Employers seeking to introduce random alcohol and drug testing in hazardous work environments have reason to be optimistic following the recently released decision in Suncor Energy Inc. v. Unifor Local 707A. Justice D.B. Nixon held that the decision of the three-member arbitration panel (Panel) to reject Suncor Energy Inc.’s (Suncor) proposed implementation of random alcohol and drug testing in its operations in the Regional Municipality of Wood Buffalo was unreasonable and cannot stand. As a result, the Panel’s decision has been quashed and the matter has been directed to be reconsidered by a new arbitration panel.


In June 2012, following a prolonged period of concern regarding the safety hazards posed by alcohol and drug use in the workplace, Suncor informed the union that it would be introducing random alcohol and drug testing for employees in safety-sensitive and specified positions at its operations in the Regional Municipality of Wood Buffalo. Random alcohol testing would be conducted by breathalyzer and random drug testing by urinalysis (consistent with Suncor’s post-incident, reasonable cause, follow-up and return to work testing practices).

In July 2012, the union filed a grievance challenging the random testing program scheduled for October 2012. The union subsequently obtained an interim injunction delaying the implementation of the program pending the outcome of the grievance arbitration. The injunction was upheld by the Alberta Court of Appeal.

The majority of the Panel were in favour of the union on the basis that: the harm to employee privacy rights from random testing outweighed the safety benefit to be gained by Suncor; and the imposition of random testing was not warranted because Suncor failed to show sufficient evidence of a problem with alcohol and drugs in its workplace.

In a strongly worded dissent, Panel member David Laird, Q.C., criticized numerous aspects of the Panel majority decision. In particular, he argued that the majority misconstrued and misapplied the test for imposition of random testing as enunciated in Communications, Energy and Paperworkers Union of Canada, Local 30 v. Irving Pulp & Paper, Ltd. (Irving). Further, he observed that the evidence adduced by Suncor regarding the detection of alcohol and drug use in the workplace was “profound” and “significantly more compelling than the evidence in any other decision in Canada considering random alcohol and drug testing.” Mr. Laird’s dissent was positively and frequently cited by Justice Nixon in his quashing of the Panel majority decision.


Justice Nixon identified three main issues where the Panel majority got it wrong.

Issue 1: Inappropriately raised evidentiary threshold required to establish a workplace alcohol and drug problem

In Irving, the Supreme Court of Canada (SCC) held that random drug testing in a hazardous workplace environment may be justified where there is “evidence of a general problem with substance abuse in the workplace.” The Panel majority’s decision, however, incorrectly raised the threshold from evidence of a general problem to evidence of a significant, extreme or serious problem. For Justice Nixon, this amounted to an unwarranted elevation of the Irving test, as noted by the Panel dissent: “In none of the cases which we are aware…have we seen language requiring evidence of a ‘significant’ or ‘serious’ problem. Rather, the standard has been that evidence of a problem.”

For further information on the Supreme Court of Canada’s decision in Irving, please see our June 2013 Blakes Bulletin: Sober Second Thought: Supreme Court Rules on Random Drug and Alcohol Testing.

Justice Nixon also rejected the Panel majority’s requirement that Suncor establish a causal link between alcohol and drug use and incidents in the workplace:

Again, the Court does not read Irving as imposing such a causal connection. Indeed, the majority judgment in Irving makes no reference to such a causal connection. Rather, both the majority and dissent in Irving referred to a balancing exercise to be undertaken on a case-by-case basis.

Accordingly, Justice Nixon found that the Panel majority misconstrued the Irving test by applying more rigorous requirements than those articulated by the SCC.

Issue 2: Erred in considering only evidence within the bargaining unit

Justice Nixon found that the Panel majority erred in its decision to only consider evidence concerning alcohol and drug problems within the union bargaining unit. While the Panel majority was correct to point out that its decision would only be binding upon members of the union, it was not appropriate to only consider evidence directly tied to such workers. Rather, the Panel majority should have considered evidence relating to the entire workplace, including Suncor’s union, non-union and contract workers.

This approach is consistent with Irving where the SCC directed that the test is whether there is a general workplace problem with alcohol and drugs, not a bargaining unit problem. By reducing the scope of the analysis, the Panel majority failed to consider Suncor’s statutory obligations under Alberta’s Occupational Health and Safety Act and the Criminal Code to ensure a safe workplace for all of its employees, not just those in the bargaining unit. “Safety,” Justice Nixon wrote, “is an aggregate concept, especially in a dangerous work environment.”

Justice Nixon also stressed two additional factors that called for the broader admission of evidence. First, there was nothing to suggest that alcohol and drug use within the bargaining unit differed in some meaningful way from the rest of Suncor’s workforce. Second, the only “workplace[s]” relevant to the case are the locations of Suncor’s oil sands operations in the Regional Municipality of Wood Buffalo that are well defined and by their nature dangerous. Further, the random testing program applied only to workers in safety sensitive positions working within those locations.

Issue 3: Failed to properly consider all of the evidence before it

The failure to consider all of the relevant evidence Suncor presented was a major shortcoming of the Panel majority’s decision. Indeed, Justice Nixon explicitly acknowledged that he reviewed the record before him, stating at paragraph 88: “After considering the evidence carefully, the Court finds that it has a reasoned belief that the Majority ignored or misunderstood the evidence in a manner that affected its decision.”

By way of example, the Panel majority’s decision to consider only evidence within the bargaining unit resulted in the exclusion of all but 12 of the 2,276 security incidents reported between 2004 and August 2013, on the basis that they were not directly attributable to bargaining unit employees. According to Justice Nixon:

The relevance of that “security incident” evidence was lost because it was taken out of context as a result of the overly narrow analytical approach applied by the Majority. For example, the Majority ignored the fact that the camp at Firebag is within the security perimeter of the operating environment, and thereby eliminated the security incidents that had been identified within that camp.

Justice Nixon went on to hold that the Panel majority’s “wholesale dismissal” of Suncor’s evidence regarding security incidents further affected the reasonableness of its decision. The Panel majority’s sidelining of relevant evidence was also a major theme in the Panel dissent, where Laird characterized the evidence of problems with alcohol and drugs as “overwhelming.” He went on to note that Suncor had almost 250 “for cause” positive tests since the introduction of alcohol and drug testing in 2003, the bulk of those positive tests being union members.

Justice Nixon further highlighted the internal inconsistencies in the Panel majority’s decision. For example, the Panel majority emphasized “other, more advanced methods” of drug testing, even though the union failed to provide evidence of a single alternative to urinalysis. Moreover, also with respect to urinalysis, Justice Nixon pointed out that the Panel majority appeared to irreconcilably embrace and reject the use of breathalyzers. On the one hand, the Panel majority suggested that oral fluid testing for drug use was better than urinalysis, being “closely aligned to the long accepted breathalyzer.” On the other hand, the Panel majority seemed to simultaneously reject the use of breathalyzers for alcohol testing, which it characterized as a “significant inroad” on privacy. These logical inconsistencies further undermined the reasonableness of the Panel majority decision.


Justice Nixon found that the door to random testing is still open in Canada. Part of the reason the Panel majority decision was quashed and found to be unreasonable was because it “virtually foreclose[d] any possibility of random testing, regardless of circumstances.” This outcome is plainly not what the SCC intended in Irving: “If it represents a proportionate response in light of both legitimate safety concerns and privacy interests, it may well be justified.” We note that this decision may also call into question other decisions that have cited or considered the Panel decision.

Interestingly, Justice Nixon also directed at paragraph 101 that “[t]o avoid any apprehension of bias, the matter should be considered by a fresh panel.” The acknowledgment of potential bias suggests that perhaps Justice Nixon detected an inappropriate degree of partiality by the Panel majority. Justice Nixon also appears to have gone to considerable lengths to protect the decision upon appeal by providing a thorough judicial review analysis. The union has already stated it intends to appeal the decision but, given Justice Nixon’s reasons, it may have an uphill battle on that front.

Ultimately, this decision is encouraging for Suncor and other employers who strive to improve and ensure the safety of their employees in dangerous workplaces. Pending the outcome of any appeal by the union, a newly appointed arbitration panel will be required to apply the Irving test in a more balanced way without improperly elevating thresholds or overlooking relevant evidence. This matter will continue to generate considerable media coverage.