489 U.S. 602
109 S.Ct. 1402
103 L.Ed.2d 639
Samuel K. SKINNER, Secretary of Transportation, et al., Petitioners
RAILWAY LABOR EXECUTIVES' ASSOCIATION et al.
Argued Nov. 2, 1988.
Decided March 21, 1989.
Upon the basis of evidence indicating that alcohol and drug abuse by railroad employees had caused or contributed to a number of significant train accidents, the Federal Railroad Administration (FRA) promulgated regulations under petitioner Secretary of Transportation's statutory authority to adopt safety standards for the industry. Among other things, Subpart C of the regulations requires railroads to see that blood and urine tests of covered employees are conducted following certain major train accidents or incidents, while Subpart D authorizes, but does not require, railroads to administer breath or urine tests or both to covered employees who violate certain safety rules. Respondents, the Railway Labor Executives' Association and various of its member labor organizations, brought suit in the Federal District Court to enjoin the regulations. The court granted summary judgment for petitioners, concluding that the regulations did not violate the Fourth Amendment. The Court of Appeals reversed, ruling, inter alia, that a requirement of particularized suspicion is essential to a finding that toxicological testing of railroad employees is reasonable under the Fourth Amendment. The court stated that such a requirement would ensure that the tests, which reveal the presence of drug metabolites that may remain in the body for weeks following ingestion, are confined to the detection of current impairment.
1. The Fourth Amendment is applicable to the drug and alcohol testing mandated or authorized by the FRA regulations. Pp. 614-618.
(a) The tests in question cannot be viewed as private action outside the reach of the Fourth Amendment. A railroad that complies with Subpart C does so by compulsion of sovereign authority and therefore must be viewed as an instrument or agent of the Government. Similarly, even though Subpart D does not compel railroads to test, it cannot be concluded, in the context of this facial challenge, that such testing will be primarily the result of private initiative, since specific features of the regulations combine to establish that the Government has actively encouraged, endorsed, and participated in the testing. Specifically, since the regulations pre-empt state laws covering the same subject matter and are intended to supersede collective-bargaining and arbitration-award provisions, the Government has removed all legal barriers to the testing authorized by Subpart D. Moreover, by conferring upon the FRA the right to receive biological samples and test results procured by railroads, Subpart D makes plain a strong preference for testing and a governmental desire to share the fruits of such intrusions. In addition, the regulations mandate that railroads not bargain away their Subpart D testing authority and provide that an employee who refuses to submit to such tests must be withdrawn from covered service. Pp. 614-616.
(b) The collection and subsequent analysis of the biological samples required or authorized by the regulations constitute searches of the person subject to the Fourth Amendment. This Court has long recognized that a compelled intrusion into the body for blood to be tested for alcohol content and the ensuing chemical analysis constitute searches. Similarly, subjecting a person to the breath test authorized by Subpart D must be deemed a search, since it requires the production of "deep lung" breath and thereby implicates concerns about bodily integrity. Moreover, although the collection and testing of urine under the regulations do not entail any intrusion into the body, they nevertheless constitute searches, since they intrude upon expectations of privacy as to medical information and the act of urination that society has long recognized as reasonable. Even if the employer's antecedent interference with the employee's freedom of movement cannot be characterized as an independent Fourth Amendment seizure, any limitation on that freedom that is necessary to obtain the samples contemplated by the regulations must be considered in assessing the intrusiveness of the searches affected by the testing program. Pp. 616-618.
2. The drug and alcohol tests mandated or authorized by the FRA regulations are reasonable under the Fourth Amendment even though there is no requirement of a warrant or a reasonable suspicion that any particular employee may be impaired, since, on the present record, the compelling governmental interests served by the regulations outweigh employees' privacy concerns. Pp. 618-633.
(a) The Government's interest in regulating the conduct of railroad employees engaged in safety-sensitive tasks in order to ensure the safety of the traveling public and of the employees themselves plainly justifies prohibiting such employees from using alcohol or drugs while on duty or on call for duty and the exercise of supervision to assure that the restrictions are in fact observed. That interest presents "special needs" beyond normal law enforcement that may justify departures from the usual warrant and probable-cause requirements. Pp. 618-621.
(b) Imposing a warrant requirement in the present context is not essential to render the intrusions at issue reasonable. Such a requirement would do little to further the purposes of a warrant, since both the circumstances justifying toxicological testing and the permissible limits of such intrusions are narrowly and specifically defined by the regulations and doubtless are well known to covered employees, and since there are virtually no facts for a neutral magistrate to evaluate, in light of the standardized nature of the tests and the minimal discretion vested in those charged with administering the program. Moreover, imposing a warrant requirement would significantly hinder, and in many cases frustrate, the objectives of the testing program, since the delay necessary to procure a warrant could result in the destruction of valuable evidence, in that alcohol and drugs are eliminated from the bloodstream at a constant rate, and since the railroad supervisors who set the testing process in motion have little familiarity with the intricacies of Fourth Amendment jurisprudence. Pp. 621-624.
(c) Imposing an individualized suspicion requirement in the present context is not essential to render the intrusions at issue reasonable. The testing procedures contemplated by the regulations pose only limited threats to covered employees' justifiable privacy expectations, particularly since they participate in an industry subject to pervasive safety regulation by the Federal and State Governments. Moreover, because employees ordinarily consent to significant employer-imposed restrictions on their freedom of movement, any additional interference with that freedom that occurs in the time it takes to procure a sample from a railroad employee is minimal. Furthermore, Schmerber v. California, 384 U.S. 757, 86 S.Ct. 1826, 16 L.Ed.2d 908, established that governmentally imposed blood tests do not constitute an unduly extensive imposition on an individual's privacy and bodily integrity, and the breath tests authorized by Subpart D are even less intrusive than blood tests. And, although urine tests require employees to perform an excretory function traditionally shielded by great privacy, the regulations reduce the intrusiveness of the collection process by requiring that samples be furnished in a medical environment without direct observation. In contrast, the governmental interest in testing without a showing of individualized suspicion is compelling. A substance-impaired railroad employee in a safety-sensitive job can cause great human loss before any signs of the impairment become noticeable, and the regulations supply an effective means of deterring such employees from using dru § or alcohol by putting them on notice that they are likely to be discovered if an accident occurs. An individualized suspicion requirement would also impede railroads' ability to obtain valuable information about the causes of accidents or incidents and how to protect the public, since obtaining evidence giving rise to the suspicion that a particular employee is impaired is impracticable in the chaotic aftermath of an accident when it is difficult to determine which employees contributed to the occurrence and objective indicia of impairment are absent. The Court of Appeals' conclusion that the regulations are unreasonable because the tests in question cannot measure current impairment is flawed. Even if urine test results disclosed nothing more specific than the recent use of controlled substances, this information would provide the basis for a further investigation and might allow the FRA to reach an informed judgment as to how the particular accident occurred. More importantly, the court overlooked the FRA's policy of placing principal reliance on blood tests, which unquestionably can identify recent drug use, and failed to recognize that the regulations are designed not only to discern impairment but to deter it. Pp. 624-632.
839 F.2d 575 (CA9 1988), reversed.
KENNEDY, J., delivered the opinion of the Court, in which REHNQUIST, C.J., and WHITE, BLACKMUN, O'CONNOR, and SCALIA, JJ., joined, and in all but portions of Part III of which STEVENS, J., joined. STEVENS, J., filed an opinion concurring in part and concurring in the judgment, post, p. 634. MARSHALL, J., filed a dissenting opinion, in which BRENNAN, J., joined, post, p. 635.
Atty. Gen. Dick Thornburgh, Washington, D.C., for petitioners.
Lawrence M. Mann, Washington, D.C., for respondents.
Justice KENNEDY delivered the opinion of the Court.
The Federal Railroad Safety Act of 1970 authorizes the Secretary of Transportation to "prescribe, as necessary, appropriate rules, regulations, orders, and standards for all areas of railroad safety." 84 Stat. 971, 45 U.S.C. § 431(a). Finding that alcohol and drug abuse by railroad employees poses a serious threat to safety, the Federal Railroad Administration (FRA) has promulgated regulations that mandate blood and urine tests of employees who are involved in certain train accidents. The FRA also has adopted regulations that do not require, but do authorize, railroads to administer breath and urine tests to employees who violate certain safety rules. The question presented by this case is whether these regulations violate the Fourth Amendment.
The problem of alcohol use on American railroads is as old as the industry itself, and efforts to deter it by carrier rules began at least a century ago. For many years, railroads have prohibited operating employees from possessing alcohol or being intoxicated while on duty and from consuming alcoholic beverages while subject to being called for duty. More recently, these proscriptions have been expanded to forbid possession or use of certain drugs. These restrictions are embodied in "Rule G," an industry-wide operating rule promulgated by the Association of American Railroads, and are enforced, in various formulations, by virtually every railroad in the country. The customary sanction for Rule G violations is dismissal.
In July 1983, the FRA expressed concern that these industry efforts were not adequate to curb alcohol and drug abuse by railroad employees. The FRA pointed to evidence indicating that on-the-job intoxication was a significant problem in the railroad industry.1 The FRA also found, after a review of accident investigation reports, that from 1972 to 1983 "the nation's railroads experienced at least 21 significant train accidents involving alcohol or drug use as a probable cause or contributing factor," and that these accidents "resulted in 25 fatalities, 61 non-fatal injuries, and property damage estimated at $19 million (approxima ely $27 million in 1982 dollars)." 48 Fed.Reg. 30726 (1983). The FRA further identified "an additional 17 fatalities to operating employees working on or around rail rolling stock that involved alcohol or drugs as a contributing factor." Ibid. In light of these problems, the FRA solicited comments from interested parties on various regulatory approaches to the problems of alcohol and drug abuse throughout the Nation's railroad system.
Comments submitted in response to this request indicated that railroads were able to detect a relatively small number of Rule G violations, owing, primarily, to their practice of relying on observation by supervisors and co-workers to enforce the rule. 49 Fed.Reg. 24266-24267 (1984). At the same time, "industry participants . . . confirmed that alcohol and drug use [did] occur on the railroads with unacceptable frequency," and available information from all sources "suggest[ed] that the problem includ[ed] 'pockets' of drinking and drug use involving multiple crew members (before and during work), sporadic cases of individuals reporting to work impaired, and repeated drinking and drug use by individual employees who are chemically or psychologically dependent on those substances." Id., at 24253-24254. "Even without the benefit of regular post-accident testing," the FRA "identified 34 fatalities, 66 injuries and over $28 million in property damage (in 1983 dollars) that resulted from the errors of alcohol and drug-impaired employees in 45 train accidents and train incidents during the period 1975 through 1983." Id., at 24254. Some of these accidents resulted in the release of hazardous materials and, in one case, the ensuing pollution required the evacuation of an entire Louisiana community. Id., at 24254, 24259. In view of the obvious safety hazards of drug and alcohol use by railroad employees, the FRA announced in June 1984 its intention to promulgate federal regulations on the subject.
After reviewing further comments from representatives of the railroad industry, labor groups, and the general public, the FRA, in 1985, promulgated regulations addressing the problem of alcohol and drugs on the railroads. The final regulations apply to employees assigned to perform service subject to the Hours of Service Act, ch. 2939, 34 Stat. 1415, as amended, 45 U.S.C. § 61 et seq. The regulations prohibit covered employees from using or possessing alcohol or any controlled substance. 49 CFR § 219.101(a)(1) (1987). The regulations further prohibit those employees from reporting for covered service while under the influence of, or impaired by, alcohol, while having a blood alcohol concentration of .04 or more, or while under the influence of, or impaired by, any controlled substance. § 219.101(a)(2). The regulations do not restrict, however, a railroad's authority to impose an absolute prohibition on the presence of alcohol or any drug in the body fluids of persons in its employ, § 219.101(c), and, accordingly, they do not "replace Rule G or render it unenforceable." 50 Fed.Reg. 31538 (1985).
To the extent pertinent here, two subparts of the regulations relate to testing. Subpart C, which is entitled "Post-Accident Toxicological Testing," is mandator . It provides that railroads "shall take all practicable steps to assure that all covered employees of the railroad directly involved . . . provide blood and urine samples for toxicological testing by FRA," § 219.203(a), upon the occurrence of certain specified events. Toxicological testing is required following a "major train accident," which is defined as any train accident that involves (i) a fatality, (ii) the release of hazardous material accompanied by an evacuation or a reportable injury, or (iii) damage to railroad property of $500,000 or more. § 219.201(a)(1). The railroad has the further duty of collecting blood and urine samples for testing after an "impact accident," which is defined as a collision that results in a reportable injury, or in damage to railroad property of $50,000 or more. § 219.201(a)(2). Finally, the railroad is also obligated to test after "[a]ny train incident that involves a fatality to any on-duty railroad employee." § 219.201(a)(3).
After occurrence of an event which activates its duty to test, the railroad must transport all crew members and other covered employees directly involved in the accident or incident to an independent medical facility, where both blood and urine samples must be obtained from each employee.2 After the samples have been collected, the railroad is required to ship them by prepaid air freight to the FRA laboratory for analysis. § 219.205(d). There, the samples are analyzed using "state-of-the-art equipment and techniques" to detect and measure alcohol and drugs.3 The FRA proposes to place primary reliance on analysis of blood samples, as blood is "the only available body fluid . . . that can provide a clear indication not only of the presence of alcohol and drugs but also their current impairment effects." 49 Fed.Reg. 24291 (1984). Urine samples are also necessary, however, because drug traces remain in the urine longer than in blood, and in some cases it will not be possible to transport employees to a medical facility before the time it takes for certain drugs to be eliminated from the bloodstream. In those instances, a "positive urine test, taken with specific information on the pattern of elimination for the particular drug and other information on the behavior of the employee and the circumstances of the accident, may be crucial to the determination of" the cause of an accident. Ibid.
The regulations require that the FRA notify employees of the results of the tests and afford them an opportunity to respond in writing before preparation of any final investigative report. See § 219.211(a)(2). Employees who refuse to provide required blood or urine samples may not perform covered service for nine months, but they are entitled to a hearing concerning their refusal to take the test, § 219.213
Subpart D of the regulations, which is entitled "Authorization to Test for Cause," is permissive. It authorizes railroads to require covered employees to submit to breath or urine tests in certain circumstances not addressed by Subpart C. Breath or urine tests, or both, may be ordered (1) after a reportable accident or incident, where a supervisor has a "reasonable suspicion" that an employee's acts or omissions contributed to the occurrence or severity of the accident or incident, § 219.301(b)(2); or (2) in the event of certain specific rule violations, including noncompliance with a signal and excessive speeding, § 219.301(b)(3). A railroad also may require breath tests where a supervisor has a "reasonable suspicion" that an employee is under the influence of alcohol, based upon specific, personal observations concerning the appearance, behavior, speech, or body odors of the employee. § 219.301(b)(1). Where impairment is suspected, a railroad, in addition, may require urine tests, but only if two supervisors make the appropriate determination, § 219.301(c)(2)(i), and, where the supervisors suspect impairment due to a substance other than alcohol, at least one of those supervisors must have received specialized training in detecting the signs of drug intoxication. § 219.301(c)(2)(ii).
Subpart D further provides that whenever the results of either breath or urine tests are intended for use in a disciplinary proceeding, the employee must be given the opportunity to provide a blood sample for analysis at an independent medical facility. § 219.303(c). If an employee declines to give a blood sample, the railroad may presume impairment, absent persuasive evidence to the contrary, from a positive showing of controlled substance residues in the urine. The railroad must, however, provide detailed notice of this presumption to its employees, and advise them of their right to provide a contemporaneous blood sample. As in the case of samples procured under Subpart C, the regulations set forth procedures for the collection of samples, and require that samples "be analyzed by a method that is reliable within known tolerances." § 219.307(b).
Respondents, the Railway Labor Executives' Association and various of its member labor organizations, brought the instant suit in the United States District Court for the Northern District of California, seeking to enjoin the FRA's regulations on various statutory and constitutional grounds. In a ruling from the bench, the District Court granted summary judgment in petitioners' favor. The court concluded that railroad employees "have a valid interest in the integrity of their own bodies" that deserved protection under the Fourth Amendment. App. to Pet. for Cert. 53a. The court held, however, that this interest was outweighed by the competing "public and governmental interest in the . . . promotion of . . . railway safety, safety for employees, and safety for the general public that is involved with the transportation." Id., at 52a. The District Court found respondents' other constitutional and statutory arguments meritless.
A divided panel of the Court of Appeals for the Ninth Circuit reversed. Railway Labor Executives' Assn. v. Burnley, 839 F.2d 575 (1988). The court held, first, that tests mandated by a railroad in reliance on the authority conferred by Subpart D involve sufficient Government action to implicate the Fourth Amendment, and that the breath, blood, and urine tests contemplated by the FRA regulations are Fourth Amendment searches. The court also "agre[ed] that the exigencies of testing for the presence of alcohol and drugs in blood, urine or breath require prompt action which precludes obtaining a warrant." Id., at 583. The court further held that "accommodation of railroad employees' privacy interest with the significant safety concerns of the government does not require adherence to a probable cause requirement," and, accordingly, that the legality of the searches cont mplated by the FRA regulations depends on their reasonableness under all the circumstances. Id., at 587.
The court concluded, however, that particularized suspicion is essential to a finding that toxicological testing of railroad employees is reasonable. Ibid. A requirement of individualized suspicion, the court stated, would impose "no insuperable burden on the government," id., at 588, and would ensure that the tests are confined to the detection of current impairment, rather than to the discovery of "the metabolites of various drugs, which are not evidence of current intoxication and may remain in the body for days or weeks after the ingestion of the drug." Id., at 588-589. Except for the provisions authorizing breath and urine tests on a "reasonable suspicion" of drug or alcohol impairment, 49 CFR §§ 219.301(b)(1) and (c)(2) (1987), the FRA regulations did not require a showing of individualized suspicion, and, accordingly, the court invalidated them.
Judge Alarcon dissented. He criticized the majority for "fail[ing] to engage in [a] balancing of interests" and for focusing instead "solely on the degree of impairment of the workers' privacy interests." 839 F.2d, at 597. The dissent would have held that "the government's compelling need to assure railroad safety by controlling drug use among railway personnel outweighs the need to protect privacy interests." Id., at 596.
We granted the federal parties' petition for a writ of certiorari, 486 U.S. 1042, 108 S.Ct. 2033, 100 L.Ed.2d 618 (1988), to consider whether the regulations invalidated by the Court of Appeals violate the Fourth Amendment. We now reverse.
The Fourth Amendment provides that "[t]he right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated. . . ." The Amendment guarantees the privacy, dignity, and security of persons against certain arbitrary and invasive acts by officers of the Government or those acting at their direction. Camara v. Municipal Court of San Francisco, 387 U.S. 523, 528, 87 S.Ct. 1727, 1730, 18 L.Ed.2d 930 (1967). See also Delaware v. Prouse, 440 U.S. 648, 653-654, 99 S.Ct. 1391, 1395-1396, 59 L.Ed.2d 660 (1979); United States v. Martinez-Fuerte, 428 U.S. 543, 554, 96 S.Ct. 3074, 3081, 49 L.Ed.2d 1116 (1976). Before we consider whether the tests in question are reasonable under the Fourth Amendment, we must inquire whether the tests are attributable to the Government or its agents, and whether they amount to searches or seizures. We turn to those matters.
Although the Fourth Amendment does not apply to a search or seizure, even an arbitrary one, effected by a private party on his own initiative, the Amendment protects against such intrusions if the private party acted as an instrument or agent of the Government. See United States v. Jacobsen, 466 U.S. 109, 113-114, 104 S.Ct. 1652, 1656-1657, 80 L.Ed.2d 85 (1984); Coolidge v. New Hampshire, 403 U.S. 443, 487, 91 S.Ct. 2022, 2048, 29 L.Ed.2d 564 (1971). See also Burdeau v. McDowell, 256 U.S. 465, 475, 41 S.Ct. 574, 575-576, 65 L.Ed. 1048 (1921). A railroad that complies with the provisions of Subpart C of the regulations does so by compulsion of sovereign authority, and the lawfulness of its acts is controlled by the Fourth Amendment. Petitioners contend, however, that the Fourth Amendment is not implicated by Subpart D of the regulations, as nothing in Subpart D compels any testing by private railroads.
We are unwilling to conclude, in the context of this facial challenge, that breath and urine tests required by private railroads in reliance on Subpart D will not implicate the Fourth Amendment. Whether a private party should be deemed an agent or instrument of the Government for Fourth Amendment purposes necessarily turns on the degree of the Government's participation in the private party's activities, cf. Lustig . United States, 338 U.S. 74, 78-79, 69 S.Ct. 1372, 1373-1374, 93 L.Ed. 1819 (1949) (plurality opinion); Byars v. United States, 273 U.S. 28, 32-33, 47 S.Ct. 248, 249-250, 71 L.Ed. 520 (1927), a question that can only be resolved "in light of all the circumstances," Coolidge v. New Hampshire, supra, 403 U.S., at 487, 91 S.Ct., at 2026. The fact that the Government has not compelled a private party to perform a search does not, by itself, establish that the search is a private one. Here, specific features of the regulations combine to convince us that the Government did more than adopt a passive position toward the underlying private conduct.
The regulations, including those in Subpart D, pre-empt state laws, rules, or regulations covering the same subject matter, 49 CFR § 219.13(a) (1987), and are intended to supersede "any provision of a collective bargaining agreement, or arbitration award construing such an agreement," 50 Fed.Reg. 31552 (1985). They also confer upon the FRA the right to receive certain biological samples and test results procured by railroads pursuant to Subpart D. § 219.11(c). In addition, a railroad may not divest itself of, or otherwise compromise by contract, the authority conferred by Subpart D. As the FRA explained, such "authority . . . is conferred for the purpose of promoting the public safety, and a railroad may not shackle itself in a way inconsistent with its duty to promote the public safety." 50 Fed.Reg. 31552 (1985). Nor is a covered employee free to decline his employer's request to submit to breath or urine tests under the conditions set forth in Subpart D. See § 219.11(b). An employee who refuses to submit to the tests must be withdrawn from covered service. See 4 App. to Field Manual 18.
In light of these provisions, we are unwilling to accept petitioners' submission that tests conducted by private railroads in reliance on Subpart D will be primarily the result of private initiative. The Government has removed all legal barriers to the testing authorized by Subpart D and indeed has made plain not only its strong preference for testing, but also its desire to share the fruits of such intrusions. In addition, it has mandated that the railroads not bargain away the authority to perform tests granted by Subpart D. These are clear indices of the Government's encouragement, endorsement, and participation, and suffice to implicate the Fourth Amendment.
Our precedents teach that where, as here, the Government seeks to obtain physical evidence from a person, the Fourth Amendment may be relevant at several levels. See, e.g., United States v. Dionisio, 410 U.S. 1, 8, 93 S.Ct. 764, 768, 35 L.Ed.2d 67 (1973). The initial detention necessary to procure the evidence may be a seizure of the person, Cupp v. Murphy, 412 U.S. 291, 294-295, 93 S.Ct. 2000, 2003, 36 L.Ed.2d 900 (1973); Davis v. Mississippi, 394 U.S. 721, 726-727, 89 S.Ct. 1394, 1397, 22 L.Ed.2d 676 (1969), if the detention amounts to a meaningful interference with his freedom of movement. INS v. Delgado, 466 U.S. 210, 215, 104 S.Ct. 1758, 1762, 80 L.Ed.2d 247 (1984); United States v. Jacobsen, supra, 466 U.S., at 113, n. 5, 104 S.Ct., at 1656, n. 5. Obtaining and examining the evidence may also be a search, see Cupp v. Murphy, supra, 412 U.S., at 295, 93 S.Ct., at 2003, United States v. Dionisio, supra, 410 U.S., at 8, 13-14, 93 S.Ct., at 768, 771-772, if doing so infringes an expectation of privacy that society is prepared to recognize as reasonable, see, e.g., California v. Greenwood, 486 U.S. 35, 43, 108 S.Ct. 1625, 1630, 100 L.Ed.2d 30 (1988); United States v. Jacobsen, supra, 466 U.S., at 113, 104 S.Ct., at 1656.
We have long recognized that a "compelled intrusio[n] into the body for blood to be analyzed for alcohol content" must be deemed a Fourth Amendment search. See Schmerber v. California, 384 U.S. 757, 767-768, 86 S.Ct. 1826, 1833-1834, 16 L.Ed.2d 908 (1966). See also Winston v. Lee, 470 U.S. 753, 760, 105 S.Ct. 1611, 1616, 84 L.Ed.2d 662 (1985). In light of our society's concern for the security of one's person, see, e.g., Terry v. Ohio, 392 U.S. 1, 9, 88 S.Ct. 1868, 1873, 20 L.Ed.2d 889 (1968), it is obvious that this physical intrusion, penetrating beneath the skin, infringes an expectation of privacy that society is prepared to recognize as reasonable. The ensuing chemical analysis of the sample to obtain physiological data is a further invasion of the tested employee's privacy interests. Cf. Arizona v. Hicks, 480 U.S. 321, 324-325, 107 S.Ct. 1149, 1152-1153, 94 L.Ed.2d 347 (1987). Much the same is true of the breath-testing procedures required under Subpart D of the regulations. Subjecting a person to a breathalyzer test, which generally requires the production of alveolar or "deep lung" breath for chemical analysis, see, e.g., California v. Trombetta, 467 U.S. 479, 481, 104 S.Ct. 2528, 2530, 81 L.Ed.2d 413 (1984), implicates similar concerns about bodily integrity and, like the blood-alcohol test we considered in Schmerber, should also be deemed a search, see 1 W. LaFave, Search and Seizure § 2.6(a), p. 463 (1987). See also Burnett v. Anchorage, 806 F.2d 1447, 1449 (CA9 1986); Shoemaker v. Handel, 795 F.2d 1136, 1141 (CA3), cert. denied, 479 U.S. 986, 107 S.Ct. 577, 93 L.Ed.2d 580 (1986).
Unlike the blood-testing procedure at issue in Schmerber, the procedures prescribed by the FRA regulations for collecting and testing urine samples do not entail a surgical intrusion into the body. It is not disputed, however, that chemical analysis of urine, like that of blood, can reveal a host of private medical facts about an employee, including whether he or she is epileptic, pregnant, or diabetic. Nor can it be disputed that the process of collecting the sample to be tested, which may in some cases involve visual or aural monitoring of the act of urination, itself implicates privacy interests. As the Court of Appeals for the Fifth Circuit has stated:
"There are few activities in our society more personal or private than the passing of urine. Most people describe it by euphemisms if they talk about it at all. It is a function traditionally performed without public observation; indeed, its performance in public is generally prohibited by law as well as social custom." National Treasury Employees Union v. Von Raab, 816 F.2d 170, 175 (CA5 1987).
Because it is clear that the collection and testing of urine intrudes upon expectations of privacy that society has long recognized as reasonable, the Federal Courts of Appeals have concluded unanimously, and we agree, that these intrusions must be deemed searches under the Fourth Amendment.4
In view of our conclusion that the collection and subsequent analysis of the requisite biological samples must be deemed Fourth Amendment searches, we need not characterize the employer's antecedent interference with the employee's freedom of movement as an independent Fourth Amendment seizure. As our precedents indicate, not every governmental interference with an individual's freedom of movement raises such constitutional concerns that there is a seizure of the person. See United States v. Dionisio, supra, 410 U.S., at 9-11, 93 S.Ct., at 769-770 (grand jury subpoena, though enforceable by contempt, does not effect a seizure of the person); United States v. Mara, 410 U.S. 19, 21, 93 S.Ct. 774, 775, 35 L.Ed.2d 99 (1973) (same). For present purposes, it suffices to note that any limitation on an employee's freedom of movement that is necessary to obtain the blood, urine, or breath samples contemplated by the regulations must be considered in assessing the intrusiveness of the searches effected by the Government's testing program. Cf. United States v. Place, 462 U.S. 696, 707-709, 103 S.Ct. 2637, 2644-2645, 77 L.Ed.2d 110 (1983).
To hold that the Fourth Amendment is applicable to the drug and alcohol testing prescribed by the FRA regulations is only to begin the inquiry into the standards governing such intrusions. O'Connor v. Ortega, 480 U.S. 709, 719, 107 S.Ct. 1492, 1499, 94 L.Ed.2d 714 (1987) (plurality opinion); New Jersey v. T.L.O., 469 U.S. 325, 337, 105 S.Ct. 733, 741, 83 L.Ed.2d 720 (1985). For the Fourth Amendment does not proscribe all searches and seizures, but only those that are unreasonable. United States v. Sharpe, 470 U.S. 675, 682, 105 S.Ct. 1568, 1573, 84 L.Ed.2d 605 (1985); Schmerber v. California, 384 U.S., at 768, 86 S.Ct., at 1834. What is reasonable, of course, "depends on all of the circumstances surrounding the search or seizure and the nature of the search or seizure itself." United States v. Montoya de Hernandez, 473 U.S. 531, 537, 105 S.Ct. 3304, 3308, 87 L.Ed.2d 381 (1985). Thus, the permissibility of a particular practice "is judged by balancing its intrusion on the individual's Fourth Amendment interests against its promotion of legitimate governmental interests." Delaware v. Prouse, 440 U.S., at 654, 99 S.Ct., at 1396; United States v. Martinez-Fuerte, 428 U.S. 543, 96 S.Ct. 3074, 49 L.Ed.2d 1116 (1976).
In most criminal cases, we strike this balance in favor of the procedures described by the Warrant Clause of the Fourth Amendment. See United States v. Place, supra, 462 U.S., at 701, and n. 2, 103 S.Ct., at 2641 and n. 2; United States v. United States District Court, 407 U.S. 297, 315, 92 S.Ct. 2125, 2135-2136, 32 L.Ed.2d 752 (1972). Except in certain well-defined circumstances, a search or seizure in such a case is not reasonable unless it is accomplished pursuant to a judicial warrant issued upon probable cause. See, e.g., Payton v. New York, 445 U.S. 573, 586, 100 S.Ct. 1371, 1380, 63 L.Ed.2d 639 (1980); Mincey v. Arizona, 437 U.S. 385, 390, 98 S.Ct. 2408, 2412, 57 L.Ed.2d 290 (1978). We have recognized exceptions to this rule, however, "when 'special needs, beyond the normal need for law enforcement, make the warrant and probable-cause requirement impracticable.' " Griffin v. Wisconsin, 483 U.S. 868, 873, 107 S.Ct. 3164, 3168, 97 L.Ed.2d 709 (1987), quoting New Jersey v. T.L.O., supra, 469 U.S., at 351, 105 S.Ct., at 748 (BLACKMUN, J., concurring in judgment). When faced with such special needs, we have not hesitated to balance the governmental and privacy interests to assess the practicality of the warrant and probable-cause requirements in the particular context. See, e.g., Griffin v. Wisconsin, supra, 483 U.S , at 873, 107 S.Ct., at 3168 (search of probationer's home); New York v. Burger, 482 U.S. 691, 699-703, 107 S.Ct. 2636, 2642-2644, 96 L.Ed.2d 601 (1987) (search of premises of certain highly regulated businesses); O'Connor v. Ortega, supra, 480 U.S., at 721-725, 107 S.Ct., at 1499-1502 (work-related searches of employees' desks and offices); -New Jersey v. T.L.O., supra, 469 U.S., at 337-342, 105 S.Ct., at 740-743 (search of student's property by school officials); Bell v. Wolfish, 441 U.S. 520, 558-560, 99 S.Ct. 1861, 1884-1885, 60 L.Ed.2d 447 (1979) (body cavity searches of prison inmates).
The Government's interest in regulating the conduct of railroad employees to ensure safety, like its supervision of probationers or regulated industries, or its operation of a government office, school, or prison, "likewise presents 'special needs' beyond normal law enforcement that may justify departures from the usual warrant and probable-cause requirements." Griffin v. Wisconsin, supra, 483 U.S., at 873-874, 107 S.Ct., at 3168. The hours of service employees covered by the FRA regulations include persons engaged in handling orders concerning train movements, operating crews, and those engaged in the maintenance and repair of signal systems. 50 Fed.Reg. 31511 (1985). It is undisputed that these and other covered employees are engaged in safety-sensitive tasks. The FRA so found, and respondents conceded the point at oral argument. Tr. of Oral Arg. 46-47. As we have recognized, the whole premise of the Hours of Service Act is that "[t]he length of hours of service has direct relation to the efficiency of the human agencies upon which protection [of] life and property necessarily depends." Baltimore & Ohio R. Co. v. ICC, 221 U.S. 612, 619, 31 S.Ct. 621, 625, 55 L.Ed. 878 (1911). See also Atchison, T. & S.F.R. Co. v. United States, 244 U.S. 336, 342, 37 S.Ct. 635, 637, 61 L.Ed. 1175 (1917) ("[I]t must be remembered that the purpose of the act was to prevent the dangers which must necessarily arise to the employee and to the public from continuing men in a dangerous and hazardous business for periods so long as to render them unfit to give that service which is essential to the protection of themselves and those entrusted to their care").
The FRA has prescribed toxicological tests, not to assist in the prosecution of employees, but rather "to prevent accidents and casualties in railroad operations that result from impairment of employees by alcohol or drugs." 49 CFR § 219.1(a) (1987).5 This governmental interest in ensuring the safety of the traveling public and of the employees themselves plainly justifies prohibiting covered employees from using alcohol or drugs on duty, or while subject to being called for duty. This interest also "require[s] and justif[ies] the exercise of supervision to assure that the restrictions are in fact observed." Griffin v. Wisconsin, supra, 483 U.S., at 875, 107 S.Ct., at 3169. The question that remains, then, is whether the Government's need to monitor compliance with these restrictions justifies the privacy intrusions at issue absent a warrant or individualized suspicion.
An essential purpose of a warrant requirement is to protect privacy interests by assuring citizens subject to a search or seizure that such intrusions are not the random or arbitrary acts of government agents. A warrant assures the citizen that the intrusion is authorized by law, and that it is narrowly limited in its objectives and scope. See, e.g., New York v. Burger, supra, 482 U.S., at 703, 107 S.Ct., at 2644; United States v. Chadwick, 433 U.S. 1, 9, 97 S.Ct. 2476, 2482, 53 L.Ed.2d 538 (1977); Camara v. Municipal Court of San Francisco, 387 U.S., at 532, 87 S.Ct., at 1732. A warrant also provides the detached scrutiny of a neutral magistrate, and thus ensures an objective determination whether an intrusion is justified in any given case. See United States v. Chadwick, supra, 433 U.S., at 9, 97 S.Ct., at 2482. In the present context, however, a warrant would do little to further these aims. Both the circumstances justifying toxicological testing and the permissible limits of such intrusions are defined narrowly and specifically in the regulations that authorize them, and doubtless are well known to covered employees. Cf. United States v. Biswell, 406 U.S. 311, 316, 92 S.Ct. 1593, 1596, 32 L.Ed.2d 87 (1972). Indeed, in light of the standardized nature of the tests and the minimal discretion vested in those charged with administering the program, there are virtually no facts for a neutral magistrate to evaluate. Cf. Colorado v. Bertine, 479 U.S. 367, 376, 107 S.Ct. 738, 743, 93 L.Ed.2d 739 (1987) (BLACKMUN, J., concurring).6
We have recognized, moreover, that the government's interest in dispensing with the warrant requirement is at its strongest when, as here, "the burden of obtaining a warrant is likely to frustrate the governmental purpose behind the search." Camara v. Municipal Court of San Francisco, supra, 387 U.S., at 533, 87 S.Ct., at 1733. See also New Jersey v. T.L.O., 469 U.S., at 340, 105 S.Ct., at 742; Donovan v. Dewey, 452 U.S. 594, 603, 101 S.Ct. 2534, 2540, 69 L.Ed.2d 262 (1981). As the FRA recognized, alcohol and other drugs are eliminated from the bloodstream at a constant rate, see 49 Fed.Reg. 24291 (1984), and blood and breath samples taken to measure whether these substances were in the bloodstream when a triggering event occurred must be obtained as soon as possible. See Schmerber v. California, 384 U.S., at 770-771, 86 S.Ct., at 1835-1836. Although the metabolites of some drugs remain in the urine for longer periods of time and may enable the FRA to estimate whether the employee was impaired by those drugs at the time of a covered accident, incident, or rule violation, 49 Fed.Reg. 24291 (1984), the delay necessary to procure a warrant nevertheless may result in the destruction of valuable evidence.
The Government's need to rely on private railroads to set the testing process in motion also indicates that insistence on a warrant requirement would impede the achievement of the Government's objective. Railroad supervisors, like school officials, see New Jersey v. T.L.O., supra, 469 U.S., at 339-340, 105 S.Ct., at 742, and hospital administrators, see O'Connor v. Ortega, 480 U.S., at 722, 107 S.Ct., at 1500; are not in the business of investigating violations of the criminal laws or enforcing administrative codes, and otherwise have little occasion to become familiar with the intricacies of this Court's Fourth Amendment jurisprudence. "Imposing unwieldy warrant procedures . . . upon supervisors, who would otherwise have no reason to be familiar with such procedures, is simply unreasonable." Ibid.
In sum, imposing a warrant requirement in the present context would add little to the assurances of certainty and regularity already afforded by the regulations, while significantly hindering, and in many cases frustrating, the objectives of the Government's testing program. We do not believe that a warrant is essential to render the intrusions here at issue reasonable under the Fourth Amendment.
Our cases indicate that even a search that may be performed without a warrant must be based, as a general matter, on probable cause to believe that the person to be searched has violated the law. See New Jersey v. T.L.O., supra, 469 U.S., at 340, 105 S.Ct., at 742. When the balance of interests precludes insistence on a showing of probable cause, we have usually required "some quantum of individualized suspicion" before concluding that a search is reasonable. See, e.g., United States v. Martinez-Fuerte, 428 U.S., at 560, 96 S.Ct., at 3084. We made it clear, however, that a showing of individualized suspicion is not a constitutional floor, below which a search must be presumed unreasonable. Id., at 561, 96 S.Ct., at 3084. In limited circumstances, where the privacy interests implicated by the search are minimal, and where an important governmental interest furthered by the intrusion would be placed in jeopardy by a requirement of individualized suspicion, a search may be reasonable despite the absence of such suspicion. We believe this is true of the intrusions in question here.
By and large, intrusions on privacy under the FRA regulations are limited. To the extent transportation and like restrictions are necessary to procure the requisite blood, breath, and urine samples for testing, this interference alone is minimal given the employment context in which it takes place. Ordinarily, an employee consents to significant restrictions in his freedom o movement where necessary for his employment, and few are free to come and go as they please during working hours. See, e.g., INS v. Delgado, 466 U.S., at 218, 104 S.Ct., at 1763. Any additional interference with a railroad employee's freedom of movement that occurs in the time it takes to procure a blood, breath, or urine sample for testing cannot, by itself, be said to infringe significant privacy interests.
Our decision in Schmerber v. California, supra, indicates that the same is true of the blood tests required by the FRA regulations. In that case, we held that a State could direct that a blood sample be withdrawn from a motorist suspected of driving while intoxicated, despite his refusal to consent to the intrusion. We noted that the test was performed in a reasonable manner, as the motorist's "blood was taken by a physician in a hospital environment according to accepted medical practices." Id., at 771, 86 S.Ct., at 1836. We said also that the intrusion occasioned by a blood test is not significant, since such "tests are a commonplace in these days of periodic physical examinations and experience with them teaches that the quantity of blood extracted is minimal, and that for most people the procedure involves virtually no risk, trauma, or pain." Ibid. Schmerber thus confirmed "society's judgment that blood tests do not constitute an unduly extensive imposition on an individual's privacy and bodily integrity." Winston v. Lee, 470 U.S., at 762, 105 S.Ct., at 1617. See also South Dakota v. Neville, 459 U.S. 553, 563, 103 S.Ct. 916, 922, 74 L.Ed.2d 748 (1983) ("The simple blood-alcohol test is . . . safe, painless, and commonplace"); Breithaupt v. Abram, 352 U.S. 432, 436, 77 S.Ct. 408, 410, 1 L.Ed.2d 448 (1957) ("The blood test procedure has become routine in our everyday life").
The breath tests authorized by Subpart D of the regulations are even less intrusive than the blood tests prescribed by Subpart C. Unlike blood tests, breath tests do not require piercing the skin and may be conducted safely outside a hospital environment and with a minimum of inconvenience or embarrassment. Further, breath tests reveal the level of alcohol in the employee's bloodstream and nothing more. Like the blood-testing procedures mandated by Subpart C, which can be used only to ascertain the presence of alcohol or controlled substances in the bloodstream, breath tests reveal no other facts in which the employee has a substantial privacy interest. Cf. United States v. Jacobsen, 466 U.S., at 123, 104 S.Ct., at 1661-1662; United States v. Place, 462 U.S., at 707, 103 S.Ct., at 2644-2645. In all the circumstances, we cannot conclude that the administration of a breath test implicates significant privacy concerns.
A more difficult question is presented by urine tests. Like breath tests, urine tests are not invasive of the body and, under the regulations, may not be used as an occasion for inquiring into private facts unrelated to alcohol or drug use.7 We recognize, however, that the procedures for collecting the necessary samples, which require employees to perform an excretory function traditionally shielded by great privacy, raise concerns not implicated by blood or breath tests. While we would not characterize these additional privacy concerns as minimal in most contexts, we note that the regulations endeavor to reduce the intrusiveness of the collection process. The regulations do not require that samples be furnished under the direct observation of a monitor, despite the desirability of such a procedure to ensure the integrity of the sample. See 50 Fed.Reg. 31555 (1985). See also Field Manual B-15, D-1. The sample is also collected in a medical environment, by personnel unrelated to the railroad employer, and is thus not unlike similar procedures encountered often in the context of a regular physical examination.
More importantly, the expectations of privacy of covered employees are diminished by reason of their participation in an industry that is regulated pervasively to ensure safety, a goal dependent, in substantial part, on the health and fitness of covered employees. This relation between safety and employee fitness was recognized by Congress when it enacted the Hours of Service Act in 1907, Baltimore & Ohio R. Co. v. ICC, 221 U.S., at 619, 31 S.Ct., at 625, and also when it authorized the Secretary to "test . . . railroad facilities, equipment, rolling stock, operations, or persons, as he deems necessary to carry out the provisions" of the Federal Railroad Safety Act of 1970. 45 U.S.C. § 437(a) (emphasis added). It has also been recognized by state governments,8 and has long been reflected in industry practice, as evidenced by the industry's promulgation and enforcement of Rule G. Indeed, the FRA found, and the Court of Appeals acknowledged, see 839 F.2d, at 585, that "most railroads require periodic physical examinations for train and engine employees and certain other employees." 49 Fed.Reg. 24278 (1984). See also Railway Labor Executives Assn. v. Norfolk & Western R. Co., 833 F.2d 700, 705-706 (CA7 1987); Brotherhood of Maintenance of Way Employees, Lodge 16 v. Burlington Northern R. Co., 802 F.2d 1016, 1024 (CA8 1986).
We do not suggest, of course, that the interest in bodily security enjoyed by those employed in a regulated industry must always be considered minimal. Here, however, the covered employees have long been a principal focus of regulatory concern. As the dissenting judge below noted: "The reason is obvious. An idle locomotive, sitting in the roundhouse, is harmless. It becomes lethal when operated negligently by persons who are under the influence of alcohol or drugs." 839 F.2d, at 593. 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. We conclude, therefore, that the testin procedures contemplated by Subparts C and D pose only limited threats to the justifiable expectations of privacy of covered employees.
By contrast, the Government 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, see, e.g., Rushton v. Nebraska Public Power Dist., 844 F.2d 562, 566 (CA8 1988); Alverado v. Washington Public Power Supply System, 111 Wash.2d 424, 436, 759 P.2d 427, 433-434 (1988), cert. pending, No. 88-645, 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. An impaired employee, the FRA found, will seldom display any outward "signs detectable by the lay person or, in many cases, even the physician." 50 Fed.Reg. 31526 (1985). This view finds ample support in the railroad industry's experience with Rule G, and in the judgment of the courts that have examined analogous testing schemes. See, e.g., Brotherhood of Maintenance Way Employees, Lodge 16 v. Burlington Northern R. Co., supra, at 1020. Indeed, while respondents posit that impaired employees might be detected without alcohol or drug testing,9 the premise of respondents' lawsuit is that even the occurrence of a major calamity will not give rise to a suspicion of impairment with respect to any particular employee.
While no procedure can identify all impaired employees with ease and perfect accuracy, the FRA regulations supply an effective means of deterring employees engaged in safety-sensitive tasks from using controlled substances or alcohol in the first place. 50 Fed.Reg. 31541 (1985). The railroad industry's experience with Rule G persuasively shows, and common sense confirms, that the customary dismissal sanction that threatens employees who use drugs or alcohol while on duty cannot serve as an effective deterrent unless violators know that they are likely to be discovered. By ensuring that employee in safety-sensitive positions know they will be tested upon the occurrence of a triggering event, the timing of which no employee can predict with certainty, the regulations significantly increase the deterrent effect of the administrative penalties associated with the prohibited conduct, cf. Griffin v. Wisconsin, 483 U.S., at 876, 107 S.Ct., at 3170, concomitantly increasing the likelihood that employees will forgo using drugs or alcohol while subject to being called for duty.
The testing procedures contemplated by Subpart C also help railroads obtain invaluable information about the causes of major accidents, see 50 Fed.Reg. 31541 (1985), and to take appropriate measures to safeguard the general public. Cf. Michigan v. Tyler, 436 U.S. 499, 510, 98 S.Ct. 1942, 1950, 56 L.Ed.2d 486 (1978) (noting that prompt investigation of the causes of a fire may uncover continuing dangers and thereby prevent the fire's recurrence); Michigan v. Clifford, 464 U.S. 287, 308, 104 S.Ct. 641, 654, 78 L.Ed.2d 477 (1984) (REHNQUIST, J., dissenting) (same). Positive test results would point toward drug or alcohol impairment on the part of members of the crew as a possible cause of an accident, and may help to establish whether a particular accident, otherwise not drug related, was made worse by the inability of impaired employees to respond appropriately. Negative test results would likewise furnish invaluable clues, for eliminating drug impairment as a potential cause or contributing factor would help establish the significance of equipment failure, inadequate training, or other potential causes, and suggest a more thorough examination of these alternatives. Tests performed following the rule violations specified in Subpart D likewise can provide valuable information respecting the causes of those transgressions, which the FRA found to involve "the potential for a serious train accident or grave personal injury, or both." 50 Fed.Reg. 31553 (1985).
A requirement of particularized suspicion of drug or alcohol use would seriously impede an employer's ability to obtain this information, despite its obvious importance. Experience confirms the FRA's judgment that the scene of a serious rail accident is chaotic. Investigators who arrive at the scene shortly after a major accident has occurred may find it difficult to determine which members of a train crew contributed to its occurrence. Obtaining evidence that might give rise to the suspicion that a particular employee is impaired, a difficult endeavor in the best of circumstances, is most impracticable in the aftermath of a serious accident. While events following the rule violations that activate the testing authority of Subpart D may be less chaotic, objective indicia of impairment are absent in these instances as well. Indeed, any attempt to gather evidence relating to the possible impairment of particular employees likely would result in the loss or deterioration of the evidence furnished by the tests. Cf. Michigan v. Clifford, supra, at 293, n. 4, 104 S.Ct., at 647, n. 4 (plurality opinion); Michigan v. Tyler, supra, 436 U.S., at 510, 98 S.Ct., at 1950. It would be unrealistic, and inimical to the Government's goal of ensuring safety in rail transportation, to require a showing of individualized suspicion in these circumstances.
Without quarreling with the importance of these governmental interests, the Court of Appeals concluded that the postaccident testing regulations were unreasonable because "[b]lood and urine tests intended to establish drug use other than alcohol . . . cannot measure current drug intoxication or degree of impairment." 839 F.2d, at 588. The court based its conclusion on its reading of certain academic journals that indicate that the testing of urine can disclose only drug metabolites, which "may remain in the body for days or weeks after the ingestion of the drug." Id., at 589. We find this analysis flawed for several reasons.
As we emphasized in New Jersey v. T.L O., "it is universally recognized that evidence, to be relevant to an inquiry, need not conclusively prove the ultimate fact in issue, but only have 'any tendency to make the existence of any fact that is of consequence to the determination [of the point in issue] more probable or less probable than it would be without the evidence.' " 469 U.S., at 345, 105 S.Ct., at 746, quoting Fed.Rule Evid. 401. Even if urine test results disclosed nothing more specific than the recent use of controlled substances by a covered employee, this information would provide the basis for further investigative work designed to determine whether the employee used drugs at the relevant times. See Field Manual B-4. The record makes clear, for example, that a positive test result, coupled with known information concerning the pattern of elimination for the particular drug and information that may be gathered from other sources about the employee's activities, may allow the FRA to reach an informed judgment as to how a particular accident occurred. See supra, at 609-610.
More importantly, the Court of Appeals overlooked the FRA's policy of placing principal reliance on the results of blood tests, which unquestionably can identify very recent drug use, see, e.g., 49 Fed.Reg. 24291 (1984), while relying on urine tests as a secondary source of information designed to guard against the possibility that certain drugs will be eliminated from the bloodstream before a blood sample can be obtained. The court also failed to recognize that the FRA regulations are designed not only to discern impairment but also to deter it. Because the record indicates that blood and urine tests, taken together, are highly effective means of ascertaining on-the-job impairment and of deterring the use of drugs by railroad employees, we believe the Court of Appeals erred in concluding that the postaccident testing regulations are not reasonably related to the Government objectives that support them.10
We conclude that the compelling Government interests served by the FRA's regulations would be significantly hindered if railroads were required to point to specific facts giving rise to a reasonable suspicion of impairment before testing a given employee. In view of our conclusion that, on the present record, the toxicological testing contemplated by the regulations is not an undue infringement on the justifiable expectations of privacy of covered employees, the Government's compelling interests outweigh privacy concerns.
The possession of unlawful drugs is a criminal offense that the Government may punish, but it is a separate and far more dangerous wrong to perform certain sensitive tasks while under the influence of those substances. Performing those tasks while impaired by alcohol is, of course, equally dangerous, though consumption of alcohol is legal in most other contexts. The Government may take all necessary and reasonable regulatory steps to prevent or deter that hazardous conduct, and since the gravamen of the evil is performing certain functions while concealing the substance in the body, it may be necessary, as in the case before us, to examine the body or its fluids to accomplish the regulatory purpose. The necessity to perform that regulator function with respect to railroad employees engaged in safety-sensitive tasks, and the reasonableness of the system for doing so, have been established in this case.
Alcohol and drug tests conducted in reliance on the authority of Subpart D cannot be viewed as private action outside the reach of the Fourth Amendment. Because the testing procedures mandated or authorized by Subparts C and D effect searches of the person, they must meet the Fourth Amendment's reasonableness requirement. In light of the limited discretion exercised by the railroad employers under the regulations, the surpassing safety interests served by toxicological tests in this context, and the diminished expectation of privacy that attaches to information pertaining to the fitness of covered employees, we believe that it is reasonable to conduct such tests in the absence of a warrant or reasonable suspicion that any particular employee may be impaired. We hold that the alcohol and drug tests contemplated by Subparts C and D of the FRA's regulations are reasonable within the meaning of the Fourth Amendment. The judgment of the Court of Appeals is accordingly reversed.
It is so ordered.
Justice STEVENS, concurring in part and concurring in the judgment.
In my opinion the public interest in determining the causes of serious railroad accidents adequately supports the validity of the challenged regulations. I am not persuaded, however, that the interest in deterring the use of alcohol or drugs is either necessary or sufficient to justify the searches authorized by these regulations.
I think it a dubious proposition that the regulations significantly deter the use of alcohol and drugs by hours of service employees. Most people—and I would think most railroad employees as well—do not go to work with the expectation that they may be involved in a major accident, particularly one causing such catastrophic results as loss of life or the release of hazardous material requiring an evacuation. Moreover, even if they are conscious of the possibilities that such an accident might occur and that alcohol or drug use might be a contributing factor, if the risk of serious personal injury does not deter their use of these substances, it seems highly unlikely that the additional threat of loss of employment would have any effect on their behavior.
For this reason, I do not join the portions of Part III of the Court's opinion that rely on a deterrence rationale; I do, however, join the balance of the opinion and the Court's judgment.
Justice MARSHALL, with whom Justice BRENNAN joins, dissenting.
The issue in this case is not whether declaring a war on illegal drugs is good public policy. The importance of ridding our society of such drugs is, by now, apparent to all. Rather, the issue here is whether the Government's deployment in that war of a particularly Draconian weapon—the compulsory collection and chemical testing of railroad workers' blood and urine—comports with the Fourth Amendment. Precisely because the need for action against the drug scourge is manifest, the need for vigilance against unconstitutional excess is great. History teaches that grave threats to liberty often come in times of urgency, when constitutional rights seem too extravagant to endure. The World War II relocation-camp cases, Hirabayashi v. United States, 320 U.S. 81, 63 S.Ct. 1375, 87 L.Ed. 1774 (1943); Korematsu v. United States, 323 U.S. 214, 65 S.Ct. 193, 89 L.Ed. 194 (1944), and the Red scare and McCarthy-era internal subversion cases, Schenck v. United States, 249 U.S. 47, 39 S.Ct. 247, 63 L.Ed. 470 (1919); Dennis v. United States, 341 U.S. 494, 71 S.Ct. 857, 95 L.Ed. 1137 (1951), are only the most extreme reminders that when we allow fundamental freedoms to be sacrificed in the name of real or perceived exigency, we invariably come to regret it.
In permitting the Government to force entire railroad crews to submit to invasive blood and urine tests, even when it lacks any vidence of drug or alcohol use or other wrongdoing, the majority today joins those shortsighted courts which have allowed basic constitutional rights to fall prey to momentary emergencies. The majority holds that the need of the Federal Railroad Administration (FRA) to deter and diagnose train accidents outweighs any "minimal" intrusions on personal dignity and privacy posed by mass toxicological testing of persons who have given no indication whatsoever of impairment. Ante, at 624. In reaching this result, the majority ignores the text and doctrinal history of the Fourth Amendment, which require that highly intrusive searches of this type be based on probable cause, not on the evanescent cost-benefit calculations of agencies or judges. But the majority errs even under its own utilitarian standards, trivializing the raw intrusiveness of, and overlooking serious conceptual and operational flaws in, the FRA's testing program. These flaws cast grave doubts on whether that program, though born of good intentions, will do more than ineffectually symbolize the Government's opposition to drug use.
The majority purports to limit its decision to postaccident testing of workers in "safety-sensitive" jobs, ante, at 620, much as it limits its holding in the companion case to the testing of transferees to jobs involving drug interdiction or the use of firearms. Treasury Employees v. Von Raab, 489 U.S. 656, 664, 109 S.Ct. 1384, 1390, 103 L.Ed.2d 685 (1989). But the damage done to the Fourth Amendment is not so easily cabined. The majority's acceptance of dragnet blood and urine testing ensures that the first, and worst, casualty of the war on drugs will be the precious liberties of our citizens. I therefore dissent.
* The Court today takes its longest step yet toward reading the probable-cause requirement out of the Fourth Amendment. For the fourth time in as many years, a majority holds that a " 'special nee[d], beyond the normal need for law enforcement,' " makes the " 'requirement' " of probable cause " 'impracticable.' " Ante, at 619 (citations omitted). With the recognition of "[t]he Government's interest in regulating the conduct of railroad employees to ensure safety" as such a need, ante, at 620, the Court has now permitted "special needs" to displace constitutional text in each of the four categories of searches enumerated in the Fourth Amendment: searches of "persons," ante, at 613-614; "houses," Griffin v. Wisconsin, 483 U.S. 868, 107 S.Ct. 3164, 97 L.Ed.2d 709 (1987); "papers," O'Connor v. Or- tega, 480 U.S. 709, 107 S.Ct. 1492, 94 L.Ed.2d 714 (1987); and "effects," New Jersey v. T.L.O., 469 U.S. 325, 105 S.Ct. 733, 83 L.Ed.2d 720 (1985).
The process by which a constitutional "requirement" can be dispensed with as "impracticable" is an elusive one to me. The Fourth Amendment provides that "[t]he right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated; and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized." The majority's recitation of the Amendment, remarkably, leaves off after the word "violated," ante, at 613, but the remainder of the Amendment—the Warrant Clause—is not so easily excised. As this Court has long recognized, the Framers intended the provisions of that Clause—a warrant and probable cause—to "provide the yardstick against which official searches and seizures are to be measured." T.L.O., supra, at 359-360, 105 S.Ct., at 753 (opinion of BRENNAN, J.). Without the content which those provisions give to the Fourth Amendment's overarching command that searches and seizures be "reasonable," the Amendment lies virtually devoid of meaning, subject to whatever content shifting judicial majorities, concerned about the problems of the day, choose o give to that supple term. See Dunaway v. New York, 442 U.S. 200, 213, 99 S.Ct. 2248, 2257, 60 L.Ed.2d 824 (1979) ("[T]he protections intended by the Framers could all too easily disappear in the consideration and balancing of the multifarious circumstances presented by different cases"). Constitutional requirements like probable cause are not fair-weather friends, present when advantageous, conveniently absent when "special needs" make them seem not.
Until recently, an unbroken line of cases had recognized probable cause as an indispensable prerequisite for a full-scale search, regardless of whether such a search was conducted pursuant to a warrant or under one of the recognized exceptions to the warrant requirement. T.L.O., supra, 469 U.S., at 358 and 359, n. 3, 105 S.Ct., at 752 and 753, n. 3 (opinion of BRENNAN, J.); see also Chambers v. Maroney, 399 U.S. 42, 51, 90 S.Ct. 1975, 1981, 26 L.Ed.2d 419 (1970). Only where the government action in question had a "substantially less intrusive" impact on privacy, Dunaway, 442 U.S., at 210, 99 S.Ct., at 2255, and thus clearly fell short of a full-scale search, did we relax the probable-cause standard. Id., at 214, 99 S.Ct., at 2257 ("For all but those narrowly defined intrusions, the requisite 'balancing' . . . is embodied in the principle that seizures are 'reasonable' only if supported by probable cause"); see also T.L.O., supra, 469 U.S., at 360, 105 S.Ct., at 753 (opinion of BRENNAN, J.). Even in this class of cases, we almost always required the government to show some individualized suspicion to justify the search.1 The few searches which we upheld in the absence of individualized justification were routinized, fleeting, and nonintrusive encounters conducted pursuant to regulatory programs which entailed no contact with the person.2
In the four years since this Court, in T.L.O., first began recognizing "special needs" exceptions to the Fourth Amendment, the clarity of Fourth Amendment doctrine has been badly distorted, as the Court has eclipsed the probable-cause req irement in a patchwork quilt of settings: public school principals' searches of students' belongings, T.L.O.; public employers' searches of employees' desks, O'Connor; and probation officers' searches of probationers' homes, Griffin.3 Tellingly, each time the Court has found that "special needs" counseled ignoring the literal requirements of the Fourth Amendment for such full-scale searches in favor of a formless and unguided "reasonableness" balancing inquiry, it has concluded that the search in question satisfied that test. I have joined dissenting opinions in each of these cases, protesting the "jettison[ing of] . . . the only standard that finds support in the text of the Fourth Amendment" and predicting that the majority's "Rohrschach-like 'balancing test' " portended "a dangerous weakening of the purpose of the Fourth Amendment to protect the privacy and security of our citizens." T.L.O., supra, 469 U.S., at 357-358, 105 S.Ct., at 752 (opinion of BRENNAN, J.).
The majority's decision today bears out that prophecy. After determining that the Fourth Amendment applies to the FRA's testing regime, the majority embarks on an extended inquiry into whether that regime is "reasonable," an inquiry in which it balances " 'all the circumstances surrounding the search or seizure and the nature of the search or seizure itself.' " Ante, at 619, quoting United States v. Montoya de Hernandez, 473 U.S. 531, 537, 105 S.Ct. 3304, 3308-3309, 87 L.Ed.2d 381 (1985). The result is "special needs" balancing analysis' deepest incursion yet into the core protections of the Fourth Amendment. Until today, it was conceivable that, when a government search was aimed at a person and not simply the person's possessions, balancing analysis had no place. No longer: with nary a word of explanation or acknowledgment of the novelty of its approach, the majority extends the "special needs" framework to a regulation involving compulsory blood withdrawal and urinary excretion, and chemical testing of the bodily fluids collected through these procedures. And until today, it was conceivable that a prerequisite for surviving "special needs" analysis was the existence of individualized suspicion. No longer: in contrast to the searches in T.L.O., O'Connor, and Griffin, which were supported by individualized evidence suggesting the culpability of the persons whose property was searched,4 the regulatory regime upheld today requires the postaccident collection and testing of the blood and urine of all covered employees—even if every member of this group gives every indication of sobriety and attentiveness.
In widening the "special needs" exception to probable cause to authorize searches of the human body unsupported by any evidence of wrongdoing, the majority today completes the process begun in T.L.O. of eliminating altogether the probable-cause requirement for c vil searches—those undertaken for reasons "beyond the normal need for law enforcement." Ante, at 619 (citations omitted). In its place, the majority substitutes a manipulable balancing inquiry under which, upon the mere assertion of a "special need," even the deepest dignitary and privacy interests become vulnerable to governmental incursion. See ibid. (distinguishing criminal from civil searches). By its terms, however, the Fourth Amendment—unlike the Fifth and Sixth—does not confine its protections to either criminal or civil actions. Instead, it protects generally "[t]he right of the people to be secure."5
The fact is that the malleable "special needs" balancing approach can be justified only on the basis of the policy results it allows the majority to reach. The majority's concern with the railroad safety problems caused by drug and alcohol abuse is laudable; its cavalier disregard for the text of the Constitution is not. There is no drug exception to the Constitution, any more than there is a communism exception or an exception for other real or imagined sources of domestic unrest. Coolidge v. New Hampshire, 403 U.S. 443, 455, 91 S.Ct. 2022, 2032, 29 L.Ed.2d 564 (1971). Because abandoning the explicit protections of the Fourth Amendment seriously imperils "the right to be let alone—the most comprehensive of rights and the right most valued by civilized men," Olmstead v. United States, 277 U.S. 438, 478, 48 S.Ct. 564, 578, 72 L.Ed. 944 (1928) (Brandeis, J., dissenting), I reject the majority's "special needs" rationale as unprincipled and dangerous.
The proper way to evaluate the FRA's testing regime is to use the same analytic framework which we have traditionally used to appraise Fourth Amendment claims involving full-scale searches, at least until the recent "special needs" cases. Under that framework, we inquire, serially, whether a search has taken place, see, e.g., Katz v. United States, 389 U.S. 347, 350-353, 88 S.Ct. 507, 510-512, 19 L.Ed.2d 576 (1967); whether the search was based on a valid warrant or undertaken pursuant to a recognized exception to the warrant requirement, see, e.g., Welsh v. Wisconsin, 466 U.S. 740, 748-750, 104 S.Ct. 2091, 2096-2098, 80 L.Ed.2d 732 (1984); whether the search was based on probable cause or validly based on lesser suspicion because it was minimally intrusive, see, e.g., Dunaway, 442 U.S., at 208-210, 99 S.Ct., at 2254-2255 and, finally, whether the search was conducted in a reasonable manner, see, e.g., Winston v. Lee, 470 U.S. 753, 763-766, 105 S.Ct. 1611, 1618-1619, 84 L.Ed.2d 662 (1985). See also T.L.O., 469 U.S., at 354-355, 105 S.Ct., at 749-750 (opinion of BRENNAN, J.) (summarizing analytic framework).
The majority's threshold determination that "covered" railroad employees have been searched under the FRA's testing program is certainly correct. Ante, at 616-618. Who among us is not prepared to consider reasonable a person's expectation of privacy with respect to the extraction of his blood, the collection of his urine, or the chemical testing of these fluids? United States v. Jacobsen, 466 U.S. 109, 113, 104 S.Ct. 1652, 1656, 80 L.Ed.2d 85 (1984).6 The majority's ensuing conclusion that the warrant requirement may be dispensed with, however, conveniently overlooks the fact that there are three distinct searches at issue. Although the importance of collecting blood and urine samples before drug or alcohol metabolites disappear justifies waiving the warrant requirement for those two searches under the narrow "exigent circumstances" exception, see Schmerber v. California, 384 U.S. 757, 770, 86 S.Ct. 1826, 1835, 16 L.Ed.2d 908 (1966) ("[T]he delay necessary to obtain a warrant . . . threaten[s] 'the destruction of evidence' "), no such exigency prevents railroad officials from securing a warrant before chemically testing the samples they obtain. Blood and urine do not spoil if properly collected and preserved, and there is no reason to doubt the ability of railroad officials to grasp the relatively simple procedure of obtaining a warrant authorizing, where appropriate, chemical analysis of the extracted fluids. It is therefore wholly unjustified to dispense with the warrant requirement for this final search. See Chimel v. California, 395 U.S. 752, 761-764, 89 S.Ct. 2034, 2039-2041, 23 L.Ed.2d 685 (1969) (exigency exception permits warrantless searches only to the extent that exigency exists).
It is the probable-cause requirement, however, that the FRA's testing regime most egregiously violates, a fact which explains the majority's ready acceptance and expansion of the countertextual "special needs" exception. By any measure, the FRA's highly intrusive collection and testing procedures qualify as full-scale personal searches. Under our precedents, a showing of probable cause is therefore clearly required. But even if these searches were viewed as entailing only minimal intrusions on the order, say, of a police stop-and-frisk, the FRA's program would still fail to pass constitutional muster, for we have, without exception, demanded that even minimally intrusive searches of the person be founded on individualized suspicion. See supra, at 638, and n. 1. The federal parties concede it does not satisfy this standard. Brief for Federal Parties 18. Only if one construes the FRA's collection and testing procedures as akin to the routinized and fleeting regulatory interactions which we have permitted in the absence of individualized suspicion, see n. 2, supra, might these procedures survive constitutional scrutiny. Presumably for this reason, the majority likens this case to United States v. Martinez-Fuerte, 428 U.S. 543, 96 S.Ct. 3074, 49 L.Ed.2d 1116 (1976), which upheld brief automobile stops at the border to ascertain the validity of motorists' residence in the United States. Ante, at 624. Case law and common sense reveal both the bankruptcy of this absurd analogy and the constitutional imperative of adhering to the textual standard of probable cause to evaluate the FRA's multifarious full-scale searches.
Compelling a person to submit to the piercing of his skin by a hypodermic needle so that his blood may be extracted significantly intrudes on the "personal privacy and dignity against unwarranted intrusion by the State" against which the Fourth Amendment protects. Schmerber, supra, 384 U.S., at 767, 86 S.Ct., at 2042. As we emphasized in Terry v. Ohio, 392 U.S. 1, 24-25, 88 S.Ct. 1868, 1881-1882, 20 L.Ed.2d 889 (1968), "Even a limited search of the outer clothing . . . constitutes a severe, though brief, intrusion upon cherished personal security, and it must surely be an annoying, frightening, and perhaps humiliating experience." 392 U.S., at 24-25, 88 S.Ct., at 1881-1882. We have similarly described the taking of a suspect's fingernail scrapings as a " 'severe, though brief, intrusion upon cherished personal security.' " Cupp v. Murphy, 412 U.S. 291, 295, 93 S.Ct. 2000, 2003, 36 L.Ed.2d 900 (1973) (quoting Terry, supra, 392 U.S., at 24-25, 88 S.Ct., at 1882, a d upholding this procedure upon a showing of probable cause). The government-compelled withdrawal of blood, involving as it does the added aspect of physical invasion, is surely no less an intrusion. The surrender of blood on demand is, furthermore, hardly a quotidian occurrence. Cf. Martinez-Fuerte, supra, 428 U.S., at 557, 96 S.Ct. at 3082 (routine stops involve "quite limited" intrusion).
In recognition of the intrusiveness of this procedure, we specifically required in Schmerber that police have evidence of a drunken-driving suspect's impairment before forcing him to endure a blood test:
"The interests in human dignity and privacy which the Fourth Amendment protects forbid any such intrusions on the mere chance that desired evidence might be obtained. In the absence of a clear indication that in fact such evidence will be found, these fundamental human interests require law officers to suffer the risk that such evidence may disappear. . . ." 384 U.S., at 769-770, 86 S.Ct., at 1835.
Schmerber strongly suggested that the "clear indication" needed to justify a compulsory blood test amounted to a showing of probable cause, which "plainly" existed in that case. Id., at 768, 86 S.Ct., at 1834. Although subsequent cases interpreting Schmerber have differed over whether a showing of individualized suspicion would have sufficed, compare Winston, 470 U.S., at 760, 105 S.Ct. at 1616 (Schmerber "noted the importance of probable cause"), with Montoya de Hernandez, 473 U.S., at 540, 105 S.Ct., at 3310 (Schmerber "indicate[d] the necessity for particularized suspicion"), by any reading, Schmerber clearly forbade compulsory blood tests on any lesser showing than individualized suspicion. Exactly why a blood test which, if conducted on one person, requires a showing of at least individualized suspicion may, if conducted on many persons, be based on no showing whatsoever, the majority does not—and cannot explain.7
Compelling a person to produce a urine sample on demand also intrudes deeply on privacy and bodily integrity. Urination is among the most private of activities. It is generall forbidden in public, eschewed as a matter of conversation, and performed in places designed to preserve this tradition of personal seclusion. Cf. Martinez-Fuerte, 428 U.S., at 560, 96 S.Ct., at 3084 (border-stop questioning involves no more than "some annoyance" and is neither "frightening" nor "offensive"). The FRA, however, gives scant regard to personal privacy, for its Field Manual instructs supervisors monitoring urination that railroad workers must provide urine samples "under direct observation by the physician/technician." Federal Railroad Administration, United States Dept. of Transportation, Field Manual: Control of Alcohol and Drug Use in Railroad Operations D-5 (1986) (emphasis added).8 That the privacy interests offended by compulsory and supervised urine collection are profound is the overwhelming judgment of the lower courts and commentators. As Professor—later Solicitor General—Charles Fried has written:
"[I]n our culture the excretory functions are shielded by more or less absolute privacy, so much so that situations in which this privacy is violated are experienced as extremely distressing, as detracting from one's dignity and self esteem." Privacy, 77 Yale L.J. 475, 487 (1968).9
The majority's characterization of the privacy interests implicated by urine collection as "minimal," ante, at 624, is nothing short of startling. This characterization is, furthermore, belied by the majority's own prior explanation of why compulsory urination constitutes a search for the purposes of the Fourth Amendment:
" 'There are few activities in our society more personal or private than the passing of urine. Most people describe it by euphemisms if they talk about it at all. It is a function traditionally performed without public observation; indeed, its performance in public is generally prohibited by law as well as social custom.' " Ante, at 617, quoting National Treasury Employees Union v. Von Raab, 816 F.2d 170, 175 (CA5 1987).
The fact that the majority can invoke this powerful passage in the context of deciding that a search has occurred, and then ignore it in deciding that the privacy interests this search implicates are "minimal," underscores the shameless manipulability of its balancing approach.
Finally, the chemical analysis the FRA performs upon the blood and urine samples implicates strong privacy interests apart from those intruded upon by the collection of bodily fluids. Technological advances have made it possible to uncover, through analysis of chemical compounds in these fluids, not only drug or alcohol use, but also medical disorders such as epilepsy, diabetes, and clinical depression. Cf. Martinez-Fuerte, supra, at 558, 96 S.Ct., at 3083, quoting United States v. Brignoni-Ponce, 422 U.S. 873, 880, 95 S.Ct 2574, 2580, 45 L.Ed.2d 607 (1975) (check-point inquiry involves only " 'a brief question or two' " about motorist's residence). As the Court of Appeals for the District of Columbia Circuit has observed: "[S]uch tests may provide Government officials with a periscope through which they can peer into an individual's behavior in her private life, even in her own home." Jones v. McKenzie, 266 U.S.App.D.C. 85, 89, 833 F.2d 335, 339 (1987); see also Capua v. Plainfield, 643 F.Supp. 1507, 1511 (NJ 1986) (urine testing is "form of surveillance" which "reports on a person's off-duty activities just as surely as someone had been present and watching"). The FRA's requirement that workers disclose the medications they have taken during the 30 days prior to chemical testing further impinges upon the confidentiality customarily attending personal health secrets.
By any reading of our precedents, the intrusiveness of these three searches demands that they—like other full-scale searches—be justified by probable cause. It is no answer to suggest, as does the majority, that railroad workers have relinquished the protection afforded them by this Fourth Amendment requirement, either by "participat[ing] in an industry that is regulated pervasively to ensure safety" or by undergoing periodic fitness tests pursuant to state law or to collective-bargaining agreements. Ante, at 627.
Our decisions in the regulatory search area refute the suggestion that the heavy regulation of the railroad industry eclipses workers' rights under the Fourth Amendment to insist upon a showing of probable cause when their bodily fluids are being extracted. This line of cases has exclusively involved searches of employer property, with respect to which "[c]ertain industries have such a history of government oversight that no reasonable expectation of privacy could exist for a proprietor over the stock of such an enterprise." Marshall v. Barlow's, Inc., 436 U.S. 307, 313, 98 S.Ct. 1816, 1821, 56 L.Ed.2d 305 (1978) (emphasis added; citation omitted), quoted in New York v. Burger, 482 U.S. 691, 700, 107 S.Ct. 2636, 2642, 96 L.Ed.2d 601 (1987). Never have we intimated that regulatory searches reduce employees' rights of privacy in their persons. See Camara v. Municipal Court of San Francisco, 387 U.S. 523, 537, 87 S.Ct. 1727, 1735, 18 L.Ed.2d 930 (1967) ("[T]he inspections are [not] personal in nature"); cf. Donovan v. Dewey, 452 U.S. 594, 598-599, 101 S.Ct. 2534, 2538, 69 L.Ed.2d 262 (1981); Marshall, supra, 436 U.S. at 313, 98 S.Ct. at 1821. As the Court pointed out in O'Connor, individuals do not lose Fourth Amendment rights at the workplace gate, 480 U.S., at 716-718, 107 S.Ct., at 1497-1498; see also Oliver v. United States, 466 U.S. 170, 178, n. 8, 104 S.Ct. 1735, 1741, n. 8, 80 L.Ed.2d 214 (1984), any more than they relinquish these rights at the schoolhouse door, T.L.O., 469 U.S., at 333, 105 S.Ct., at 739, or the hotel room threshold, Hoffa v. United States, 385 U.S. 293, 301, 87 S.Ct. 408, 413, 17 L.Ed.2d 374 (1966). These rights mean little indeed if, having passed through these portals, an individual may remain subject to a suspicionless search of his person justified solely on the grounds that the government already is permitted to conduct a search of the inanimate contents of the surrounding area. In holding that searches of persons may fall within the category of regulatory searches permitted in the absence of probable cause or even individualized suspicion, the majority sets a dangerous and ill-conceived precedent.
The majority's suggestion that railroad workers' privacy is only minimally invaded by the collection and testing of their bodily fluids because they undergo periodic fitness tests, ante, at 624-625, is equally baseless. As an initial matter, even if participation in these fitness tests did render "minimal" an employee's "interest in bodily security," ante, at 628, such minimally intrusive searches of the person require, under our precedents, a justificatory showing of individualized suspicion. See supra, at 637. More fundamentally, railroad employees are not routinely required to submit to blood or urine tests to gain or to maintain employment, and railroad employers do not ordinarily have access to employees' blood or urine, and certainly not for the purpose of ascertaining drug or alcohol usage. That railroad employees sometimes undergo tests of eyesight, hearing, skill, intelligence, and agility, ante, at 627, n. 8, hardly prepares them for Government demands to submit to the extraction of blood, to excrete under supervision, or to have these bodily fluids tested for the physiological and psychological secrets they may contain. Surely employees who release basic information about their financial and personal history so that employers may ascertain their "ethical fitness" do not, by so doing, relinquish their expectations of privacy with respect to their personal letters and diaries, revealing though these papers may be of their character.
I recognize that invalidating the full-scale searches involved in the FRA's testing regime for failure to comport with the Fourth Amendment's command of probable cause may hinder the Government's attempts to make rail transit as safe as humanly possible. But constitutional rights have their consequences, and one is that efforts to maximize the public welfare, no matter how well intentioned, must always be pursued within constitutional boundaries. Were the police freed from the constraints of the Fourth Amendment for just one day to seek out evidence of criminal wrongdoing, the resulting convictions and incarcerations would probably prevent thousands of fatalities. Our refusal to tolerate this specter reflects our shared belief that even beneficent governmental power—whether exercised to save money, save lives, or make the trains run on time—must always yield to "a resolute loyalty to constitutional safeguards." Almeida-Sanchez v. United States, 413 U.S. 266, 273, 93 S.Ct. 2535, 2539-2540, 37 L.Ed.2d 596 (1973). The Constitution demands no less loyalty here.
Even accepting the majority's view that the FRA's collection and testing program is appropriately analyzed under a multifactor balancing test, and not under the literal terms of the Fourth Amendment, I would still find the program invalid. The benefits of suspicionless blood and urine testing are far outstripped by the costs imposed on personal liberty by such sweeping searches. Only by erroneously deriding as "minimal" the privacy and dignity interests at stake, and by uncritically inflating the likely efficacy of the FRA's testing program, does the majority strike a different balance.
For the reasons stated above, I find nothing minimal about the intrusion on individual liberty that occurs whenever the Government forcibly draws and analyzes a person's blood and urine. Several aspects of the FRA's testing program exacerbate the intrusiveness of these procedures. Most strikingly, the agency's regulations not only do not forbid, but, in fact, appear to invite criminal prosecutors to obtain the blood and urine samples drawn by the FRA and use them as the basis of criminal investigations and trials. See 49 CFR § 219.211(d) (1987) ("Each sample . . . may be made available to . . . a party in litigation upon service of appropriate compulsory process on the custodian of the sample . . ."). This is an unprecedented invitation, leaving open the possibility of criminal prosecutions based on suspicionless searches of the human body. Cf. Treasury Employees, 489 U.S., at 666, 109 S.Ct., at 1390 (Customs Service drug-testing program prohibits use of test results in criminal prosecutions); Camara, 387 U.S., at 537, 87 S.Ct., at 1735.
To be sure, the majority acknowledges, in passing, the possibility of criminal prosecutions, ante, at 621, n. 5, but it refuses to factor this possibility into its Fourth Amendment balancing process, stating that "the record does not disclose that [49 CFR § 219.211(d) (1987) ] was intended to be, or actually has been, so used." Ibid. This demurrer is highly disingenuous. The federal parties concede that they find "no prohibition on the release of FRA testing results to prosecutors." Brief for Federal Parties 10, n. 15. The absence of prosecutions to date—which is likely due to the fact that the FRA's regulations have been held invalid for much of their brief history—hardly proves that prosecutors will not avail themselves of the FRA's invitation in the future. If the majority really views the impact of FRA testing on privacy interests as minimal even if these tests generate criminal prosecutions, it should say so. If the prospect of prosecutions would lead the majority to reassess the validity of the testing program with prosecutions as part of the balance, it should say so, too, or condition its approval of that program on the nonrelease of test results to prosecutors. In ducking this important issue, the majority gravely disserves both the values served by the Fourth Amendment and the rights of those persons whom the FRA searches. Furthermore, the majority's refusal to restrict the release of test results casts considerable doubt on the conceptual basis of its decision—that the "special need" of railway safety is one "beyond the normal need for law enforcement." Ante, at 619 (citations omitted).10
The majority also overlooks needlessly intrusive aspects of the testing process itself. Although the FRA requires the collection and testing of both blood and urine, the agency concedes that mandatory urine tests—unlike blood tests—do not measure current impairment and therefore cannot differentiate on-duty impairment from prior drug or alcohol use which has ceased to affect the user's behavior. See 49 CFR § 219.309(2) (1987) (urine test may reveal use of drugs or alcohol as much as 60 days prior to sampling). Given that the FRA's stated goal is to ascertain current impairment, and not to identify persons who have used substances in their spare time sufficiently in advance of their railroad duties to pose no risk of on-duty impairment, § 219.101(a), mandatory urine testing seems wholly excessive. At the very least, the FRA could limit its use of urinalysis to confirming findings of current impairment suggested by a person's blood tests. The additional invasion caused by automatically testing urine as well as blood hardly ensures that privacy interests "will be invaded no more than is necessary." T.L.O., 469 U.S., at 343, 105 S.Ct., at 744.
The majority's trivialization of the intrusions on worker privacy posed by the FRA's testing program is matched at the other extreme by its blind acceptance of the Government's assertion that testing will "dete[r] employees engaged in safety-sensitive tasks from using controlled substances or alcohol," and "help railroads obtain invaluable information about the causes of major accidents." Ante, at 629-630. With respect, first, to deterrence, it is simply implausible that testing employees after major accidents occur, 49 CFR § 219.201(a)(1) (1987), will appreciably discourage them from using drugs or alcohol. As Justice STEVENS observes in his oncurring opinion:
"Most people—and I would think most railroad employees as well—do not go to work with the expectation that they may be involved in a major accident, particularly one causing such catastrophic results as loss of life or the release of hazardous material requiring an evacuation. Moreover, even if they are conscious of the possibilities that such an accident might occur and that alcohol or drug use might be a contributing factor, if the risk of serious personal injury does not deter their use of these substances, it seems highly unlikely that the additional threat of loss of employment would have any effect on their behavior." Ante, at 634.
Under the majority's deterrence rationale, people who skip school or work to spend a sunny day at the zoo will not taunt the lions because their truancy or absenteeism might be discovered in the event they are mauled. It is, of course, the fear of the accident, not the fear of a postaccident revelation, that deters. The majority's credulous acceptance of the FRA's deterrence rationale is made all the more suspect by the agency's failure to introduce, in an otherwise ample administrative record, any studies explaining or supporting its theory of accident deterrence.
The poverty of the majority's deterrence rationale leaves the Government's interest in diagnosing the causes of major accidents as the sole remaining justification for the FRA's testing program. I do not denigrate this interest, but it seems a slender thread from which to hang such an intrusive program, particularly given that the knowledge that one or more workers were impaired at the time of an accident falls far short of proving that substance abuse caused or exacerbated that accident. See 839 F.2d 575, 587 (CA9 1988). Some corroborative evidence is needed: witness or co-worker accounts of a worker's misfeasance, or at least indications that the cause of the accident was within a worker's area of responsibility. Such particularized facts are, of course, the very essence of the individualized suspicion requirement which the respondent railroad workers urge, and which the Court of Appeals found to "pos[e] no insuperable burden on the government." Id., at 588. Furthermore, reliance on the importance of diagnosing the causes of an accident as a critical basis for upholding the FRA's testing plan is especially hard to square with our frequent admonition that the interest in ascertaining the causes of a criminal episode does not justify departure from the Fourth Amendment's requirements. "[T]his Court has never sustained a search upon the sole ground that officers reasonably expected to find evidence of a particular crime. . . ." Katz, 389 U.S., at 356, 88 S.Ct., at 514. Nor should it here.
In his first dissenting opinion as a Member of this Court, Oliver Wendell Holmes observed:
"Great cases, like hard cases, make bad law. For great cases are called great, not by reason of their real importance in shaping the law of the future, but because of some accident of immediate overwhelming interest which appeals to the feelings and distorts the judgment. These immediate interests exercise a kind of hydraulic pressure which makes what previously was clear seem doubtful, and before which even well settled principles of law will bend." Northern Securities Co. v. United States, 193 U.S. 197, 400-401, 24 S.Ct. 436, 468, 48 L.Ed. 679 (1904).
A majority of this Court, swept away by society's obsession with stopping the scourge of illegal drugs, today succumbs to the popular pressures described by Justice Holmes. In upholding the FRA's plan for blood and urine testing, the majority bends time-honored and textually based principles of the Fourth Amendment—principles the Framers of the Bill of Rights designed to ensure that the Government has a strong and individualized justification when it seeks to invade an individual's privacy. I believe the Framers would be appalled by the vision of mass government l intrusions upon the integrity of the human body that the majority allows to become reality. The immediate victims of the majority's constitutional timorousness will be those railroad workers whose bodily fluids the Government may now forcibly collect and analyze. But ultimately, today's decision will reduce the privacy all citizens may enjoy, for, as Justice Holmes understood, principles of law, once bent, do not snap back easily. I dissent.
The FRA noted that a 1979 study examining the scope of alcohol abuse on seven major railroads found that "[a]n estimated one out of every eight railroad workers drank at least once while on duty during the study year." 48 Fed.Reg. 30724 (1983). In addition, "5% of workers reported to work 'very drunk' or got 'very drunk' on duty at least once in the study year," and "13% of workers reported to work at least 'a little drunk' one or more times during that period." Ibid. The study also found that 23% of the operating personnel were "problem drinkers," but that only 4% of these employees "were receiving help through an employee assistance program, and even fewer were handled through disciplinary procedures." Ibid.
The regulations provide a limited exception from testing "if the railroad representative can immediately determine, on the basis of specific information, that the employee had no role in the cause(s) of the accident/incident." 49 CFR § 219.203(a)(3)(i) (1987). No exception may be made, however, in the case of a "major train accident." Ibid. In promulgating the regulations, the FRA noted that, while it is sometimes possible to exonerate crew members in other situations calling for testing, it is especially difficult to assess fault and degrees of fault in the aftermath of the more substantial accidents. See 50 Fed.Reg. 31544 (1985).
See Federal Railroad Administration, United States Dept. of Transportation Field Manual: Control of Alcohol and Drug Use in Railroad Operations B-12 (1986) (Field Manual). Ethyl alcohol is measured by gas chromatography. Ibid. In addition, while drug screens may be conducted by immunoassays or other techniques, "[p]ositive drug findings are confirmed by gas chromatography/mass spectrometry." Ibid. These tests, if properly conducted, identify the presence of alcohol and drugs in the biological samples tested with great accuracy.
See, e.g., Lovvorn v. Chattanooga, 846 F.2d 1539, 1542 (CA6 1988); Copeland v. Philadelphia Police Dept., 840 F.2d 1139, 1143 (CA3 1988), cert. pending No. 88-66; Railway Labor Executives' v. Burnley, 839 F.2d 575, 580 (CA9 1988) (case below); Everett v. Napper, 833 F.2d 1507, 1511 (CA11 1987); Jones v. McKenzie, 266 U.S.App.D.C. 85, 88, 833 F.2d 335, 338 (CADC 1987); National Treasury Employees Union v. Von Raab, 816 F.2d 170, 176 (CA5 1987), aff'd in pertinent part, 489 U.S. 656, 109 S.Ct. 1384, 103 L.Ed.2d 685; McDonell v. Hunter, 809 F.2d 1302, 1307 (CA8 1987); Division 241 Amalgamated Transit Union v. Suscy, 538 F.2d 1264, 1266-1267 (CA 7), cert. denied, 429 U.S. 1029, 97 S.Ct. 653, 50 L.Ed.2d 632 (1976). See also Alverado v. Washington Public Power Supply System, 111 Wash.2d 424, 434, 759 P.2d 427, 432-433 (1988), cert. pending, No. 88-645.
Taking a blood or urine sample might also be characterized as a Fourth Amendment seizure, since it may be viewed as a meaningful interference with the employee's possessory interest in his bodily fluids. Cf. United States v. Jacobsen, 466 U.S. 109, 113, 104 S.Ct. 1652, 1656, 80 L.Ed.2d 85 (1984). It is not necessary to our analysis in this case, however, to characterize the taking of bl od or urine samples as a seizure of those bodily fluids, for the privacy expectations protected by this characterization are adequately taken into account by our conclusion that such intrusions are searches.
The regulations provide that "[e]ach sample provided under [Subpart C] is retained for not less than six months following the date of the accident or incident and may be made available to . . . a party in litigation upon service of appropriate compulsory process on the custodian. . . ." 49 CFR § 219.211(d) (1987). The FRA explained, when it promulgated this provision, that it intends to retain such samples primarily "for its own purposes (e.g., to permit reanalysis of a sample if another laboratory reported detection of a substance not tested for in the original procedure)." 50 Fed.Reg. 31545 (1985). While this provision might be read broadly to authorize the release of biological samples to law enforcement authorities, the record does not disclose that it was intended to be, or actually has been, so used. Indeed, while respondents aver generally that test results might be made available to law enforcement authorities, Brief for Respondents 24, they do not seriously contend that this provision, or any other part of the administrative scheme, was designed as "a 'pretext' to enable law enforcement authorities to gather evidence of penal law violations." New York v. Burger, 482 U.S. 691, 716-717, n. 27, 107 S.Ct. 2636, 2651, n. 27, 96 L.Ed.2d 601 (1987). Absent a persuasive showing that the FRA's testing program is pretextual, we assess the FRA's scheme in light of its obvious administrative purpose. We leave for another day the question whether routine use in criminal prosecutions of evidence obtained pursuant to the administrative scheme would give rise to an inference of pretext, or otherwise impugn the administrative nature of the FRA's program.
Subpart C of the regulations, for example, does not permit the exercise of any discretion in choosing the employees who must submit to testing, except in limited circumstances and then only if warranted by objective criteria. See n. 2, supra. Subpart D, while conferring some discretion to choose those who may be required to submit to testing, also imposes specific constraints on the exercise of that discretion. Covered employees may be required to submit to breath or urine tests only if they have been directly involved in specified rule violations or errors, or if their acts or omissions contributed to the occurrence or severity of specified accidents or incidents. To be sure, some discretion necessarily must be used in determining whether an employee's acts or omissions contributed to the occurrence or severity of an event, but this limited assessment of the objective circumstances surrounding the event does not devolve unbridled discretion upon the supervisor in the field. Cf. Marshall v. Barlow's, Inc., 436 U.S. 307, 323, 98 S.Ct. 1816, 1825-1826, 56 L.Ed.2d 305 (1978).
In addition, the regulations contain various safeguards against any possibility that discretion will be abused. A railroad that requires post-accident testing in bad faith, 49 CFR § 219.201(c) (1987), or that willfully imposes a program of authorized testing that does not comply with Subpart D, § 219.9(a)(3), or that otherwise fails to follow the regulations, § 219.9(a)(5), is subject to civil penalties, see pt. 219, App. A, p. 105, in addition to whatever damages may be awarded through the arbitration process.
When employees produce the blood and urine samples required by § bpart C, they are asked by medical personnel to complete a form stating whether they have taken any medications during the preceding 30 days. The completed forms are shipped with the samples to the FRA's laboratory. See Field Manual B-15. This information is used to ascertain whether a positive test result can be explained by the employee's lawful use of medications. While this procedure permits the Government to learn certain private medical facts that an employee might prefer not to disclose, there is no indication that the Government does not treat this information as confidential, or that it uses the information for any other purpose. Under the circumstances, we do not view this procedure as a significant invasion of privacy. Cf. Whalen v. Roe, 429 U.S. 589, 602, 97 S.Ct. 869, 878, 51 L.Ed.2d 64 (1977).
See, e.g., Ala.Code § 37-2-85 (1977) (requiring that persons to be employed as dispatchers, engineers, conductors, brakemen, and switchmen be subjected to a "thorough examination" respecting, inter alia, their skill, sobriety, eyesight, and hearing); Mass.Gen.Laws §§ 160:178-160:181 (1979) (prescribing eyesight examination and experience requirements for railroad engineers and conductors); N.Y.R.R. Law § 63 (McKinney 1952) (requiring that all applicants for positions as motormen or gripmen "be subjected to a thorough examination . . . as to their habits, physical ability, and intelligence"). See also Nashville, C. & S.L.R. Co. v. Alabama, 128 U.S. 96, 98-99, 9 S.Ct. 28, 28- 29, 32 L.Ed. 352 (1888) (noting, in upholding a predecessor of Alabama's fitness-for-duty statute against a Commerce Clause challenge, that a State may lawfully require railway employees to undergo eye examinations in the interests of safety).
Respondents offer a list of "less drastic and equally effective means" of addressing the Government's concerns, including reliance on the private proscriptions already in force, and training supervisory personnel "to effectively detect employees who are impaired by drug or alcohol use without resort to such intrusive procedures as blood and urine tests." Brief for Respondents 40-43. We have repeatedly stated, however, that "[t]he reasonableness of any particular government activity does not necessarily or invariably turn on the existence of alternative 'less intrusive' means." Illinois v. Lafayette, 462 U.S. 640, 647, 103 S.Ct. 2605, 2610, 77 L.Ed.2d 65 (1983). See also Colorado v. Bertine, 479 U.S. 367, 373-374, 107 S.Ct. 738, 742, 93 L.Ed.2d 739 (1987). It is obvious that "[t]he logic of such elaborate less-restrictive-alternative arguments could raise insuperable barriers to the exercise of virtually all search-and-seizure powers," United States v. Martinez-Fuerte, 428 U.S., at 556-557, n. 12, 96 S.Ct., at 3082, n. 12, because judges engaged in post hoc evaluations of government conduct " 'can almost always imagine some alternative means by which the objectives of the [government] might have been accomplished.' " United States v. Montoya de Hernandez, 473 U.S. 531, 542, 105 S.Ct. 3304, 3311, 87 L.Ed.2d 381 (1985), quoting United States v. Sharpe, 470 U.S. 675, 686-687, 105 S.Ct. 1568, 1575-1576, 84 L.Ed.2d 605 (1985). Here, the FRA expressly considered various alternatives to its drug-screening program and reasonably found them wanting. At bottom, respondents' insistence on less drastic alternatives would require us to second-guess the reasonable conclusions drawn by the FRA after years of investigation and study. This we decline to do.
The Court of Appeals also expressed concern that the tests might be quite unreliable, and thus unreasonable. 839 F.2d, at 589. The record compiled by the FRA after years of investigation and study does not support this conclusion. While it is impossible to guarantee that no mistakes will ever be made in isolated cases, respondents have challenged the administrative scheme on its face. We deal therefore with whether the tests contemplated by the regulations can ever be conducted. Cf. Bell v. Wolfish, 441 U.S. 520, 560, 99 S.Ct. 1861, 1885, 60 L.Ed.2d 447 (1979). Respondents have provided us with no reason for doubting the FRA's conclusion that the tests at issue here are accurate in the overwhelming majority of cases.
The first, and leading, case of a minimally intrusive search held valid when based on suspicion short of probable cause is Terry v. Ohio, 392 U.S. 1, 30, 88 S.Ct. 1868, 1884, 20 L.Ed.2d 889 (1968), where we held that a police officer who observes unusual conduct suggesting criminal activity by persons he reasonably suspects are armed and presently dangerous may "conduct a carefully limited search of the outer clothing of such persons." See also United States v. Hensley, 469 U.S. 221, 105 S.Ct. 675, 83 L.Ed.2d 604 (1985) (upholding brief stop of person described on wanted flyer while police ascertain if arrest warrant has been issued); Delaware v. Prouse, 440 U.S. 648, 99 S.Ct. 1391, 59 L.Ed.2d 660 (1979) (invalidating discretionary stops of motorists to check licenses and registrations when not based on reasonable suspicion that the motorist is unlicensed, the automobile is unregistered, or that the vehicle or an occupant should otherwise be detained); Pennsylvania v. Mimms, 434 U.S. 106, 98 S.Ct. 330, 54 L.Ed.2d 331 (1977) (upholding limited search where officers who had lawfully stopped car saw a large bulge under the driver's jacket); United States v. Brignoni-Ponce, 422 U.S. 873, 95 S.Ct. 2574, 45 L.Ed.2d 607 (1975) (upholding brief stops by roving border patrols where officers reasonably believe car may contain illegal aliens); Adams v. Williams, 407 U.S. 143, 92 S.Ct. 1921, 32 L.Ed.2d 612 (1972) (upholding brief stop to interrogate suspicious individual believed to be carrying narcotics and gun).
See, e.g., United States v. Martinez-Fuerte, 428 U.S. 543, 96 S.Ct. 3074, 49 L.Ed.2d 1116 (1976) (brief interrogative stop at permanent border checkpoint to ascertain motorist's residence status); Camara v. Municipal Court of San Francisco, 387 U.S. 523, 87 S.Ct. 1727, 18 L.Ed.2d 930 (1967) (routine annual inspection by city housing department).
The "special needs" the Court invoked to justify abrogating the probable-cause requirement were, in New Jersey v. T.L.O., 469 U.S., at 341, 105 S.Ct., at 743-744, "the substantial need of teachers and administrators for freedom to maintain order in the schools"; in O'Connor v. Ortega, 480 U.S., at 725, 107 S.Ct., at 1502, "the efficient and proper operation of the workplace"; and in Griffin v. Wisconsin, 483 U.S., at 878, 107 S.Ct., at 3171, the need to preserve "the deterrent effect of the supervisory arrangement" of probation.
See T.L.O., supra, 469 U.S., at 346, 105 S.Ct., at 746 (teacher's report that student had been smoking provided reasonable suspicion that purse contained cigarettes); O'Connor, supra, 480 U.S., at 726, 107 S.Ct., at 1503 (charges of specific financial improprieties gave employer individualized suspicion of misconduct by employee); Griffin, supra, 483 U.S., at 879-880, 107 S.Ct., at 3171-3172 (tip to police officer that probationer was storing guns in his apartment provided reasonable suspicion).
That the Fourth Amendment applies equally to criminal and civil searches was emphasized, ironically enough, in the portion of T.L.O. holding the Fourth Amendment applicable to schoolhouse searches. 469 U.S., at 335, 105 S.Ct., at 739. The malleability of "special needs" balancing thus could not be clearer: the majority endorses the applicability of the Fourth Amendment to civil searches in determining whether a search has taken place, but then wholly ignores it in the subsequent inquiry into the validity of that search.
The FRA's breath-testing procedures also constitute searches subject to constitutional safeguards. See ante, at 616-617 (reaching same conclusion). I focus my discussion on the collection and testing of blood and urine because those more intrusive procedures better demonstrate the excesses of the FRA's scheme.
The majority, seeking to lessen the devastating ramifications of Schmerber v. California, and to back up its assertion that Government-imposed blood extraction does not "infringe significant privacy interests," ante, at 625, emphasizes Schmerber § observation that blood tests are commonplace and can be performed with " 'virtually no risk, trauma, or pain,' " quoting 384 U.S., at 771, 86 S.Ct., at 1836. The majority, however, wrenches this statement out of context. The Schmerber Court made this statement only after it established that the blood test fell within the "exigent circumstances" exception to the warrant requirement, and that the test was supported by probable cause. Indeed, the statement was made only in the context of the separate inquiry into whether the compulsory blood test was conducted in a reasonable manner. 384 U.S., at 768-772, 86 S.Ct., at 1834-1836; see also Winston v. Lee, 470 U.S. 753, 760-761, 105 S.Ct. 1611, 1612-1613, 84 L.Ed.2d 662 (1985) ("Schmerber recognized that the ordinary requirements of the Fourth Amendment would be the threshold requirements for conducting this kind of surgical search and seizure. . . . Beyond these standards, Schmerber's inquiry considered a number of other factors in determining the 'reasonableness' of the blood test") (emphasis added). The majority also cites South Dakota v. Neville, 459 U.S. 553, 103 S.Ct. 916, 74 L.Ed.2d 748 (1983), and Breithaupt v. Abram, 352 U.S. 432, 77 S.Ct. 408, 1 L.Ed.2d 448 (1957), for the proposition that blood tests are commonplace. Ante, at 625. In both those cases, however, the police officers who attempted to impose blood tests on drunken-driving suspects had exceptionally strong evidence of the driver's inebriation. 459 U.S., at 554-556, 103 S.Ct., at 917-918; 352 U.S., at 433, 77 S.Ct., at 409.
The majority dismisses as nonexistent the intrusiveness of such "direct observation," on the ground that FRA regulations state that such observation is not "require[d]." 50 Fed.Reg. 31555 (1985), cited ante, at 626. The majority's dismissal is too hasty, however, for the regulations—in the very same sentence—go on to state: "[B]ut observation is the most effective means of ensuring that the sample is that of the employee and has not been diluted." 50 Fed.Reg. 31555 (1985). Even if this were not the case, the majority's suggestion that officials monitoring urination will disregard the clear commands of the Field Manual with which they are provided is dubious, to say the least.
See, e.g., National Treasury Employees Union v. Von Raab, 816 F.2d 170, 175 (CA5 1987), aff'd in pertinent part, 489 U.S. 656, 109 S.Ct. 1384, 103 L.Ed.2d 685; Taylor v. O'Grady, 669 F.Supp. 1422, 1433-1434 (ND Ill.1987); Feliciano v. Cleveland, 661 F.Supp. 578, 586 (ND Ohio 1987); American Federation of Government Employees, AFL-CIO v. Weinberger, 651 F.Supp. 726, 732-733 (SD Ga.1986); Capua v. Plainfield, 643 F.Supp. 1507, 1514 (NJ 1986).
As a result of the majority's extension of the regulatory search doctrine to searches of the person, individuals the FRA finds to have used drugs may face criminal prosecution, even if their impairment had nothing to do with causing an accident. The majority observes that evidence of criminal behavior unearthed during an otherwise valid regulatory search is not excludible unless the search is shown to be a "pretext" for obtaining evidence for a criminal trial, ante, at 621, n. 5, citing New York v. Burger, 482 U.S. 691, 716-717, n. 27, 107 S.Ct. 2636, 2651, n. 27, 96 L.Ed.2d 601 (1987)—a defense the majority belittles but, mercifully, preserves for another day.