On April 13, the government of Canada introduced Bill C-45, the
Cannabis Act, which, once passed, will remove from the country’s
Criminal Code incidental marijuana consumption and possession. Expected
to become law by July 1, 2018, the act will also regulate marijuana
production, distribution and retail sale.
The legalization of marijuana has heightened employer concerns that
employees will come to work impaired. Now, more than ever, employers are
asking, “When and how can an employer test for drugs and alcohol in the
The answer: While testing may still be rare, scientific advances,
driven largely by law enforcement seeking effective ways to roadside
test for drug-related impairment, have meant some of the more compelling
historical and technological barriers to drug testing are (or will soon
be) a thing of the past.
Traditionally, drug and alcohol testing in Canadian workplaces has
been permitted sparingly as courts and arbitrators tried to balance the
competing interests of privacy, human rights and safety. Testing has
been allowed where the position at issue is considered safety-sensitive
• the employer has reasonable cause to believe the employee is impaired at work
• the employee has been involved in a significant workplace incident or “near miss”
• it is a component of a return-to-work arrangement following an employee’s treatment for drug or alcohol addiction.
Pre-employment testing has been considered impermissible because a
failed test does not mean a candidate will attend work impaired, only
that she has used alcohol or a controlled substance in the past — and
screening on this basis alone could violate human rights legislation.
Random testing was considered impermissible as an unnecessary and
unjustified infringement on an individual’s privacy rights. Drug testing
(as opposed to alcohol testing), in particular, has been criticized
because traditional methods were seen as intrusive (such as a blood
test) and could not determine current, or level of, impairment.
In 2013, the Supreme Court of Canada gave employers some leeway. In Communications, Energy and Paperworkers Union of Canada, Local 30 v. Irving Pulp & Paper, Ltd,
the court confirmed an employer may conduct random alcohol testing for a
safety-sensitive position if it can establish it operates a dangerous
workplace and there is a general workplace problem with alcohol abuse.
Fast-forward to 2017 and the law has begun to catch up with technology.
TTC takes a stand
In October 2010, the Toronto Transit Commission (TTC) implemented a
“Fitness for Duty Policy” designed to “ensure the health and safety of
employees and the safety of customers and members of the public.” The
policy provides for drug and alcohol testing of employees in
safety-sensitive and specified management/executive positions, under the
• Where there is reasonable cause to believe drug or alcohol use resulted in an employee being unfit for duty.
• As part of an investigation into a significant work-related incident or accident.
• In the context of a return-to-work plan following treatment or after a violation of the policy.
• As a final condition of employment in safety-sensitive positions.
The union for TTC employees alleged the policy was contrary to the
collective agreement and Human Rights Code of Ontario. A grievance was
brought to arbitration (which is still ongoing).
In October 2011, in response to a culture of drug and alcohol use at
some of its locations, the TTC announced its intention to expand the
policy to include random drug and alcohol testing of employees in
safety-sensitive and specified management/executive positions.
The union brought a motion for an interlocutory injunction preventing
the TTC from implementing the program pending a final determination of
the issue through the arbitration. The party requesting the injunction
must demonstrate three things:
• There is a serious issue to be tried.
• If the injunction is not granted, the party (in this case, the
union) will incur irreparable harm that cannot be compensated in
• The balance of convenience favours granting the relief, taking into account the public interest.
In refusing the injunction, the court responded to the three issues as follows:
• Yes, there is a serious issue to be tried.
• No, the union will not suffer irreparable harm if the injunction is
not granted. If the policy, or its random component, is found to
contravene the collective agreement or Human Rights Code, the law of
Ontario provides for the payment of monetary damages to those employees
who have been wronged.
• No, the balance of convenience does not favour granting the
injunction. If random testing proceeds, it will increase the likelihood
an employee in a safety-
sensitive position, prone to using drugs or alcohol, too close in time
to coming to work, will either be detected or deterred by the prospect
of being detected. This will enhance public safety.
The court commented favourably on aspects of the policy designed to
address historic concerns with testing, including methods, the impact on
privacy and human rights, and the stigma of being selected. The court
also accepted scientific evidence advanced by the TTC to support and
underscore the necessity and reasonableness of random testing.
• There is a culture of drug and alcohol use at the TTC. This is factually different from Irving Pulp and Paper
in which the arbitration board concluded the employer exceeded the
scope of its management rights under a collective agreement by imposing
random alcohol testing in the absence of evidence of a workplace
• It is very likely a worker with a substance use disorder will report to work in an impaired condition.
• Many cases of drug and alcohol-related activity among TTC employees
at work go undetected and unverified due to difficulties in detecting
drug and alcohol misconduct.
• Between 2010 and 2016, about 2.4 per cent of external applicants —
who knew they would be tested for drugs — nevertheless returned a
• Statistically, random workplace testing results in a significant decline in the rate of positive drug tests of employees.
• An external candidate interested in working for the TTC in a
safety-sensitive position must pass a pre-employment urinalysis test for
drug use. Hence, the idea of ongoing testing should be a reasonable
• The TTC distributed comprehensive information to all workers about the policy and intention to implement random testing.
Minimal intrusion into privacy
• A breathalyzer measures a person’s breath alcohol level at the time
of the test; it is not invasive and does not reveal other personal
• A positive test result would be followed by a review with a medical
review officer who would discuss the results with the employee to
determine if there was a legitimate, medical explanation. If so, the
officer would have the discretion to report the test as negative.
• General stigma and psychological and reputational damage, as a
result of being tested, is unlikely given that 20 per cent of employees
would be randomly tested each year.
Reliability of results
• Threshold levels for a positive test would be higher than in other
internationally recognized programs, ensuring a greater likelihood of
impairment at the time of testing based on recent drug use, and
minimizing intrusion into an employee’s life choices by screening out
results that detect previous drug use unlikely to amount to impairment
at the time.
• Oral fluid testing gives a better indication of recent use and likely impairment than urinalysis.
• There’s less chance of a false-positive due to second-hand smoke.
By denying the injunction, the court permitted the TTC to implement
random testing pending the outcome of the arbitration and, so far, it
would appear the court got it right; of the first group of employees
randomly tested, two of eight failed the test.
What does this mean for employers?
The court’s decision did not determine the permissibility of the
TTC’s testing protocol; this will be decided in the arbitration.
However, it does provide insight into the analysis a court may undertake
when assessing a testing program.
An effective and defensible policy must be tailored to the workplace
and appropriately address important factual and legal considerations. At
a minimum, an employer should consult with counsel and consider the
• Are there sufficient factual grounds to implement a testing policy,
such as a culture of drug or alcohol use among workers engaged in
• Is the policy broad enough to capture impairment not only from the
use of illegal drugs, but also prescription and legal recreational
• Does the policy appropriately address safety, privacy and human rights issues?
• Does the policy clearly outline how testing will be undertaken?
• Does the policy identify consequences in the event of a breach
(meaning discipline and termination), including failure to participate
Shana French and Brian Wasyliw are lawyers at Sherrard Kuzz LLP,
one of Canada’s leading employment and labour law firms, representing
management. They can be reached at (416) 603-0700 (main), (416) 420-0738
(24-hour) or by visiting www.sherrardkuzz.com.
Case law on testing
Communications, Energy and Paperworkers Union of Canada, Local 30 v. Irving Pulp & Paper Ltd.,
2013 SCC 34 (S.C.C.). Irving’s random testing policy was struck down by
an arbitration board as a significant encroachment into employee
privacy that was “out of proportion to any benefit.” The company’s
alcohol-related incidents were insufficient to demonstrate a “problem in
the workplace.” That decision was overturned, but the Supreme Court of
Canada later found random testing to be unreasonable: “The expected
safety gains to the employer were found by the board to range from
uncertain to minimal, while the impact on employee privacy was severe.”
Hotz Environmental and TC, Local 879 (B. (G.)), Re, 2016
CarswellOnt 1824 (Ont. Arb.). The employer had a drug and alcohol policy
that allowed employees who tested positive for drugs or alcohol to
return to work with random followup testing. Employees who refused a
test were also subject to the followup testing. The arbitrator found
that while a refusal wasn’t the same as a positive test, the employer
would have no other way of knowing if someone was a recreational user or
addict. However, the policy’s description of the testing process was
vague. While the concept was reasonable, the wording should be amended
to talk about individualized assessments.
Suncor Energy Inc. v. Unifor, Local 707A, 2016 CarswellAlta
921 (Alta. Q.B.). The arbitration board found Suncor had failed to show
an existing problem in the bargaining unit that justified the intrusive
nature of random testing. The court disagreed, finding the Irving
test only required evidence of a general “workplace” problem with drugs
and alcohol. The court held that workplace safety is an aggregate
concept, particularly in dangerous environments. A broader focus on the
overall problem, as opposed to a narrow focus on the bargaining unit,
was consistent with an employer’s obligation to ensure the safety of its