The ADA requires that an entity subject to the act cannot require a medical examination and cannot make inquiries of an employee as to whether that employee is an individual with a disability or as to the nature of the severity of the employee’s disability unless that examination or inquiry is job-related and consistent with business necessity. See 42 U.S.C. § 12112(d)(4)(A). That is all fine and dandy, but that doesn’t answer the question as to whether some kind of preliminary showing is necessary before the employer can conduct such a postemployment medical exam. That is, must the employer have reasonable suspicion, probable cause, good cause, reasonable cause, or some other standard before they can demand a postemployment medical exam of a particular employee? Or is it possible that the employer doesn’t need any of that, and they can demand a postemployment medical exam of an employee regardless of whether they are on notice as to a potential problem. The Western District of Pennsylvania in EEOC v. United States Steel Corporation 2013 WL 625315 (W.D. Pennsylvania, February 20, 2013), was faced with this very question. In that case, probationary employees (employees that had worked under 1,040 hours), were subject to random drug testing. It didn’t matter whether the employer was on any notice as to a particular problem. All employees on probation were subject to random drug testing. The random drug testing was something agreed to by the Corporation and the union that represented the workers in the factory and enshrined in their basic labor agreement. The new employees at the plant performed either of three functions (door cleaners, lidman, and Coke oven laborer). All three of those job were very close to or on the coke batteries, which contain molten Coke and reach a temperature of up to 2100°F. These employees deal with narrow work areas, dangerous heights, extremely large moving machinery, toxic and combustibles superheated gases, biosludge, as well as a number of mobile vehicles containing hot Coke or bituminous coal. As a result, these employees must wear layers of protective clothing, equipment, and masks covering the entire body. On January 29, 2008, the plaintiff was terminated after she failed both an initial and confirmatory breath alcohol test. She filed a claim with the EEOC complaining that the test didn’t account for her diabetes. From there, the case becomes quite messy in terms of how the EEOC went about the investigation and conciliation process prior to filing suit. The EEOC did file suit saying that this policy violated the ADA with respect to when a postemployment medical exam could be performed. The EEOC ‘s view was that medical testing is not permitted unless the employer has an individualized reasonable suspicion of a safety concern. Therefore, any random alcohol testing policy must be per se invalid.
Before moving on to the issue of what kind of cause is needed, if any, before conducting a medical exam, some of the other issues can be squared away first. First, before the EEOC can bring a case against someone, there are certain procedural formalities that they have to go through. United States Steel alleged that those formalities were not done in this case. The court, after detailing the sequence of events, said that it was a close call, but ultimately it was a question of fact as to whether the formalities had been observed. Second, United States Steel also maintained that the random drug testing provision was within the safe harbor contained in employee benefit plans, but the court was having none of it. The court said that this was a medical exam and not the type of thing that you would see in an employee benefit plan. Furthermore, the random drug testing was part of the collective bargaining agreement and was not voluntary. Accordingly, that safe harbor exception didn’t apply. Finally, the EEOC argued that their enforcement guidance, which required some kind of reasonable suspicion before doing a medical exam postemployment, was binding on the matter. However, the court was having none of it. Since the guidance did not go through proper rulemaking, the guidance was not binding on the courts and could be used by the court or not at its own discretion.
Turning to the merits of the case, the court first set forth the paradigm for figuring out how the job-related and business necessity defense works. That paradigm works as follows. First, it is the employer that has the burden of showing a medical examination is job-related and consistent with business necessity. Second, whether a medical examination is consistent with business necessity is an objective test. Third, the employee does not need to have a disability under the ADA to challenge a medical examination or inquiry. Fourth, to determine whether an employer has shown that a medical examination or inquiry is job-related and consistent with business necessity, the employer has to show that a business necessity exists and that the policy at issue serves the asserted business necessity. Fifth, any business necessity warranting a medical examination has to be vital to the business, such as ensuring that the workplace is safe and secure or reducing egregious absenteeism. Sixth, in order to show that the policy serves the asserted business necessity, an employer must demonstrate the examination or inquiry genuinely serves the asserted business necessity and is no broader nor more intrusive than necessary. Finally, if an employer argues that safety is the business necessity, then all the employer has to do is demonstrate that the medical test or inquiry is consistent with business necessity is to show that the medical test or inquiry makes even a small (emphasis mine) contribution to reducing the risk posed by unfit employees.
With that in mind, the court was ready to take on the merits of the case and they approached it in the following way. First, there is nothing in the text of § 12112(d)(4)(A) specifically requiring employers to have suspicion, reasonable or otherwise, that an employee presents a safety hazard before they can conduct a random alcohol test on new employees.
Second, the court noted that the case law is all over the place on whether some form of reasonable suspicion or otherwise is required first before such a test can be conducted.
Third, the court looked to an analogous statute in the ADA §12112(b)(6), dealing with qualifications that screen out people with disabilities, and said that nothing in that particular statutory provision would suggest that across-the-board medical screenings are per se violations of the ADA.
Fourth, the fact that only new employees were subject to this random testing did mean that the employer has the burden of justifying the necessity of performing the medical examination only upon the new employees. The court believed that United States Steel more than met their burden with respect to that in several ways. First, new employees work in very dangerous jobs that included molten hot Coke, toxic waste products, massive moving machinery and so no level of intoxication could be acceptable under those circumstances. Second, the random drug testing provision was negotiated between the union representing the workers and the Corporation and was made part of the collective bargaining agreement. Third, many employers, federal and private, use random drug and alcohol testing for employees working in safety sensitive or mission-critical position and their common theme is that random testing can and should be used by employees working in positions where even the smallest miscalculation could lead to dire consequences. Finally, random testing deters other employees from bad behavior leading to bad consequences.
Fifth, the court explored the viability of limiting the random drug testing to new employees only. They, in essence, said that there was a rational basis for limiting the testing to new employees only. That basis included: regular employees that have been on the job for a long period of time have proven they can follow appropriate safety standards and adequately perform their daily job, which is not the case with new employees; new employees are less skilled at performing their jobs, less familiar with company rules, and have not fully internalized the importance of workplace safety; factory workers in training, a.k.a. new employees, are inherently more dangerous to themselves and others because of the lack of training and familiarity with the job and have an elevated chance that they will arrive at work in an unfit state; new hires, until they fully can appreciate the risk in the workplace, are more likely to engage in risky behavior than regular employees, and there is no room for error when the dangers are as numerous as they are in this plant; there is no way for the employer to tell who may be engaging in behavior demanding further investigation when everybody has to be completely dressed up and covered due to the hazardous conditions they work in. Finally, even a 4 to 5 month differential (the policy applied to employees that had worked under 1040 hours-about 4 to 5 months full-time), can become critical given the stakes as high as they are. Therefore, probationary employees are not in the same place as regular employees and can be treated differently, and the random testing was warranted on new employees because they were less likely to comply with the alcohol policy. All this said, since the court reasoned this way with respect to new employees, they really had little choice but to say that to do it on non-probationary employees would result in a policy that would be broader and more intrusive than necessary and therefore, would fail the business necessity test with respect to nonprobationary employees.
Takeaways: There are numerous things we can take away from this case. First, many practitioners probably assumed based upon an implied reading of the statute, which is certainly supported by the EEOC enforcement guidance, that some form of reasonable suspicion was necessary before being able to do a post employment medical exam. This case makes clear that we should no longer assume that necessarily. Second, in the labor context, it really helps if a random drug or alcohol testing policy is part of the collective bargaining agreement. Third, the EEOC does have certain procedural formalities they have to go through before being able to pursue a claim. It is certainly worth for the defense to claim that the formalities were not observed, but such an argument may be a tough one to carry. Fourth, enforcement guidances and the like are just persuasive authority and an employer is free to argue that they should not apply. Such an argument, as was seen here, may indeed carry the day (such an argument was also successful in Toyota Motor Manufacturing, Kentucky V. Williams, 534 U.S. 184 (2002), where the United States Supreme Court gave no deference to the EEOC’s view on what constituted a substantial limitation on a major life activity)). Fifth, this case does not really tell us what job-related means, but instead focuses on business necessity. From this case, if an employer has, for the lack of a better word, a “rational basis,” for its policy, they may well prevail. Sixth, if the employer has a “rational basis,” for treating different kinds of employees differently that may also carry the day as well. That said, this “rational basis,” approach only goes so far. For example, this court, as noted above, said that a random drug testing policy for employees no longer new would fail. Finally, the employee benefit plan safe harbor is not something that is a catchall category, rather a voluntary employee welfare plan must be involved.
I would expect that the EEOC would appeal this decision. That said, I do not want to hazard a guess as to what the United States Court of Appeals for the Third Circuit will do with this case. Especially if United States Steel loses in the Third Circuit, I would expect United States Steel to appeal this to the United States Supreme Court. With respect to the United States Supreme Court, I most certainly would not want to hazard a guess as to what they would do in this situation should they take the case, though I can say that it has been extremely difficult for an employee to prevail at the United States Supreme Court level with respect to the ADA (title I cases. Persons with disabilities have fared better at the United States Supreme Court when it comes to claims arising under title II or title III of the ADA).
If this particular blog entry looks a little bit different to you (that is, all the citations are missing), that is intentional on my part. Over the months, I have read many blogs, including those that would be considered fairly academic in nature, and none of them adopt the formal writing convention of citing any time that it is not your idea. Also, I have become very worried that putting the citations in the document, especially since it is a blog, unnecessarily slows the reader down. Finally, while lawyers may appreciate the citations per the formal legal writing convention, people whom are not lawyers (such as disability professionals, businessmen, businesswomen, consumers with disabilities, employees, the media in general, etc.), may find it inaccessible. When the fourth edition of my book comes out, you will see that many of my blog entries have made their way into the fourth edition of the book. In the book, you will see the formal legal writing conventions of when to cite observed, but, at least for now, not in the blog. Hopefully, you will find the reading easier and more accessible. Again, thanks so much for your readership!