Northern Exposure

Supreme Court rejects random alcohol testing policy in dangerous workplace

By Kyla Stott-Jess, Katie Clayton, and Hannah Roskey

Canada’s highest court has ruled that random drug and alcohol testing in the workplace violates privacy rights. In Communications, Energy and Paperworkers Union of Canada, Local 30 v. Irving Pulp & Paper Ltd., the Supreme Court of Canada (SCC) considered the validity of a random alcohol testing policy in a unionized workplace. In a 6-3 decision, the SCC agreed with the original arbitration board decision to strike down the employer’s mandatory drug and alcohol testing policy.


Irving operates a paper mill in Saint John, New Brunswick. In 2006, it unilaterally adopted a drug and alcohol policy under the management rights clause of the collective agreement. This policy imposed random alcohol testing for employees in “safety sensitive” positions. Under the policy, 10 percent of the employees in safety sensitive positions were randomly selected for breathalyser testing over the course of a year. A positive test attracted significant disciplinary action, including dismissal. Failure to submit to testing was grounds for immediate dismissal.

Perley Day, a member of the union, was tested under this policy. Although he passed, his union filed a grievance on his behalf to challenge the legitimacy of the policy. The union alleged that, in the absence of an alcohol-related incident at the workplace, there were no reasonable grounds to randomly test employees for alcohol.

Arbitration board sides with union

The arbitration board that heard the union’s case allowed the grievance. Crucial to its decision was the absence of evidence to indicate any real risk associated with alcohol. There were only eight documented incidents of alcohol consumption or impairment at the mill in the 15 years prior to the implementation of the policy. And there were no accidents or injuries associated with alcohol use at all. Further, by the time the arbitration was heard in December 2008, not one employee had tested positive under the policy.

The arbitration board also emphasized the importance of protecting employee privacy. It said that alcohol testing involves bodily intrusion and public embarrassment. In its opinion, the gains likely to result to Irving from random alcohol testing were minimal at best, and the inroads into employee privacy were out of proportion to any benefit. As a result, the policy was unreasonable and was set aside.

Arbitration board’s decision initially overturned

Irving applied for judicial review of the board’s decision. On review, the decision was set aside as unreasonable because of the inherent dangerousness of the workplace.

The union appealed, but the New Brunswick Court of Appeal dismissed the appeal. It said that employers are allowed to unilaterally impose alcohol testing in dangerous workplaces, unionized or not, without having to provide evidence of an existing problem with alcohol use.

Supreme Court allows appeal

A majority of the Supreme Court disagreed with the Court of Appeal and remarked that a substantial body of arbitral jurisprudence provided a valuable benchmark for assessing the arbitration board’s decision. The SCC determined that the Court of Appeal had relied too heavily on the “dangerousness” of the workplace. Rather, random testing requires “reasonable cause,” such as evidence of a general problem of substance abuse. Safety risks, it said, are only the beginning of the inquiry.

The majority of the SCC found that the arbitration board had rightly determined that there was insufficient evidence of an alcohol-related problem at the mill. In its view, then, the impact of random testing on employee privacy wasn’t justified. The expected safety gains to the employer were found to range from uncertain to minimal, while the impact on an employee’s privacy was severe. The arbitration board’s decision was within the range of reasonable outcomes and wasn’t disturbed.

The three dissenting SCC judges would have agreed with the Court of Appeal’s decision. Although they agreed that there must be some evidence of alcohol use in the workplace to justify random testing, they argued that evidence of a “serious” or “significant” problem isn’t necessary. Unfortunately for employers, their decision didn’t rule the day.

Takeaway for employers

Irving establishes that random alcohol and drug testing is generally not allowed in Canada. This decision will have a significant effect on any Canadian employers who have been randomly testing employees for drug and alcohol use. Although the court recognized that random testing may be imposed where it is a “proportionate response” to safety and privacy interests, this case leaves Canadian employers with a much higher bar to clear.

1 thought on “Supreme Court rejects random alcohol testing policy in dangerous workplace”

  1. This is a bad decision since it places employee privacy rights above the employer’s duty to provide a safe work environment for all workers and the state’s responsibility to ensure all workers in every workplace are safe from employees who are irresponsible and thus more susceptible to placing themselves and others of risk of injury. Why should any employee be forced to team with or work alongside who comes to work under the influence of alcohol or controlled substances that impair their judgement? What about their rights to be safe from the irresponsible behaviors of the alcohol and/or substance abuser? Even if the abuser imbibed before coming to work, what is to say they won’t imbibe wihile at work? This kind of reasoning regarding privacy rights totally neglects the reality of the workplace and the dangers inherent in disregarding the rights of the majority of employees to be free from harm, or potential harm, brought about by fellow workers under the influence of illegal substances and/or alcohol. We don’t allow drunk driving. How is that any less invasive of privacy using the reasoning of the majority in this instance? As Dickens aptly stated: “The law is an ass.”

Leave a Reply

Your email address will not be published. Required fields are marked *