In-House Advisor: Cannabis Quandary

Rulings on random alcohol and drug tests in the workplace have helped to clarify the issues, but questions remain about accommodating workers who use medical marijuana

Just when In-House Counsel thought the issue of random drug and alcohol testing by employers had been largely settled in law, the related question of whether medicinal marijuana can be used in the workplace is raising its head.

In the past two years, two key rulings have clarified the circumstances in which employers are allowed to do random alcohol and drug testing of their workers in a unionized workplace. In June 2013, the Supreme Court of Canada upheld a New Brunswick arbitration board ruling that quashed a policy at an Irving Pulp & Paper Ltd. mill requiring all workers in “safety-sensitive positions” to submit to random alcohol tests.

In March 2014, an Alberta arbitration board ruled against the introduction by Suncor Energy Inc. of random alcohol and drug testing of safety-sensitive employees at its oil sands operations. In both rulings, the union contention that such testing violated workers' right to privacy was upheld.

The Alberta board was critical that Suncor's drug testing was done by urinalysis, which is more invasive than a breathalyzer used for alcohol. And the panel noted that just because a worker tests positive, that only proves they recently used drugs, not that they are currently impaired.

“I haven't seen much of an impact from the Suncor decision,” says Melanie McNaught, a partner at Filion Wakely Thorup Angeletti LLP in Toronto. “It's still early days. I have had questions from clients flowing from the Irving Pulp and Paper decision.

Irving dealt [only] with alcohol testing,” says McNaught. “Prior to the Supreme Court of Canada decision, people thought random alcohol testing was more likely to be permitted than random drug testing. Obviously, it's pretty much the final word.”

However, notes Robert Dupont, a partner at Fasken Martineau DuMoulin LLP in Montreal, the SCC also held in Irving that if there is an alcohol problem in the workplace, random testing may be justified. “But there are a number of expressions used by the court which make it difficult for an employer to say what evidence they need to show an arbitrator.

“You need more than a dangerous work environment,” says Dupont, “and you need more than safety-sensitive positions. You need to show that there is ‘evidence of a general problem with substance abuse in the workplace' that needs to be addressed and cannot be addressed differently.”

Dupont says clients whom he has represented have decided not to move ahead with random testing following the Irving decision. “But I'm not aware of any employers who rolled back such a policy.” (A 2006 survey found that 10 per cent of Canadian workplaces with more than 100 employees do random testing.)

Irving provided validation for employers to do certain types of alcohol and drug testing: reasonable cause testing; testing following a significant incident or near miss in the workplace; and random testing of a specific employee as part of their return-to-work program after treatment for substance abuse.

The Supreme Court did not address the question of pre-employment testing — the subject of previous rulings that practitioners have found difficult to reconcile.

For non-unionized workplaces, the implications of the Irving and Suncor decisions are uncertain. While the rulings recognize privacy rights for union employees, most employees don't have the protection of a collective agreement and arguably don't have those privacy rights.

It is unclear whether the analytical framework that the courts and arbitrators applied in the unionized context will now be applied more regularly in the non-union context in order to develop a more consistent analytical approach to testing.

If you've got someone with a permit who thinks they have the right to use marijuana, that may be fine from a criminal-law perspective, but it doesn't mean they can bring the drug into the workplace, particularly if it's a safety-sensitive workplace.

One sector in which random testing has continued is cross-border transportation. Canadian trucking companies that cross into the US test their drivers randomly in Canada to ensure they're clean if randomly tested in the US. For the same reason, Transit Windsor randomly tests its bus drivers who have signed up to drive routes that take them through neighbouring Detroit.

Meanwhile, Teck Coal Ltd., British Columbia's largest mining company, is pushing back against the Irving and Suncor rulings. Teck implemented random drug and alcohol testing in late 2012 in BC. The United Steel Workers immediately grieved, and applied to the arbitrator for an interim order staying the new policy until the arbitrator had ruled on whether the policy itself was permissible.

The union contended that the employer had not shown “the existence at the work sites of alcohol- or drug-induced safety problems,” as required by the Irving decision. In May 2013, however, the arbitrator declined to stay implementation of the new testing policy.

The BC Labour Relations Board upheld the interim arbitral decision. The arbitrator will now rule on the merits of the policy and the impact of Irving and Suncor will then become apparent.

An emerging issue, meanwhile, is the use of medical marijuana in the workplace. Under new rules that took effect in April 2014, medicinal users now seek a licence from their doctor instead of applying through Health Canada, making the process of obtaining the drug easier.

The number of medicinal pot users could grow from the current 40,000 across Canada to between 300,000 and 400,000 within a decade, according to Health Canada's estimate. It does not endorse the use of medicinal marijuana, saying there is inadequate data about pot's benefits. However, Canadian courts have ruled that there must be reasonable access to a legal source of marijuana when authorized by an MD.

“If you've got someone with a permit who thinks they have the right to use marijuana, that may be fine from a criminal-law perspective,” says McNaught, “but it doesn't mean they can bring the drug into the workplace, particularly if it's a safety-sensitive workplace.”

Employment-law practitioners agree that a licence for medical marijuana is not a licence for impairment on the job. They say workers who are prescribed pot should be treated like any other worker with prescribed medication that could impact their ability to carry out their duties on the job.

If an ailing employee needs to take medicine while at work, the employer has a duty under occupational health and safety law to accommodate them up to the point of hardship. “But at the same time,” says McNaught, “the employer should not have to accommodate employees who are stoned driving forklifts.”

Companies with safety-sensitive workplaces seem, however grudgingly, to have absorbed the court rulings on random testing, but they are barely aware of the cannabis challenge.

Mississauga, Ont.-based construction conglomerate EllisDon Corp. does no drug and alcohol testing at its traditional commercial construction sites, but has testing policies in place for its industrial and heavy civil projects in Alberta and Saskatchewan. It's also working on policies to cover Manitoba and BC as it begins such projects in those provinces.

“EllisDon's core business has traditionally been in the ICI [industrial, commercial, institutional] sector – social infrastructure predominantly – and we're now expanding into these other sectors, which are more typical for drug and alcohol testing,” says Jody Becker, Vice President and General Counsel at EllisDon Corp.

“The policy has to be crafted based on the legislation for each province. The most challenging thing [for in-house counsel] is to keep the policy up-to-date with changing decisions through arbitrations and legislative reform.” The company's Saskatchewan policy is almost identical to its Alberta policy, except the latter adopts the drug guidelines published by the Construction Owners Association of Alberta.

“For our industrial projects, we don't do random testing unless it's an owner requirement,” says Becker. EllisDon does targeted testing approved by the courts and arbitrators. This includes “pre-access” testing: the work force assigned to a particular site is tested prior to entry. “This means the vast majority of our workers in industrial and heavy civil operations would be tested,” she says.

It also includes testing employees when there is “reasonable cause” to believe they may be under the influence. They also do mandatory “post-incident” and “near-miss” testing, when accidents have occurred or almost occurred. They also test employees who have self-identified as substance abusers and have returned from a stint in rehab.

Becker notes that EllisDon's heavy civil division in Alberta was the result of acquisition in 2012, and she inherited their existing testing policy. “Those policies are monitored and updated by our health and safety group, with oversight by Legal and HR.” The industrial divisions' policies are also recent, the group having been formed in 2012. Their policies were based on those of the heavy civil group.

EllisDon has yet to face a claim of medical marijuana in the workplace. “Our policy currently prohibits anyone from being on-site who is under the influence of marijuana, whether it's for medical purposes or not,” says Becker. She foresees that the issue “may be used as a defence that someone raises in response to a non-negative test.”

As for the need to accommodate workers who use medical marijuana, says Becker, “I suspect we won't be amending the policy to deal with that issue, but we'll certainly look at the individual in light of the policy. If they're using heavy machinery, it's going to be very difficult to accommodate them in that exact type of work, but that's not to say we can't accommodate them in another way.”

EllisDon is considering a return to the US market, where random drug testing is much more permissible. “If the client is looking to have random testing on their site – to the extent that it is in accordance with applicable laws – I think we'd put that type of procedure in place,” says Becker.

Another major construction conglomerate, Aecon Group Inc., is not doing random testing either. “The Irving case ... laid down some very narrow rules for drug and alcohol testing,” says Brian Swartz, Executive Vice President Legal and Commercial Services. “The Suncor decision narrows things further in Alberta where courts and arbitrators used to be more tolerant of drug and alcohol testing.”

Aecon, like EllisDon, follows the guidelines of the Construction Owners of Alberta, and concentrates on “pre-access” and “post-incident” testing. “There's never been a grievance with respect to how the test has been applied in our case, so we haven't had to make any changes,” says Swartz.

He believes the use of medical marijuana in the workplace will be “problematic.” He recalls an arbitration case where a worker who was prescribed cannabis as medication was found, in pre-access testing, to be above the threshold for impairment set out in the construction industry guidelines. Denied entry to the site, the worker grieved but lost at arbitration.

“Medical marijuana is commonly used to treat pain,” says Swartz. “If they're in pain, then they probably have an underlying injury where it may affect their ability to do the work in any event. But every case is different. They would have to be evaluated on the basis of being able to do the work in a safe and productive manner without endangering themselves or anybody else at the work site.”

The Toronto Transit Commission (TTC) decided it needed a drug and alcohol testing policy after a subway system worker died accidentally in 2007 after having consumed a high level of THC (a cannabis compound) shortly before, or during, his shift. This followed several incidents in which bus drivers were found to be impaired by alcohol.

TTC General Counsel Brian Leck and an in-house employment lawyer drafted a testing policy, which was rolled out in early 2010. It covers employees in safety-sensitive positions — not only drivers and mechanics, but also their supervisors and even senior management, right up to the CEO.

The testing is based on “reasonable cause” or “post-incident” circumstances. When an employee tests positive and is not fired, they are given a chance to get clean in a “last-chance agreement. In that case,” says Leck, “periodically and randomly, we would monitor that individual's situation for a period of time. The threshold would be essentially zero.”

Where an employee self-identifies as a problem drug or alcohol user, TTC sends them to a substance-abuse specialist, who assesses them for an addiction. If they are found to have this disability, they are sent to rehab and then become subject to a “last-chance agreement.”

While a breathalyzer is used to test for alcohol, the TTC uses a saliva test for drugs. “We were one of the earlier organizations to proceed with oral-fluid testing,” says Leck. Unlike the traditional urinalysis test, the saliva test “shows at that time what the level of drugs in your system is and is a strong indicator of the level of impairment. If someone is a casual marijuana user on the weekend, that's their business.”

The one exception is the TTC's mandatory testing of applicants for driver and other safety-sensitive jobs, which is done by urinalysis. “They're rejected as an applicant if the test is positive,” says Leck. “The hope is that if people have those issues, they will not apply for those positions because they know in advance that we're conducting these tests.”

TTC reserved the right to introduce a random testing policy for its employees. TTC commissioners did in fact vote to go that route in 2011 after a TTC bus crash killed a passenger. (The driver was charged with negligence and possession of marijuana, but did not appear to be impaired, police said.) “At this point, we're not proceeding to implement it,” says Leck. The Irving and Suncor decisions “caused us to pause. They both create an uphill battle.”

Leck, however, cites studies showing that, in the US, random testing “does act as a real deterrent to employees engaging in alcohol and drug consumption prior to entering the workplace.”

The transit workers' union was not impressed. In October 2010, Amalgamated Transit Union Local 113 filed a policy grievance over the existing TTC testing policy. A ruling is not expected for another year.

The policy has to be crafted based on the legislation for each province. The most challenging thing [for in-house counsel] is to keep the policy up to date with changing decisions through arbitrations and legislative reform.

SaskPower, the province-wide electrical utility in Saskatchewan, did not have a testing policy in place until March 2013. The policy was jointly developed between SaskPower and its two unions, Unifor and the International Brotherhood of Electrical Workers. But that doesn't mean the unions were happy with the result, says John Phillips, Associate General Counsel at Sask-Power. “Unifor has reserved the right to grieve if they feel we're running afoul of the rights of their members.”

Even a number of SaskPower managers were uneasy about the advent of a drug and alcohol testing policy. They worried about the duties it might impose on them as supervisors. “It was fear of the unknown,” Phillips points out, “I think they've discovered it's not as scary or as onerous as they feared. We did do company-wide training on the policy.”

The SaskPower policy does not include random testing. “As a management-side lawyer, I'm a bit chagrined at where the court went” in the Irving judgment, says Phillips. “We always had some concern that [random testing] was not going to be found to be legal. Sadly, we were right. We had hoped the Supreme Court would have given employers in safety-sensitive areas more leeway than they have.”

SaskPower limits its testing to employees on a “reasonable cause” or “post-incident” basis. The utility provides a “reasonable grounds” checklist that supervisors are expected to use. Employees in those circumstances are warned of the consequences of refusing a test.

If an employee self-identifies as having used drugs or alcohol on the job, or having a problem with them, they are referred to the utility's Return to Work Office for an addictions assessment. If they are indeed substance abusers, the company pays for them to go to rehab.

On the employee's return from rehab, they may or may not be tested. That depends on whether or not the employee is subject to a “last-chance agreement,” which, as the name implies, results in termination if they are unable to maintain their recovery.

“The unions aren't a huge fan of ‘last-chance agreements,'” says Phillips. “Management can't impose a ‘last-chance agreement' on an employee without the unions being involved.” He says there were some agreements before the testing policy's introduction, but none since.

SaskPower has yet to consider the question of medical marijuana. “There might be a bit of head-scratching before we figure out how to deal with it,” says Phillips.

“The big issue is whether someone is in a position where their use of medical marijuana may impair their ability to work safely. If they're not driving or working with high voltage, it may not be as big a deal.”

If employees self-report that they were using medical marijuana, their cases would be referred to the Return to Work Office, which would determine whether or not they had to be accommodated. “If so, it could be quite difficult depending on what the employee had been doing.”

“I think it will be a legal quandary,” says McNaught. “I think it is already. An assessment will have to be done on a case-by-case basis on whether or not an employee can be accommodated. I don't think we should jump to any conclusions about what that might mean. I certainly wouldn't be encouraging my clients to set up a room where employees could smoke marijuana during their breaks.”


In-House Counsel should keep these things in mind when considering drug and alcohol testing in the workplace:

1 > Even if you have a high-risk workplace, you must determine that there is evidence of a general problem with substance abuse in the workplace before you can apply random drug and alcohol testing to employees in safety-sensitive positions.

2 > You can apply drug and alcohol testing to a group of employees on a “pre-access” basis or to specific employees when there is “reasonable cause” to do so, in a “post-incident” or “near miss” situation or in response to a “last-chance agreement.”

3 > The use of medicinal marijuana in the workplace can be treated in the same manner as the use of any other prescribed medication, accommodating the worker by shifting him or her from a safety-sensitive position to a less risky job.