This paper briefly outlines the statutes and leading case law dealing with employers' rights around drug-testing and accommodation in the workplace and then reviews the recent case law developments in the area.
I. HUMAN RIGHTS LEGISLATION
Employee alcohol and drug use and employer policies related to alcohol and drug use can raise human rights issues because both alcohol and drug dependency are recognized as disabilities. Employers are not permitted to discriminate against employees or prospective employees on the ground of alcohol or drug dependency. The Canadian Human Rights Act specifically includes drug dependency in its definition of disability, providing in section 25 of the Act that:
"disability" means any previous or existing mental or physical disability and includes…previous or existing dependence on alcohol or a drug…
Although the Ontario Human Rights Code does not expressly refer to drug or alcohol addiction in its definition of disability, it has been read and interpreted by human rights tribunals and the courts to be included. In fact, rather than "drug dependency", Ontario considers any "substance abuse", a broader term, to constitute a disability deserving of protection. Under the Canadian Human Rights Act, an employer may not refuse to employ or continue to employ a person, or differentiate adversely between employees because of disability. An employer also may not have a practice that deprives or tends to deprive people of employment opportunities because of disability. The Act makes an exception for bona fide occupational requirements. The Ontario Human Rights Code gives every person the right to equal treatment with respect to employment without discrimination on the basis of disability, except where the treatment is reasonable and bona fide in the circumstances. In addition to individuals who are actually disabled, the human rights legislation also protects individuals who are believed, incorrectly, to be disabled. The employer has a duty to accommodate disabilities up to the point of undue hardship, as determined by cost, health, and safety.
II. ENTROP V. IMPERIAL OIL
The leading case on drug-testing in the workplace is the decision of the Ontario Court of Appeal in 2000 in Entrop v. Imperial Oil.
1. Impugned Policy
The policy of the employer, Imperial Oil, an oil refinery operator, required random alcohol and drug-testing of employees in safety-sensitive positions. Imperial Oil also required testing on reinstatement into a safety-sensitive position following drug rehabilitation treatment and unannounced testing thereafter. The consequence of testing positive in any of these instances was dismissal. For all employees, whether or not they were in a safety-sensitive position, the employer also required pre-employment testing and it required testing where there were reasonable grounds to suspect impairment or where there had been an accident or near miss with potential employee fault. For positive results on these tests, the consequence was discipline up to and including dismissal. Finally, the company insisted that employees in safety sensitive positions disclose current or past substance abuse problems at any point in their life. The consequence of disclosing a past substance abuse problem was reassignment out of a safety sensitive position until the employee completed a rehabilitation program followed by a lengthy abstention period. At that point, the employee could be reinstated subject to random testing.
2. Meiorin Test
The Supreme Court explained the proper approach to evaluating a claim that an employment requirement discriminates on the ground of disability and whether it is a bona fide occupational requirement in British Columbia (Public Service Employee Relations Commission) v. British Columbia Government and Service Employees Union (Meiorin). The approach is the same whether the claim is made under federal or Ontario legislation. The preliminary, threshold question for finding discrimination calls for a simple inquiry into whether the requirement is prima facie discriminatory. That is, does it treat people differently based on a protected ground. Once that threshold is met, to justify the policy, the employer has to show it is a bona fide occupational requirement. That test has three branches. On the first branch, the requirement must be rationally connected to the performance of the job. On the second branch, the employer must have adopted the requirement in good faith. And on the third branch, the requirement must be reasonably necessary to accomplish the employer's workplace goal. Even where the practice is a bona fide occupational requirement, the employer's policy still must accommodate employees' differences to the point of undue hardship. The claimant has the onus to show discrimination, but once discrimination is demonstrated, the onus is on the employer to show that its policy is passes the three branches of the bona fide occupational requirement test and that it meets its duty of accommodation. This approach was applied by the court in Entrop.
Imperial Oil's policy described several different types of substance abusers, ranging from experimenters through dependent users. It listed all types under the heading "categories of substance abusers", and stated: "In the cycle of substance abuse, users frequently begin by experimenting with drugs and progress to the dependent user stage later on." The Court of Appeal pointed out that some of the categories of users that were listed were not protected from discrimination. The policy, however, considered them all to be substance abusers for its purposes. Imperial Oil sanctioned anyone who tested positive on the assumption, in the court's view, that they were likely to be impaired at work and not fit for duty. The court therefore concluded that the company's policy was prima facie discriminatory.
The court found that Imperial Oil's purpose in requiring drug-testing and disclosure was "to minimize the risk of impaired performance due to substance use…to ensure a safe, healthy and productive workplace." The court held that these goals are legitimate, and concluded that the purpose of Imperial Oil's drug-testing and disclosure requirements was rationally connected to the work performed.
The court also found that Imperial Oil's requirements were made in good faith. It concluded Imperial Oil acted in good faith because the company was clearly concerned about safety and drug use and had consulted widely with its own employees and with experts in the areas of occupational health and safety and substance dependency to create its policy.
3. Random Testing with Breathalyzer
The court found that freedom from impairment on the job is a bona fide occupational requirement. Since a breathalyser test shows if an employee is impaired on the job, random breathalyser testing for safety-sensitive employees was acceptable.
4. Random Testing with Urinalysis
On the other hand, the court also found that many of the drug-testing provisions of Imperial Oil's policy are not reasonably necessary to identify people who cannot perform their work safely because they are impaired by drugs. Imperial Oil's testing regime relied on urinalysis. The court held that urinalysis has a "fundamental flaw": it can not detect whether an employee is currently impaired. Urinalysis can detect whether an employee has used drugs, but not when or how much. As a result, the Court of Appeal significantly restricted its use.
The court concluded that random drug-testing of all employees in safety-sensitive positions through urinalysis is not reasonably necessary for safety and so is not a bona fide occupational requirement. The weight the court gave to the technical limitations of urinalysis is shown by its different treatment of breathalyser testing.
Imperial Oil argued that it was not using urinalysis to determine whether employees were impaired, and that the standard it demanded from its employees was not 'no impairment', which urinalysis cannot measure. Instead, the company argued, its requirement was 'no presence of drugs or drug metabolites' in the body, which urinalysis can measure. The court doubted Imperial Oil's argument, but held that even if that were the case, the company's standard would be arbitrary, since it would not demonstrate incapability to perform work safely. The problem is not random drug testing for current impairment. That would be fine. The problem is that urinalysis tests for use of drugs in the last 30 days, including weekends, evenings, vacations and so forth. Since the company couldn't convince the court that using drugs on the weekend was connected to being under the influence at work, it couldn't use random urinalysis testing.
4. Pre-employment Drug Testing
The Court of Appeal in Entrop also ruled out pre-employment drug-testing. A positive result on a pre-employment drug-test, the court held, does not in itself show impairment on the job. According to the court, pre-employment test results also do not predict likely impairment on the job. As a result, pre-employment testing is not reasonably necessary for workplace safety and is not a bona fide occupational requirement. Although, given the Court of Appeal judgment in Kellogg, below, the situation of post-offer, pre-employment testing of employees in safety sensitive positions remains unclear. In Kellogg the court expressly declined to follow Entrop on this issue, holding that post-offer, pre-employment drug testing for safety sensitive positions was not discriminatory.
5. Testing for Cause
The court did hold that drug-testing was a bona fide occupational requirement under some circumstances. The court agreed that testing is a reasonable necessity where there are reasonable grounds to believe that an employee is impaired on the job or where there has been an accident or near miss. This is also called for cause and post-incident testing. In consequence of its finding that urinalysis alone cannot show current impairment, the court restricted drug-testing in these circumstances to only situations where it is "one facet of a larger assessment of drug abuse."
6. Testing for Certification
Although the court rejected pre-employment testing for all employees, and rejected random testing for employees in safety-sensitive positions, it accepted that testing for certification into a safety-sensitive position was a bona fide occupational requirement, provided that the testing is, again, one part of a larger assessment of drug abuse. Certification also extends to reinstating an employee into a safety-sensitive position after drug rehabilitation program. In this situation, subject to the caveat that the testing not represent the sole measure of drug abuse, the court accepted that such an employee who acknowledges overcoming a past problem would have to undergo frequent, unannounced testing after reinstatement.
7. Automatic Termination
It is important to note that the court also held that automatic termination provisions for positive results on random and pre-employment drug tests failed to sufficiently accommodate individual differences. The court indicated that it would have objected to the testing policy in any event because the policy lacked the flexibility to substitute discipline short of termination.
The distinction between pre-employment testing and certification testing is an odd one. It is likely that a big part of the distinction is arose because Imperial Oil's policy treated pre-employment testing differently from certification testing. A person testing positive on a pre-employment test was automatically not hired, while a person testing positive on the certification testing had further individualized assessment to determine whether they had a substance abuse problem. If the company's policy on pre-employment testing had been different – if the policy had said that there would be an integrated assessment of whether the person had a drug problem and offer them help if they had a problem – the pre-employment testing might well have been permitted.
8. Mandatory Disclosure
The court allowed mandatory disclosure requirements in principle, but held that insisting on disclosure is only acceptable where the risk of relapse is greater than the risk of new substance abuse problems in the general population. On the evidence, the court held it permissible to demand disclosure of past drug abuse problems up to six years previous. The court also had difficulty with the sanction, holding that the requirements for rehabilitation, abstention, and testing must be reasonable and tied to the employee's actual reported history. In cases where reported past substance abuse is not recent, it would be inappropriate to require any steps at all.
Although it dealt with many different aspects of Imperial Oil's reasonably comprehensive drug-testing regime, Entrop created a number of questions of interpretation. A number of more recent decisions have considered Entrop and the issues of raised by drug-testing.
The individual claimant in Entrop had a past alcohol addiction. In many cases, however, employees who use alcohol or drugs deny that they are addicted, and may indeed not be addicted. Some labour arbitrators have found that employees cannot complain about drug-testing requirements that may discriminate on the ground of drug dependency while at the same time denying that they themselves are drug dependent. That was also what the Alberta Court of Appeal decided in Kellogg.
Milazzo v. Autocar Connaisseur Inc.
Still, this approach has not been consistently followed by recent court and tribunal decisions though. In Milazzo v. Autocar Connaisseur Inc., the Canadian Human Rights Tribunal considered a small bus company's policy of pre-employment and random drug-testing for all its drivers. The Complainant, a driver who tested positive for marijuana, insisted he was not addicted. The policy was less elaborate than the policy of Imperial Oil considered in Entrop, and it contained no language suggesting that the company viewed casual drug users as substance abusers. On the evidence, the company did not perceive the particular driver who failed a drug test to be a substance abuser. It neither knew nor cared.
The tribunal therefore did not find discrimination on the ground of perceived disability. Nonetheless, the tribunal found there was prima facie discrimination, because drug dependent individuals would be impacted more heavily by the policy. As with Entrop, there was no difficulty in finding a rational connection between the policy and the legitimate goal of safety, and no difficulty in establishing the company's good faith. Unlike Entrop, however, the tribunal found that testing is a reasonable necessity. Persuaded by the company that it was not trying to identify drivers who were dependent on drugs, the tribunal found it was trying to identify drivers with a higher risk of impairment. The tribunal agreed that a positive drug test meant that employees were more likely to be impaired. The tribunal also agreed, in contrast to a finding in Entrop, that testing does deter drug use, and that this is a valid employer goal.
Finally, the bus company required its drivers to be able to operate in the United States, where legislation requires bus drivers to undergo drug-testing. On these grounds, the tribunal distinguished the facts from Entrop and found testing to be a bona fide occupational requirement. This is one of the clearest lines in the area. If a company can't send drivers into the US without testing them, then it is allowed to test the drivers. If we had similar legislation in Canada, this issue would be much simpler for the transportation industry. But, we don't, and I don't foresee that kind of a law coming anytime soon.
Back to Milazzo. Although the tribunal agreed the testing was fine, it found that the company's sanction of automatic termination for a positive test was too harsh and not sufficiently accommodating. Taking into account the company's small size and marginal profitability, the tribunal directed that the bus company give unpaid time off to drivers testing positive to allow them to rehabilitate themselves. Drivers failing a pre-employment test, rather than having their offer of employment withdrawn, were also to receive time off, unless their hiring was urgent.
Alberta v. Elizbeth Métis Settlement
An Alberta court also recently found that random alcohol and drug-testing was a bona fide occupational requirement, though in somewhat unusual circumstances. In Alberta (Human Rights and Citizenship Commission) v. Elizabeth Métis Settlement, the council of a small, isolated Métis community imposed random drug and alcohol testing on its employees in safety-sensitive positions and for cause. The council's policy allowed discipline including dismissal for testing positive and then refusing rehabilitation services or for refusing to take the test. The policy made appropriate accommodations, providing a generous employee assistance program, disability income plans, and progressive discipline for failure or refusal to test.
Finding that the employer's position treated employees who refuse to test or test positive as if they were drug or alcohol dependent, the employer's policy was prima facie discriminatory. However, its defence succeeded. On the bona fide occupational requirement test, the employer easily cleared the hurdles of rational connection and good faith. On the reasonable necessity branch, the court noted that the council had adopted the policy after a unanimous vote of the members of the community and in response to ongoing concerns about serious drug and alcohol abuse by community members, including council employees. Testing was therefore a reasonable necessity to set a positive standard for the community, with safety and performance concerns receiving less emphasis.
These concerns, contended the court, would likely not be present in an urban area with relative anonymity and sufficient job availability that those occupying municipal clerical positions are not held in special esteem. The court accepted, without needing to hear evidence, that the testing policy would lead employees to arrive at work in a sober condition, and that sobriety among council employees would lead to sobriety among the broader community. Reasonable necessity was also shown by the fact that the council had already taken all other reasonable steps available to it to reduce substance abuse in the community, such as education. While the testing regime did not violate human rights laws, terminating an employee under the newly imposed alcohol and drug policy might give rise to a wrongful dismissal claim, the court concluded.
The lower court ruling was reversed by the Alberta Court of Appeal, on the narrow ground that the individual employee claimants had not occupied safety-sensitive positions, and that the council's policy only allowed testing for cause for those employees, and the employee in question wasn't in a safety sensitive position. This highlights another problem. Even if your policy passes the human rights issues, you need to make sure the employee has agreed to the policy and is covered by it.
Alberta v. Kellogg Brown & Root (Canada) Co.
In 2008 another significant judicial decision was released dealing with workplace drug-testing: Alberta (Human Rights and Citizenship Commission) v. Kellogg Brown & Root (Canada) Co. The employer, operated a syncrude plant in Fort McMurray and had a policy of mandatory post-offer, pre-employment drug-testing for all employees. A new employee in a safety-sensitive position took the test, and began work. Three weeks after the test, and after he had worked for nine days, his test result came back positive for marijuana. The employee was immediately terminated. The policy was one of zero tolerance, with automatic termination on a positive test and no accommodation. The employee used the drug on a recreational basis outside of work and had used it five days prior to his test. The employer did not allege that the employee was addicted.
The human rights tribunal that heard the complaint initially dismissed it, reasoning that if the employer did not perceive the employee to have a disability, and the employee did not in fact have a disability, there could not have been any discrimination.
The Human Rights Commission appealed the courts. The first court overturned the Human Rights Commission's decision, but the Alberta Court of Appeal upheld the Commission's decision.
The Court of Appeal ruled that the employer's policy did not exclude the affected individual from employment on the basis of a perceived disability. The employer's drug and alcohol policy did not perceive the employee to be drug addicted. Instead, it perceived that persons who use drugs are a safety risk in an already dangerous workplace. The court placed emphasized that the policy was directed to the actual effects of marijuana use, which are suffered by recreational marijuana uses. The policy was not directed toward the perceived effects of marijuana addiction on users. The Court reasoned that while there may be an overlap of sorts between the effects on causal users and the effects on addicts, that does not mean that casual users are addicts or that the policy treated them as addicts. Therefore, the policy was not prima facie discriminatory. This conclusion is contrary to that reached by the Ontario Court of Appeal in Entrop. The Alberta Court acknowledges this to be the case and specifically states that where their conclusion is at odds with Entrop it declines to follow it.
The Alberta Court placed special emphasis on the safety aspect of the position at Kellogg and the employer's right to ensure a safe work environment. They viewed Kellogg's policy as identical to that of a taxi or trucking company instituting a policy whereby employees must refrain from consuming alcohol for some time prior to beginning work. As the court stated, "such a policy does not mean that the company perceives all its drivers to be alcoholics, rather, assuming it is aimed at safety, the policy perceives any level of alcohol in a driver's blood reduces his or her ability to operate the employer's vehicles safely". Crucially, the Human Rights Commission and the Court of Appeal accepted expert testimony that the lingering effects of marijuana use can impact an employee's ability.
1. Saliva Testing
Following the Entrop decision, Imperial Oil ceased random drug-testing of its employees in safety sensitive positions, though it continued random breathalyser tests for alcohol impairment. Recently, equipped with more sensitive technology utilizing a saliva test from a cheek swab that can measure current impairment, it resumed random testing for marijuana. The new technology has not been considered under the human rights legislation, but it was examined by a labour arbitration panel in late 2006.
The panel acknowledged that the test was likely not a violation of the Ontario Human Rights Code. Nonetheless, it found that the test did not treat employees with appropriate respect and dignity. The panel based its conclusion on two factors. First, it felt that the cheek swab was much more intrusive of employees' privacy and bodily integrity than the breathalyser test. Second, the panel was concerned that the cheek swab took several days to analyze, in contrast to the immediate results of the breathalyser. So, while the cheek swab could ultimately lead to an employee facing discipline for impairment on the job, no employee could ever be suspended for present drug impairment on the basis of the cheek swab. The tested employees were always permitted to return to work after a drug test, because it took several days to get the result. This compromised Imperial Oil's claim of reasonable necessity. The panel relied on arbitral principles rather than human rights principles to reach its decision, and hung its hat on the issue that there was no authority in the collective agreement for the testing. The fairly strong suggestion by the arbitration panel that the cheek swab testing probably isn't a violation of the Human Rights Code was merely a suggestion.
2. Marijuana Use Outside of Work
Other arbitration decisions have considered the boundaries of reasonable grounds for requesting a drug test from employees in safety-sensitive positions. In Hamilton Street Railway and A.T.U., Local 127 (Haines) (Re), the employer became suspicious after the employee accessed a marijuana-themed website from work. The employee admitted to using marijuana infrequently outside of work. The employer tried to require a drug-testing regime for the employee, a bus driver, on the basis of cause. The arbitrator considered Entrop and concluded that possession and use of marijuana outside the workplace without any other exacerbating circumstances likely does not justify discipline. If it did justify discipline, the arbitrator reasoned, Entrop would have found that drug-testing similar to urinalysis, which showed use but not impairment, was relevant and permissible. The larger point for the arbitrator was that evidence of drug use outside of work is not reasonable grounds for suspecting that an employee will be impaired on the job, and hence not is reasonable grounds for testing, only evidence of drug abuse provides such grounds.
3. Possession of Marijuana at Work
Other arbitration awards have decided the point differently. The employee in Fording Coal Ltd. and U.S.W.A., Loc. 7884 (Shypitka) (Re) was found with marijuana in his possession at the work site. He admitted to the employer that he was a frequent user, but said he did not use marijuana before or at work. The arbitrator found that the employer had a right to demand testing for two years. The arbitrator noted that the employer would not be justified in terminating the employee on the grounds of a urinalysis test alone, as it does not reveal present impairment. Urinalysis is not irrelevant to determining impairment on the job, however, because it can function to corroborate or refute it. The employer has the right to demand that additional information from the employee where the position is safety-sensitive and there is evidence of use.
However, reasonable grounds were not found where an employee refused to take a drug test following a complaint from a private citizen who observed the employee bus driver holding what looked to be a cigarette while smells of marijuana were coming from his general direction. The arbitrator reasoned that since the employee acknowledged smoking marijuana during his own time and that the test would likely come back positive, the employer was not reasonable in demanding the test be taken and imposing punishment when the employee continued to refuse.
4. Reliable Sources
In Fluor Constructors Canada Ltd. and I.B.E.W., Loc. 424 (Chornyj) (Re), the arbitrator held that indicators that can constitute reasonable grounds for requesting a test include information obtained from a reliable conversation. A statement by an employee to a company nurse that he used marijuana every night was sufficient cause for requesting a drug test, even though the nurse did not think the employee posed any safety risks.
In Re Core-Mark International Inc. and United Food & Commercial Workers, Local 401, the employer was aware the employee had a past problem with cocaine, and had at one point used it twice a month. When the employee, wholly out of character, became irritable and error-prone for a two-week period, his supervisor had reasonable grounds to demand a drug test.
5. Post-Incident Testing
In the related circumstance of testing post-incident, the arbitration panel in Construction Labour Relations (Alberta Assn. Operating Engineers (Provincial) Trade Division) v. International Union of Operating Engineers, Local 955 (Graham Grievance) concluded that an employer does not have an automatic right to demand drug-testing post-accident, irrespective of the circumstances. The employer can only demand a drug test where there are reasonable grounds to believe that employee use of alcohol or drugs could have caused the accident.
What constitutes 'reasonable grounds' was recently canvassed in the decision of Arbitrator Jolliffe in the Pipefitters Grievance. In the Sarnia/Lambtom area known as 'Chemical Valley' two multi-trade employers' organisations encouraged their members to adopt the drug and alcohol guidelines and directives published by Imperial Oil and Suncor. The edicts stated that, for any independent contractor working on Imperial Oil or Suncor premises, the contractor's employer must have in place a drug and alcohol policy. Otherwise the contractor will not be admitted on site. Following a workplace accident and test demand, the union grieved all the policies adopted by the employers.
Central to the union's concern was that the criteria setting out what constitutes 'reasonable grounds' for post-incident was so broad in the most of the policies that the employers essentially gave themselves authority to demand random tests following any workplace incident. That is, even when an incident really did not give any reasonable grounds to suspect any impairment, the policies still said the employer automatically had reasonable grounds. The union argued that a thorough and comprehensive investigation had to be conducted before any request to submit a sample could be justified.
Arbitrator Jolliffe held that the "pursuit of reasonable cause requires actual supportable observations which realistically present some suspicion of working while under the influence". The incident itself might be capable of generating such reasonable cause. However, for smaller and more insignificant incidents it may not be reasonable to demand a test without requesting an explanation from the employee and conducting an immediate inquiry into what occurred in order to establish a realistic suspicion that drugs or alcohol could have played some role. 
It is interesting to note that where, following an incident, a drug test is mandatory as a result of drug and alcohol policy, the cost of conducting that test may not be recoverable in a lawsuit against the party who caused the incident. In South West Concrete v. Koszta the employer sought to recover such damages from an individual who drove his motorcycle into the side of a cement truck driven by an employee. The company had a policy whereby all serious workplace incidents required a test be completed. The court held that the defendant would not have foreseen that drug testing would occur and it was therefore unreasonable that he compensate the company for this expense.
6. Post-reinstatement Testing – Last Chance Agreements
Lastly, post-reinstatement drug-testing was considered in Kimberly-Clark Forest Products Inc. and P.A.C.E. International Union, Loc. 7- 0665 (Re). The employee was found with marijuana in his possession at work and admitted to the employer that he had regularly smoked it at work. Under a last chance agreement, the employer required the employee to abstain from the use of non-prescribed drugs for three years and submit to unannounced drug-testing, with automatic discharge as penalty for positive test. Following a positive test, the requirements were challenged under the Ontario Human Rights Code. The arbitrator concluded that in the unusual and severe circumstances of the case, the employer had already accommodated the employee up to the point of undue hardship through the last chance agreement. As a result, he concluded the discharge was appropriate.
In the 2008 case of Canadian Union of Postal Workers v. Canada Post Corp. the British Columbia Supreme Court upheld an arbitrator decision that stayed the termination of employment but imposed conditions that would remain in force and affect for the remainder of the employee's service with the company without the possibility of change. The conditions centred around maintaining a specific level of attendance. The employee had been habitually late and absent from work as a result of alcohol and cocaine addictions. Over a period of five years the employer offered varying accommodations and assistance and finally warned of termination of employment. The court held the arbitrator was reasonable in concluding that it was impossible to accommodate the employee through conditions of a more limited duration, without imposing undue hardship on the employer.
Despite a wealth of case law further defining and refining the decision in Entrop, excepting Kellogg, many of the major principles have not substantially changed. The decision by the court of Appeal in Kellogg has lead to a divergence in pre-employment testing. Only further time will tell how this decision will be received in Ontario. It's also not clear whether the court in Kellogg would have reached the same decision had the complainant been more than a recreational drug user and actually had an addiction.
Exactly what the Ontario Court of Appeal meant in Entrop by the "larger assessment of drug abuse" or the "larger process of assessment" of drug abuse that is necessary to permit drug-testing in certification for safety-sensitive positions and other circumstances, remains unclear.. The best insight here may come from the policies of the Canadian Human Rights Commission and the Ontario Human Rights Commission. The policies, drafted in 2002 and 2000, respectively, largely summarize the decision of Entrop. The Canadian Human Rights Commission policy states, however, that "[a]n employer can generally establish that "reasonable cause" and "post-incident" testing is reasonably necessary to ensure the heightened safety standard that is necessary in risk-sensitive environments, if testing is part of a broader program of medical assessment, monitoring and support." The Ontario policy suggests that the larger assessment "could include a broader medical assessment under a physician's care," and "employee assistance programs, peer reviews and supervisory reviews."
The subsequent decisions and awards have shed some light on a few areas though. It is fairly clear that the claimants under the human rights legislation not only do not need to be drug dependent themselves, but they do not even need to be perceived by the employer as drug dependent. The employer policy in Entrop declared that all drug users were substance abusers and would generally progress to a situation of drug dependence. That was sufficient to create a finding of discrimination on perceived disability in that case. In Milazzo and the decision of Madam Justice Martin in Kellogg, however, the employers made it clear that they did not perceive the individuals involved were drug dependent. Nonetheless, the employers' policies violated the applicable human rights legislation. In Milazzo, the policy was off-side because it had a differential impact on drug dependent drivers. Madam Justice Martin's decision in Kellogg, in addition to finding the policy would have a heavier impact on drug dependent employees, found the policy problematic because it treated all drug users as if they were likely to be impaired at work, thereby implicitly, in her view, treating them as substance abusers. Although the employer's position was less clear, the same analytical point was made in Elizabeth Métis. As a result of these decisions, is seems clear that it will be very difficult for the employer to avoid a finding of prima facie discrimination. Though, the decision of the Court of Appeal in Kellogg may indicate a retreat from these employee-favoured decisions.
A number of new arguments to support reasonable necessity have been considered. One important factor, the need to comply with American legislation, has been approved. Decisions have gone both ways on the hangover effects of marijuana use. Other arguments have been rejected. It has been decided that the fact that testing won't catch everyone does not undermine the employer's claim of reasonable necessity, an issue raised prior to Entrop by the Federal Court of Appeal, which has been reaffirmed.
The decisions related to drug-testing since 2000 have grown more comfortable with Entrop and the existing framework, but also added some new concerns for employers. As a result, the need for careful attention to best meet an employer's goals while respecting the human rights framework is undiminished.