An Evaluation of Fitness-for-Duty Testing: Part II

Interviews with current users of fitness-for-duty testing

Based on my conversations with representatives of companies that had abandoned their fitness-for-duty testing program, as well as the asserted assets and drawbacks of such testing discussed earlier, I designed an interview for representatives of organi zations currently conducting impairment testing (see Appendix A).

Although all questions were asked of all respondents in the sample, the order varied depending on each respondent's particular comments.

As reported earlier, only a handful of organizations conduct fitness-for-duty testing. Although half of the organizations I contacted were found in trade publications (some organizations, apparently, are happy to gain publicity for their innovativenes s), I depended on the test manufacturers for the names of the others. I spoke with representatives (two presidents, two human resources directors, and two managers) of four West Coast clients of Performance Factors's Factor 1000 (a municipality, a petrole um distributor, an ambulance service, and a resort) and two mid-Atlantic clients of Essex's Delta-WP (a contractor of security guards and a pharmaceuticals concern).

In four of the six organizations in the sample, only those employees in safety-sensitive positions take fitness-for-duty tests. In one of the other two, the executive staff also takes the test, to show support, and in the other, everyone in the organi zation takes the test, in the name of solidarity and fairness. Testtakers at these organizations include emergency medical technicians, paramedics, security personnel, drivers of motorized vehicles, machinery/equipment operators, maintenance workers, sci entists, inspectors, and those responsible for children.

At five of the six organizations in the sample, even employees with outstanding records take a fitness-for-duty test every day. Respondents explained that daily testing guarantees safety, and, as such, is the cornerstone of this type of program. The f ollowing comments, from three respondents, are representative:

• We need to know if they're fit for duty.
• We have to be sure.
• It wouldn't be an effective program otherwise. Judging future performance test results from past ones is an useless as using pre-employment testing and never testing again.

At one organization, management considered retesting employees who work 24-hour shifts in the middle of a long shift, but decided it would be too much of a burden.

These remarks raise questions about the adequacy and advisability of the other organization's random fitness-for-duty testing system. Here, not every employee is tested every day. Instead, because employees, who are contracted out to other businesses , work at many different locations, "roving" supervisors, armed with laptop computers, are sent to about 10 to 15 sites on a given day.

These sites are chosen at random; thus, some employees could be tested several times per week, and others, only once every several weeks. It has been asserted that "a random testing protocol for fitness-for-duty analysis may hinder the overall objecti ve of conducting fitness-for-duty analysis" (Transportation Research Associates, Inc., 1994, p. 86). Likewise, a representative of the test manufacturer told me that he does not endorse this random testing. First, although all employees there are "eligi ble" every day for testing, any of those not randomly selected on a given day may, of course, go to work in an impaired state. Second, some of those who are selected for testing take the test after their shift has already begun. Third, if a long period passes between an employee's consecutive random tests, lack of practice at the test will affect the employee's ability to meet his or her baseline and thus complicate interpretation of test results.

All organizations in the sample test job applicants for drug use and test employees for cause. One also conducts post-accident testing, and two organizations are also required to conduct random drug testing for their employees in transportation-relate d jobs.

Nonetheless, at three organizations, representatives told me they perceived behavior-based testing as more effective than drug testing because of the former's ability to detect impaired behavior every day before it can cause harm. (One of these repres entatives commented, further, that drug testing did not make sense in light of research findings that most on-the-job accidents result from stress and fatigue rather than from drug use.) They also perceived fitness-for-duty testing as less intrusive than drug testing.

Reasons for implementing fitness-for-duty testing. At one organization, fitness-for-duty testing was implemented in hopes of reducing the high rate of accidents and workers' compensation claims. In contrast, another organization, with a low accident record, testing was adopted pro-actively to ensure continuing workplace safety. At a third organization, whose representative asserted, "There's no room for error in our industry," fitness-for-duty testing was initially implemented as a pro-active measur e for one group of employees, and was later expanded to other groups after a customer rejected a flawed product. At a fourth organization, fitness-for-duty testing was initiated to appeal to customers and was seen as a way to gain a competitive edge. Th e other two organizations wanted to enhance safety, and viewed behavior-based testing as less intrusive than drug testing.

Impact of fitness-for-duty testing.

Although fitness-for-duty testing is conducted to ensure safety, its impact is generally not monitored systematically. Instead, respondents pointed to anecdotal evidence that fitness-for-duty testing is working.

Only one of the six respondents gave me any "hard" data about the effect of his organization's fitness-for-duty testing program. He pointed to steadily decreasing insurance premiums, workers' compensation claims, and minor accidents, all of which he at tributes to increased alertness resulting from fitness-for-duty testing. When I pointed out that his company has had urine drug testing for most of the time that fitness-for-duty testing has been in place, he claimed confidently that the latter was the tr ue boon to safety and productivity. When asked how he knew, however, he replied only with, "You can just tell."

Most of the other respondents pointed to anecdotal evidence that fitness-for-duty testing is working. As one commented, "Our best feedback comes from employees." Employees who are less sharp than usual, yet not impaired enough to fail their test, hav e told her that because they are focused on safety now, they take pains to concentrate at work. A second explained that the test "gets their attention, gets people to refrain from risky behavior the night before" that could make them show up tired for wo rk. A third also alluded to the deterrent quality of fitness-for-duty testing, going so far as to say that "It seems as if people have changed their lifestyles because of the test." He and another respondent indicated that because employees are entitled to a maximum number of unexcused absences, they cannot afford to fail the test repeatedly and will therefore come to work ready to perform. This other respondent also told me that the fitness-for-duty test had saved an employee's life at her organizatio n. (After failing the test, the employee was discovered to have kidney failure.)

At the remaining organization in the sample, not even anecdotal signs of the impact of fitness-for-duty testing are collected: "Just keeping track of who passes or fails each day is enough to do," the manager expressed in an exasperated tone. Testing policies. Some organizations have very specific and precise policies; others use more flexible case-by-case approaches to decide what to do when employees fail to meet their baseline. The latter grant greater discretion to supervisors, to the extent that an employee's actual test result becomes less important for determining how it is interpreted.

At one organization, employees who fail their test the first or second time within a 90-day period go home, taking a sick day because there is no safety-nonsensitive work to do; another employee takes over their safety-sensitive duties. The employees may be advised, but are not required, to visit the Employee Assistance Program (EAP) counselor. After the third failure within that 90-day period, an EAP visit is mandatory. After the fourth failure, the employee must submit to a drug test.

The policy is similar at another organization. However, an employee who fails is usually reassigned to work that is not safety-sensitive. Also, EAP counseling is not mandatory until the fourth failure, at which time the EAP counselor may also ask for a drug or alcohol test. Failure to comply with the EAP results in suspension or termination. The respondent explained: Human Resources has full discretion. This eliminates favoritism. Our goal is not to punish but to help; the EAP is the step between failing and being possibly fired.

At a third organization, there is broader supervisory discretion. If an employee fails all the allowed trials, the supervisor consults with the employee to decide if the employee should be granted yet another trial, after a half-hour to an hour has pa ssed.

Upon failing this extra trial, the employee is sent home without pay (there are no sick days) and a part-timer or substitute takes over. Although the respondent conceded that "it's a bit of a pain to juggle the schedule when someone fails and can't wo rk," he hastened to add, "but it's better than having an accident." Nobody at this organization has failed more than once a month, and failures are usually due to fatigue. However, employees understand that they can be asked to take a drug test after "e xcessive" failures and that they can be disciplined up to termination.

Counseling is not recommended, "because that would get into the specifics of why the employee failed," but it is provided upon the employee's request.

At a fourth organization, someone who fails is interviewed by a human resources representative, who first checks that the employee is doing the test correctly. An employee who fails either takes vacation or sick time, goes home without pay (if no sick days remain), goes home to sleep a few hours and return later, or is reassigned to a position that is not safety-sensitive (possibly for less pay): "No impaired person should be doing safety-sensitive work." The employee's work is either divided among co-workers or assigned to a replacement. Counseling may be recommended, especially for a first failure. Nobody has yet been tested for drug or alcohol test use after failing the fitness-the-duty test, but the organization is considering having someone t ested after three failures within 30 days.

At the organization that conducts random testing, roving supervisors have great discretion. Even if an employee fails, if the supervisor believes the employee can do the job and the employee has a good track record, the employee will be "given the ben efit of the doubt" and allowed to work. If the employee is new, he or she will more likely be sent home, replaced easily by someone on the long backup list.

The remaining organization in the sample has no policy whatsoever. Rather, anyone who cannot meet his or her baseline is handled on a case by case basis. Such flexibility gives total discretion to the supervisor. For example, it is up to the supervi sor to decide whether an employee who fails his or her test should be given another opportunity to pass. At this lean-staffed organization, sending someone home would create logistical difficulties, so supervisors try not to exercise this option (which h as caused trouble). Instead, those who fail are given safety-nonsensitive responsibilities.

Employees' responses to fitness-for-duty testing. As described by their representatives, employees' responses to fitness-for-duty testing depend on effective communication of these programs. At one organization, employees reacted positively, seeming ly because they were educated and convinced about the need for testing and because their input was solicited and used. Plus, their top management depicted behavioral testing as preferable to biochemical testing. Similarly, at two other organizations, ap propriate preimplementation training emphasized the necessity and safety benefits (for employees) of testing.

The other organizations might also have benefitted from careful pre-implementation campaigning. Some employees at the fourth organization complained ("Why are we doing this? Why don't we get a raise for learning new skills?") until they became accust omed to testing. This same group of employees had difficulty with the left-right spatial relations involved in Delta-WP's manikin test. The program administrator eliminated this test, replacing three minute-long tests with two 90-second tests.

Although employees are no longer confused, they are bored by their daily test session. The representative of the fifth organization implied that employees are still ambivalent about testing: "They feel they're above [daily testing]" even though they welcome "a system that assures them that the guy [working] right next to them is okay." And the respondent from the sixth organization told me that "the ones who stand to get caught don't like it," but that the others have not complained.

Supervisors' responses to fitness-for-duty testing. Supervisors' responses to fitness-for-duty testing, like those of employees, seem to be affected by the nature of program implementation. At four of the organizations, supervisors had wholly positiv e views of testing. As one respondent offered, "Everyone's accepted it, probably because top management fully supports it....We started the program by getting support from the top and working down."

Likewise, another commented, "Everything's worked out smoothly because we spent five months planning it and introducing it." In contrast, supervisors at a fifth organization, according to the representative's report, initially viewed testing negativel y. They complained that daily testing consumed too much of their time, worried about their subordinates' stress, and disliked having to deal with subordinates who failed. However, after the organization provided follow-up training, installed more comput er stations, and established training coordinators and interviewers to administer the daily testing and handle test failures, supervisors developed more positive attitudes. At the sixth organization, although some supervisors have belatedly come to appre ciate the benefits of fitness-for-duty testing for ensuring a safe workplace, they continue to view the program as a "put-on" -- something the president put on supervisors without first asking if they thought it would be useful. Furthermore, the represen tative told me that one employee had always been able to pass her fitness-for-duty test -- even though she seemed impaired -- because she was an expert at video games. However, this employee was unable to perform her job satisfactorily and was terminated .

Questionnaire responses of employees currently taking fitness-for-duty tests

The questionnaire sent to employees in the fitness-for-duty testing organizations began by asking respondents to provide general information about themselves: number of years worked at the organization, job title, gender, age, and education. Employee s then answered 16 five-point Likert-type questions assessing their views of performance testing (see Appendix B). These items, pretested with employees who had taken Factor 1000 until their organization discontinued performance testing, assessed employe es' views of a) the ability of fitness-for-duty tests to measure impairment, b) their fairness, c) their ability to create a safe workplace, and d) their cost-effectiveness; as well as e) the impact of these tests on employees' job satisfaction and f) emp loyees' experience of taking tests. Respondents were also provided with blank space, which they were invited to use to explain their responses and/or record any other comments about their experiences with performance testing.

Of the six organizations in the sample, one had been conducting behavior-based testing and drug testing for nearly five years at the time of my telephone interview. The remaining five had been testing their employees for a mean of only five months. At the company where testing had been going on a while, it would have been appropriate to administer a questionnaire to employees without any delay. The company president took a long while to read the brief questionnaire I had sent for his perusal. (The original version of the questionnaire for this organization, which must conduct random urine drug testing, included several items asking respondents to compare performance testing with drug testing.) Each of the several times I called him, he professed he had not yet had an opportunity to read it. When finally, I politely assured him that it would take him less time to read the questionnaire than to tell me he hadn't yet read it, he promised to read it and decide by the end of the week. While I awaite d his response, I received a call =66rom a marketing representative of Performance Factors, Inc., who chided me for intruding on his client with sensitive questions (and presumptuously advised me on the fine points of conducting organizational research). At that point, I concluded that I might have inadvertently pressured the company president to the extent that he could not confront me directly with his decision not to participate in my study. When I called him to explain that it was, of course, his pr erogative not to administer my questionnaire to his employees, and to apologize for any discomfort I might have caused him, he assured me that he did not feel uncomfortable and that his reason for contacting the test manufacturer was to seek advice as to whether he should administer the questionnaire. It would appear that the test manufacturer advised the client to deny me access to his employees, for fear that the questionnaire would stir up negative feelings about fitness-for-duty testing.

A half year later, I called each of the representatives of the five other organizations to ascertain how fitness-for-duty testing was progressing and to seek permission to distribute my questionnaire to assess employees' views toward testing. I learne d that one organization had postponed its fitness-for-duty program.

Moreover, according to the new safe workplace coordinator (my initial contact, the deputy director of personnel, had since accepted a position elsewhere in the organization), before the postponement, employees at the organization's primary site had bee n taking daily fitness-for-duty tests on a "dry-run" basis only. That is, employees who did not meet their baselines could not perform safety-sensitive work that day, but faced no disciplinary repercussions.

The program was suspended in response to DOT drug-and-alcohol testing regulations, which the organization must address for its employees in transportation-related positions by January of 1996. (Apparently, both the fitness-for-duty and drug-and-alcohol policies described to me during the interview phase of the research had been planned but were not formally in place.) The organization decided to defer fitness-for-duty testing until its biochemical testing program was underway for necessary employee gr oups. Both types of testing will constitute its "safe workplace plan."

Meanwhile, the safe workplace coordinator described to me some problems he foresees with fitness-for-duty testing. Although he praised fitness-for-duty testing for obviating the "awkward" situation of having to administer a drug or alcohol test for re asonable suspicion, he complained that such testing required too many resources -- trainers and coordinators, as well as computers.

He also explained that travel between the organization's primary site and its scores of widely dispersed testing sites, to conduct train-the-trainer sessions to expand the testing program, had all too frequently been pre-empted by inclement Arctic weat her. I asked the safe workplace coordinator if I could distribute my questionnaire to employees who had been taking fitness-for-duty tests before the program's abeyance. He refused to give me access, replying that he wanted to "monitor the program inter nally."

Three organizational representatives who participated in the interview phases of the study immediately consented to distribute my questionnaire. One neither responded to my letter nor returned my numerous phone calls. I decided to reinitiate contact with the company president whom the test manufacturer had advised against administering my questionnaire. I told him I had removed the questions asking employees to compare performance-based testing with urine drug testing (as such questions were not per tinent to any of the other organizations) and asked him if he would review the new version of the questionnaire. Not long after I sent him a copy of the questionnaire, he agreed to distribute it if I would remove Question 7 (because he felt employees wou ld not be in a position to judge whether fitness-for-duty testing is a sound investment) and Question 12 (because he wanted to avoid the "legal" issues he thought it raised). He was also worried that asking employees their gender and age might be constru ed by them as discriminatory. I explained to him that researchers typically gather demographic information about their sample, but offered to add the following phrase: "The purpose of Questions 1-5 is to gather some basic information about the individual s responding to this survey. However, if there is any question you would prefer not to answer, please leave it blank."

The data from this organization are not included in the analysis to follow, because employees had not yet received questionnaires when this manuscript was prepared.

Sample

Responses from 62 individuals are included in the data analysis. At one organization, 42 of 100 individuals returned the questionnaire; at another, nine of 35 (25.7%) did. These 51 individuals responded about Factor 1000. At the third organization, which uses Delta-WP, the representative informed me that nine (one whole unit) of the 35 individuals who were sent the questionnaire had stopped taking the performance test before the questionnaires arrived, and one supervisor had refused to distribute th e questionnaires to her unit of six, for fear of the impact on her job security of her subordinates' participation in my research.

Further, my representative told me she had overheard some employees' complaining: "Why should we help anyone connected with this test?" (Of course, it would have behooved them to express their negative views to an outsider.) Still, eleven of the 20 ( 55%) receiving the questionnaire responded. In sum, the questionnaire had a 40% response rate across the three organizations.

The mean age of the 19 females and 43 males in the sample was 36.95 years (s.d. 11.57). All but three (95.16%) had at least a high school diploma or G.E.D. Forty-five (72.6%) had completed at least some college, and 17 (27.4%) had earned a bachelor's degree. Data analysis. The scree plot of a principal components factor analysis indicated that one factor, accounting for 49.5% of the variance, comprised the items in the questionnaire (Eigenvalue 7.92 and Kaiser's Measure of Sampling Adequacy .84). A ll items loaded onto this factor except Question 6. The mean loading for the other 15 items was .72.

Responses to these 15 items were averaged to form a scale of employees' attitudes toward fitness-for-duty testing (coefficient .93). Values ranged from 1 to 4.533, with a median of 2.57 and a mean of 2.60 (s.d. .90). That is, employees generally had somewhat negative views toward fitness-for-duty testing. It is notable that three of the five highest (most positive) scores toward fitness-for-duty testing came from one company's human resources director and vice president of operations, as well as an employee there who commented that her own experience with fitness-for-duty testing was limited because she had worked for the organization for only five weeks and was therefore "not positive" that testing "indicates level of work ability." Attitudes towa rd fitness-for-duty testing were not significantly correlated with gender, age, education, or years in the organization.

Forty-nine (79.03%) of the respondents wrote comments in the space provided. Only one comment was purely positive: "It is not annoying or unfair to take a performance test because it is good to know if you are aware of what you are doing on the job" ( female, 33, about Delta-WP).

Nine comments were neutral or mixed. For example: "We have not had it in operation very long to give real accurate answers to the questions above" (male, 36, about Factor 1000). "Although I'm not real fond of performance testing, I do think there is a need for it....[W]hether it works or not, [it] is at least a means of trying to be proactive in safety concerns (male, 37, about Factor 1000)." "Performance testing is a valid way to assess my fitness for work and is much more palatable than drug testing . As with any method, I don't think it is 100% foolproof" (male, 40, about Factor 1000).

Whereas a total of three respondents commented that fitnessfor-duty testing is preferable to drug testing, two expressed the opposite opinion.

Six commented that fitness-for-duty testing is stressful: "Taking the test every day causes more stress than the job" (male, 42, about Factor 1000). "The test itself is stressful in that if one doesn't pass right way, we tend to get upset, which seems to affect performance. In other words, if I wasn't impaired to begin with, I may be by the end of the tries of the test" (female, 42, about Factor 1000).

Six expressed that performance testing is intrusive and/or that it signals a lack of trust: "I feel it is an invasion of privacy for me to take this test" (female, 32, about Delta-WP).

"I have worked for this organization for 23 years and have always had satisfactory to exemplary performance evaluations. I feel "untrusted" [sic] now that I have to proved my fitness for work before every shift" (male, 45, about Factor 1000).

Four respondents complained that fitness-for-duty testing is too time-consuming: "My co-workers and I are already faced with limited time. The ten minutes it takes for the test alone is very valuable, not to mention waiting to take it and having to ret ake it if the printer jams, etc." (male, 24, about Delta-WP). "Personally, I've had difficulty passing Factor 1000 if I have a lot of work issues to deal with early in the morning. I resent having to take valuable time out to find a free machine and then deal with it if I don't pass on the first try" (female, 33).

Twelve respondents commented on the unfairness of performance testing. Many deemed it unfair because not everyone in their organization is tested. Additionally, some consider testing unfair because a) it does not measure what it purports to measure, b) individuals with higher baselines must always perform better, or c) the test is personally discriminating: "Where I work, not all employees have to take the test -- a double standard which...causes a bad attitude" (male, 29, about Factor 1000). "It's b een really hard for me. I have mixed dominance. My left eye is dominant and my right hand is dominant... I've never been good at video games because my hand-eye coordination wasn't up to par....[A]fter you fail three sets of eight we have to take a drug test. Which I think is wrong. Because I've failed it more times than anyone [else] at the company" (male, 23, about Factor 1000).

The most common topic of respondents' comments about fitnessfor-duty testing was their doubts as to its validity. Whereas three individuals indicated their uncertainty about the validity of performance testing, 33 have decided more definitively that t esting does not work. Several reported that they have seen impaired individuals pass and fit individuals fail: "I have failed the computerized fitness test when there was no reason for me to be "unfit" -- no controlled substances, plenty of rest, no undu e stress. I have also seen people come to work extremely hung over and pass. I have seen people drink three beers and two shots of hard liquor and pass. This is not a foolproof system" (male, 45, about Factor 1000). "As a drug user, I can assure you th at Factor 1000 does not work" (male, 27).

In addition to commenting about performance tests' inability to detect the presence of risk factors, employees also question its job-related validity: "I have seen workers hung over to the extent that [they] can't do their jobs well, but can pass Facto r 1000" (male, 60). "I don't feel [Delta-WP] is an accurate representation, mainly because it is not directly job related....[T]he only way to evaluate a worker and deem him/her "fit to work" is simply by being observed by the supervisor, or by judging th e quality and quantity of work (male, 24)."

Respondents also observed that the difficulty of Factor 1000 varies: "I believe that the test is inconsistent. [It] has let me in when it shouldn't have, with a considerable amount of ease. I myself have never been locked out. Some days it is more d ifficult and other days it lets me in with little effort" (male, 21). "Performance testing is inconsistent with respect to failure region. For example, some tests take as much as two minutes to complete with the marker remaining in the white area for mos t of the test time and a failure result, while some tests are passed with only 15 seconds of test time" (male, 23).

Moreover, contrary to test manufacturers' assurances, some employees assert that they can pass or fail intentionally. "I have failed because I wanted to. The reps said there is no way that a person could alter the results. Not true" (male, 44, about Factor 1000). "You can get a majority of answers wrong but do it very quickly and get a very good score. I know this to be true because many people I work with don't pay any attention to the questions or simply keep one key depressed for the entire test and end up with very good scores" (female, 32, about Delta-WP).

Revisiting questions about fitness-for-duty testing validity.

Factor 1000 and Delta-WP are designed to have both riskfactor validity. Factor 1000 is also intended to have job-relevant validity. Many respondents, however, reported that individuals too impaired to work can pass while fit individuals often fail.</ p>

According to employees' accounts of their own and others' behavior, individuals under the influence of drugs pass their tests. It is possible that these individuals have built up tolerance to their drug of choice, enabling them to perform their well-l earned fitness-for-duty test and the more routine facets of their jobs.

But they may be too impaired to perform unfamiliar tasks. It is also plausible that individuals who are too impaired even to perform the most basic aspects of their job can concentrate for the moments required to pass a fitness-for-duty test. Still, despite these validity problems, some believe that fitness-for-duty testing may deter a proportion of employees from engaging in risky behavior. Testing Policies. Organizations have handily devised policies to administer fitness-for-duty tests and addres s failures. Typically, employees can fail their test only a few times within a given interval of time before they must visit an EAP counselor and/or take a drug test. In some organizations, someone who fails a test on a given day is, according to human resource policy, removed from safety-sensitive responsibilities and either sent home or given other work. In other organizations, the employee's supervisor has greater input in interpreting a test failure, and may determine that the employee is still rea dy to perform, even if he or she has failed. No evidence was found to suggest that supervisors abuse this prerogative. Administrative costs. According to reports from organizations that have discontinued fitness-for-duty testing, as well as organization s that currently test their employees, testing entails more expenses than simply purchasing software. Organizations must decide whether it is more costly to set up additional computer testing stations and thereby reduce employees' waiting time to take da ily tests, or to save on equipment costs but increase workers' downtime. The daily administration and interpretation of these tests also require time and effort. Organizations with smaller staffs especially encounter difficulties when they need to repla ce an employee who is found unfit to perform safety-sensitive work. Falsing. Despite test manufacturers' that test-takers cannot purposely depress their scores without being noticed, some employees reported that it is possible to fail or pass deliberatel y. Perhaps the test manufacturers are correct, but falsing goes undetected because organizations cannot afford the time needed to monitor employees' test patterns as comprehensively as they should. Employees' views. In general, employees' attitudes towa rd fitnessfor-duty testing are somewhat negative. Employees consider these tests stressful, intrusive, time-consuming, unfair, and invalid.

Conclusion

Performance-based fitness-for-duty testing would seem to be a promising tool to prevent substandard work and unsafe behavior. The popular press has sung the virtues of fitness-for-duty tests, and laboratory research has supplied evidence that they are psychometrically solid. But what has been conspicuously absent from the literature is a systematic investigation of fitness-for-duty testing programs in organizations.

The study presented in this paper reviewed archival data from two commercial test manufacturers, interview data with managers of former and current clients of fitness-for-duty tests, and questionnaires completed by employees who take fitness-for-duty t ests. The results suggest that fitness-for-duty testing loses something in the translation from theory to practice. These tests emerge valid and reliable in laboratory investigations; yet current test-takers and managers of former test-takers assert tha t fit individuals fail while obviously impaired individuals pass, and employees claim they can pass or fail their performance tests at will. Further, introducing and administering a fitness-for-duty testing program require an ongoing commitment, which, a pparently, has left most organizations no time to monitor whether fitness-for-duty testing actually enhances organizational safety and productivity.

These results are personally disappointing to me. Although I, like Dennis Attwood and some of my respondents, still advocate fitness-for-duty testing in theory, I am not convinced that Factor 1000 or Delta-WP is accomplishing what its clients expect.

References

Abbasi, S.M., Hollman, K.W., & Murrey, J.H., Jr. (1988). Drug testing: The moral, constitutional, and accuracy issues. Journal of Collective Negotiations, 17 (3), 221-235.

Alcohol-related impairment. (1994, July). Alcohol Alert, No. 25, PH 351, 1-3.

Allen, R.W., Jex, H.R., & Stein, A.C. (1984). A cybernetic test for the detection and deterrence of impaired human operators. Systems technology inc. Paper # 340, Paper presented at the International Federation of Automatic Control, 19th triennial world congress.

Allen, R.W., Stein, A.C., & Jex, H.R. (1981, October). Detecting human operator impairment with a psychomotor task. Paper presented at the 17th annual conference on manual control, Los Angeles, CA.

Allen, R.W., Stein, A.C., & Miller, J.C. (1990, October). Performance testing as a determinant of fitness for duty. Paper presented at the Society for Automotive Engineers Aerotech conference, Warrendale, PA.

Attwood, D.A., Nicolich, M.J., & Muise, E.C. (1994, January). The effects of shift schedules on the performance of control room operators: Implications for a fitness-for- duty test. Paper presented at the Second International Conference on Health, Safety and Environment in Oil and Gas Exploration and Production, Society of Petroleum Engineers, Jakarta, Indonesia.

Axel, H. (1990). Corporate experiences with drug testing programs. Research report no. 941, NY: Conference Board. Bearman, D. (1988). The medical case against drug testing. Harvard Business Review, January-February, 126.

Belleville, J.P., Dorey, F., & Bellville, J.W. (1979). Effects of nefopam on visual tracking. Clinical Pharmacology Therapy, 26, 457-463.

Blanton, A.E., Kidwell, R.E., Jr., & Bennett, N. (1992). Application of performance tests to identifying workplace drug users: A panacea or a familiar set of problems? Journal of Employee Assistance Research, 1 (2), 350-361.

Burke, M.J., Normand, J., & Raju, N.S. (1987). Examinee attitudes toward computer-administered ability testing. Computers in Human Behavior, 3, 95-107.

Burns, M., & Moskowitz, H. (1980). Effects of diphenhydramine and alcohol on skills performance. European Journal of Clinical Pharmacology, 17, 259-266.

Butler, B. (1993). Alcohol and drugs in the workplace. Toronto: Butterworths.

Caste, N.J. (1992). Drug testing and productivity. Journal of Business Ethics, 11, 301-306.

Cavanaugh, J.M., & Prasad, P. (1994). Drug testing as symbolic managerial action: In response to "A case against workplace drug testing". Organization Science, 5 (2), 267- 271.

Chesher, G., & Greeley, J. (1992). Tolerance to the effects of alcohol. Alchohol, drugs and driving, 8 (2), 93-106.

Clark, K. (1994, August 21). To work, you must beat video 'game'. The Sun, D1, 8. Comer, D.R. (1994). A case against workplace drug testing. Organization Science, 5 (2), 259-267.

Comer, D.R., & Buda, R. (in press). Drug-testers vs. non- testers: Human resources managers' perceptions and organizational characteristics. Employee Responsibilities and Rights Journal.

Crant, J.M., & Bateman, T.S. (1990). An experimental test of the impact of drug-testing programs on potential job applicants' attitudes and intentions. Journal of Applied Psychology, 75 (2), 127-131.

Cropanzano, R., & Konovsky, M. (1993). Drug use and its implications for employee drug testing. In G. Ferris & M. Rowland, (Eds.), Research in Personnel and Human Resources Management, Vol. 11 (pp. 207-257). Greenwich, CT: JAI Press.

Crown, D.F., & Ross=82, J.G. (1991). Critical issues in drug testing. In Jones, J.W., Steffy, B.D., & Bray, D.W. (Eds.), Applying Psychology in Business (pp. 260-274). Lexington, MA: D.C. Heath and Company.

Delta-WP manager's manual (1994). Columbia, MD: Essex Corporation.

DesJardins, J., & Duska, R. (1987). Drug testing in employment. Business & Professional Ethics Journal, 6 (3), 3- 21.

Drasgow, F., Olson, J.B., Keenan, P.A., Moberg, P., & Mead, A.D. (1993). Computerized assessment. In G. Ferris, & M. Rowland, (Eds.), Research in Personnel and Human Resources Management, Vol. 11 (pp. 163-206). Greenwich, CT: JAI Press.

Factor 1000 fitness-for-work assessment program technology overview and technical validation. (1991-1994). Golden, CO: Performance Factors, Inc.

Factor 1000 fitness for work program policy and procedure development workbook (1994). Golden, CO: Performance Factors, Inc.

Factor 1000 fitness for work program common questions and answers (1993). Golden, CO: Performance Factors, Inc.

Fine, C.R. 1992. Video tests are the new frontier in drug detection. Personnel Journal (6), 149-161.

Flaig, J.M. (1990). Preserving employee rights during the war on drugs. Pacific Law Journal, 21, 995-1033.

Frieden, J. (1990). Corporate America's response to substance abuse. Business & Health, 8 (7), 32-42.

Gill, A.M., & Michaels, R.J. (1992). Does drug use lower wages? Industrial and Labor Relations Review, 45 (3), 419- 434.

Gilliland, K., & Schlegel, R.E. (1993, August). Readiness to perform testing: A critical analysis of the concept and current practices. Technical report DOT/FAA/AM-93/13. University of Oklahoma.

Greenberg, E.R., Canzoneri, C., & Straker, T. (1994). 1994 AMA survey on workplace drug testing and drug abuse policies. NY: American Management Association.

Grinspoon, L., & Balakar, J.B. (1993). Marihuana, the forbidden medicine. New Haven, CT: Yale University Press.

Gross, M.R. (1991, June). A report on Factor 1000, a critical tracking test, and its application as a fitness for job duty test. Labor Center Reporter, No. 275.

Guthrie, J.P., & Olian, J.D. (1991). Drug and alcohol testing programs: Do firms consider their operating environments? Human Resource Planning, 14 (3), 221-232.

Haas, K.C. (1990, Winter). The Supreme Court enters the "jar wars": Drug testing, public employees, and the fourth amendment. Dickinson Law Review, 94, 305-371.

Hanson, F. A. (1988). Some social implications of drug testing. The University of Kansas Law Review, 36 (4), 899- 917.

Hanson, A. (1990). What employees say about drug testing. Personnel, 67 (7), 32-36.

Harris, M.M., & Heft, L.L. (1992). Alcohol and drug use in the workplace: Issues, controversies, and directions for future research. Journal of Management, 18 (2), 239-266.

Hoffman, R.G., & Lovler, R. (1989, Fall). Drug testing: A multifaceted issue. AITIA, 6-8.

Holcom, M.L., Lehman, W.E.K., & Simpson, D.D. (1993). Employee accidents: Influences of personal characteristics, job characteristics, and substance use in jobs differing in accident potential. Journal of Safety Research, 24, 205-221.

Hopkins backs off on random drug testing program. (1991). American Journal of Nursing, 91 (4), 11.

Jex, H.R. (1988). The critical-instability tracking task-- its background, development and application. In W.B. Rouse (Ed.), Advances in man-machine systems research, Vol. 5.

Karper, M.D., Donn, C.B., & Lyndaker, M.E. (1994). Drug testing in the transportation industry: The maritime case. Employee Responsibilities and Rights Journal, 7 (3), 219-233.

Karren, R.J. (1989). An analysis of the drug testing decision. Employee Responsibilities and Rights Journal, 2 (1), 27-37.

Klein, R.H., Jex, H.R. (1975). Effects of alcohol on a critical tracking task. Journal of Studies on Alcohol, 36 (1), 11-20.

Kupfer, A. (1988, December 19). Drug testing programs in industry. Professional Safety, 12-17.

Latessa, E.J., Travis, L.F., & Cullen, F.T. (1988). Public support for mandatory drug testing in the workplace. Crime and Delinquency, 34, 379-392.

Lewis, A. (1990). Drug testing: Can privacy interests be protected under the "special needs" doctrine? Brooklyn Law Review, 56, 1013-1044.

Lundberger, G. (1986). Mandatory unindicated urine drug screening still chemical McCarthyism. JAMA, 256 (21), 3003- 3004.

Malia, C.E. (1989, April 27). Drug testing in regulated industries: Public interest versus employee privacy. Public Utilities Fortnightly, 48-51.

Maltby, L.L. (1990). Put performance to the test. Personnel, 67 (7), 30-31.

McDaniel, M.A. (1988). Does pre-employment drug use predict onthe-job suitability? Personnel Psychology, 41 (4), 717- 729.

McGinley, L. (1992). "Fitness" exams help to measure worker acuity. Wall Street Journal (April 21), B1, B10.

Mendelson, S.R., & Libbin, A.E. (1988). The right to privacy at the workplace, Part 3: Employee alcoholand drug-testing programs. Personnel, 65 (9), 65-72.

Moore, J. (1989). Drug testing and corporate responsibility: The "ought implies can" argument. Journal of Business Ethics, 8 (4), 279-287.

Morgan, J.P. (1987). Urine testing for abused drugs: Technology and problems. In S.G. Kramer (Ed.), Aids and drug abuse in the work place (pp. 63-67), NY: Prentice-Hall.

Morgan, J.P. (1991). Under what influence?: A critical review of reputed benefits of workplace drug testing. Unpublished paper, NY: City University of New York Medical School.

Morrow, D., Yesavage, J., Leirer, V., Dolbert, N., Taylor, J., & Tinklenberg, J. (1993). The time-course of alcohol impairment of general aviation pilot performance in a Frasca 141 simulator. Aviation, Space, and Environmental Medicine, 64 (8), 697-705.

Murphy, K.R., Thornton, G.C., & Prue, K. (1991). Influence of job characteristics on the acceptability of employee drug testing. Journal of Applied Psychology, 76 (3), 447-453.

Murphy, K.R., Thornton, G.C., & Reynolds, D.H. (1990). College students' attitudes toward employee drug testing programs. Personnel Psychology, 43, 615-631.

Newman, A. (1994, March 15). Drug-testing firms face pluses, minuses in new rules. Wall Street Journal, B4.

Normand, J., Lempert, R.O., & O'Brien, C.P. (Eds.). (1994). Under the influence?: Drugs and the American work force. Washington, D.C.: National Academy Press.

O'Hanlon, J.F. (1981). Critical tracking test (CTT) sensitivity to fatigue in truck drivers. In B.J. Osborn & J.A. Levis (Eds.), Human factors in transportation research (pp. 143-150). London: Academic Press.

Orentlicher, D. (1990, August 22/29). Drug testing of physicians Journal of the American Medical Association, 264 (8), 1039-1040.

Pavlovich, J.S. (1989). Just say yes to drug-testing legislation: The Skinner and Von Raab decisions. DePaul Law Review, 39, 161198.

Register, C.A., & Williams, D.R. (1992). Labor market effects of marijuana and cocaine use among young men. Industrial and Labor Relations Review, 45 (3), 435-448.

Ross, J.G., Ringer, R.C., & Miller, J.L. (1992). Personality and drug testing: An exploration of the perceived fairness of alternatives to urinalysis. Paper presented at the Annual Meeting of the Academy of Management, Las Vegas.

Rynes, S.L., & Connerley, M.L. (1993). Applicant reactions to alternative selection procedures. Journal of Business and Psychology, 7 (3), 261-277.

Schuler, H. (1993). Social validity of selection situations: A concept and some empirical results. In H. Schuler, J.L. Farr, & M. Smith (Eds.), Personnel selection and assessment: Individual and organizational perspectives (pp. 11-21). Hillsdale, NJ: Erlbaum.

Sharit, J., & Czaja, S.J. (1994). Ageing, computer-based task performance, and stress: Issues and challenges. Ergonomics, 37 (4), 559-577.

Stevens, W.K. (1990, March 6). Measuring workplace impairment. The New York Times, B5, B8.

Stoller, K., & Bellville, J.W. (1976). Effects on visual tracking of delta-9 tetrahydrocannabinol and pentobarbital. Journal of Clinical Pharmacology, 16 (5-6), 271-275.

Stone, D.L., & Kotch, D.A. (1989). Individuals' attitudes toward organizational drug testing policies and practices. Journal of Applied Psychology, 74 (3), 518-521.

Stone, D.L., & Vine, P.L. (1989). Some procedural determinants of attitudes toward drug testing. Paper presented at the Annual Conference of the Society for Industrial/Organizational Psychology, Boston.

Thompson, F.J., Riccucci, N.M., & Ban, C. (1991). Drug testing in the federal workplace: An instrumental and symbolic assessment. Public Administration Review, 51 (6), 515-525.

Transportation Research Associates, Inc. (1994, April 29). Transportation Cooperative Research Program project on feasibility of applying and evaluating fitness-for-duty testing in the transit workplace: Task 4. Columbus, OH: Battelle Memorial Institute.

Turnage, J.J., & Kennedy, R.S. (1992). The development and use of a computerized human performance test battery for repeatedmeasures applications. Human Performance, 5 (4), 265-301.

Turnage, J.J., Kennedy, R.S., Smith, M.G., Baltzley, D.R., & Lane, N.E. (1992). Development of microcomputer- based mental acuity tests. Ergonomics, 35 (10), 1271-1295.

Vodanovich, S.J., & Reyna, M. (1988). Alternatives to workplace testing. Personnel Administrator, 33, 5, 78-84.

Warshauer, W. (1991). Performance testing. The New York Times (September 17), A-20.

Yes, test to prevent tragedy. (1991). The New York Times (August 30), A-22.

Zimmer, L., & Jacobs, J.B. (1992, Spring). The business of drug testing: Technological innovation and social control. Contemporary Drug Problems, 1-26.

Appendix A: Interview Questions Asked of Representatives of Current Users of Fitness-for-Duty Tests

Why was impairment testing implemented?

In what ways would you say impairment testing has enhanced safety and/or productivity? reduced accidents, mistakes, and injuries? affected absenteeism? How do you know?

What kinds of human resources policies have been adopted to deal with those employees who fail their tests -- the first time or repeatedly? Who determines the outcomes for those who fail? Why?

In which (if any) situations is counseling recommended? Why (not)?

In which (if any) situations is drug or alcohol testing recommended? Why (not)?

If an employee has sustained an excellent performance record over a period of time, does the employee still take a daily impairment test? Why is that? How have employees reacted to this? In general, how have employees responded to impairment testing?

In general, how have supervisors responded to impairment testing of their subordinates?

Appendix B: Items Assessing Views toward Fitness-for-Duty Tests

1. Performance tests can measure if an employee is fit to perform his or her job. 2.47 (1.29)

2. Taking a performance test is fun. 2.186 (1.21)

3. Performance testing is an effective way to cut the costs of risky behavior in the workplace. 2.45 (1.18)

4. Performance tests are fair. 2.50 (1.28)

*5. Performance testing is an invasion of an employee's privacy. 3.39 (1.36)

*6. Taking a performance test can be stressful. 2.13 (1.19)

7. The cost of administering and interpreting performance tests is a sound investment that pays off in the long run. 2.47 (1.17)

*8. I would be more satisfied with my job if I didn't have to take any performance tests. 2.47 (1.35)

9. Knowing that my company conducts performance testing makes me feel safe when I'm at work. 2.32 (1.28)

10. Taking a performance test is a pleasant experience. 2.00 (1.00)

*11. It is unfair for an organization to administer performance tests to employees. 3.29 (1.23)

*12. The results of a performance test reveal personal information that someone's employer has no right to know. 3.68 (1.15)

*13. It is annoying to take a performance test before every shift. 2.22 (1.35)

14. The results of an individual's performance test indicate whether or not the individual is too impaired to work. 2.35 (1.35)

15. I feel good about working for an organization that conducts performance testing. 2.77 (1.31)

16. I feel more secure because my company conducts performance testing. 2.45 (1.26)