Employee Responsibilities and Rights Journa~ VoL 7, No. 3, 1994 Drug Testing in the Transportation Industry: The Maritime Case M a r k D. Karper, 1 Clifford B. Donn, 1 and Marie E. Lyndaker 2 The maritime industry provides an interesting case study o f the design and implementation of drug-testing programs in the transportation industry during the 1980s. It is clear that such programs were designed and implemented less because of empirical evidence of safety problems than for political reasons. The results hz the maritime industry are indicative o f a program that is expensive to operate, intrusive with regard to employee privacy, and which will have little or no impact on safety. Focush~g such programs on the issue of impairment rather than on the issue o f drag usage per se is likely to improve the outcome of the programs. KEY WORDS: drug testing; maritime; substance abuse; transportation. INTRODUCTION During the 1980s, a program of mandatory random and "for cause" drug testing was imposed on all federal government employees (although legal challenges eventually led to the exclusion of employees in nonsensitive positions). Many private employers also developed drug-testing programs which either were implemented unilaterally or were negotiated with the unions that represented their employees) By the end of the decade, the National Institute on Drug Abuse estimated that approximately 40% of "Fortune 500" companies were using drug tests to screen job applicants, while the U.S. Department of Labor estimated that 20% of the labor force in the private sector was covered by employer drug-testing policies (Ackerman, 1991, 16; Walsh & Trumble, 1991, 26). Eventually the controversy surrounding this issue caused the General Counsel of the National Labor Relations Board to express the view that drug testing is a "mandatory" subject of bargaining, i.e., one that unionized employers would have to negotiate with their unions rather than 1Department of Industrial Relations and Human Resource Management, Le Moyne College, Syracuse, New York 13214-1399. 2Lewis County Opportunities. 3Ahhough, despite common belief, there is little evidence that drug usage or the "drug problem" in U.S. society as a whole had gotten worse in the preceding years (Schonsheck, 1989, 250-251). 219 0892-7545/94,~090ff0219507.00,'00 1994PlenumPublishingCorporation 220 Karper, Donn, and Lyndaker impose unilaterally, and one over which unions could lawfully strike, a view that the Board itself affirmed in 1989 (Collyer, 1987; Cerrito et al., 1990, 290). Well-publicized rail and truck accidents focused attention on the transportation industry. In October 1982, • . . during a discussion on alcohol and drug use in transportation held at the annual meeting of the National Safety Council, a representative of the Greyhound Corporation indicated that his company had begun regularly scheduled drug testing of bus drivers . . . . The announcement by Greyhound and subsequent reports of unacceptable rates of positive drug tests in its drivers and applicants brought considerable pressure on companies that provided service to the traveling public to implement testing programs. (Walsh & Trumble, 1991, 27-28) Then in January 1987, a train filled with passengers including many college students returning from holidays crashed in Maryland killing 16 passengers and injuring 174. Tests revealed that the engineer and brakeman had been smoking marijuana (Walsh & Trumble, 1991, 40). As a result, Secretary of Transportation James Burnley proposed that employees throughout the transportation industry be subject to testing (American Shipper, August 1988). Unlike other testing programs in private industry, this one was mandated by government, and thus neither the existence of the program itself nor its general outlines were subject to negotiation between transportation employers and unions, a position confirmed by the Supreme Court in 1989 (Cerrito et al., 1990, 290). The maritime industry is a particularly interesting setting to examine the drugtesting debate. Merchant seafarers live and work in isolated locales away from their families, which may increase the temptation to use controlled substances while at the same time complicating the procedures fo~"reliable testing. There has been only one published study of drug testing in the maritime industry (Wright, 1989), but that study is concerned purely with legal issues, in particular, with the constitutionality of the program the initial Coast Guard regulations seemed to require. It was written before the court decisions specific to the maritime drug-testing program were made and before the current version of the regulations was issued. This article discusses and analyzes the specific drug-testing program implemented by the Department of Transportation through the Coast Guard in the maritime industry. The governmental regulations will be explained along with the responses of maritime employers and unions to those regulations. The legal challenges to the program will be evaluated and the efficacy of maritime drug testing will be reviewed. Finally, we shall offer a critique and a proposal. BACKGROUND Over the years, there has emerged a voluminous literature on drug use, drug testing, and their implications for the workplace (Wiilette, 1991; American Management Association, 1987; Dwyer, 1989; Crant & Bateman, 1989). Indeed, reading from some of the path-breaking works in this area can reveal how little has changed Maritime Drug Testing 221 in terms of the need to deal with the "drug crisis" over the years. Thus in the early 1970s, Harrison Trice and Paul Roman wrote that, This publicity has led to a "felt need" in many work organizations to establish programs to combat drug abuse . . . . In some organizations programs to deal with the deviant drinker are unfortunately neglected or down-pedaled as a consequence while elsewhere drug abuse programs are being formulated without sufficient knowledge of the extent of the problem and in a mood bordering on hysteria and panic. (Trice & Roman, 1972, 40) As we shall see below, this is an extremely accurate description of how drug testing emerged in the maritime industry in the 1980s. There are several aspects of the maritime industry that make it unique as a setting for examination of substance abuse and drug-testing issues. The first and most obvious point is that the nature of maritime transportation involves a unique kind of isolation. Vessels at sea are isolated for days or even weeks at a time. Crew members live where they work, and thus any drugs that they use on a daily basis will have to be used at and located at the work place with them. There is no behavior analogous to going home after work and having a drink or smoking a marijuana cigarette. The ordinary separation of workplace and home does not exist during periods when seafarers are aboard ship. This is important because U.S.-flag vessels are subject to random and intensive searches by the U.S. Customs Service not only in U.S. ports but in foreign ports and even while at sea. When the workplace is subject to search, in effect the seafarer's home is subject to search as well, and privacy rights "at work" become more complex than is the case for most other types of workers (Journal of Commerce, September 20, 1988). 4 There are also severe limitations on the ability of the employer to replace impaired crew members once the vessel is at sea, and doing so can be extremely expensive. The implications of this limitation have become more serious in the 1980s as crew sizes have declined from the forties and fifties to twenty-one and even lower, implying very little redundancy in vessel crews (Donn, 1988, 35-38; 1989, 195-197; Committee on the Effect of Smaller Crews on Maritime Safety, 1990, 44). A second aspect of seafaring is that ship-board personnel on a voyage are isolated from medical laboratories and other shore-side facilities through which drug testing could be carried out. Thus, blood and urine samples need to be collected and stored aboard ship and certain testing procedures may actually need to be performed aboard ship if testing is to be timely and accurate. The economic competitiveness of the U.S.-flag maritime industry is also relevant in that a very high proportion of the maritime traffic in U.S. waters consists of vessels that are not under the U.S. flag and which are thus not subject to U.S. government control with regard to crewing and on-board conditions in the same way and to the same extent as are vessels of the U.S.-flag fleet. Thus, for a variety of reasons, only approximately 4% of U.S. international waterborne commerce is carried in U.S.-flag vessels (Donn, 1988, 39). This implies, to the extent that drug testing is safety related and to the extent that it applies only to seafarers in the 4Such searches are designed to prevent the importation of illegal goods or the untaxed importation of goods subject to tariff. 222 Karper, Donn, and Lyndaker U.S.-fiag fleet, that the safety-related efficacy of the program will be reduced since a high percentage of vessels in U.S. waters will remain unregulated. REGULATIONS AND INDUSTRY RESPONSE The proposed "final regulations" under which the maritime drug-testing program was to be carried out were issued by the U.S. Coast Guard on July 8, 1988 (Coast Guard, 1988; Williams, 1988; DiBenedetto, September 23, 1988a). The program that the Coast Guard regulations contemplated was far more extensive than that put in place in any of the other transportation sectors, requiring testing under five separate scenarios (Seafarers' Lo~ September 1988; DiBenedetto, November 21, 1988b; Abruzzese, October 13, 1989; Wright, 1989, 543-547). The regulations required preemployment testing, periodic testing (at each occasion where license or certificate renewal requires a medical examination), random testing, postaccident testing, and reasonable cause testing. Only the last requires that the parties ordering the testing have any specific grounds to believe the seafarer in question has used any controlled substances The substances to be tested for were amphetamines, cocaine, opiates, phencyclidine (PCP), and marijuana. Only in the case of postaccident testing was the use of alcohol to be subject to testing (46 C.F.R. at 16.105, Pilot, May 1989; Wright, 1989, 547). This summary of the regulations simplifies a highly complex and detailed set of rules governing each type of testing. For example, preemployment testing is required unless the employee has passed a preemployment or periodic drug test in the past six months or has been subject to random testing during the past 12 months. Random testing must involve 50% of an employer's work force over the course of a given year (originally the Coast Guard proposed an annual random testing rate of 125% of employees) (Seham, June 26, 1990). Postaccident testing is required after any "serious maritime incident," and the definition of this term is quite detailed as are the rules as to who must be tested after such an accident (Wright, 1989, 545-547). The cost of the testing program is to be borne by the employers. The consequences to seafarers of positive drug tests are potentially severe. The seafarer will have her/his license or certificate suspended or withdrawn until he/she demonstrates the successful completion of a rehabilitation program. However, there is no legal requirement that the employer offer the seafarer the opportunity for rehabilitation. (Seafarers' Log, September 1988) These consequences do not depend on the seafarer being shown to have used the controlled substance while on duty or even while on board ship, nor do they depend on the seafarer exhibiting any "impairment." 5For the time being, there is to be no analogous testing requirement applicable to seafarers on foreign-flag vessels operating in U.S. territorial waters (Maritime Newsletter, September 1988). It seems likely, given past court decisions and recent changes in international law increasing the scope for "port state control" (the right of the nation where a vessel is berthed to regulate safety and labor decisions) that Congress could impose such a requirement if it chose. Maritime Drug Testing 223 Maritime employers and unions responded to the Coast Guard's issuance of its "final" regulations with almost unanimous disapproval, clearly disappointed that so little of their input during the rule-making process had been heeded. 6 Indeed, the Journal of Commerce noted that drug testing provided almost the only issue in recent years that united virtually the entire maritime industry as well as many of the congressional representatives whose committees directly oversee the industry (DiBenedetto, September 23, 1988a). Almost every aspect of the program attracted criticism. The criticisms generally fell into several somewhat overlapping categories. First, it was noted that there was no hard evidence either indicating a significant substance abuse problem in the maritime industry or tying substance abuse to maritime mishaps (Journal of Commerce, September 20, 1988; DiBenedetto, September 23, 1988a). In this context, it was sometimes noted that the high average age of seafarers in the U.S.-flag fleet made drug abuse (at least of the five illegal drugs being tested for) less likely than in industries with younger work forces (Joltrhal of Commerce, September 20, 1988). Thus, it was often asserted or implied that the transportation industry had been unfairly singled out for political reasons and not because of evidence of a serious substance-abuse-related safety problem in the industry. In the words of the President of the United Shipowners of America, "There's no empirical data that drug usage affects our industry" (Journal of Commerce, September 20, 1988). Second, the program was criticized on grounds of cost. While the government estimated that the program would cost somewhere in excess of $60 million in the first year, maritime employer groups projected that costs would run perhaps three times that level (Pilot, May 1989). The related point was also frequently made that these costs would provide yet a further competitive disadvantage to U.S.-flag vessel operators already at a serious cost disadvantage vis-a-vis their foreign-flag competitors. In addition, there is the issue of the cost of repatriating and replacing crew members who fail drug tests given in distant parts of the globe (Vail, 1988). Some critics noted that the technology of the maritime industry created costs well above those of other transportation industries because testing equipment had to be carried aboard ship and sea-going personnel had to be trained in collection and testing procedures. In addition, the testing procedures themselves are highly detailed and complex and the failure of nonexpert shipboard personnel to follow them completely and accurately would invalidate the tests (Vail, 1988; Prat, 1989). Finally, questions were raised that reflected concerns common to all industries about drug testing. These included concerns about constitutionality and invasion of privacy as well as about the accuracy of the tests (DiBenedetto, November 21, 1988b). These questions were not completely unrelated to those of cost because some critics predicted that employers would open themselves to law suits through various aspects of the drug-testing program (Williams, 1988). Unions in the maritime industry, as elsewhere, have also argued strongly that employees who test posi6Both employers and unions had submitted substantial volumes of testimony and other material during the rule-making process. The employers had sought ways to make the program, if there was to be a program, as inexpensive as possible while the unions had objected to testing in the absence of individualized evidence of drug use on the job. 224 Karper, Donn, and Lyndaker tively for controlled substances or alcohol should be offered rehabilitation, not punishment (Marithne Newsletter, January 19890). Various parts of the drug-testing program were scheduled to go into effect during 1989, although there were several delays during that year as the Coast Guard tried to respond to industry objections and to lawsuits by adjusting various aspects of the administration of the program. Preemployment testing went into effect in July 1989, while periodic, accident, and probable cause testing were implemented in December of that year (Seafarers'Log, January 1990). As will be discussed below, random testing was further delayed by court order. Accordingly, the industry responded not only with criticism but also with action, putting into place the mechanisms that would permit compliance with the regulations. The discussion here will focus particularly on the program implemented by the American Maritime Association (AMA), the employer organization that generally represents operators who have contracts with the Seafarers International Union. Some other segments of the industry chose to join in this same program while others created their own, broadly similar, arrangements. On May 1, 1989, preemployment testing of members of the Seafarers International Union began in all ports. This operated through a mechanism called American Maritime Safety, Inc. (AMS). AMS is a nonprofit corporation which is a creation of the AMA companies and exists purely to administer the drug-testing program. Since its creation, some non-AMA companies, including even some nonunion companies, have joined AMS to administer their drug-testing programs. Urine samples are collected at Seafarers Welfare Fund clinics or at the offices of contracted physicians (Seafarers' Log, May 1989). Costs are paid by the Seafarers Welfare Fund, in effect, by the employers. Periodic testing was implemented directly as part of the medical examination requirement for renewal of licenses and certificates required of crew members on U.S.-flag vessels. As noted above, random testing was delayed by court order. As will be noted below, maritime labor responded by suing to prevent implementation of the Coast Guard regulations. However, the employers concentrated their attention on altering the regulations so as to reduce the costs of compliance. One of the employers' key objectives here focused on the preemployment testing part of the program. The maritime industry relies almost exclusively on union hiring halls to provide crews for private commercial vessels. Except at the level of the ship's master, few crew members have any permanent attachment to a given vessel or even to any given vessel operating company.7 Typically, seafarers complete a voyage, spend some time ashore, then report to their hiring hall and use their seniority and length of time ashore to claim their next billet (job). Thus, the employee in this industry has a "permanent" tie to the industry and to a particular hiring hall but not to any particular vessel or employer. 7Recent collective bargaining agreements between American President Lines and the Masters, Mates and Pilots Union have begun to change that arrangement for deck officers, giving them a regular attachment to particular vessels of that company. Maritime Drug Testing 225 The Cost Guard has interpreted every such change of vessel and vessel operator by a seafarer as requiring a new preemployment test unless a preemployment or periodic test has been performed in the previous six months or the employee has been subject to random testing during the last year. The A M A interprets this, in effect, as requiring that the typical employee be "preemployment" tested twice each year. Together with a random testing rate of 50%, this would result in 250% of the employees in the industry being tested on an annual basis (i.e., on average each employee would be tested two-and-a-half times per year), in addition to periodic, postaccident, and reasonable cause testing (Seham, June 26, 1990). The employers have argued that this is excessive and that such preemployment testing will account for 80% of total testing costs. The employers have also argued that this regime of preemployment testing, while extremely costly, is not efficacious because employees schedule it themselves. They argue that the early results do not justify these costs. AMS results for the period from April 1, 1989 to March 31, 1990 indicated positive test results--that is, evidence of usage of one of the five indicated drugs less than 2% of the time. Later AMS testing up to June of 1991 indicated even lower rates of positive tests. For the over 35,000 tests administered by AMS for the period from April 1989 through June 1991, fewer than 2% overall were positive, and among the licensed officers tested, fewer than one-half of one percent were positive, s However, although maritime employers have been vocal in their misgivings about drug testing, they have left it to the unions in the industry to take legal action. This may be for political reasons, but it also may be because the features of the Coast Guard regulations that most disturb the employers (those that produced the extremely high rate of preemployment testing) also seem relatively immune to legal challenge in that they raise no constitutional issue of privacy or unreasonable search and seizure. So while the unions challenged the regulations in court, the employers followed the administrative avenue of seeking a waiver of the sections of the regulations that they found most onerous. LEGAL ISSUES The Seafarers International Union (SIU) together with the Transportation Institute (a maritime research and promotion group with ties to the SIU) and two 8The AMS testing data referenced in this article are contained in the written version of the testimony given by Lee R. A. Seham, American Maritime Safety, Inc., before the U.S. House of Representatives, Subcommittee on Coast Guard and Navigation, June 26, 1990 and in other data provided to the authors by Mr. Seham. Exhibit D of the testimony includes an "updated statistical overview" of the preemployment tests provided by Greystone Health Sciences Corporation, the group that acts as "medical review officer" for the AMS testing program. The data in the testimony indicate the percentage of "positive" findings, the distribution of those findings across the controlled substances tested for, and the distribution of positive findings by age and by licensed versus unlicensed crew status for the period April 1, 1989 to March 31, 1990. The other material provided by Mr. Seham was also prepared by Greystone Health Sciences Corporation and covers tests administered between April 1989 and June 1991. Data will be provided by the authors upon request. 226 Karper, Donn, and Lyndaker individual seafarers joined to file suit against the Coast Guard drug-testing regulations (Transportation Institute et al. v. United States Coast Guard et aL, 1989, 883429). Marine Engineers Beneficial Association II (MEBA II) filed a similar suit (District #2 MEBA v. Burnley, 1989, 89-1519), and these two suits were joined by the federal district court for hearing in May 1989 (Seafarers' Log, January 1990). 9 Before the disposition of those suits is discussed, mention must be made of the leading legal precedents in the drug-testing area. The two key drug-testing decisions by the Supreme Court, both issued in 1989, have dealt with agents of the U.S. Customs Service (National Treasury Employees Union v. Von Raab, 1989, 109 S.Ct. 1384) and with railroad employees (Skinner v. Railway Labor Executives Association, 1989, 109 S.Ct. 1402). Both of these have been widely discussed, so only brief analysis is required here. 1° Von Raab and Skinner both dealt with a wide variety of issues. In Von Raab, employees seeking promotion into areas involving access to classified information, drug interdiction, or involving carrying firearms were required to undergo drug testing. This testing was of three types: prepromotion testing, reasonable cause testing, and random testing. By a vote of 5 to 4, the Supreme Court upheld this program. While the legality of the entire program was challenged, it was the components involving testing without individualized suspicion of drug use that two of the dissenters argued violated guarantees against unreasonable search and seizure as provided in the Fourth Amendment of the United States Constitution (in their view, searches must be grounded in individual suspicion to be reasonable). The other two dissenters argue that the government failed to prove any special need for any form of drug testing in the Customs Service because of the lack of evidence of a generalized problem of substance abuse in that agency (Wright, 1989, 551-553). The Skinner case is more relevant to the maritime experience because it also involves government mandated drug testing in the predominantly private transportation industry. The drug-testing program involved in the railroad industry was much less extensive than that proposed for the maritime industry, involving principally postaccident drug and alcohol testing and various types of reasonable cause drug testing (Skinner, at 1407). The Court quickly accepted the argument that constitutional guarantees are relevant here despite the private nature of employment in the industry since the drug-testing program was adopted pursuant to government regulation (Skinner, at 1412). However, the Court upheld the railroad drug-testing program 7 to 2. The two justices who dissented in Von Raab on the grounds of the Fourth Amendment to the constitution did so here as well because some of the forms of testing required still did not involve individualized suspicion of drug or alcohol use. The other two Von Raab dissenters joined the majority this time because they believed that the government had indeed demonstrated a special need for railroad drug testing (Skinner, at 1422). 9The SIU principally represents unlicensed seafarers (those who are not officers) in the deck, engine, and steward departments. MEBA II principally represents licensed officers in the engine department. l°See Wright for a discussion of the implications of these two cases for the maritime industry. Maritime Drug Testing 227 In these two cases, the principal reservations the Supreme Court expressed about drug-testing programs were whether a special need for those programs had been demonstrated in the particular industries/occupations subject to regulation, and in addition whether a search of an individual can be constitutionally justified without reasonable cause to suspect that individual of a violation. In applying the principles of these cases to the maritime industry, two points need to be made. First, neither involved a testing program as extensive as that proposed for the maritime industry. In particular, neither addressed the issue of random testing and whether such testing could meet the reasonable search and seizure standard. Second, in Skinner, the government presented substantial evidence of widespread abuse of controlled substances in the railroad industry, equivalent evidence not being produced in the maritime industry cases. The absence of equivalent evidence in Von Raab led to two additional dissenters and thus to a very close decision. The decision of the Federal District Court for the District of Columbia in the combined maritime suits (referred to here as Transportation Institute), issued December 18, 1989, represents an application of the principles set forth in both Von Raab and Skinner. However, it also represents an extension of those principles since it deals with a drug-testing program much more extensive than that examined in either of the Supreme Court cases. The District Court took as given that urinalysis represents a search and that drug tests pursuant to government regulation represent governmental action regulated by the constitution. It also took as given the Supreme Court majority view in Von Raab that such general regulations can be issued without evidence of a widespread drug problem. The heart of the decision is the District Court's consideration of the issue of how to define "reasonable search" given the Fourth Amendment's prohibition of unreasonable searches. The specific issue with respect to drug testing is whether, since it occurs without a judicial warrant, it falls within the "Special needs" category which traditionally allows government agents to conduct searches without judicial due process--that is, without a search warrant. "Special needs" normally means that there is a compelling public interest that outweighs the individual's right to privacy in a context in which securing a warrant prior to the gathering of evidence is impractical (Skinner, at 1414). Since the issuance of a warrant would often be impractical where drug use is suspected, and since four of the five types of testing do not involve individualized suspicion of drug use where a warrant would be relevant anyway, the court focused on the compelling public interest criterion. This is based on the theory that if there is a compelling public interest that justifies testing individuals without probable cause to suspect drug use, then drug testing for cause can also be justified without elaborate tests to determine what constitutes probable cause. As in Skinner, the court in Transportation b~stitute defined the compelling public interest in terms of whether the individuals to be tested would represent a threat to public health and safety should their job performance be impaired by drug usage. The key to this analysis is that the program is designed to prevent drug usage, not just to apprehend and punish the guilty. It is this purpose of prevention that seemed to allow the court to accept testing techniques that do not measure impairment and to permit warrantless testing. 228 Karper, Donn, and Lyndaker Accordingly, the court analyzed all five types of testing called for in the regulations. The court found a compelling public interest present in preemployment, periodic, post-incident, and probable causes testing because in every case it found a nexus with public safety protection. However, the District Court came to a different conclusion with regard to random testing. In this case, the Coast Guard regulations made the assumption that all members of the crew may be required to perform safety-related functions in case of emergency (e.g., handle hoses in case of fire) and therefore made all onboard personnel, regardless of job title, subject to random testing. The District Court took exception to this line of reasoning. Citing Von Raab, the court said that a compelling public interest justification for a warrantless search must involve a "clear direct nexus between the nature of the employee's duty and the nature of the feared violation" (Transportation Institute, 1989, at 658). The District Court went on to state that the government had established no direct connection between the duties of some crew members and safety in such a way as to justify generalized random testing. The District Court gave examples from the steward department (cook) and the engine department (wiper) of positions where mistakes were unlikely to cause a significant threat to public health and safety. Accordingly, the District Court found the random testing part of the regulations unconstitutional and enjoined the random testing program. The response of the Coast Guard to the decision in Transportation Institute was to spend the year of 1990 reformulating the regulations. While rumor in the industry was that the new regulations would still mandate random testing for all members of the crew but would try to draw a more direct nexus between public safety and the emergency duties of all of the crew members, this did not turn out to be the case. Instead, the notice of proposed rule making, issued in February 1991, and the final rules issued July 8, 1991, focused random testing on personnel engaged in the mechanical and steerage operations of the vessel, especially on deck and engineering officers. In addition, unlicensed deck personnel, unlicensed engine room personnel except for the relatively low-skilled wiper, and steward department personnel except for the cook must be subject to random testing, excluding less than 10% of the crew on most vessels. In effect, random testing is to apply to all crew members required by the Certificate of Inspection issued for each vessel by the Coast Guard and to any other employee who provides emergency services to passengers. 11 The February notice of proposed rule-making also waived preemployment and periodic testing if the seafarer shows that he/she passed a drug test in the past six months and was subject to random testing for at least 90 days during the previous six months (Individual Employee Rights, January 1989, 635; Journal of Commerce, 1991c, DiBenedetto, July 9, 1991a). t2 Given the logic of the TranspornEach vessel has a Coast Guard certificate that indicates the minimum crew complement for safe operation. Deck, engineering, and radio staff are included, but steward department staff is not. The typical ocean-going U.S.-flag vessel will have few staff or no more staff in the deck, engineering, or radio departments than the Coast Guard requires. lZThe requirements with regard to onboard testing of crews of U.S.-flag vessels while in the territorial waters of other nations also had their implementation delayed, from January 1992 to January 1993 (lournal of Commerce, April 30, 1991). Maritime Drug Testing 229 tation htstitute decision, it is hard to see how these changes could fail to pass muster with the District Court. Presuming the District Court upholds the new regulations, further appeals would then be possible but, given the decisions in Von Raab and Skinner and the change in the composition of the Supreme Court with both of the dissenters in the two cases no longer on the Court, Justice Brennan, being replaced by Justice Souter, and Justice Marshall being replaced by Justice Thomas, the likelihood of ultimately successful appeal by the unions would seem small. 13 As of this writing, the maritime unions seem to agree that further court challenges would be futile and seem disinclined to follow that route (DiBenedetto, July 17, 1991b). Given the District Court decision in Transportation hlstitute, we now have a much more complete understanding, in terms of industry generally, of the conditions under which drug-testing programs mandated by government will pass constitutional muster. In particular, random drug testing will only be permitted where there can be shown to be a clear nexus with public health and safety. However, in demonstrating that nexus, no actual drug problem in the given industry or occupation needs to be shown, only that a public health or safety concern would exist if drug use were to occur. Read together with Von Raab and Skinner, Transportation btstitute makes clear how extensive drug-testing programs can be without exceeding constitutional limitations. ANALYSIS OF THE PROBLEM It is difficult to assess the efficacy of the drug-testing program in the maritime industry because there is no solid basis for determining the extent, if any, of the drug problem in the industry. The AMS data for the year of preemployment testing to June 1991, referred to above, do not seem to support the view that a significant safety problem exists. When these data are further analyzed, the safety issue becomes even more problematical. Thus while the overall rate of "positive" test results was less than 2%, the positive rate among licensed personnel (officers) who have direct responsibility for propulsion and navigation was less than .4%. These data are consistent with low levels of positives found in the random testing program covering the Brotherhood of Locomotive Engineers and show lower levels than found in the transportation industry as a whole for the year 1990 (Journal of Commerce, 1991d; Wall Street Journal, 1991). Indeed, on the railroads in 1990 even "reasonable cause" testing produced positive test results less than 3% of the time (Brown, 1991). This rate is so small as to lead one commentator to speculate that reasonable cause drug tests in the railroad industry are being used by supervisors to victimize workers (Connors, 1991). 14 In the words of a joint statement adopted t3In addition, the 9th Circuit U.S. Appeals Court has since upheld the random testing program in the railroad industry, which had been challenged in a lawsuit by railway unions (Journalof Commerce,June 13, 1991). 14It should be noted that the 3% rate cited here is lower than it might be under a "reasonable cause" testing procedure because it includes test results from those who violate certain work rules and those involved in accidents too minor to be included in the postaccident testing program. That is, this set of "reasonable cause" drug tests includes test results for some employees for whom there was no real "reasonable cause," as commonly defined, to suspect drug use. 230 Karper, Donn, and L~'ndaker by four maritime unions with regard to the original Coast Guard regulations, the government "has proceeded in significant haste to propose a regulatory regime to correct and deter a problem that essentially has no factual b a s i s . . . " (DiBenedetto, 1988a). If the government were primarily interested in maritime safety (and probably transportation safety generally) the focus of testing would be much more heavily on alcohol and much less heavily on other drugs, an approach that was scheduled for implementation at the end of 1992 (Solomon, 1992). Those familiar with the maritime industry are aware of the concerns expressed by many participants and knowledgeable observers of the impact of alcohol use on safety aboard ship. There has been discussion of additional legislation which focuses more directly on alcohol use, but, as of this writing, there are still no requirements with regard to alcohol nearly as structured or as stringent as those outlined in the Coast Guard regulations governing drug testing (Abruzzese, 1989; Journal of Commerce, 1989). Is This is despite the fact that careful observers have argued that, "alcohol abuse is arguably the single greatest cause of traumatic injury in the U.S. transportation industry," and despite the fact that the use of alcohol is easy to detect and to measure with breath testing (Pollard et al., 1990, 9-10). Vessels must have alcohol test kits aboard and a blood alcohol content of 0.04% (a level that some criticize as too high for safety) is defined by the Coast Guard as intoxicated, but there is no detailed set of testing requirements (except postaccident as noted above) analogous to the testing for the five substances covered in preemployment, periodic, and random testing requirements. What can be concluded about the U.S. government's transportation drug-testing policy as it applies to the maritime industry? Clearly, preemployment and periodic testing are not particularly efficacious deterrents since they can be anticipated far in advance by seafarers. Nor is postincident testing effective since seafarers who are not deterred from drug use by the threat of an accident are unlikely to be deterred by the additional threat of testing after an accident. If the railroad experience is any guide, probable cause testing is likely to provide little more information than random testing and principally provides supervisors with a weapon to harass employees with whom they do not get along (Connors, 1991). This leaves random testing, the most intrusive type. In random testing, the employee's expectation of privacy (a significant factor in the Von Raab, Skinner, and Transportation Institute decisions) is greatest in that random testing occurs when employees have the least reason to expect themselves to be disturbed and when they have committed no act that would seem to justify such an intrusion. Despite the position taken by the Supreme Court majority in Von Raab and Skinner, we find it hard to disagree with the position of dissenters Brennan and Marshall in 15In late 1991 Congress was considering mandating alcohol testing for transportation employees, most of the pressure for passage coming in the wake of a New York City subway accident initially reported to be alcohol related. A number of concerns have been expressed by opponents including many of the same ones raised with regard to the drug-testing program. There is an additional set of issues with regard to privacy and constitutional rights in the case of alcohol testing since possession and consumption of alcoholic beverages are not illegal (Solomon, October 13, 1991, Solomon, October 15, 1991). Maritime Drug Testing 231 those cases that a significant drug problem is no reason to abandon the protections of the Fourth Amendment. This is especially true when there are no data to indicate that drug-related safety problem exists in the maritime industry and when there is no evidence that use of drugs or alcohol has gotten any worse on board vessels in the last 30 years (Committee on the Effect of Smaller Crews on Maritime Safety, 1990, 44). In addition, this discussion abstracts from concerns with regard to test accuracy and chain of custody, issues that are particularly difficult for the maritime industry because of the difficulties in securing and preserving samples in the absence of nearby laboratory facilities or trained medical personnel. CONCLUSIONS It is our position that the focus on drug use and drug testing is misplaced. As has been widely discussed, evidence of many controlled substances remains in the blood stream long after any conceivable impairment has ended. The key word here is "impairment." If the concern is not with deterring drug use or apprehending drug users but with maritime safety, then the focus should be on impairment. Indeed there are many factors that can make a normally competent employee unfit for work on a given day, and drug testing focuses on only a small number of such factors. Depression or illness can be equally debilitating. Maximizing safety suggests focusing less attention on the causes and more on levels of impairment. Technology is now being developed to test impairment regardless of cause. Software can be developed to provide a "precise, performance-related basis for action" (Hamilton, 1991). This leaves open the question of what action will be taken, but at least impaired seafarers can be kept away from safety sensitive tasks (e.g., lookout, navigation, monitoring of engine functions) while the impairment lasts or until they can be replaced. 16 This can be done in a manner far less intrusive than drug testing and it gets at the real work- and safety-related issue. That issue is not what the employee may or may not have ingested in recent hours, days, or even weeks, but rather whether the employee is impaired, no matter what the cause. We believe that any maritime drug-testing program should be principally motivated by safety concerns and not by political or other symbolic considerations. Maritime employers have much to gain from improved safety, yet they have been almost as vociferous in their opposition to the Coast Guard regulations as have the unions. Could it be that labor and management in the maritime industry are 161"he nature of this testing involves the creation of "computer games," which test, inter alia, eye-hand coordination and response times. Baseline parameters are developed by the employee playing the game a number of times, and then the software compares later attempts against those baseline parameters. Interestingly, early nonmaritime experience with such technologies as reported in the popular press seems to indicate two clear results: First, they are effective in at least some environments in reducing accidents and errors. Second, most of the people who fail the test on a given day seem not to be drug or even alcohol impaired. Fatigue and illness are more frequent culprits (Hamilton, 1991). Smaller vessel crew sizes in recent years may or may not lead to increased substance abuse aboard ship, but there is anecdotal evidence that fatigue and stress are greater than they were on the more heavily staffed vessels of earlier times. Smaller crews and less port time have also been hypothesized to lead to greater boredom, more strained family relationships, and other social dysfunctions that could potentially lead to stress and ultimately to impairment (Donn, 1988). 232 Karper, Donn, and Lyndaker correct in their assertion that this program will not only intrude on the privacy of seafarers and impose substantial costs on the carriers, but that it will also have minimal impact on safety? Indeed, given the design and focus of the program, it is hard to imagine any other outcome. In his dissent in the Von Raab decision, Justice Scalia was scathing and ironic in his commentary on a government drug-testing policy that he saw as dearly intended to have symbolic content rather than to improve safety or efficiency. He pointed out in that case that there was not "even a single instance in which any of the speculated horrors occurred." (The horrors in question involved drug-abusing border guards themselves smuggling drugs.) He went on to ask, "What better way to say that the government is serious about the 'war on drugs' than to subject its employees on the front line to this invasion of their privacy and affront to their dignity?" (Individual Employee Rights, January 1990, 258-259). Until the government focuses on impairment, rather than on drug usage per se, the maritime drug-testing program will incur unnecessarily high costs with relatively few benefits in terms of increased maritime safety. Unfortunately, focusing on accidents caused by fatigue and illness carries little symbolic content and thus fewer political benefits than focusing on drugs. ACKNOWLEDGMENTS The authors would like to acknowledge the helpful information and insights provided by Lee R.A. Seham and Martin Seham as well as research assistance from Chris Tangorra. Financial assistance from the Fund for Labor Relations Studies is gratefully acknowledged. The usual caveats apply. REFERENCES Abruzzese, L. (1989). Mandatory drug tests may become law yet. Journal of Commerce, October 13. Ackerman, D. L. (1991). A history of drug testing. In Coombs, R. H., & West, L. J. (Eds.), Dntg Testing: Issues and Options, 3-21. New York: Oxford University Press. American Shipper. (1988, August). Burnley Proposes Drug Testing, 90. American Management Association (1987). Drug Abuse: The Workplace Issues. New York: AMA Membership Publications Division. Brown, G. H. (1991). Rail agency says drug use continues to fall. Journal of Commerce, May 7. Cerrito, J. C., Hendrickson, T. A., & Pletscher, D. (1990). Confronting the issue of substance abuse in the workplace: When organized labor and management work together. Employee Responsibilities and Rights Journal, 3(4), 285-290. Coast Guard (1988). Programs for Chemical Testing of Commercial Vessel Personnel (CGD86-067). Collyer, R. M. (1987, September 8). Guideline Memorandum Concerning Drug and Alcohol Testing of Employees, Office of the General Counsel, National Labor Relations Board, Washington, DC. Committee on the Effect of Smaller Crews on Maritime Safety. (1990). Crew Size and Maritime Safety, Washington, DC: National Academy Press. Connors, T. (1991). Drugs: Unreasonable testing. Journal of Commerce, May 9. Crant, J. M., & Bateman, T. S. (1989). A model of employee responses to drug-testing programs. Employee Responsibilities and Rights Journal, 2(3), 173-190. DiBenedetto, W. (1991a). Coast Guard reimposes random drug test rule. Journal of Commerce, July 9. DiBenedetto, W. (1991b). Ship unions not likely to challenge new drug rule. Journal of Commerce, July 17. Maritime Drug Testing 233 DiBenedetto, W. (1988a). Drug testing under fire. 1ournal of Commerce, September 23. DiBenedetto, W. (1988b). Industry still not happy with final drug rules. Journal of Commerce, November 21. Donn, C. B. (1989). Concession bargaining in the ocean-going Maritime Industry. Industrial and Labor Relations Review, 42(2), 189-200. Donn, C. B. (1988). Foreign competition, technological change, and the decline in U.S. Maritime employment. Transportation Journal, 27(4), 31-41. Dwyer, D. J. (1989). Don't 'Just Say No' to drug testing in the workplace: Some recommendations for developing and implementing policy. Employee Responsibilities and Rights Journal, 2(4), 275-288. Hamilton, J. O'C. (1991). A video game that tells if employees are fit for work. Busbzess Week. June 3. Individual Employee Rights. (1990). Bureau of National Affairs, January 1989, January 1990. Joutvzal of Commerce. (1991a). Appeals court backs random drug testing of rail workers, June 13. Journal of Commerce. (1991b). Coast Guard delays portion of drug-testing regulations, April 30. Journal of Commerce. (1991c). Coast Guard proposes drug-testing changes, February 22. .lournal of Commerce. (1991d). 1% of rail workers test positive for drug use, May 1. Journal of Commerce. (1989). Mariners and alcohol, November 17. Journal of Commerce. (1988). Shipowners oppose drug test proposal, September 20. Marithne Newsletter. Maritime Trades Department, AFL-CIO, September 1988, January 1989. Pilot. National Maritime Union, May 1989. National Treasury Employees Union v. Von Raab 109 S.Ct. 1384. (1989). [cited as Von Raab]. Pollard, J. K., Stearns, M., & Sussman, E. D. (1990). Shipboard crew, fatigue, safety and reduced manning, Draft Final Report, U.S. Department of Transportation, Research and Special Programs Administration. Prat, M. (1989). Maritime drug test problems cited. Journal of Commerce, July 27. Schonsheck, J. (1989). On various hypocrisies of the "drugs" in sports scandal. Philosophical Forum, 20(4), 247-285. Seafarers' Log, Seafarers' International Union, September 1988, May 1989, January 1990. Seham, L. R. A. (1990). Testimony before the House Subcommittee on Coast Guard and Navigation, June 26. Skinner v. Railway Labor Executives Association, 109 S.Ct. 1402. (1989). [cited as Skinner]. Solomon, M. B. (1991a). Alcohol testing likely to pass hill hurdle. Journal of Commerce, October 13. Solomon, M. B. (1991b). Senate vote may come today on drug, alcohol testing..Iourttal of Commerce, October 15. Solomon, M. B. (1992). Transport braces for alcohol tests, frets over cost, privacy concerns. Journal of Commerce, February 5. Transportation Institute et al. v. United States Coast Guard et al. and District #2 MEBA v. Burnley, Nos. 88-3429, 89-1519 (1989). [cited as Transportation Institute] Trice, H. M., & Roman, P. M. (1972). Spirits and Demons at Work: Alcohol and Other Drugs on the Job. Ithaca, NY: ILR Paperback No. 11. Vail, B. (1988). Drug tests held costly. Journal of Commerce, December 13. Wall Street Journal (1991). Drug abuse by employees is declining, tests show. February 13. Walsh, J. M., & Trumble, J. G. (1991). The politics of drug testing. In Coombs, R. H., & West, J. L. (Eds.), Drug Testing: Issues and Options, 22-49. New York: Oxford University Press. Willette, R. E. (1991). Techniques of reliable drug testing. In Coombs, R. H., and West, L. J. (Eds.), Drug Testing: Issues and Options, 67-91. New York: Oxford University Press. Williams, J. (1988). Drug testing attacked as costly to ship lines. Journal of Commerce, September 15. Wright, J. P. (1989). Drug and alcohol testing of Maritime personnel. Journal of Marithne law and Commerce, 20(2), 541-557.