by federal regulations are essentially free to test any employee at any time, without guarantees of the accuracy of the testing process and with few restrictions on the use or disclosure of the test results. For employers who invest little in training, have little concern with morale, and cannot afford confirmatory tests or treatment programs, a high probability that an employee uses illegal drugs may be reason enough to fire him. If such quick-and-dirty drug testing becomes widespread, it may provoke another round of restrictive state statutes, not to supersede federal guidelines, but to extend them to a wider range of workplaces.

New legislation may also come with new technologies of drug and performance testing, and such developments might also reopen constitutional issues that for the moment appear settled. For example, hair assays are less intrusive than urine testing and so might justify testing in situations that now fail to pass constitutional muster when the need for and intrusiveness of testing are balanced. At the same time, hair assays may raise new legal and constitutional problems due to their capacity to detect drug metabolites long after they have left the urine. The nexus between detected drug use and occupational impairment will be weaker than it is with urine testing, and the provisions of the Americans With Disabilities Act of 1991, which excludes current drug users but protects recovering addicts, might be more likely to be at issue when workers test positive.

Additional problems might also be posed with advances in performance testing. To the extent that new performance tests, such as tests of hand-eye coordination, can cheaply and accurately measure job-relevant psychomotor skills at the start of each workday, the apparent necessity of drug testing will diminish1 and the relative intrusiveness of urine testing may appear to increase.

Ultimately, however, the legal future of drug testing is likely to turn less on the development of new technologies than it is on widespread social attitudes. The emerging judicial consensus in favor of drug testing appears to be built on four widely shared perceptions: (1) that drug abuse is one of the nation's most serious problems; (2) that drug-taking workers, at least in certain job categories, pose a threat to themselves, their coworkers, and the public and to the integrity and productivity of their operations; (3) that drug testing is reliable and accurate and can be protected from arbitrary and discriminatory implementation; and (4) that drug testing can be carried out in a way that minimizes intrusions on privacy and dignity. So long as judges continue to have these perceptions, courts are unlikely to interfere greatly with either legally mandated or private drug-testing programs.

A Word About Ethics

The fact that the law allows the government to test its employees in many circumstances and imposes few or no limits on applicant drug testing or on drug testing in the private sector does not, however, mean that drug testing is a wise or proper thing to do. Policies that are legally permissible may be ethically wrong. We do not mean to canvass ethical issues here, for there is neither a body of authority to summarize nor a scientific consensus to restate. But we do wish to emphasize that, when a drug-testing program is challenged on ethical grounds, it is no answer to say, "But the law allows it." Similarly the fact that the law forbids drug testing in certain circumstances does not mean that testing in those circumstances would, apart from the legal violation, be ethically unjustified.

To give just one example, preemployment screening is ordinarily allowed without a strong nexus between revealed drug use and likely job performance. It does not follow that it is right to screen for drugs where no job-related nexus can be shown. The privacy of many is invaded to screen out a few whose drug use would have caused no harm had they not been screened out. The situation would be even more problematic if testing for non-job-related drug use disproportionately screened out members of a particular group and was known to do so. Indeed, in those circumstances even some job-relatedness might not be enough to ethically justify drug screening. At the same time, the lack of a job-related nexus does not necessarily mean that a drug-testing program is ethically wrong. If such testing programs discourage the use of drugs associated with non-job-related harms, this deterrent effect can form the basis for a coherent ethical argument supporting both applicant and employee drug testing. The strength of the argument from deterrence and its weight relative to the ethical case against generalized drug screening are matters that can be and have been disputed.

An interesting feature of the different ethical stances that can be maintained toward ethically problematic policies, like some preemployment drug screening, is that an evaluation of their persuasive force depends in large measure on empirical judgments. If widespread preemployment drug screening offended few people, discouraged many people from using dangerous drugs, did not systematically disadvantage people of a particular social status, and had a strong job nexus, then the ethical status of generalized screening programs would, for most people, be different from what it would be if the characteristics of the screening program were just the reverse. Thus empirical research is important not only to develop more effective drug-testing programs or to tell organizations whether proposed programs make economic or safety sense. It is also important to inform the serious ethical conversation that should take place whenever there is a proposal to institute, cancel, or substantially change a program of workplace drug testing.


  • Hopson, E.S. 1986. Alcohol and drug abuse cases in arbitration. Pp. 275-290 in W. Dolson, editor. , ed., Annual Labor and Employment Law Institute. Littleton, Colo.: Fred B. Rothman & Co.
  • Labor Relations Week 1989. Workplace drug testing: evolving law and employer practice. Labor Relations Week (suppl) 9:3-32.
  • Silverstein, M. 1989. Privacy rights in state consitutions: models for Illinois? University of Illinois Law Review 215-296.
  • Veglahn, P.A. 1988. What is a reasonable drug testing program?: insight from arbitration decisions. Labor Law Journal 39:688-695.



To the extent that drugs use substantially degrades job performance, performance tests will even serve some of the deterrence functions of chemical drug testing, for a worker using drugs can expect to fail the performance test and be sent home, losing a day's pay. Also not doing well on a performance test might serve to establish cause for a drug test.

Copyright 1994 by the National Academy of Sciences. All rights reserved.
Bookshelf ID: NBK236253


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (2.6M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...